diff options
Diffstat (limited to 'net')
495 files changed, 27573 insertions, 11637 deletions
diff --git a/net/802/fc.c b/net/802/fc.c index 34cf1ee014b..1e49f2d4ea9 100644 --- a/net/802/fc.c +++ b/net/802/fc.c @@ -70,7 +70,7 @@ static int fc_header(struct sk_buff *skb, struct net_device *dev, if(daddr) { memcpy(fch->daddr,daddr,dev->addr_len); - return(hdr_len); + return hdr_len; } return -hdr_len; } diff --git a/net/802/fddi.c b/net/802/fddi.c index 3ef0ab0a543..94b3ad08f39 100644 --- a/net/802/fddi.c +++ b/net/802/fddi.c @@ -82,10 +82,10 @@ static int fddi_header(struct sk_buff *skb, struct net_device *dev, if (daddr != NULL) { memcpy(fddi->daddr, daddr, dev->addr_len); - return(hl); + return hl; } - return(-hl); + return -hl; } @@ -108,7 +108,7 @@ static int fddi_rebuild_header(struct sk_buff *skb) { printk("%s: Don't know how to resolve type %04X addresses.\n", skb->dev->name, ntohs(fddi->hdr.llc_snap.ethertype)); - return(0); + return 0; } } @@ -162,7 +162,7 @@ __be16 fddi_type_trans(struct sk_buff *skb, struct net_device *dev) /* Assume 802.2 SNAP frames, for now */ - return(type); + return type; } EXPORT_SYMBOL(fddi_type_trans); @@ -170,9 +170,9 @@ EXPORT_SYMBOL(fddi_type_trans); int fddi_change_mtu(struct net_device *dev, int new_mtu) { if ((new_mtu < FDDI_K_SNAP_HLEN) || (new_mtu > FDDI_K_SNAP_DLEN)) - return(-EINVAL); + return -EINVAL; dev->mtu = new_mtu; - return(0); + return 0; } EXPORT_SYMBOL(fddi_change_mtu); diff --git a/net/802/hippi.c b/net/802/hippi.c index cd3e8e92952..91aca8780fd 100644 --- a/net/802/hippi.c +++ b/net/802/hippi.c @@ -152,7 +152,7 @@ int hippi_change_mtu(struct net_device *dev, int new_mtu) if ((new_mtu < 68) || (new_mtu > 65280)) return -EINVAL; dev->mtu = new_mtu; - return(0); + return 0; } EXPORT_SYMBOL(hippi_change_mtu); diff --git a/net/802/tr.c b/net/802/tr.c index 1c6e596074d..5e20cf8a074 100644 --- a/net/802/tr.c +++ b/net/802/tr.c @@ -145,7 +145,7 @@ static int tr_header(struct sk_buff *skb, struct net_device *dev, { memcpy(trh->daddr,daddr,dev->addr_len); tr_source_route(skb, trh, dev); - return(hdr_len); + return hdr_len; } return -hdr_len; diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c index a2ad1525057..05b867e4375 100644 --- a/net/8021q/vlan.c +++ b/net/8021q/vlan.c @@ -44,9 +44,6 @@ int vlan_net_id __read_mostly; -/* Our listing of VLAN group(s) */ -static struct hlist_head vlan_group_hash[VLAN_GRP_HASH_SIZE]; - const char vlan_fullname[] = "802.1Q VLAN Support"; const char vlan_version[] = DRV_VERSION; static const char vlan_copyright[] = "Ben Greear <greearb@candelatech.com>"; @@ -59,40 +56,6 @@ static struct packet_type vlan_packet_type __read_mostly = { /* End of global variables definitions. */ -static inline unsigned int vlan_grp_hashfn(unsigned int idx) -{ - return ((idx >> VLAN_GRP_HASH_SHIFT) ^ idx) & VLAN_GRP_HASH_MASK; -} - -/* Must be invoked with RCU read lock (no preempt) */ -static struct vlan_group *__vlan_find_group(struct net_device *real_dev) -{ - struct vlan_group *grp; - struct hlist_node *n; - int hash = vlan_grp_hashfn(real_dev->ifindex); - - hlist_for_each_entry_rcu(grp, n, &vlan_group_hash[hash], hlist) { - if (grp->real_dev == real_dev) - return grp; - } - - return NULL; -} - -/* Find the protocol handler. Assumes VID < VLAN_VID_MASK. - * - * Must be invoked with RCU read lock (no preempt) - */ -struct net_device *__find_vlan_dev(struct net_device *real_dev, u16 vlan_id) -{ - struct vlan_group *grp = __vlan_find_group(real_dev); - - if (grp) - return vlan_group_get_device(grp, vlan_id); - - return NULL; -} - static void vlan_group_free(struct vlan_group *grp) { int i; @@ -111,8 +74,6 @@ static struct vlan_group *vlan_group_alloc(struct net_device *real_dev) return NULL; grp->real_dev = real_dev; - hlist_add_head_rcu(&grp->hlist, - &vlan_group_hash[vlan_grp_hashfn(real_dev->ifindex)]); return grp; } @@ -151,7 +112,7 @@ void unregister_vlan_dev(struct net_device *dev, struct list_head *head) ASSERT_RTNL(); - grp = __vlan_find_group(real_dev); + grp = real_dev->vlgrp; BUG_ON(!grp); /* Take it out of our own structures, but be sure to interlock with @@ -173,11 +134,10 @@ void unregister_vlan_dev(struct net_device *dev, struct list_head *head) if (grp->nr_vlans == 0) { vlan_gvrp_uninit_applicant(real_dev); - if (real_dev->features & NETIF_F_HW_VLAN_RX) + rcu_assign_pointer(real_dev->vlgrp, NULL); + if (ops->ndo_vlan_rx_register) ops->ndo_vlan_rx_register(real_dev, NULL); - hlist_del_rcu(&grp->hlist); - /* Free the group, after all cpu's are done. */ call_rcu(&grp->rcu, vlan_rcu_free); } @@ -196,18 +156,13 @@ int vlan_check_real_dev(struct net_device *real_dev, u16 vlan_id) return -EOPNOTSUPP; } - if ((real_dev->features & NETIF_F_HW_VLAN_RX) && !ops->ndo_vlan_rx_register) { - pr_info("8021q: device %s has buggy VLAN hw accel\n", name); - return -EOPNOTSUPP; - } - if ((real_dev->features & NETIF_F_HW_VLAN_FILTER) && (!ops->ndo_vlan_rx_add_vid || !ops->ndo_vlan_rx_kill_vid)) { pr_info("8021q: Device %s has buggy VLAN hw accel\n", name); return -EOPNOTSUPP; } - if (__find_vlan_dev(real_dev, vlan_id) != NULL) + if (vlan_find_dev(real_dev, vlan_id) != NULL) return -EEXIST; return 0; @@ -222,7 +177,7 @@ int register_vlan_dev(struct net_device *dev) struct vlan_group *grp, *ngrp = NULL; int err; - grp = __vlan_find_group(real_dev); + grp = real_dev->vlgrp; if (!grp) { ngrp = grp = vlan_group_alloc(real_dev); if (!grp) @@ -252,8 +207,11 @@ int register_vlan_dev(struct net_device *dev) vlan_group_set_device(grp, vlan_id, dev); grp->nr_vlans++; - if (ngrp && real_dev->features & NETIF_F_HW_VLAN_RX) - ops->ndo_vlan_rx_register(real_dev, ngrp); + if (ngrp) { + if (ops->ndo_vlan_rx_register) + ops->ndo_vlan_rx_register(real_dev, ngrp); + rcu_assign_pointer(real_dev->vlgrp, ngrp); + } if (real_dev->features & NETIF_F_HW_VLAN_FILTER) ops->ndo_vlan_rx_add_vid(real_dev, vlan_id); @@ -264,7 +222,6 @@ out_uninit_applicant: vlan_gvrp_uninit_applicant(real_dev); out_free_group: if (ngrp) { - hlist_del_rcu(&ngrp->hlist); /* Free the group, after all cpu's are done. */ call_rcu(&ngrp->rcu, vlan_rcu_free); } @@ -321,7 +278,7 @@ static int register_vlan_device(struct net_device *real_dev, u16 vlan_id) if (new_dev == NULL) return -ENOBUFS; - new_dev->real_num_tx_queues = real_dev->real_num_tx_queues; + netif_copy_real_num_queues(new_dev, real_dev); dev_net_set(new_dev, net); /* need 4 bytes for extra VLAN header info, * hope the underlying device can handle it. @@ -428,7 +385,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event, dev->netdev_ops->ndo_vlan_rx_add_vid(dev, 0); } - grp = __vlan_find_group(dev); + grp = dev->vlgrp; if (!grp) goto out; @@ -439,7 +396,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event, switch (event) { case NETDEV_CHANGE: /* Propagate real device state to vlan devices */ - for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) { + for (i = 0; i < VLAN_N_VID; i++) { vlandev = vlan_group_get_device(grp, i); if (!vlandev) continue; @@ -450,7 +407,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event, case NETDEV_CHANGEADDR: /* Adjust unicast filters on underlying device */ - for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) { + for (i = 0; i < VLAN_N_VID; i++) { vlandev = vlan_group_get_device(grp, i); if (!vlandev) continue; @@ -464,7 +421,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event, break; case NETDEV_CHANGEMTU: - for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) { + for (i = 0; i < VLAN_N_VID; i++) { vlandev = vlan_group_get_device(grp, i); if (!vlandev) continue; @@ -478,7 +435,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event, case NETDEV_FEAT_CHANGE: /* Propagate device features to underlying device */ - for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) { + for (i = 0; i < VLAN_N_VID; i++) { vlandev = vlan_group_get_device(grp, i); if (!vlandev) continue; @@ -490,7 +447,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event, case NETDEV_DOWN: /* Put all VLANs for this dev in the down state too. */ - for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) { + for (i = 0; i < VLAN_N_VID; i++) { vlandev = vlan_group_get_device(grp, i); if (!vlandev) continue; @@ -508,7 +465,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event, case NETDEV_UP: /* Put all VLANs for this dev in the up state too. */ - for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) { + for (i = 0; i < VLAN_N_VID; i++) { vlandev = vlan_group_get_device(grp, i); if (!vlandev) continue; @@ -525,10 +482,14 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event, break; case NETDEV_UNREGISTER: + /* twiddle thumbs on netns device moves */ + if (dev->reg_state != NETREG_UNREGISTERING) + break; + /* Delete all VLANs for this dev. */ grp->killall = 1; - for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) { + for (i = 0; i < VLAN_N_VID; i++) { vlandev = vlan_group_get_device(grp, i); if (!vlandev) continue; @@ -536,7 +497,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event, /* unregistration of last vlan destroys group, abort * afterwards */ if (grp->nr_vlans == 1) - i = VLAN_GROUP_ARRAY_LEN; + i = VLAN_N_VID; unregister_vlan_dev(vlandev, &list); } @@ -742,8 +703,6 @@ err0: static void __exit vlan_cleanup_module(void) { - unsigned int i; - vlan_ioctl_set(NULL); vlan_netlink_fini(); @@ -751,10 +710,6 @@ static void __exit vlan_cleanup_module(void) dev_remove_pack(&vlan_packet_type); - /* This table must be empty if there are no module references left. */ - for (i = 0; i < VLAN_GRP_HASH_SIZE; i++) - BUG_ON(!hlist_empty(&vlan_group_hash[i])); - unregister_pernet_subsys(&vlan_net_ops); rcu_barrier(); /* Wait for completion of call_rcu()'s */ diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h index 8d9503ad01d..db01b3181fd 100644 --- a/net/8021q/vlan.h +++ b/net/8021q/vlan.h @@ -72,23 +72,6 @@ static inline struct vlan_dev_info *vlan_dev_info(const struct net_device *dev) return netdev_priv(dev); } -#define VLAN_GRP_HASH_SHIFT 5 -#define VLAN_GRP_HASH_SIZE (1 << VLAN_GRP_HASH_SHIFT) -#define VLAN_GRP_HASH_MASK (VLAN_GRP_HASH_SIZE - 1) - -/* Find a VLAN device by the MAC address of its Ethernet device, and - * it's VLAN ID. The default configuration is to have VLAN's scope - * to be box-wide, so the MAC will be ignored. The mac will only be - * looked at if we are configured to have a separate set of VLANs per - * each MAC addressable interface. Note that this latter option does - * NOT follow the spec for VLANs, but may be useful for doing very - * large quantities of VLAN MUX/DEMUX onto FrameRelay or ATM PVCs. - * - * Must be invoked with rcu_read_lock (ie preempt disabled) - * or with RTNL. - */ -struct net_device *__find_vlan_dev(struct net_device *real_dev, u16 vlan_id); - /* found in vlan_dev.c */ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev, struct packet_type *ptype, struct net_device *orig_dev); diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c index 01ddb0472f8..69b2f79800a 100644 --- a/net/8021q/vlan_core.c +++ b/net/8021q/vlan_core.c @@ -4,50 +4,29 @@ #include <linux/netpoll.h> #include "vlan.h" -/* VLAN rx hw acceleration helper. This acts like netif_{rx,receive_skb}(). */ -int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp, - u16 vlan_tci, int polling) +bool vlan_hwaccel_do_receive(struct sk_buff **skbp) { + struct sk_buff *skb = *skbp; + u16 vlan_id = skb->vlan_tci & VLAN_VID_MASK; struct net_device *vlan_dev; - u16 vlan_id; - - if (netpoll_rx(skb)) - return NET_RX_DROP; - - if (skb_bond_should_drop(skb, ACCESS_ONCE(skb->dev->master))) - skb->deliver_no_wcard = 1; - - skb->skb_iif = skb->dev->ifindex; - __vlan_hwaccel_put_tag(skb, vlan_tci); - vlan_id = vlan_tci & VLAN_VID_MASK; - vlan_dev = vlan_group_get_device(grp, vlan_id); - - if (vlan_dev) - skb->dev = vlan_dev; - else if (vlan_id) - goto drop; - - return (polling ? netif_receive_skb(skb) : netif_rx(skb)); + struct vlan_rx_stats *rx_stats; -drop: - dev_kfree_skb_any(skb); - return NET_RX_DROP; -} -EXPORT_SYMBOL(__vlan_hwaccel_rx); - -int vlan_hwaccel_do_receive(struct sk_buff *skb) -{ - struct net_device *dev = skb->dev; - struct vlan_rx_stats *rx_stats; + vlan_dev = vlan_find_dev(skb->dev, vlan_id); + if (!vlan_dev) { + if (vlan_id) + skb->pkt_type = PACKET_OTHERHOST; + return false; + } - skb->dev = vlan_dev_info(dev)->real_dev; - netif_nit_deliver(skb); + skb = *skbp = skb_share_check(skb, GFP_ATOMIC); + if (unlikely(!skb)) + return false; - skb->dev = dev; - skb->priority = vlan_get_ingress_priority(dev, skb->vlan_tci); + skb->dev = vlan_dev; + skb->priority = vlan_get_ingress_priority(vlan_dev, skb->vlan_tci); skb->vlan_tci = 0; - rx_stats = this_cpu_ptr(vlan_dev_info(dev)->vlan_rx_stats); + rx_stats = this_cpu_ptr(vlan_dev_info(vlan_dev)->vlan_rx_stats); u64_stats_update_begin(&rx_stats->syncp); rx_stats->rx_packets++; @@ -64,12 +43,13 @@ int vlan_hwaccel_do_receive(struct sk_buff *skb) * This allows the VLAN to have a different MAC than the * underlying device, and still route correctly. */ if (!compare_ether_addr(eth_hdr(skb)->h_dest, - dev->dev_addr)) + vlan_dev->dev_addr)) skb->pkt_type = PACKET_HOST; break; } u64_stats_update_end(&rx_stats->syncp); - return 0; + + return true; } struct net_device *vlan_dev_real_dev(const struct net_device *dev) @@ -84,68 +64,27 @@ u16 vlan_dev_vlan_id(const struct net_device *dev) } EXPORT_SYMBOL(vlan_dev_vlan_id); -static gro_result_t -vlan_gro_common(struct napi_struct *napi, struct vlan_group *grp, - unsigned int vlan_tci, struct sk_buff *skb) +/* VLAN rx hw acceleration helper. This acts like netif_{rx,receive_skb}(). */ +int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp, + u16 vlan_tci, int polling) { - struct sk_buff *p; - struct net_device *vlan_dev; - u16 vlan_id; - - if (skb_bond_should_drop(skb, ACCESS_ONCE(skb->dev->master))) - skb->deliver_no_wcard = 1; - - skb->skb_iif = skb->dev->ifindex; __vlan_hwaccel_put_tag(skb, vlan_tci); - vlan_id = vlan_tci & VLAN_VID_MASK; - vlan_dev = vlan_group_get_device(grp, vlan_id); - - if (vlan_dev) - skb->dev = vlan_dev; - else if (vlan_id) - goto drop; - - for (p = napi->gro_list; p; p = p->next) { - NAPI_GRO_CB(p)->same_flow = - p->dev == skb->dev && !compare_ether_header( - skb_mac_header(p), skb_gro_mac_header(skb)); - NAPI_GRO_CB(p)->flush = 0; - } - - return dev_gro_receive(napi, skb); - -drop: - return GRO_DROP; + return polling ? netif_receive_skb(skb) : netif_rx(skb); } +EXPORT_SYMBOL(__vlan_hwaccel_rx); gro_result_t vlan_gro_receive(struct napi_struct *napi, struct vlan_group *grp, unsigned int vlan_tci, struct sk_buff *skb) { - if (netpoll_rx_on(skb)) - return vlan_hwaccel_receive_skb(skb, grp, vlan_tci) - ? GRO_DROP : GRO_NORMAL; - - skb_gro_reset_offset(skb); - - return napi_skb_finish(vlan_gro_common(napi, grp, vlan_tci, skb), skb); + __vlan_hwaccel_put_tag(skb, vlan_tci); + return napi_gro_receive(napi, skb); } EXPORT_SYMBOL(vlan_gro_receive); gro_result_t vlan_gro_frags(struct napi_struct *napi, struct vlan_group *grp, unsigned int vlan_tci) { - struct sk_buff *skb = napi_frags_skb(napi); - - if (!skb) - return GRO_DROP; - - if (netpoll_rx_on(skb)) { - skb->protocol = eth_type_trans(skb, skb->dev); - return vlan_hwaccel_receive_skb(skb, grp, vlan_tci) - ? GRO_DROP : GRO_NORMAL; - } - - return napi_frags_finish(napi, skb, - vlan_gro_common(napi, grp, vlan_tci, skb)); + __vlan_hwaccel_put_tag(napi->skb, vlan_tci); + return napi_gro_frags(napi); } EXPORT_SYMBOL(vlan_gro_frags); diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 3d59c9bf8fe..14e3d1fa07a 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -158,7 +158,7 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev, vlan_id = vlan_tci & VLAN_VID_MASK; rcu_read_lock(); - vlan_dev = __find_vlan_dev(dev, vlan_id); + vlan_dev = vlan_find_dev(dev, vlan_id); /* If the VLAN device is defined, we use it. * If not, and the VID is 0, it is a 802.1p packet (not @@ -177,8 +177,8 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev, } else { skb->dev = vlan_dev; - rx_stats = per_cpu_ptr(vlan_dev_info(skb->dev)->vlan_rx_stats, - smp_processor_id()); + rx_stats = this_cpu_ptr(vlan_dev_info(skb->dev)->vlan_rx_stats); + u64_stats_update_begin(&rx_stats->syncp); rx_stats->rx_packets++; rx_stats->rx_bytes += skb->len; @@ -226,12 +226,14 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev, } netif_rx(skb); + rcu_read_unlock(); return NET_RX_SUCCESS; err_unlock: rcu_read_unlock(); err_free: + atomic_long_inc(&dev->rx_dropped); kfree_skb(skb); return NET_RX_DROP; } @@ -510,7 +512,8 @@ static int vlan_dev_open(struct net_device *dev) if (vlan->flags & VLAN_FLAG_GVRP) vlan_gvrp_request_join(dev); - netif_carrier_on(dev); + if (netif_carrier_ok(real_dev)) + netif_carrier_on(dev); return 0; clear_allmulti: @@ -842,7 +845,7 @@ static struct rtnl_link_stats64 *vlan_dev_get_stats64(struct net_device *dev, st accum.rx_packets += rxpackets; accum.rx_bytes += rxbytes; accum.rx_multicast += rxmulticast; - /* rx_errors is an ulong, not protected by syncp */ + /* rx_errors is ulong, not protected by syncp */ accum.rx_errors += p->rx_errors; } stats->rx_packets = accum.rx_packets; diff --git a/net/9p/client.c b/net/9p/client.c index dc6f2f26d02..83bf0541d66 100644 --- a/net/9p/client.c +++ b/net/9p/client.c @@ -61,13 +61,13 @@ static const match_table_t tokens = { inline int p9_is_proto_dotl(struct p9_client *clnt) { - return (clnt->proto_version == p9_proto_2000L); + return clnt->proto_version == p9_proto_2000L; } EXPORT_SYMBOL(p9_is_proto_dotl); inline int p9_is_proto_dotu(struct p9_client *clnt) { - return (clnt->proto_version == p9_proto_2000u); + return clnt->proto_version == p9_proto_2000u; } EXPORT_SYMBOL(p9_is_proto_dotu); @@ -331,8 +331,10 @@ static void p9_tag_cleanup(struct p9_client *c) } } - if (c->tagpool) + if (c->tagpool) { + p9_idpool_put(0, c->tagpool); /* free reserved tag 0 */ p9_idpool_destroy(c->tagpool); + } /* free requests associated with tags */ for (row = 0; row < (c->max_tag/P9_ROW_MAXTAG); row++) { @@ -669,7 +671,7 @@ static void p9_fid_destroy(struct p9_fid *fid) kfree(fid); } -int p9_client_version(struct p9_client *c) +static int p9_client_version(struct p9_client *c) { int err = 0; struct p9_req_t *req; @@ -728,7 +730,6 @@ error: return err; } -EXPORT_SYMBOL(p9_client_version); struct p9_client *p9_client_create(const char *dev_name, char *options) { @@ -885,54 +886,6 @@ error: } EXPORT_SYMBOL(p9_client_attach); -struct p9_fid * -p9_client_auth(struct p9_client *clnt, char *uname, u32 n_uname, char *aname) -{ - int err; - struct p9_req_t *req; - struct p9_qid qid; - struct p9_fid *afid; - - P9_DPRINTK(P9_DEBUG_9P, ">>> TAUTH uname %s aname %s\n", uname, aname); - err = 0; - - afid = p9_fid_create(clnt); - if (IS_ERR(afid)) { - err = PTR_ERR(afid); - afid = NULL; - goto error; - } - - req = p9_client_rpc(clnt, P9_TAUTH, "dss?d", - afid ? afid->fid : P9_NOFID, uname, aname, n_uname); - if (IS_ERR(req)) { - err = PTR_ERR(req); - goto error; - } - - err = p9pdu_readf(req->rc, clnt->proto_version, "Q", &qid); - if (err) { - p9pdu_dump(1, req->rc); - p9_free_req(clnt, req); - goto error; - } - - P9_DPRINTK(P9_DEBUG_9P, "<<< RAUTH qid %x.%llx.%x\n", - qid.type, - (unsigned long long)qid.path, - qid.version); - - memmove(&afid->qid, &qid, sizeof(struct p9_qid)); - p9_free_req(clnt, req); - return afid; - -error: - if (afid) - p9_fid_destroy(afid); - return ERR_PTR(err); -} -EXPORT_SYMBOL(p9_client_auth); - struct p9_fid *p9_client_walk(struct p9_fid *oldfid, int nwname, char **wnames, int clone) { @@ -944,6 +897,7 @@ struct p9_fid *p9_client_walk(struct p9_fid *oldfid, int nwname, char **wnames, int16_t nwqids, count; err = 0; + wqids = NULL; clnt = oldfid->clnt; if (clone) { fid = p9_fid_create(clnt); @@ -994,9 +948,11 @@ struct p9_fid *p9_client_walk(struct p9_fid *oldfid, int nwname, char **wnames, else fid->qid = oldfid->qid; + kfree(wqids); return fid; clunk_fid: + kfree(wqids); p9_client_clunk(fid); fid = NULL; diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c index c85109d809c..078eb162d9b 100644 --- a/net/9p/trans_fd.c +++ b/net/9p/trans_fd.c @@ -222,7 +222,7 @@ static void p9_conn_cancel(struct p9_conn *m, int err) } } -static unsigned int +static int p9_fd_poll(struct p9_client *client, struct poll_table_struct *pt) { int ret, n; diff --git a/net/9p/trans_rdma.c b/net/9p/trans_rdma.c index 0ea20c30466..17c5ba7551a 100644 --- a/net/9p/trans_rdma.c +++ b/net/9p/trans_rdma.c @@ -426,8 +426,10 @@ static int rdma_request(struct p9_client *client, struct p9_req_t *req) /* Allocate an fcall for the reply */ rpl_context = kmalloc(sizeof *rpl_context, GFP_KERNEL); - if (!rpl_context) + if (!rpl_context) { + err = -ENOMEM; goto err_close; + } /* * If the request has a buffer, steal it, otherwise @@ -445,8 +447,8 @@ static int rdma_request(struct p9_client *client, struct p9_req_t *req) } rpl_context->rc = req->rc; if (!rpl_context->rc) { - kfree(rpl_context); - goto err_close; + err = -ENOMEM; + goto err_free2; } /* @@ -458,11 +460,8 @@ static int rdma_request(struct p9_client *client, struct p9_req_t *req) */ if (atomic_inc_return(&rdma->rq_count) <= rdma->rq_depth) { err = post_recv(client, rpl_context); - if (err) { - kfree(rpl_context->rc); - kfree(rpl_context); - goto err_close; - } + if (err) + goto err_free1; } else atomic_dec(&rdma->rq_count); @@ -471,8 +470,10 @@ static int rdma_request(struct p9_client *client, struct p9_req_t *req) /* Post the request */ c = kmalloc(sizeof *c, GFP_KERNEL); - if (!c) - goto err_close; + if (!c) { + err = -ENOMEM; + goto err_free1; + } c->req = req; c->busa = ib_dma_map_single(rdma->cm_id->device, @@ -499,9 +500,15 @@ static int rdma_request(struct p9_client *client, struct p9_req_t *req) return ib_post_send(rdma->qp, &wr, &bad_wr); error: + kfree(c); + kfree(rpl_context->rc); + kfree(rpl_context); P9_DPRINTK(P9_DEBUG_ERROR, "EIO\n"); return -EIO; - + err_free1: + kfree(rpl_context->rc); + err_free2: + kfree(rpl_context); err_close: spin_lock_irqsave(&rdma->req_lock, flags); if (rdma->state < P9_RDMA_CLOSING) { diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c index dcfbe99ff81..b88515936e4 100644 --- a/net/9p/trans_virtio.c +++ b/net/9p/trans_virtio.c @@ -329,7 +329,8 @@ p9_virtio_create(struct p9_client *client, const char *devname, char *args) mutex_lock(&virtio_9p_lock); list_for_each_entry(chan, &virtio_chan_list, chan_list) { - if (!strncmp(devname, chan->tag, chan->tag_len)) { + if (!strncmp(devname, chan->tag, chan->tag_len) && + strlen(devname) == chan->tag_len) { if (!chan->inuse) { chan->inuse = true; found = 1; diff --git a/net/Kconfig b/net/Kconfig index e330594d370..55fd82e9ffd 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -217,7 +217,7 @@ source "net/dns_resolver/Kconfig" config RPS boolean - depends on SMP && SYSFS + depends on SMP && SYSFS && USE_GENERIC_SMP_HELPERS default y menu "Network testing" @@ -293,6 +293,7 @@ source "net/wimax/Kconfig" source "net/rfkill/Kconfig" source "net/9p/Kconfig" source "net/caif/Kconfig" +source "net/ceph/Kconfig" endif # if NET diff --git a/net/Makefile b/net/Makefile index ea60fbce9b1..6b7bfd7f141 100644 --- a/net/Makefile +++ b/net/Makefile @@ -68,3 +68,4 @@ obj-$(CONFIG_SYSCTL) += sysctl_net.o endif obj-$(CONFIG_WIMAX) += wimax/ obj-$(CONFIG_DNS_RESOLVER) += dns_resolver/ +obj-$(CONFIG_CEPH_LIB) += ceph/ diff --git a/net/atm/br2684.c b/net/atm/br2684.c index 651babdfab3..ad2b232a205 100644 --- a/net/atm/br2684.c +++ b/net/atm/br2684.c @@ -399,12 +399,6 @@ static void br2684_push(struct atm_vcc *atmvcc, struct sk_buff *skb) unregister_netdev(net_dev); free_netdev(net_dev); } - read_lock_irq(&devs_lock); - if (list_empty(&br2684_devs)) { - /* last br2684 device */ - unregister_atmdevice_notifier(&atm_dev_notifier); - } - read_unlock_irq(&devs_lock); return; } @@ -675,7 +669,6 @@ static int br2684_create(void __user *arg) if (list_empty(&br2684_devs)) { /* 1st br2684 device */ - register_atmdevice_notifier(&atm_dev_notifier); brdev->number = 1; } else brdev->number = BRPRIV(list_entry_brdev(br2684_devs.prev))->number + 1; @@ -815,6 +808,7 @@ static int __init br2684_init(void) return -ENOMEM; #endif register_atm_ioctl(&br2684_ioctl_ops); + register_atmdevice_notifier(&atm_dev_notifier); return 0; } @@ -830,9 +824,7 @@ static void __exit br2684_exit(void) #endif - /* if not already empty */ - if (!list_empty(&br2684_devs)) - unregister_atmdevice_notifier(&atm_dev_notifier); + unregister_atmdevice_notifier(&atm_dev_notifier); while (!list_empty(&br2684_devs)) { net_dev = list_entry_brdev(br2684_devs.next); diff --git a/net/atm/clip.c b/net/atm/clip.c index 95fdd118506..ff956d1115b 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -310,9 +310,9 @@ static int clip_constructor(struct neighbour *neigh) return 0; } -static u32 clip_hash(const void *pkey, const struct net_device *dev) +static u32 clip_hash(const void *pkey, const struct net_device *dev, __u32 rnd) { - return jhash_2words(*(u32 *) pkey, dev->ifindex, clip_tbl.hash_rnd); + return jhash_2words(*(u32 *) pkey, dev->ifindex, rnd); } static struct neigh_table clip_tbl = { diff --git a/net/atm/common.c b/net/atm/common.c index 940404a73b3..1b9c52a02cd 100644 --- a/net/atm/common.c +++ b/net/atm/common.c @@ -792,7 +792,7 @@ int vcc_getsockopt(struct socket *sock, int level, int optname, default: if (level == SOL_SOCKET) return -EINVAL; - break; + break; } if (!vcc->dev || !vcc->dev->ops->getsockopt) return -EINVAL; diff --git a/net/atm/lec.c b/net/atm/lec.c index d98bde1a0ac..181d70c73d7 100644 --- a/net/atm/lec.c +++ b/net/atm/lec.c @@ -220,7 +220,6 @@ static unsigned char *get_tr_dst(unsigned char *packet, unsigned char *rdesc) static int lec_open(struct net_device *dev) { netif_start_queue(dev); - memset(&dev->stats, 0, sizeof(struct net_device_stats)); return 0; } diff --git a/net/atm/mpc.c b/net/atm/mpc.c index 622b471e14e..74bcc662c3d 100644 --- a/net/atm/mpc.c +++ b/net/atm/mpc.c @@ -778,7 +778,7 @@ static void mpc_push(struct atm_vcc *vcc, struct sk_buff *skb) eg->packets_rcvd++; mpc->eg_ops->put(eg); - memset(ATM_SKB(skb), 0, sizeof(struct atm_skb_data)); + memset(ATM_SKB(new_skb), 0, sizeof(struct atm_skb_data)); netif_rx(new_skb); } diff --git a/net/atm/proc.c b/net/atm/proc.c index 6262aeae398..f85da0779e5 100644 --- a/net/atm/proc.c +++ b/net/atm/proc.c @@ -38,6 +38,7 @@ static ssize_t proc_dev_atm_read(struct file *file, char __user *buf, static const struct file_operations proc_atm_dev_ops = { .owner = THIS_MODULE, .read = proc_dev_atm_read, + .llseek = noop_llseek, }; static void add_stats(struct seq_file *seq, const char *aal, diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c index cfdfd7e2a17..26eaebf4aaa 100644 --- a/net/ax25/af_ax25.c +++ b/net/ax25/af_ax25.c @@ -1103,7 +1103,7 @@ done: out: release_sock(sk); - return 0; + return err; } /* diff --git a/net/ax25/ax25_ds_timer.c b/net/ax25/ax25_ds_timer.c index 2ce79df0068..c7d81436213 100644 --- a/net/ax25/ax25_ds_timer.c +++ b/net/ax25/ax25_ds_timer.c @@ -112,8 +112,8 @@ void ax25_ds_heartbeat_expiry(ax25_cb *ax25) if (sk) { sock_hold(sk); ax25_destroy_socket(ax25); - sock_put(sk); bh_unlock_sock(sk); + sock_put(sk); } else ax25_destroy_socket(ax25); return; diff --git a/net/ax25/ax25_route.c b/net/ax25/ax25_route.c index 7805945a5fd..a1690845dc6 100644 --- a/net/ax25/ax25_route.c +++ b/net/ax25/ax25_route.c @@ -412,7 +412,7 @@ int ax25_rt_autobind(ax25_cb *ax25, ax25_address *addr) { ax25_uid_assoc *user; ax25_route *ax25_rt; - int err; + int err = 0; if ((ax25_rt = ax25_get_route(addr, NULL)) == NULL) return -EHOSTUNREACH; @@ -453,7 +453,7 @@ int ax25_rt_autobind(ax25_cb *ax25, ax25_address *addr) put: ax25_put_route(ax25_rt); - return 0; + return err; } struct sk_buff *ax25_rt_build_path(struct sk_buff *skb, ax25_address *src, diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c index 421c45bd1b9..c4cf3f59500 100644 --- a/net/bluetooth/af_bluetooth.c +++ b/net/bluetooth/af_bluetooth.c @@ -265,6 +265,115 @@ int bt_sock_recvmsg(struct kiocb *iocb, struct socket *sock, } EXPORT_SYMBOL(bt_sock_recvmsg); +static long bt_sock_data_wait(struct sock *sk, long timeo) +{ + DECLARE_WAITQUEUE(wait, current); + + add_wait_queue(sk_sleep(sk), &wait); + for (;;) { + set_current_state(TASK_INTERRUPTIBLE); + + if (!skb_queue_empty(&sk->sk_receive_queue)) + break; + + if (sk->sk_err || (sk->sk_shutdown & RCV_SHUTDOWN)) + break; + + if (signal_pending(current) || !timeo) + break; + + set_bit(SOCK_ASYNC_WAITDATA, &sk->sk_socket->flags); + release_sock(sk); + timeo = schedule_timeout(timeo); + lock_sock(sk); + clear_bit(SOCK_ASYNC_WAITDATA, &sk->sk_socket->flags); + } + + __set_current_state(TASK_RUNNING); + remove_wait_queue(sk_sleep(sk), &wait); + return timeo; +} + +int bt_sock_stream_recvmsg(struct kiocb *iocb, struct socket *sock, + struct msghdr *msg, size_t size, int flags) +{ + struct sock *sk = sock->sk; + int err = 0; + size_t target, copied = 0; + long timeo; + + if (flags & MSG_OOB) + return -EOPNOTSUPP; + + msg->msg_namelen = 0; + + BT_DBG("sk %p size %zu", sk, size); + + lock_sock(sk); + + target = sock_rcvlowat(sk, flags & MSG_WAITALL, size); + timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); + + do { + struct sk_buff *skb; + int chunk; + + skb = skb_dequeue(&sk->sk_receive_queue); + if (!skb) { + if (copied >= target) + break; + + if ((err = sock_error(sk)) != 0) + break; + if (sk->sk_shutdown & RCV_SHUTDOWN) + break; + + err = -EAGAIN; + if (!timeo) + break; + + timeo = bt_sock_data_wait(sk, timeo); + + if (signal_pending(current)) { + err = sock_intr_errno(timeo); + goto out; + } + continue; + } + + chunk = min_t(unsigned int, skb->len, size); + if (memcpy_toiovec(msg->msg_iov, skb->data, chunk)) { + skb_queue_head(&sk->sk_receive_queue, skb); + if (!copied) + copied = -EFAULT; + break; + } + copied += chunk; + size -= chunk; + + sock_recv_ts_and_drops(msg, sk, skb); + + if (!(flags & MSG_PEEK)) { + skb_pull(skb, chunk); + if (skb->len) { + skb_queue_head(&sk->sk_receive_queue, skb); + break; + } + kfree_skb(skb); + + } else { + /* put message back and return */ + skb_queue_head(&sk->sk_receive_queue, skb); + break; + } + } while (size); + +out: + release_sock(sk); + return copied ? : err; +} +EXPORT_SYMBOL(bt_sock_stream_recvmsg); + static inline unsigned int bt_accept_poll(struct sock *parent) { struct list_head *p, *n; @@ -297,13 +406,12 @@ unsigned int bt_sock_poll(struct file * file, struct socket *sock, poll_table *w mask |= POLLERR; if (sk->sk_shutdown & RCV_SHUTDOWN) - mask |= POLLRDHUP; + mask |= POLLRDHUP | POLLIN | POLLRDNORM; if (sk->sk_shutdown == SHUTDOWN_MASK) mask |= POLLHUP; - if (!skb_queue_empty(&sk->sk_receive_queue) || - (sk->sk_shutdown & RCV_SHUTDOWN)) + if (!skb_queue_empty(&sk->sk_receive_queue)) mask |= POLLIN | POLLRDNORM; if (sk->sk_state == BT_CLOSED) diff --git a/net/bluetooth/cmtp/core.c b/net/bluetooth/cmtp/core.c index d4c6af082d4..ec0a1347f93 100644 --- a/net/bluetooth/cmtp/core.c +++ b/net/bluetooth/cmtp/core.c @@ -321,14 +321,10 @@ static int cmtp_session(void *arg) int cmtp_add_connection(struct cmtp_connadd_req *req, struct socket *sock) { struct cmtp_session *session, *s; - bdaddr_t src, dst; int i, err; BT_DBG(""); - baswap(&src, &bt_sk(sock->sk)->src); - baswap(&dst, &bt_sk(sock->sk)->dst); - session = kzalloc(sizeof(struct cmtp_session), GFP_KERNEL); if (!session) return -ENOMEM; @@ -347,7 +343,7 @@ int cmtp_add_connection(struct cmtp_connadd_req *req, struct socket *sock) BT_DBG("mtu %d", session->mtu); - sprintf(session->name, "%s", batostr(&dst)); + sprintf(session->name, "%s", batostr(&bt_sk(sock->sk)->dst)); session->sock = sock; session->state = BT_CONFIG; diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index c52f091ee6d..bc2a052e518 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -562,7 +562,6 @@ static int hci_dev_do_close(struct hci_dev *hdev) hci_dev_lock_bh(hdev); inquiry_cache_flush(hdev); hci_conn_hash_flush(hdev); - hci_blacklist_clear(hdev); hci_dev_unlock_bh(hdev); hci_notify(hdev, HCI_DEV_DOWN); diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index 8fb967beee8..5fce3d6d07b 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -37,9 +37,7 @@ static ssize_t show_link_type(struct device *dev, struct device_attribute *attr, static ssize_t show_link_address(struct device *dev, struct device_attribute *attr, char *buf) { struct hci_conn *conn = dev_get_drvdata(dev); - bdaddr_t bdaddr; - baswap(&bdaddr, &conn->dst); - return sprintf(buf, "%s\n", batostr(&bdaddr)); + return sprintf(buf, "%s\n", batostr(&conn->dst)); } static ssize_t show_link_features(struct device *dev, struct device_attribute *attr, char *buf) @@ -196,8 +194,8 @@ static inline char *host_typetostr(int type) switch (type) { case HCI_BREDR: return "BR/EDR"; - case HCI_80211: - return "802.11"; + case HCI_AMP: + return "AMP"; default: return "UNKNOWN"; } @@ -238,9 +236,7 @@ static ssize_t show_class(struct device *dev, struct device_attribute *attr, cha static ssize_t show_address(struct device *dev, struct device_attribute *attr, char *buf) { struct hci_dev *hdev = dev_get_drvdata(dev); - bdaddr_t bdaddr; - baswap(&bdaddr, &hdev->bdaddr); - return sprintf(buf, "%s\n", batostr(&bdaddr)); + return sprintf(buf, "%s\n", batostr(&hdev->bdaddr)); } static ssize_t show_features(struct device *dev, struct device_attribute *attr, char *buf) @@ -408,10 +404,8 @@ static int inquiry_cache_show(struct seq_file *f, void *p) for (e = cache->list; e; e = e->next) { struct inquiry_data *data = &e->data; - bdaddr_t bdaddr; - baswap(&bdaddr, &data->bdaddr); seq_printf(f, "%s %d %d %d 0x%.2x%.2x%.2x 0x%.4x %d %d %u\n", - batostr(&bdaddr), + batostr(&data->bdaddr), data->pscan_rep_mode, data->pscan_period_mode, data->pscan_mode, data->dev_class[2], data->dev_class[1], data->dev_class[0], @@ -445,13 +439,10 @@ static int blacklist_show(struct seq_file *f, void *p) list_for_each(l, &hdev->blacklist) { struct bdaddr_list *b; - bdaddr_t bdaddr; b = list_entry(l, struct bdaddr_list, list); - baswap(&bdaddr, &b->bdaddr); - - seq_printf(f, "%s\n", batostr(&bdaddr)); + seq_printf(f, "%s\n", batostr(&b->bdaddr)); } hci_dev_unlock_bh(hdev); diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index bfe641b7dfa..c0ee8b3928e 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -758,7 +758,6 @@ static int hidp_setup_hid(struct hidp_session *session, struct hidp_connadd_req *req) { struct hid_device *hid; - bdaddr_t src, dst; int err; session->rd_data = kzalloc(req->rd_size, GFP_KERNEL); @@ -781,9 +780,6 @@ static int hidp_setup_hid(struct hidp_session *session, hid->driver_data = session; - baswap(&src, &bt_sk(session->ctrl_sock->sk)->src); - baswap(&dst, &bt_sk(session->ctrl_sock->sk)->dst); - hid->bus = BUS_BLUETOOTH; hid->vendor = req->vendor; hid->product = req->product; @@ -791,8 +787,8 @@ static int hidp_setup_hid(struct hidp_session *session, hid->country = req->country; strncpy(hid->name, req->name, 128); - strncpy(hid->phys, batostr(&src), 64); - strncpy(hid->uniq, batostr(&dst), 64); + strncpy(hid->phys, batostr(&bt_sk(session->ctrl_sock->sk)->src), 64); + strncpy(hid->uniq, batostr(&bt_sk(session->ctrl_sock->sk)->dst), 64); hid->dev.parent = hidp_get_device(session); hid->ll_driver = &hidp_hid_driver; diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c index 3e3cd9d4e52..daa7a988d9a 100644 --- a/net/bluetooth/l2cap.c +++ b/net/bluetooth/l2cap.c @@ -1008,10 +1008,20 @@ static int l2cap_sock_bind(struct socket *sock, struct sockaddr *addr, int alen) goto done; } - if (la.l2_psm && __le16_to_cpu(la.l2_psm) < 0x1001 && - !capable(CAP_NET_BIND_SERVICE)) { - err = -EACCES; - goto done; + if (la.l2_psm) { + __u16 psm = __le16_to_cpu(la.l2_psm); + + /* PSM must be odd and lsb of upper byte must be 0 */ + if ((psm & 0x0101) != 0x0001) { + err = -EINVAL; + goto done; + } + + /* Restrict usage of well-known PSMs */ + if (psm < 0x1001 && !capable(CAP_NET_BIND_SERVICE)) { + err = -EACCES; + goto done; + } } write_lock_bh(&l2cap_sk_list.lock); @@ -1190,6 +1200,13 @@ static int l2cap_sock_connect(struct socket *sock, struct sockaddr *addr, int al goto done; } + /* PSM must be odd and lsb of upper byte must be 0 */ + if ((__le16_to_cpu(la.l2_psm) & 0x0101) != 0x0001 && + sk->sk_type != SOCK_RAW) { + err = -EINVAL; + goto done; + } + /* Set destination address and psm */ bacpy(&bt_sk(sk)->dst, &la.l2_bdaddr); l2cap_pi(sk)->psm = la.l2_psm; @@ -1441,33 +1458,23 @@ static inline void l2cap_do_send(struct sock *sk, struct sk_buff *skb) static void l2cap_streaming_send(struct sock *sk) { - struct sk_buff *skb, *tx_skb; + struct sk_buff *skb; struct l2cap_pinfo *pi = l2cap_pi(sk); u16 control, fcs; - while ((skb = sk->sk_send_head)) { - tx_skb = skb_clone(skb, GFP_ATOMIC); - - control = get_unaligned_le16(tx_skb->data + L2CAP_HDR_SIZE); + while ((skb = skb_dequeue(TX_QUEUE(sk)))) { + control = get_unaligned_le16(skb->data + L2CAP_HDR_SIZE); control |= pi->next_tx_seq << L2CAP_CTRL_TXSEQ_SHIFT; - put_unaligned_le16(control, tx_skb->data + L2CAP_HDR_SIZE); + put_unaligned_le16(control, skb->data + L2CAP_HDR_SIZE); if (pi->fcs == L2CAP_FCS_CRC16) { - fcs = crc16(0, (u8 *)tx_skb->data, tx_skb->len - 2); - put_unaligned_le16(fcs, tx_skb->data + tx_skb->len - 2); + fcs = crc16(0, (u8 *)skb->data, skb->len - 2); + put_unaligned_le16(fcs, skb->data + skb->len - 2); } - l2cap_do_send(sk, tx_skb); + l2cap_do_send(sk, skb); pi->next_tx_seq = (pi->next_tx_seq + 1) % 64; - - if (skb_queue_is_last(TX_QUEUE(sk), skb)) - sk->sk_send_head = NULL; - else - sk->sk_send_head = skb_queue_next(TX_QUEUE(sk), skb); - - skb = skb_dequeue(TX_QUEUE(sk)); - kfree_skb(skb); } } @@ -1645,7 +1652,7 @@ static inline int l2cap_skbuff_fromiovec(struct sock *sk, struct msghdr *msg, in *frag = bt_skb_send_alloc(sk, count, msg->msg_flags & MSG_DONTWAIT, &err); if (!*frag) - return -EFAULT; + return err; if (memcpy_fromiovec(skb_put(*frag, count), msg->msg_iov, count)) return -EFAULT; @@ -1671,7 +1678,7 @@ static struct sk_buff *l2cap_create_connless_pdu(struct sock *sk, struct msghdr skb = bt_skb_send_alloc(sk, count + hlen, msg->msg_flags & MSG_DONTWAIT, &err); if (!skb) - return ERR_PTR(-ENOMEM); + return ERR_PTR(err); /* Create L2CAP header */ lh = (struct l2cap_hdr *) skb_put(skb, L2CAP_HDR_SIZE); @@ -1700,7 +1707,7 @@ static struct sk_buff *l2cap_create_basic_pdu(struct sock *sk, struct msghdr *ms skb = bt_skb_send_alloc(sk, count + hlen, msg->msg_flags & MSG_DONTWAIT, &err); if (!skb) - return ERR_PTR(-ENOMEM); + return ERR_PTR(err); /* Create L2CAP header */ lh = (struct l2cap_hdr *) skb_put(skb, L2CAP_HDR_SIZE); @@ -1737,7 +1744,7 @@ static struct sk_buff *l2cap_create_iframe_pdu(struct sock *sk, struct msghdr *m skb = bt_skb_send_alloc(sk, count + hlen, msg->msg_flags & MSG_DONTWAIT, &err); if (!skb) - return ERR_PTR(-ENOMEM); + return ERR_PTR(err); /* Create L2CAP header */ lh = (struct l2cap_hdr *) skb_put(skb, L2CAP_HDR_SIZE); @@ -1944,6 +1951,9 @@ static int l2cap_sock_recvmsg(struct kiocb *iocb, struct socket *sock, struct ms release_sock(sk); + if (sock->type == SOCK_STREAM) + return bt_sock_stream_recvmsg(iocb, sock, msg, len, flags); + return bt_sock_recvmsg(iocb, sock, msg, len, flags); } @@ -1960,6 +1970,11 @@ static int l2cap_sock_setsockopt_old(struct socket *sock, int optname, char __us switch (optname) { case L2CAP_OPTIONS: + if (sk->sk_state == BT_CONNECTED) { + err = -EINVAL; + break; + } + opts.imtu = l2cap_pi(sk)->imtu; opts.omtu = l2cap_pi(sk)->omtu; opts.flush_to = l2cap_pi(sk)->flush_to; @@ -2705,8 +2720,9 @@ done: case L2CAP_MODE_ERTM: pi->remote_tx_win = rfc.txwin_size; pi->remote_max_tx = rfc.max_transmit; - if (rfc.max_pdu_size > pi->conn->mtu - 10) - rfc.max_pdu_size = le16_to_cpu(pi->conn->mtu - 10); + + if (le16_to_cpu(rfc.max_pdu_size) > pi->conn->mtu - 10) + rfc.max_pdu_size = cpu_to_le16(pi->conn->mtu - 10); pi->remote_mps = le16_to_cpu(rfc.max_pdu_size); @@ -2723,8 +2739,8 @@ done: break; case L2CAP_MODE_STREAMING: - if (rfc.max_pdu_size > pi->conn->mtu - 10) - rfc.max_pdu_size = le16_to_cpu(pi->conn->mtu - 10); + if (le16_to_cpu(rfc.max_pdu_size) > pi->conn->mtu - 10) + rfc.max_pdu_size = cpu_to_le16(pi->conn->mtu - 10); pi->remote_mps = le16_to_cpu(rfc.max_pdu_size); @@ -2770,10 +2786,10 @@ static int l2cap_parse_conf_rsp(struct sock *sk, void *rsp, int len, void *data, case L2CAP_CONF_MTU: if (val < L2CAP_DEFAULT_MIN_MTU) { *result = L2CAP_CONF_UNACCEPT; - pi->omtu = L2CAP_DEFAULT_MIN_MTU; + pi->imtu = L2CAP_DEFAULT_MIN_MTU; } else - pi->omtu = val; - l2cap_add_conf_opt(&ptr, L2CAP_CONF_MTU, 2, pi->omtu); + pi->imtu = val; + l2cap_add_conf_opt(&ptr, L2CAP_CONF_MTU, 2, pi->imtu); break; case L2CAP_CONF_FLUSH_TO: @@ -2806,7 +2822,6 @@ static int l2cap_parse_conf_rsp(struct sock *sk, void *rsp, int len, void *data, if (*result == L2CAP_CONF_SUCCESS) { switch (rfc.mode) { case L2CAP_MODE_ERTM: - pi->remote_tx_win = rfc.txwin_size; pi->retrans_timeout = le16_to_cpu(rfc.retrans_timeout); pi->monitor_timeout = le16_to_cpu(rfc.monitor_timeout); pi->mps = le16_to_cpu(rfc.max_pdu_size); @@ -2862,7 +2877,6 @@ static void l2cap_conf_rfc_get(struct sock *sk, void *rsp, int len) done: switch (rfc.mode) { case L2CAP_MODE_ERTM: - pi->remote_tx_win = rfc.txwin_size; pi->retrans_timeout = le16_to_cpu(rfc.retrans_timeout); pi->monitor_timeout = le16_to_cpu(rfc.monitor_timeout); pi->mps = le16_to_cpu(rfc.max_pdu_size); @@ -2897,7 +2911,7 @@ static inline int l2cap_connect_req(struct l2cap_conn *conn, struct l2cap_cmd_hd struct l2cap_chan_list *list = &conn->chan_list; struct l2cap_conn_req *req = (struct l2cap_conn_req *) data; struct l2cap_conn_rsp rsp; - struct sock *parent, *uninitialized_var(sk); + struct sock *parent, *sk = NULL; int result, status = L2CAP_CS_NO_INFO; u16 dcid = 0, scid = __le16_to_cpu(req->scid); @@ -3006,7 +3020,7 @@ sendresp: L2CAP_INFO_REQ, sizeof(info), &info); } - if (!(l2cap_pi(sk)->conf_state & L2CAP_CONF_REQ_SENT) && + if (sk && !(l2cap_pi(sk)->conf_state & L2CAP_CONF_REQ_SENT) && result == L2CAP_CR_SUCCESS) { u8 buf[128]; l2cap_pi(sk)->conf_state |= L2CAP_CONF_REQ_SENT; @@ -3072,6 +3086,17 @@ static inline int l2cap_connect_rsp(struct l2cap_conn *conn, struct l2cap_cmd_hd return 0; } +static inline void set_default_fcs(struct l2cap_pinfo *pi) +{ + /* FCS is enabled only in ERTM or streaming mode, if one or both + * sides request it. + */ + if (pi->mode != L2CAP_MODE_ERTM && pi->mode != L2CAP_MODE_STREAMING) + pi->fcs = L2CAP_FCS_NONE; + else if (!(pi->conf_state & L2CAP_CONF_NO_FCS_RECV)) + pi->fcs = L2CAP_FCS_CRC16; +} + static inline int l2cap_config_req(struct l2cap_conn *conn, struct l2cap_cmd_hdr *cmd, u16 cmd_len, u8 *data) { struct l2cap_conf_req *req = (struct l2cap_conf_req *) data; @@ -3089,14 +3114,8 @@ static inline int l2cap_config_req(struct l2cap_conn *conn, struct l2cap_cmd_hdr if (!sk) return -ENOENT; - if (sk->sk_state != BT_CONFIG) { - struct l2cap_cmd_rej rej; - - rej.reason = cpu_to_le16(0x0002); - l2cap_send_cmd(conn, cmd->ident, L2CAP_COMMAND_REJ, - sizeof(rej), &rej); + if (sk->sk_state == BT_DISCONN) goto unlock; - } /* Reject if config buffer is too small. */ len = cmd_len - sizeof(*req); @@ -3136,9 +3155,7 @@ static inline int l2cap_config_req(struct l2cap_conn *conn, struct l2cap_cmd_hdr goto unlock; if (l2cap_pi(sk)->conf_state & L2CAP_CONF_INPUT_DONE) { - if (!(l2cap_pi(sk)->conf_state & L2CAP_CONF_NO_FCS_RECV) || - l2cap_pi(sk)->fcs != L2CAP_FCS_NONE) - l2cap_pi(sk)->fcs = L2CAP_FCS_CRC16; + set_default_fcs(l2cap_pi(sk)); sk->sk_state = BT_CONNECTED; @@ -3154,6 +3171,7 @@ static inline int l2cap_config_req(struct l2cap_conn *conn, struct l2cap_cmd_hdr if (!(l2cap_pi(sk)->conf_state & L2CAP_CONF_REQ_SENT)) { u8 buf[64]; + l2cap_pi(sk)->conf_state |= L2CAP_CONF_REQ_SENT; l2cap_send_cmd(conn, l2cap_get_ident(conn), L2CAP_CONF_REQ, l2cap_build_conf_req(sk, buf), buf); l2cap_pi(sk)->num_conf_req++; @@ -3226,9 +3244,7 @@ static inline int l2cap_config_rsp(struct l2cap_conn *conn, struct l2cap_cmd_hdr l2cap_pi(sk)->conf_state |= L2CAP_CONF_INPUT_DONE; if (l2cap_pi(sk)->conf_state & L2CAP_CONF_OUTPUT_DONE) { - if (!(l2cap_pi(sk)->conf_state & L2CAP_CONF_NO_FCS_RECV) || - l2cap_pi(sk)->fcs != L2CAP_FCS_NONE) - l2cap_pi(sk)->fcs = L2CAP_FCS_CRC16; + set_default_fcs(l2cap_pi(sk)); sk->sk_state = BT_CONNECTED; l2cap_pi(sk)->next_tx_seq = 0; @@ -4648,6 +4664,8 @@ static int l2cap_recv_acldata(struct hci_conn *hcon, struct sk_buff *skb, u16 fl if (flags & ACL_START) { struct l2cap_hdr *hdr; + struct sock *sk; + u16 cid; int len; if (conn->rx_len) { @@ -4658,7 +4676,8 @@ static int l2cap_recv_acldata(struct hci_conn *hcon, struct sk_buff *skb, u16 fl l2cap_conn_unreliable(conn, ECOMM); } - if (skb->len < 2) { + /* Start fragment always begin with Basic L2CAP header */ + if (skb->len < L2CAP_HDR_SIZE) { BT_ERR("Frame is too short (len %d)", skb->len); l2cap_conn_unreliable(conn, ECOMM); goto drop; @@ -4666,6 +4685,7 @@ static int l2cap_recv_acldata(struct hci_conn *hcon, struct sk_buff *skb, u16 fl hdr = (struct l2cap_hdr *) skb->data; len = __le16_to_cpu(hdr->len) + L2CAP_HDR_SIZE; + cid = __le16_to_cpu(hdr->cid); if (len == skb->len) { /* Complete frame received */ @@ -4682,6 +4702,19 @@ static int l2cap_recv_acldata(struct hci_conn *hcon, struct sk_buff *skb, u16 fl goto drop; } + sk = l2cap_get_chan_by_scid(&conn->chan_list, cid); + + if (sk && l2cap_pi(sk)->imtu < len - L2CAP_HDR_SIZE) { + BT_ERR("Frame exceeding recv MTU (len %d, MTU %d)", + len, l2cap_pi(sk)->imtu); + bh_unlock_sock(sk); + l2cap_conn_unreliable(conn, ECOMM); + goto drop; + } + + if (sk) + bh_unlock_sock(sk); + /* Allocate skb for the complete frame (with header) */ conn->rx_skb = bt_skb_alloc(len, GFP_ATOMIC); if (!conn->rx_skb) diff --git a/net/bluetooth/lib.c b/net/bluetooth/lib.c index ad2af5814e4..b826d1bf10d 100644 --- a/net/bluetooth/lib.c +++ b/net/bluetooth/lib.c @@ -51,8 +51,8 @@ char *batostr(bdaddr_t *ba) i ^= 1; sprintf(str[i], "%2.2X:%2.2X:%2.2X:%2.2X:%2.2X:%2.2X", - ba->b[0], ba->b[1], ba->b[2], - ba->b[3], ba->b[4], ba->b[5]); + ba->b[5], ba->b[4], ba->b[3], + ba->b[2], ba->b[1], ba->b[0]); return str[i]; } diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c index 7dca91bb8c5..39a5d87e33b 100644 --- a/net/bluetooth/rfcomm/core.c +++ b/net/bluetooth/rfcomm/core.c @@ -113,11 +113,10 @@ static void rfcomm_session_del(struct rfcomm_session *s); #define __get_rpn_stop_bits(line) (((line) >> 2) & 0x1) #define __get_rpn_parity(line) (((line) >> 3) & 0x7) -static inline void rfcomm_schedule(uint event) +static inline void rfcomm_schedule(void) { if (!rfcomm_thread) return; - //set_bit(event, &rfcomm_event); set_bit(RFCOMM_SCHED_WAKEUP, &rfcomm_event); wake_up_process(rfcomm_thread); } @@ -179,13 +178,13 @@ static unsigned char rfcomm_crc_table[256] = { /* FCS on 2 bytes */ static inline u8 __fcs(u8 *data) { - return (0xff - __crc(data)); + return 0xff - __crc(data); } /* FCS on 3 bytes */ static inline u8 __fcs2(u8 *data) { - return (0xff - rfcomm_crc_table[__crc(data) ^ data[2]]); + return 0xff - rfcomm_crc_table[__crc(data) ^ data[2]]; } /* Check FCS */ @@ -203,13 +202,13 @@ static inline int __check_fcs(u8 *data, int type, u8 fcs) static void rfcomm_l2state_change(struct sock *sk) { BT_DBG("%p state %d", sk, sk->sk_state); - rfcomm_schedule(RFCOMM_SCHED_STATE); + rfcomm_schedule(); } static void rfcomm_l2data_ready(struct sock *sk, int bytes) { BT_DBG("%p bytes %d", sk, bytes); - rfcomm_schedule(RFCOMM_SCHED_RX); + rfcomm_schedule(); } static int rfcomm_l2sock_create(struct socket **sock) @@ -255,7 +254,7 @@ static void rfcomm_session_timeout(unsigned long arg) BT_DBG("session %p state %ld", s, s->state); set_bit(RFCOMM_TIMED_OUT, &s->flags); - rfcomm_schedule(RFCOMM_SCHED_TIMEO); + rfcomm_schedule(); } static void rfcomm_session_set_timer(struct rfcomm_session *s, long timeout) @@ -283,7 +282,7 @@ static void rfcomm_dlc_timeout(unsigned long arg) set_bit(RFCOMM_TIMED_OUT, &d->flags); rfcomm_dlc_put(d); - rfcomm_schedule(RFCOMM_SCHED_TIMEO); + rfcomm_schedule(); } static void rfcomm_dlc_set_timer(struct rfcomm_dlc *d, long timeout) @@ -465,7 +464,7 @@ static int __rfcomm_dlc_close(struct rfcomm_dlc *d, int err) case BT_CONFIG: if (test_and_clear_bit(RFCOMM_DEFER_SETUP, &d->flags)) { set_bit(RFCOMM_AUTH_REJECT, &d->flags); - rfcomm_schedule(RFCOMM_SCHED_AUTH); + rfcomm_schedule(); break; } /* Fall through */ @@ -485,7 +484,7 @@ static int __rfcomm_dlc_close(struct rfcomm_dlc *d, int err) case BT_CONNECT2: if (test_and_clear_bit(RFCOMM_DEFER_SETUP, &d->flags)) { set_bit(RFCOMM_AUTH_REJECT, &d->flags); - rfcomm_schedule(RFCOMM_SCHED_AUTH); + rfcomm_schedule(); break; } /* Fall through */ @@ -533,7 +532,7 @@ int rfcomm_dlc_send(struct rfcomm_dlc *d, struct sk_buff *skb) skb_queue_tail(&d->tx_queue, skb); if (!test_bit(RFCOMM_TX_THROTTLED, &d->flags)) - rfcomm_schedule(RFCOMM_SCHED_TX); + rfcomm_schedule(); return len; } @@ -545,7 +544,7 @@ void __rfcomm_dlc_throttle(struct rfcomm_dlc *d) d->v24_sig |= RFCOMM_V24_FC; set_bit(RFCOMM_MSC_PENDING, &d->flags); } - rfcomm_schedule(RFCOMM_SCHED_TX); + rfcomm_schedule(); } void __rfcomm_dlc_unthrottle(struct rfcomm_dlc *d) @@ -556,7 +555,7 @@ void __rfcomm_dlc_unthrottle(struct rfcomm_dlc *d) d->v24_sig &= ~RFCOMM_V24_FC; set_bit(RFCOMM_MSC_PENDING, &d->flags); } - rfcomm_schedule(RFCOMM_SCHED_TX); + rfcomm_schedule(); } /* @@ -577,7 +576,7 @@ int rfcomm_dlc_set_modem_status(struct rfcomm_dlc *d, u8 v24_sig) d->v24_sig = v24_sig; if (!test_and_set_bit(RFCOMM_MSC_PENDING, &d->flags)) - rfcomm_schedule(RFCOMM_SCHED_TX); + rfcomm_schedule(); return 0; } @@ -816,7 +815,7 @@ static int rfcomm_queue_disc(struct rfcomm_dlc *d) cmd->fcs = __fcs2((u8 *) cmd); skb_queue_tail(&d->tx_queue, skb); - rfcomm_schedule(RFCOMM_SCHED_TX); + rfcomm_schedule(); return 0; } @@ -1415,8 +1414,8 @@ static int rfcomm_recv_rpn(struct rfcomm_session *s, int cr, int len, struct sk_ return 0; if (len == 1) { - /* This is a request, return default settings */ - bit_rate = RFCOMM_RPN_BR_115200; + /* This is a request, return default (according to ETSI TS 07.10) settings */ + bit_rate = RFCOMM_RPN_BR_9600; data_bits = RFCOMM_RPN_DATA_8; stop_bits = RFCOMM_RPN_STOP_1; parity = RFCOMM_RPN_PARITY_NONE; @@ -1431,9 +1430,9 @@ static int rfcomm_recv_rpn(struct rfcomm_session *s, int cr, int len, struct sk_ if (rpn->param_mask & cpu_to_le16(RFCOMM_RPN_PM_BITRATE)) { bit_rate = rpn->bit_rate; - if (bit_rate != RFCOMM_RPN_BR_115200) { + if (bit_rate > RFCOMM_RPN_BR_230400) { BT_DBG("RPN bit rate mismatch 0x%x", bit_rate); - bit_rate = RFCOMM_RPN_BR_115200; + bit_rate = RFCOMM_RPN_BR_9600; rpn_mask ^= RFCOMM_RPN_PM_BITRATE; } } @@ -1698,7 +1697,7 @@ static int rfcomm_recv_frame(struct rfcomm_session *s, struct sk_buff *skb) break; default: - BT_ERR("Unknown packet type 0x%02x\n", type); + BT_ERR("Unknown packet type 0x%02x", type); break; } kfree_skb(skb); @@ -1884,7 +1883,7 @@ static inline void rfcomm_accept_connection(struct rfcomm_session *s) * L2CAP MTU minus UIH header and FCS. */ s->mtu = min(l2cap_pi(nsock->sk)->omtu, l2cap_pi(nsock->sk)->imtu) - 5; - rfcomm_schedule(RFCOMM_SCHED_RX); + rfcomm_schedule(); } else sock_release(nsock); } @@ -2093,7 +2092,7 @@ static void rfcomm_security_cfm(struct hci_conn *conn, u8 status, u8 encrypt) rfcomm_session_put(s); - rfcomm_schedule(RFCOMM_SCHED_AUTH); + rfcomm_schedule(); } static struct hci_cb rfcomm_cb = { diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c index 44a62327595..aec505f934d 100644 --- a/net/bluetooth/rfcomm/sock.c +++ b/net/bluetooth/rfcomm/sock.c @@ -82,11 +82,14 @@ static void rfcomm_sk_data_ready(struct rfcomm_dlc *d, struct sk_buff *skb) static void rfcomm_sk_state_change(struct rfcomm_dlc *d, int err) { struct sock *sk = d->owner, *parent; + unsigned long flags; + if (!sk) return; BT_DBG("dlc %p state %ld err %d", d, d->state, err); + local_irq_save(flags); bh_lock_sock(sk); if (err) @@ -108,6 +111,7 @@ static void rfcomm_sk_state_change(struct rfcomm_dlc *d, int err) } bh_unlock_sock(sk); + local_irq_restore(flags); if (parent && sock_flag(sk, SOCK_ZAPPED)) { /* We have to drop DLC lock here, otherwise @@ -617,121 +621,29 @@ static int rfcomm_sock_sendmsg(struct kiocb *iocb, struct socket *sock, return sent; } -static long rfcomm_sock_data_wait(struct sock *sk, long timeo) -{ - DECLARE_WAITQUEUE(wait, current); - - add_wait_queue(sk_sleep(sk), &wait); - for (;;) { - set_current_state(TASK_INTERRUPTIBLE); - - if (!skb_queue_empty(&sk->sk_receive_queue) || - sk->sk_err || - (sk->sk_shutdown & RCV_SHUTDOWN) || - signal_pending(current) || - !timeo) - break; - - set_bit(SOCK_ASYNC_WAITDATA, &sk->sk_socket->flags); - release_sock(sk); - timeo = schedule_timeout(timeo); - lock_sock(sk); - clear_bit(SOCK_ASYNC_WAITDATA, &sk->sk_socket->flags); - } - - __set_current_state(TASK_RUNNING); - remove_wait_queue(sk_sleep(sk), &wait); - return timeo; -} - static int rfcomm_sock_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg, size_t size, int flags) { struct sock *sk = sock->sk; struct rfcomm_dlc *d = rfcomm_pi(sk)->dlc; - int err = 0; - size_t target, copied = 0; - long timeo; + int len; if (test_and_clear_bit(RFCOMM_DEFER_SETUP, &d->flags)) { rfcomm_dlc_accept(d); return 0; } - if (flags & MSG_OOB) - return -EOPNOTSUPP; - - msg->msg_namelen = 0; - - BT_DBG("sk %p size %zu", sk, size); + len = bt_sock_stream_recvmsg(iocb, sock, msg, size, flags); lock_sock(sk); + if (!(flags & MSG_PEEK) && len > 0) + atomic_sub(len, &sk->sk_rmem_alloc); - target = sock_rcvlowat(sk, flags & MSG_WAITALL, size); - timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); - - do { - struct sk_buff *skb; - int chunk; - - skb = skb_dequeue(&sk->sk_receive_queue); - if (!skb) { - if (copied >= target) - break; - - if ((err = sock_error(sk)) != 0) - break; - if (sk->sk_shutdown & RCV_SHUTDOWN) - break; - - err = -EAGAIN; - if (!timeo) - break; - - timeo = rfcomm_sock_data_wait(sk, timeo); - - if (signal_pending(current)) { - err = sock_intr_errno(timeo); - goto out; - } - continue; - } - - chunk = min_t(unsigned int, skb->len, size); - if (memcpy_toiovec(msg->msg_iov, skb->data, chunk)) { - skb_queue_head(&sk->sk_receive_queue, skb); - if (!copied) - copied = -EFAULT; - break; - } - copied += chunk; - size -= chunk; - - sock_recv_ts_and_drops(msg, sk, skb); - - if (!(flags & MSG_PEEK)) { - atomic_sub(chunk, &sk->sk_rmem_alloc); - - skb_pull(skb, chunk); - if (skb->len) { - skb_queue_head(&sk->sk_receive_queue, skb); - break; - } - kfree_skb(skb); - - } else { - /* put message back and return */ - skb_queue_head(&sk->sk_receive_queue, skb); - break; - } - } while (size); - -out: if (atomic_read(&sk->sk_rmem_alloc) <= (sk->sk_rcvbuf >> 2)) rfcomm_dlc_unthrottle(rfcomm_pi(sk)->dlc); - release_sock(sk); - return copied ? : err; + + return len; } static int rfcomm_sock_setsockopt_old(struct socket *sock, int optname, char __user *optval, unsigned int optlen) diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c index befc3a52aa0..a9b81f5dacd 100644 --- a/net/bluetooth/rfcomm/tty.c +++ b/net/bluetooth/rfcomm/tty.c @@ -183,9 +183,7 @@ static struct device *rfcomm_get_device(struct rfcomm_dev *dev) static ssize_t show_address(struct device *tty_dev, struct device_attribute *attr, char *buf) { struct rfcomm_dev *dev = dev_get_drvdata(tty_dev); - bdaddr_t bdaddr; - baswap(&bdaddr, &dev->dst); - return sprintf(buf, "%s\n", batostr(&bdaddr)); + return sprintf(buf, "%s\n", batostr(&dev->dst)); } static ssize_t show_channel(struct device *tty_dev, struct device_attribute *attr, char *buf) @@ -844,10 +842,6 @@ static int rfcomm_tty_ioctl(struct tty_struct *tty, struct file *filp, unsigned BT_DBG("TIOCMIWAIT"); break; - case TIOCGICOUNT: - BT_DBG("TIOCGICOUNT"); - break; - case TIOCGSERIAL: BT_ERR("TIOCGSERIAL is not supported"); return -ENOIOCTLCMD; diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index cf09fe591fc..17cb0b63357 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -212,6 +212,11 @@ static int br_set_tx_csum(struct net_device *dev, u32 data) return 0; } +static int br_set_flags(struct net_device *netdev, u32 data) +{ + return ethtool_op_set_flags(netdev, data, ETH_FLAG_TXVLAN); +} + #ifdef CONFIG_NET_POLL_CONTROLLER static void br_poll_controller(struct net_device *br_dev) { @@ -304,6 +309,7 @@ static const struct ethtool_ops br_ethtool_ops = { .get_ufo = ethtool_op_get_ufo, .set_ufo = ethtool_op_set_ufo, .get_flags = ethtool_op_get_flags, + .set_flags = br_set_flags, }; static const struct net_device_ops br_netdev_ops = { @@ -343,5 +349,5 @@ void br_dev_setup(struct net_device *dev) dev->features = NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HIGHDMA | NETIF_F_GSO_MASK | NETIF_F_NO_CSUM | NETIF_F_LLTX | - NETIF_F_NETNS_LOCAL | NETIF_F_GSO; + NETIF_F_NETNS_LOCAL | NETIF_F_GSO | NETIF_F_HW_VLAN_TX; } diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index c03d2c3ff03..89ad25a7620 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -61,30 +61,27 @@ static int port_cost(struct net_device *dev) } -/* - * Check for port carrier transistions. - * Called from work queue to allow for calling functions that - * might sleep (such as speed check), and to debounce. - */ +/* Check for port carrier transistions. */ void br_port_carrier_check(struct net_bridge_port *p) { struct net_device *dev = p->dev; struct net_bridge *br = p->br; - if (netif_carrier_ok(dev)) + if (netif_running(dev) && netif_carrier_ok(dev)) p->path_cost = port_cost(dev); - if (netif_running(br->dev)) { - spin_lock_bh(&br->lock); - if (netif_carrier_ok(dev)) { - if (p->state == BR_STATE_DISABLED) - br_stp_enable_port(p); - } else { - if (p->state != BR_STATE_DISABLED) - br_stp_disable_port(p); - } - spin_unlock_bh(&br->lock); + if (!netif_running(br->dev)) + return; + + spin_lock_bh(&br->lock); + if (netif_running(dev) && netif_carrier_ok(dev)) { + if (p->state == BR_STATE_DISABLED) + br_stp_enable_port(p); + } else { + if (p->state != BR_STATE_DISABLED) + br_stp_disable_port(p); } + spin_unlock_bh(&br->lock); } static void release_nbp(struct kobject *kobj) diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 826cd522153..25207a1f182 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -141,7 +141,7 @@ struct sk_buff *br_handle_frame(struct sk_buff *skb) const unsigned char *dest = eth_hdr(skb)->h_dest; int (*rhook)(struct sk_buff *skb); - if (skb->pkt_type == PACKET_LOOPBACK) + if (unlikely(skb->pkt_type == PACKET_LOOPBACK)) return skb; if (!is_valid_ether_addr(eth_hdr(skb)->h_source)) @@ -159,7 +159,7 @@ struct sk_buff *br_handle_frame(struct sk_buff *skb) goto drop; /* If STP is turned off, then forward */ - if (p->br->stp_enabled == BR_NO_STP && dest[5] == 0) + if (p->br->stp_enabled == BR_NO_STP) goto forward; if (NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, skb, skb->dev, diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index 2c911c0759c..865fd7634b6 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -64,22 +64,24 @@ static int brnf_filter_pppoe_tagged __read_mostly = 0; static inline __be16 vlan_proto(const struct sk_buff *skb) { - return vlan_eth_hdr(skb)->h_vlan_encapsulated_proto; + if (vlan_tx_tag_present(skb)) + return skb->protocol; + else if (skb->protocol == htons(ETH_P_8021Q)) + return vlan_eth_hdr(skb)->h_vlan_encapsulated_proto; + else + return 0; } #define IS_VLAN_IP(skb) \ - (skb->protocol == htons(ETH_P_8021Q) && \ - vlan_proto(skb) == htons(ETH_P_IP) && \ + (vlan_proto(skb) == htons(ETH_P_IP) && \ brnf_filter_vlan_tagged) #define IS_VLAN_IPV6(skb) \ - (skb->protocol == htons(ETH_P_8021Q) && \ - vlan_proto(skb) == htons(ETH_P_IPV6) &&\ + (vlan_proto(skb) == htons(ETH_P_IPV6) && \ brnf_filter_vlan_tagged) #define IS_VLAN_ARP(skb) \ - (skb->protocol == htons(ETH_P_8021Q) && \ - vlan_proto(skb) == htons(ETH_P_ARP) && \ + (vlan_proto(skb) == htons(ETH_P_ARP) && \ brnf_filter_vlan_tagged) static inline __be16 pppoe_proto(const struct sk_buff *skb) @@ -106,7 +108,6 @@ static struct dst_ops fake_dst_ops = { .family = AF_INET, .protocol = cpu_to_be16(ETH_P_IP), .update_pmtu = fake_update_pmtu, - .entries = ATOMIC_INIT(0), }; /* @@ -162,8 +163,8 @@ static inline struct nf_bridge_info *nf_bridge_unshare(struct sk_buff *skb) if (tmp) { memcpy(tmp, nf_bridge, sizeof(struct nf_bridge_info)); atomic_set(&tmp->use, 1); - nf_bridge_put(nf_bridge); } + nf_bridge_put(nf_bridge); nf_bridge = tmp; } return nf_bridge; @@ -209,6 +210,72 @@ static inline void nf_bridge_update_protocol(struct sk_buff *skb) skb->protocol = htons(ETH_P_PPP_SES); } +/* When handing a packet over to the IP layer + * check whether we have a skb that is in the + * expected format + */ + +static int br_parse_ip_options(struct sk_buff *skb) +{ + struct ip_options *opt; + struct iphdr *iph; + struct net_device *dev = skb->dev; + u32 len; + + iph = ip_hdr(skb); + opt = &(IPCB(skb)->opt); + + /* Basic sanity checks */ + if (iph->ihl < 5 || iph->version != 4) + goto inhdr_error; + + if (!pskb_may_pull(skb, iph->ihl*4)) + goto inhdr_error; + + iph = ip_hdr(skb); + if (unlikely(ip_fast_csum((u8 *)iph, iph->ihl))) + goto inhdr_error; + + len = ntohs(iph->tot_len); + if (skb->len < len) { + IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INTRUNCATEDPKTS); + goto drop; + } else if (len < (iph->ihl*4)) + goto inhdr_error; + + if (pskb_trim_rcsum(skb, len)) { + IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INDISCARDS); + goto drop; + } + + /* Zero out the CB buffer if no options present */ + if (iph->ihl == 5) { + memset(IPCB(skb), 0, sizeof(struct inet_skb_parm)); + return 0; + } + + opt->optlen = iph->ihl*4 - sizeof(struct iphdr); + if (ip_options_compile(dev_net(dev), opt, skb)) + goto inhdr_error; + + /* Check correct handling of SRR option */ + if (unlikely(opt->srr)) { + struct in_device *in_dev = __in_dev_get_rcu(dev); + if (in_dev && !IN_DEV_SOURCE_ROUTE(in_dev)) + goto drop; + + if (ip_options_rcv_srr(skb)) + goto drop; + } + + return 0; + +inhdr_error: + IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INHDRERRORS); +drop: + return -1; +} + /* Fill in the header for fragmented IP packets handled by * the IPv4 connection tracking code. */ @@ -549,7 +616,6 @@ static unsigned int br_nf_pre_routing(unsigned int hook, struct sk_buff *skb, { struct net_bridge_port *p; struct net_bridge *br; - struct iphdr *iph; __u32 len = nf_bridge_encap_header_len(skb); if (unlikely(!pskb_may_pull(skb, len))) @@ -578,28 +644,9 @@ static unsigned int br_nf_pre_routing(unsigned int hook, struct sk_buff *skb, nf_bridge_pull_encap_header_rcsum(skb); - if (!pskb_may_pull(skb, sizeof(struct iphdr))) - goto inhdr_error; - - iph = ip_hdr(skb); - if (iph->ihl < 5 || iph->version != 4) - goto inhdr_error; - - if (!pskb_may_pull(skb, 4 * iph->ihl)) - goto inhdr_error; - - iph = ip_hdr(skb); - if (ip_fast_csum((__u8 *) iph, iph->ihl) != 0) - goto inhdr_error; - - len = ntohs(iph->tot_len); - if (skb->len < len || len < 4 * iph->ihl) - goto inhdr_error; - - pskb_trim_rcsum(skb, len); - - /* BUG: Should really parse the IP options here. */ - memset(IPCB(skb), 0, sizeof(struct inet_skb_parm)); + if (br_parse_ip_options(skb)) + /* Drop invalid packet */ + goto out; nf_bridge_put(skb->nf_bridge); if (!nf_bridge_alloc(skb)) @@ -614,8 +661,6 @@ static unsigned int br_nf_pre_routing(unsigned int hook, struct sk_buff *skb, return NF_STOLEN; -inhdr_error: -// IP_INC_STATS_BH(IpInHdrErrors); out: return NF_DROP; } @@ -759,12 +804,19 @@ static unsigned int br_nf_forward_arp(unsigned int hook, struct sk_buff *skb, #if defined(CONFIG_NF_CONNTRACK_IPV4) || defined(CONFIG_NF_CONNTRACK_IPV4_MODULE) static int br_nf_dev_queue_xmit(struct sk_buff *skb) { + int ret; + if (skb->nfct != NULL && skb->protocol == htons(ETH_P_IP) && skb->len + nf_bridge_mtu_reduction(skb) > skb->dev->mtu && - !skb_is_gso(skb)) - return ip_fragment(skb, br_dev_queue_push_xmit); - else - return br_dev_queue_push_xmit(skb); + !skb_is_gso(skb)) { + if (br_parse_ip_options(skb)) + /* Drop invalid packet */ + return NF_DROP; + ret = ip_fragment(skb, br_dev_queue_push_xmit); + } else + ret = br_dev_queue_push_xmit(skb); + + return ret; } #else static int br_nf_dev_queue_xmit(struct sk_buff *skb) @@ -952,15 +1004,22 @@ int __init br_netfilter_init(void) { int ret; - ret = nf_register_hooks(br_nf_ops, ARRAY_SIZE(br_nf_ops)); + ret = dst_entries_init(&fake_dst_ops); if (ret < 0) return ret; + + ret = nf_register_hooks(br_nf_ops, ARRAY_SIZE(br_nf_ops)); + if (ret < 0) { + dst_entries_destroy(&fake_dst_ops); + return ret; + } #ifdef CONFIG_SYSCTL brnf_sysctl_header = register_sysctl_paths(brnf_path, brnf_table); if (brnf_sysctl_header == NULL) { printk(KERN_WARNING "br_netfilter: can't register to sysctl.\n"); nf_unregister_hooks(br_nf_ops, ARRAY_SIZE(br_nf_ops)); + dst_entries_destroy(&fake_dst_ops); return -ENOMEM; } #endif @@ -974,4 +1033,5 @@ void br_netfilter_fini(void) #ifdef CONFIG_SYSCTL unregister_sysctl_table(brnf_sysctl_header); #endif + dst_entries_destroy(&fake_dst_ops); } diff --git a/net/bridge/netfilter/ebt_vlan.c b/net/bridge/netfilter/ebt_vlan.c index 87b53b3a921..eae67bf0446 100644 --- a/net/bridge/netfilter/ebt_vlan.c +++ b/net/bridge/netfilter/ebt_vlan.c @@ -39,8 +39,6 @@ static bool ebt_vlan_mt(const struct sk_buff *skb, struct xt_action_param *par) { const struct ebt_vlan_info *info = par->matchinfo; - const struct vlan_hdr *fp; - struct vlan_hdr _frame; unsigned short TCI; /* Whole TCI, given from parsed frame */ unsigned short id; /* VLAN ID, given from frame TCI */ @@ -48,9 +46,20 @@ ebt_vlan_mt(const struct sk_buff *skb, struct xt_action_param *par) /* VLAN encapsulated Type/Length field, given from orig frame */ __be16 encap; - fp = skb_header_pointer(skb, 0, sizeof(_frame), &_frame); - if (fp == NULL) - return false; + if (vlan_tx_tag_present(skb)) { + TCI = vlan_tx_tag_get(skb); + encap = skb->protocol; + } else { + const struct vlan_hdr *fp; + struct vlan_hdr _frame; + + fp = skb_header_pointer(skb, 0, sizeof(_frame), &_frame); + if (fp == NULL) + return false; + + TCI = ntohs(fp->h_vlan_TCI); + encap = fp->h_vlan_encapsulated_proto; + } /* Tag Control Information (TCI) consists of the following elements: * - User_priority. The user_priority field is three bits in length, @@ -59,10 +68,8 @@ ebt_vlan_mt(const struct sk_buff *skb, struct xt_action_param *par) * (CFI) is a single bit flag value. Currently ignored. * - VLAN Identifier (VID). The VID is encoded as * an unsigned binary number. */ - TCI = ntohs(fp->h_vlan_TCI); id = TCI & VLAN_VID_MASK; prio = (TCI >> 13) & 0x7; - encap = fp->h_vlan_encapsulated_proto; /* Checking VLAN Identifier (VID) */ if (GET_BITMASK(EBT_VLAN_ID)) @@ -111,10 +118,10 @@ static int ebt_vlan_mt_check(const struct xt_mtchk_param *par) * 0 - The null VLAN ID. * 1 - The default Port VID (PVID) * 0x0FFF - Reserved for implementation use. - * if_vlan.h: VLAN_GROUP_ARRAY_LEN 4096. */ + * if_vlan.h: VLAN_N_VID 4096. */ if (GET_BITMASK(EBT_VLAN_ID)) { if (!!info->id) { /* if id!=0 => check vid range */ - if (info->id > VLAN_GROUP_ARRAY_LEN) { + if (info->id > VLAN_N_VID) { pr_debug("id %d is out of range (1-4096)\n", info->id); return -EINVAL; diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index bcc102e3be4..a1dcf83f0d5 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -124,16 +124,23 @@ ebt_dev_check(const char *entry, const struct net_device *device) #define FWINV2(bool,invflg) ((bool) ^ !!(e->invflags & invflg)) /* process standard matches */ static inline int -ebt_basic_match(const struct ebt_entry *e, const struct ethhdr *h, +ebt_basic_match(const struct ebt_entry *e, const struct sk_buff *skb, const struct net_device *in, const struct net_device *out) { + const struct ethhdr *h = eth_hdr(skb); + __be16 ethproto; int verdict, i; + if (vlan_tx_tag_present(skb)) + ethproto = htons(ETH_P_8021Q); + else + ethproto = h->h_proto; + if (e->bitmask & EBT_802_3) { - if (FWINV2(ntohs(h->h_proto) >= 1536, EBT_IPROTO)) + if (FWINV2(ntohs(ethproto) >= 1536, EBT_IPROTO)) return 1; } else if (!(e->bitmask & EBT_NOPROTO) && - FWINV2(e->ethproto != h->h_proto, EBT_IPROTO)) + FWINV2(e->ethproto != ethproto, EBT_IPROTO)) return 1; if (FWINV2(ebt_dev_check(e->in, in), EBT_IIN)) @@ -213,7 +220,7 @@ unsigned int ebt_do_table (unsigned int hook, struct sk_buff *skb, base = private->entries; i = 0; while (i < nentries) { - if (ebt_basic_match(point, eth_hdr(skb), in, out)) + if (ebt_basic_match(point, skb, in, out)) goto letscontinue; if (EBT_MATCH_ITERATE(point, ebt_do_match, skb, &acpar) != 0) diff --git a/net/caif/caif_dev.c b/net/caif/caif_dev.c index 0b586e9d137..b99369a055d 100644 --- a/net/caif/caif_dev.c +++ b/net/caif/caif_dev.c @@ -9,6 +9,8 @@ * and Sakari Ailus <sakari.ailus@nokia.com> */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/version.h> #include <linux/module.h> #include <linux/kernel.h> @@ -171,7 +173,7 @@ static int receive(struct sk_buff *skb, struct net_device *dev, net = dev_net(dev); pkt = cfpkt_fromnative(CAIF_DIR_IN, skb); caifd = caif_get(dev); - if (!caifd || !caifd->layer.up || !caifd->layer.up->ctrlcmd) + if (!caifd || !caifd->layer.up || !caifd->layer.up->receive) return NET_RX_DROP; if (caifd->layer.up->receive(caifd->layer.up, pkt)) @@ -214,7 +216,7 @@ static int caif_device_notify(struct notifier_block *me, unsigned long what, switch (what) { case NETDEV_REGISTER: - pr_info("CAIF: %s():register %s\n", __func__, dev->name); + netdev_info(dev, "register\n"); caifd = caif_device_alloc(dev); if (caifd == NULL) break; @@ -225,14 +227,13 @@ static int caif_device_notify(struct notifier_block *me, unsigned long what, break; case NETDEV_UP: - pr_info("CAIF: %s(): up %s\n", __func__, dev->name); + netdev_info(dev, "up\n"); caifd = caif_get(dev); if (caifd == NULL) break; caifdev = netdev_priv(dev); if (atomic_read(&caifd->state) == NETDEV_UP) { - pr_info("CAIF: %s():%s already up\n", - __func__, dev->name); + netdev_info(dev, "already up\n"); break; } atomic_set(&caifd->state, what); @@ -273,7 +274,7 @@ static int caif_device_notify(struct notifier_block *me, unsigned long what, caifd = caif_get(dev); if (caifd == NULL) break; - pr_info("CAIF: %s():going down %s\n", __func__, dev->name); + netdev_info(dev, "going down\n"); if (atomic_read(&caifd->state) == NETDEV_GOING_DOWN || atomic_read(&caifd->state) == NETDEV_DOWN) @@ -295,11 +296,10 @@ static int caif_device_notify(struct notifier_block *me, unsigned long what, caifd = caif_get(dev); if (caifd == NULL) break; - pr_info("CAIF: %s(): down %s\n", __func__, dev->name); + netdev_info(dev, "down\n"); if (atomic_read(&caifd->in_use)) - pr_warning("CAIF: %s(): " - "Unregistering an active CAIF device: %s\n", - __func__, dev->name); + netdev_warn(dev, + "Unregistering an active CAIF device\n"); cfcnfg_del_phy_layer(get_caif_conf(), &caifd->layer); dev_put(dev); atomic_set(&caifd->state, what); @@ -307,7 +307,7 @@ static int caif_device_notify(struct notifier_block *me, unsigned long what, case NETDEV_UNREGISTER: caifd = caif_get(dev); - pr_info("CAIF: %s(): unregister %s\n", __func__, dev->name); + netdev_info(dev, "unregister\n"); atomic_set(&caifd->state, what); caif_device_destroy(dev); break; @@ -391,7 +391,7 @@ static int __init caif_device_init(void) int result; cfg = cfcnfg_create(); if (!cfg) { - pr_warning("CAIF: %s(): can't create cfcnfg.\n", __func__); + pr_warn("can't create cfcnfg\n"); goto err_cfcnfg_create_failed; } result = register_pernet_device(&caif_net_ops); diff --git a/net/caif/caif_socket.c b/net/caif/caif_socket.c index 8ce90478611..2eca2dd0000 100644 --- a/net/caif/caif_socket.c +++ b/net/caif/caif_socket.c @@ -4,6 +4,8 @@ * License terms: GNU General Public License (GPL) version 2 */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/fs.h> #include <linux/init.h> #include <linux/module.h> @@ -15,7 +17,6 @@ #include <linux/poll.h> #include <linux/tcp.h> #include <linux/uaccess.h> -#include <linux/mutex.h> #include <linux/debugfs.h> #include <linux/caif/caif_socket.h> #include <asm/atomic.h> @@ -28,9 +29,6 @@ MODULE_LICENSE("GPL"); MODULE_ALIAS_NETPROTO(AF_CAIF); -#define CAIF_DEF_SNDBUF (4096*10) -#define CAIF_DEF_RCVBUF (4096*100) - /* * CAIF state is re-using the TCP socket states. * caif_states stored in sk_state reflect the state as reported by @@ -157,9 +155,7 @@ static int caif_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) if (atomic_read(&sk->sk_rmem_alloc) + skb->truesize >= (unsigned)sk->sk_rcvbuf && rx_flow_is_on(cf_sk)) { - trace_printk("CAIF: %s():" - " sending flow OFF (queue len = %d %d)\n", - __func__, + pr_debug("sending flow OFF (queue len = %d %d)\n", atomic_read(&cf_sk->sk.sk_rmem_alloc), sk_rcvbuf_lowwater(cf_sk)); set_rx_flow_off(cf_sk); @@ -172,9 +168,7 @@ static int caif_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) return err; if (!sk_rmem_schedule(sk, skb->truesize) && rx_flow_is_on(cf_sk)) { set_rx_flow_off(cf_sk); - trace_printk("CAIF: %s():" - " sending flow OFF due to rmem_schedule\n", - __func__); + pr_debug("sending flow OFF due to rmem_schedule\n"); dbfs_atomic_inc(&cnt.num_rx_flow_off); caif_flow_ctrl(sk, CAIF_MODEMCMD_FLOW_OFF_REQ); } @@ -275,8 +269,7 @@ static void caif_ctrl_cb(struct cflayer *layr, break; default: - pr_debug("CAIF: %s(): Unexpected flow command %d\n", - __func__, flow); + pr_debug("Unexpected flow command %d\n", flow); } } @@ -536,8 +529,7 @@ static int transmit_skb(struct sk_buff *skb, struct caifsock *cf_sk, /* Slight paranoia, probably not needed. */ if (unlikely(loopcnt++ > 1000)) { - pr_warning("CAIF: %s(): transmit retries failed," - " error = %d\n", __func__, ret); + pr_warn("transmit retries failed, error = %d\n", ret); break; } @@ -827,6 +819,7 @@ static int caif_connect(struct socket *sock, struct sockaddr *uaddr, long timeo; int err; int ifindex, headroom, tailroom; + unsigned int mtu; struct net_device *dev; lock_sock(sk); @@ -896,15 +889,22 @@ static int caif_connect(struct socket *sock, struct sockaddr *uaddr, cf_sk->sk.sk_state = CAIF_DISCONNECTED; goto out; } - dev = dev_get_by_index(sock_net(sk), ifindex); + + err = -ENODEV; + rcu_read_lock(); + dev = dev_get_by_index_rcu(sock_net(sk), ifindex); + if (!dev) { + rcu_read_unlock(); + goto out; + } cf_sk->headroom = LL_RESERVED_SPACE_EXTRA(dev, headroom); + mtu = dev->mtu; + rcu_read_unlock(); + cf_sk->tailroom = tailroom; - cf_sk->maxframe = dev->mtu - (headroom + tailroom); - dev_put(dev); + cf_sk->maxframe = mtu - (headroom + tailroom); if (cf_sk->maxframe < 1) { - pr_warning("CAIF: %s(): CAIF Interface MTU too small (%d)\n", - __func__, dev->mtu); - err = -ENODEV; + pr_warn("CAIF Interface MTU too small (%d)\n", dev->mtu); goto out; } @@ -1123,10 +1123,6 @@ static int caif_create(struct net *net, struct socket *sock, int protocol, /* Store the protocol */ sk->sk_protocol = (unsigned char) protocol; - /* Sendbuf dictates the amount of outbound packets not yet sent */ - sk->sk_sndbuf = CAIF_DEF_SNDBUF; - sk->sk_rcvbuf = CAIF_DEF_RCVBUF; - /* * Lock in order to try to stop someone from opening the socket * too early. diff --git a/net/caif/cfcnfg.c b/net/caif/cfcnfg.c index 1c29189b344..41adafd1891 100644 --- a/net/caif/cfcnfg.c +++ b/net/caif/cfcnfg.c @@ -3,6 +3,9 @@ * Author: Sjur Brendeland/sjur.brandeland@stericsson.com * License terms: GNU General Public License (GPL) version 2 */ + +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/kernel.h> #include <linux/stddef.h> #include <linux/slab.h> @@ -78,7 +81,7 @@ struct cfcnfg *cfcnfg_create(void) /* Initiate this layer */ this = kzalloc(sizeof(struct cfcnfg), GFP_ATOMIC); if (!this) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return NULL; } this->mux = cfmuxl_create(); @@ -106,7 +109,7 @@ struct cfcnfg *cfcnfg_create(void) layer_set_up(this->ctrl, this); return this; out_of_mem: - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); kfree(this->mux); kfree(this->ctrl); kfree(this); @@ -194,7 +197,7 @@ int cfcnfg_disconn_adapt_layer(struct cfcnfg *cnfg, struct cflayer *adap_layer) caif_assert(adap_layer != NULL); channel_id = adap_layer->id; if (adap_layer->dn == NULL || channel_id == 0) { - pr_err("CAIF: %s():adap_layer->id is 0\n", __func__); + pr_err("adap_layer->dn == NULL or adap_layer->id is 0\n"); ret = -ENOTCONN; goto end; } @@ -204,9 +207,8 @@ int cfcnfg_disconn_adapt_layer(struct cfcnfg *cnfg, struct cflayer *adap_layer) layer_set_up(servl, NULL); ret = cfctrl_linkdown_req(cnfg->ctrl, channel_id, adap_layer); if (servl == NULL) { - pr_err("CAIF: %s(): PROTOCOL ERROR " - "- Error removing service_layer Channel_Id(%d)", - __func__, channel_id); + pr_err("PROTOCOL ERROR - Error removing service_layer Channel_Id(%d)", + channel_id); ret = -EINVAL; goto end; } @@ -216,18 +218,14 @@ int cfcnfg_disconn_adapt_layer(struct cfcnfg *cnfg, struct cflayer *adap_layer) phyinfo = cfcnfg_get_phyinfo(cnfg, phyid); if (phyinfo == NULL) { - pr_warning("CAIF: %s(): " - "No interface to send disconnect to\n", - __func__); + pr_warn("No interface to send disconnect to\n"); ret = -ENODEV; goto end; } if (phyinfo->id != phyid || phyinfo->phy_layer->id != phyid || phyinfo->frm_layer->id != phyid) { - pr_err("CAIF: %s(): " - "Inconsistency in phy registration\n", - __func__); + pr_err("Inconsistency in phy registration\n"); ret = -EINVAL; goto end; } @@ -276,21 +274,20 @@ int cfcnfg_add_adaptation_layer(struct cfcnfg *cnfg, { struct cflayer *frml; if (adap_layer == NULL) { - pr_err("CAIF: %s(): adap_layer is zero", __func__); + pr_err("adap_layer is zero\n"); return -EINVAL; } if (adap_layer->receive == NULL) { - pr_err("CAIF: %s(): adap_layer->receive is NULL", __func__); + pr_err("adap_layer->receive is NULL\n"); return -EINVAL; } if (adap_layer->ctrlcmd == NULL) { - pr_err("CAIF: %s(): adap_layer->ctrlcmd == NULL", __func__); + pr_err("adap_layer->ctrlcmd == NULL\n"); return -EINVAL; } frml = cnfg->phy_layers[param->phyid].frm_layer; if (frml == NULL) { - pr_err("CAIF: %s(): Specified PHY type does not exist!", - __func__); + pr_err("Specified PHY type does not exist!\n"); return -ENODEV; } caif_assert(param->phyid == cnfg->phy_layers[param->phyid].id); @@ -330,9 +327,7 @@ cfcnfg_linkup_rsp(struct cflayer *layer, u8 channel_id, enum cfctrl_srv serv, struct net_device *netdev; if (adapt_layer == NULL) { - pr_debug("CAIF: %s(): link setup response " - "but no client exist, send linkdown back\n", - __func__); + pr_debug("link setup response but no client exist, send linkdown back\n"); cfctrl_linkdown_req(cnfg->ctrl, channel_id, NULL); return; } @@ -374,13 +369,11 @@ cfcnfg_linkup_rsp(struct cflayer *layer, u8 channel_id, enum cfctrl_srv serv, servicel = cfdbgl_create(channel_id, &phyinfo->dev_info); break; default: - pr_err("CAIF: %s(): Protocol error. " - "Link setup response - unknown channel type\n", - __func__); + pr_err("Protocol error. Link setup response - unknown channel type\n"); return; } if (!servicel) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return; } layer_set_dn(servicel, cnfg->mux); @@ -418,7 +411,7 @@ cfcnfg_add_phy_layer(struct cfcnfg *cnfg, enum cfcnfg_phy_type phy_type, } } if (*phyid == 0) { - pr_err("CAIF: %s(): No Available PHY ID\n", __func__); + pr_err("No Available PHY ID\n"); return; } @@ -427,7 +420,7 @@ cfcnfg_add_phy_layer(struct cfcnfg *cnfg, enum cfcnfg_phy_type phy_type, phy_driver = cfserl_create(CFPHYTYPE_FRAG, *phyid, stx); if (!phy_driver) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return; } @@ -436,7 +429,7 @@ cfcnfg_add_phy_layer(struct cfcnfg *cnfg, enum cfcnfg_phy_type phy_type, phy_driver = NULL; break; default: - pr_err("CAIF: %s(): %d", __func__, phy_type); + pr_err("%d\n", phy_type); return; break; } @@ -455,7 +448,7 @@ cfcnfg_add_phy_layer(struct cfcnfg *cnfg, enum cfcnfg_phy_type phy_type, phy_layer->type = phy_type; frml = cffrml_create(*phyid, fcs); if (!frml) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return; } cnfg->phy_layers[*phyid].frm_layer = frml; diff --git a/net/caif/cfctrl.c b/net/caif/cfctrl.c index 563145fdc4c..08f267a109a 100644 --- a/net/caif/cfctrl.c +++ b/net/caif/cfctrl.c @@ -4,6 +4,8 @@ * License terms: GNU General Public License (GPL) version 2 */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/stddef.h> #include <linux/spinlock.h> #include <linux/slab.h> @@ -36,7 +38,7 @@ struct cflayer *cfctrl_create(void) struct cfctrl *this = kmalloc(sizeof(struct cfctrl), GFP_ATOMIC); if (!this) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return NULL; } caif_assert(offsetof(struct cfctrl, serv.layer) == 0); @@ -132,9 +134,7 @@ struct cfctrl_request_info *cfctrl_remove_req(struct cfctrl *ctrl, list_for_each_entry_safe(p, tmp, &ctrl->list, list) { if (cfctrl_req_eq(req, p)) { if (p != first) - pr_warning("CAIF: %s(): Requests are not " - "received in order\n", - __func__); + pr_warn("Requests are not received in order\n"); atomic_set(&ctrl->rsp_seq_no, p->sequence_no); @@ -177,7 +177,7 @@ void cfctrl_enum_req(struct cflayer *layer, u8 physlinkid) int ret; struct cfpkt *pkt = cfpkt_create(CFPKT_CTRL_PKT_LEN); if (!pkt) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return; } caif_assert(offsetof(struct cfctrl, serv.layer) == 0); @@ -189,8 +189,7 @@ void cfctrl_enum_req(struct cflayer *layer, u8 physlinkid) ret = cfctrl->serv.layer.dn->transmit(cfctrl->serv.layer.dn, pkt); if (ret < 0) { - pr_err("CAIF: %s(): Could not transmit enum message\n", - __func__); + pr_err("Could not transmit enum message\n"); cfpkt_destroy(pkt); } } @@ -208,7 +207,7 @@ int cfctrl_linkup_request(struct cflayer *layer, char utility_name[16]; struct cfpkt *pkt = cfpkt_create(CFPKT_CTRL_PKT_LEN); if (!pkt) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return -ENOMEM; } cfpkt_addbdy(pkt, CFCTRL_CMD_LINK_SETUP); @@ -253,13 +252,13 @@ int cfctrl_linkup_request(struct cflayer *layer, param->u.utility.paramlen); break; default: - pr_warning("CAIF: %s():Request setup of bad link type = %d\n", - __func__, param->linktype); + pr_warn("Request setup of bad link type = %d\n", + param->linktype); return -EINVAL; } req = kzalloc(sizeof(*req), GFP_KERNEL); if (!req) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return -ENOMEM; } req->client_layer = user_layer; @@ -276,8 +275,7 @@ int cfctrl_linkup_request(struct cflayer *layer, ret = cfctrl->serv.layer.dn->transmit(cfctrl->serv.layer.dn, pkt); if (ret < 0) { - pr_err("CAIF: %s(): Could not transmit linksetup request\n", - __func__); + pr_err("Could not transmit linksetup request\n"); cfpkt_destroy(pkt); return -ENODEV; } @@ -291,7 +289,7 @@ int cfctrl_linkdown_req(struct cflayer *layer, u8 channelid, struct cfctrl *cfctrl = container_obj(layer); struct cfpkt *pkt = cfpkt_create(CFPKT_CTRL_PKT_LEN); if (!pkt) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return -ENOMEM; } cfpkt_addbdy(pkt, CFCTRL_CMD_LINK_DESTROY); @@ -300,8 +298,7 @@ int cfctrl_linkdown_req(struct cflayer *layer, u8 channelid, ret = cfctrl->serv.layer.dn->transmit(cfctrl->serv.layer.dn, pkt); if (ret < 0) { - pr_err("CAIF: %s(): Could not transmit link-down request\n", - __func__); + pr_err("Could not transmit link-down request\n"); cfpkt_destroy(pkt); } return ret; @@ -313,7 +310,7 @@ void cfctrl_sleep_req(struct cflayer *layer) struct cfctrl *cfctrl = container_obj(layer); struct cfpkt *pkt = cfpkt_create(CFPKT_CTRL_PKT_LEN); if (!pkt) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return; } cfpkt_addbdy(pkt, CFCTRL_CMD_SLEEP); @@ -330,7 +327,7 @@ void cfctrl_wake_req(struct cflayer *layer) struct cfctrl *cfctrl = container_obj(layer); struct cfpkt *pkt = cfpkt_create(CFPKT_CTRL_PKT_LEN); if (!pkt) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return; } cfpkt_addbdy(pkt, CFCTRL_CMD_WAKE); @@ -347,7 +344,7 @@ void cfctrl_getstartreason_req(struct cflayer *layer) struct cfctrl *cfctrl = container_obj(layer); struct cfpkt *pkt = cfpkt_create(CFPKT_CTRL_PKT_LEN); if (!pkt) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return; } cfpkt_addbdy(pkt, CFCTRL_CMD_START_REASON); @@ -364,12 +361,11 @@ void cfctrl_cancel_req(struct cflayer *layr, struct cflayer *adap_layer) struct cfctrl_request_info *p, *tmp; struct cfctrl *ctrl = container_obj(layr); spin_lock(&ctrl->info_list_lock); - pr_warning("CAIF: %s(): enter\n", __func__); + pr_warn("enter\n"); list_for_each_entry_safe(p, tmp, &ctrl->list, list) { if (p->client_layer == adap_layer) { - pr_warning("CAIF: %s(): cancel req :%d\n", __func__, - p->sequence_no); + pr_warn("cancel req :%d\n", p->sequence_no); list_del(&p->list); kfree(p); } @@ -520,9 +516,8 @@ static int cfctrl_recv(struct cflayer *layer, struct cfpkt *pkt) cfpkt_extr_head(pkt, ¶m, len); break; default: - pr_warning("CAIF: %s(): Request setup " - "- invalid link type (%d)", - __func__, serv); + pr_warn("Request setup - invalid link type (%d)\n", + serv); goto error; } @@ -532,9 +527,7 @@ static int cfctrl_recv(struct cflayer *layer, struct cfpkt *pkt) if (CFCTRL_ERR_BIT == (CFCTRL_ERR_BIT & cmdrsp) || cfpkt_erroneous(pkt)) { - pr_err("CAIF: %s(): Invalid O/E bit or parse " - "error on CAIF control channel", - __func__); + pr_err("Invalid O/E bit or parse error on CAIF control channel\n"); cfctrl->res.reject_rsp(cfctrl->serv.layer.up, 0, req ? req->client_layer @@ -556,8 +549,7 @@ static int cfctrl_recv(struct cflayer *layer, struct cfpkt *pkt) cfctrl->res.linkdestroy_rsp(cfctrl->serv.layer.up, linkid); break; case CFCTRL_CMD_LINK_ERR: - pr_err("CAIF: %s(): Frame Error Indication received\n", - __func__); + pr_err("Frame Error Indication received\n"); cfctrl->res.linkerror_ind(); break; case CFCTRL_CMD_ENUM: @@ -576,7 +568,7 @@ static int cfctrl_recv(struct cflayer *layer, struct cfpkt *pkt) cfctrl->res.radioset_rsp(); break; default: - pr_err("CAIF: %s(): Unrecognized Control Frame\n", __func__); + pr_err("Unrecognized Control Frame\n"); goto error; break; } @@ -595,8 +587,7 @@ static void cfctrl_ctrlcmd(struct cflayer *layr, enum caif_ctrlcmd ctrl, case CAIF_CTRLCMD_FLOW_OFF_IND: spin_lock(&this->info_list_lock); if (!list_empty(&this->list)) { - pr_debug("CAIF: %s(): Received flow off in " - "control layer", __func__); + pr_debug("Received flow off in control layer\n"); } spin_unlock(&this->info_list_lock); break; @@ -620,7 +611,7 @@ static int handle_loop(struct cfctrl *ctrl, int cmd, struct cfpkt *pkt) if (!ctrl->loop_linkused[linkid]) goto found; spin_unlock(&ctrl->loop_linkid_lock); - pr_err("CAIF: %s(): Out of link-ids\n", __func__); + pr_err("Out of link-ids\n"); return -EINVAL; found: if (!ctrl->loop_linkused[linkid]) diff --git a/net/caif/cfdbgl.c b/net/caif/cfdbgl.c index 676648cac8d..496fda9ac66 100644 --- a/net/caif/cfdbgl.c +++ b/net/caif/cfdbgl.c @@ -4,6 +4,8 @@ * License terms: GNU General Public License (GPL) version 2 */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/stddef.h> #include <linux/slab.h> #include <net/caif/caif_layer.h> @@ -17,7 +19,7 @@ struct cflayer *cfdbgl_create(u8 channel_id, struct dev_info *dev_info) { struct cfsrvl *dbg = kmalloc(sizeof(struct cfsrvl), GFP_ATOMIC); if (!dbg) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return NULL; } caif_assert(offsetof(struct cfsrvl, layer) == 0); diff --git a/net/caif/cfdgml.c b/net/caif/cfdgml.c index ed9d53aff28..d3ed264ad6c 100644 --- a/net/caif/cfdgml.c +++ b/net/caif/cfdgml.c @@ -4,6 +4,8 @@ * License terms: GNU General Public License (GPL) version 2 */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/stddef.h> #include <linux/spinlock.h> #include <linux/slab.h> @@ -26,7 +28,7 @@ struct cflayer *cfdgml_create(u8 channel_id, struct dev_info *dev_info) { struct cfsrvl *dgm = kmalloc(sizeof(struct cfsrvl), GFP_ATOMIC); if (!dgm) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return NULL; } caif_assert(offsetof(struct cfsrvl, layer) == 0); @@ -49,14 +51,14 @@ static int cfdgml_receive(struct cflayer *layr, struct cfpkt *pkt) caif_assert(layr->ctrlcmd != NULL); if (cfpkt_extr_head(pkt, &cmd, 1) < 0) { - pr_err("CAIF: %s(): Packet is erroneous!\n", __func__); + pr_err("Packet is erroneous!\n"); cfpkt_destroy(pkt); return -EPROTO; } if ((cmd & DGM_CMD_BIT) == 0) { if (cfpkt_extr_head(pkt, &dgmhdr, 3) < 0) { - pr_err("CAIF: %s(): Packet is erroneous!\n", __func__); + pr_err("Packet is erroneous!\n"); cfpkt_destroy(pkt); return -EPROTO; } @@ -75,8 +77,7 @@ static int cfdgml_receive(struct cflayer *layr, struct cfpkt *pkt) return 0; default: cfpkt_destroy(pkt); - pr_info("CAIF: %s(): Unknown datagram control %d (0x%x)\n", - __func__, cmd, cmd); + pr_info("Unknown datagram control %d (0x%x)\n", cmd, cmd); return -EPROTO; } } diff --git a/net/caif/cffrml.c b/net/caif/cffrml.c index e86a4ca3b21..a445043931a 100644 --- a/net/caif/cffrml.c +++ b/net/caif/cffrml.c @@ -6,6 +6,8 @@ * License terms: GNU General Public License (GPL) version 2 */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/stddef.h> #include <linux/spinlock.h> #include <linux/slab.h> @@ -32,7 +34,7 @@ struct cflayer *cffrml_create(u16 phyid, bool use_fcs) { struct cffrml *this = kmalloc(sizeof(struct cffrml), GFP_ATOMIC); if (!this) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return NULL; } caif_assert(offsetof(struct cffrml, layer) == 0); @@ -83,7 +85,7 @@ static int cffrml_receive(struct cflayer *layr, struct cfpkt *pkt) if (cfpkt_setlen(pkt, len) < 0) { ++cffrml_rcv_error; - pr_err("CAIF: %s():Framing length error (%d)\n", __func__, len); + pr_err("Framing length error (%d)\n", len); cfpkt_destroy(pkt); return -EPROTO; } @@ -99,14 +101,14 @@ static int cffrml_receive(struct cflayer *layr, struct cfpkt *pkt) cfpkt_add_trail(pkt, &tmp, 2); ++cffrml_rcv_error; ++cffrml_rcv_checsum_error; - pr_info("CAIF: %s(): Frame checksum error " - "(0x%x != 0x%x)\n", __func__, hdrchks, pktchks); + pr_info("Frame checksum error (0x%x != 0x%x)\n", + hdrchks, pktchks); return -EILSEQ; } } if (cfpkt_erroneous(pkt)) { ++cffrml_rcv_error; - pr_err("CAIF: %s(): Packet is erroneous!\n", __func__); + pr_err("Packet is erroneous!\n"); cfpkt_destroy(pkt); return -EPROTO; } @@ -132,7 +134,7 @@ static int cffrml_transmit(struct cflayer *layr, struct cfpkt *pkt) cfpkt_add_head(pkt, &tmp, 2); cfpkt_info(pkt)->hdr_len += 2; if (cfpkt_erroneous(pkt)) { - pr_err("CAIF: %s(): Packet is erroneous!\n", __func__); + pr_err("Packet is erroneous!\n"); return -EPROTO; } ret = layr->dn->transmit(layr->dn, pkt); diff --git a/net/caif/cfmuxl.c b/net/caif/cfmuxl.c index 80c8d332b25..46f34b2e047 100644 --- a/net/caif/cfmuxl.c +++ b/net/caif/cfmuxl.c @@ -3,6 +3,9 @@ * Author: Sjur Brendeland/sjur.brandeland@stericsson.com * License terms: GNU General Public License (GPL) version 2 */ + +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/stddef.h> #include <linux/spinlock.h> #include <linux/slab.h> @@ -190,7 +193,7 @@ static int cfmuxl_receive(struct cflayer *layr, struct cfpkt *pkt) u8 id; struct cflayer *up; if (cfpkt_extr_head(pkt, &id, 1) < 0) { - pr_err("CAIF: %s(): erroneous Caif Packet\n", __func__); + pr_err("erroneous Caif Packet\n"); cfpkt_destroy(pkt); return -EPROTO; } @@ -199,8 +202,8 @@ static int cfmuxl_receive(struct cflayer *layr, struct cfpkt *pkt) up = get_up(muxl, id); spin_unlock(&muxl->receive_lock); if (up == NULL) { - pr_info("CAIF: %s():Received data on unknown link ID = %d " - "(0x%x) up == NULL", __func__, id, id); + pr_info("Received data on unknown link ID = %d (0x%x) up == NULL", + id, id); cfpkt_destroy(pkt); /* * Don't return ERROR, since modem misbehaves and sends out @@ -223,9 +226,8 @@ static int cfmuxl_transmit(struct cflayer *layr, struct cfpkt *pkt) struct caif_payload_info *info = cfpkt_info(pkt); dn = get_dn(muxl, cfpkt_info(pkt)->dev_info); if (dn == NULL) { - pr_warning("CAIF: %s(): Send data on unknown phy " - "ID = %d (0x%x)\n", - __func__, info->dev_info->id, info->dev_info->id); + pr_warn("Send data on unknown phy ID = %d (0x%x)\n", + info->dev_info->id, info->dev_info->id); return -ENOTCONN; } info->hdr_len += 1; diff --git a/net/caif/cfpkt_skbuff.c b/net/caif/cfpkt_skbuff.c index 01f238ff234..d7e865e2ff6 100644 --- a/net/caif/cfpkt_skbuff.c +++ b/net/caif/cfpkt_skbuff.c @@ -4,19 +4,22 @@ * License terms: GNU General Public License (GPL) version 2 */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/string.h> #include <linux/skbuff.h> #include <linux/hardirq.h> #include <net/caif/cfpkt.h> -#define PKT_PREFIX 16 +#define PKT_PREFIX 48 #define PKT_POSTFIX 2 #define PKT_LEN_WHEN_EXTENDING 128 -#define PKT_ERROR(pkt, errmsg) do { \ - cfpkt_priv(pkt)->erronous = true; \ - skb_reset_tail_pointer(&pkt->skb); \ - pr_warning("CAIF: " errmsg);\ - } while (0) +#define PKT_ERROR(pkt, errmsg) \ +do { \ + cfpkt_priv(pkt)->erronous = true; \ + skb_reset_tail_pointer(&pkt->skb); \ + pr_warn(errmsg); \ +} while (0) struct cfpktq { struct sk_buff_head head; @@ -130,13 +133,13 @@ int cfpkt_extr_head(struct cfpkt *pkt, void *data, u16 len) return -EPROTO; if (unlikely(len > skb->len)) { - PKT_ERROR(pkt, "cfpkt_extr_head read beyond end of packet\n"); + PKT_ERROR(pkt, "read beyond end of packet\n"); return -EPROTO; } if (unlikely(len > skb_headlen(skb))) { if (unlikely(skb_linearize(skb) != 0)) { - PKT_ERROR(pkt, "cfpkt_extr_head linearize failed\n"); + PKT_ERROR(pkt, "linearize failed\n"); return -EPROTO; } } @@ -156,11 +159,11 @@ int cfpkt_extr_trail(struct cfpkt *pkt, void *dta, u16 len) return -EPROTO; if (unlikely(skb_linearize(skb) != 0)) { - PKT_ERROR(pkt, "cfpkt_extr_trail linearize failed\n"); + PKT_ERROR(pkt, "linearize failed\n"); return -EPROTO; } if (unlikely(skb->data + len > skb_tail_pointer(skb))) { - PKT_ERROR(pkt, "cfpkt_extr_trail read beyond end of packet\n"); + PKT_ERROR(pkt, "read beyond end of packet\n"); return -EPROTO; } from = skb_tail_pointer(skb) - len; @@ -202,7 +205,7 @@ int cfpkt_add_body(struct cfpkt *pkt, const void *data, u16 len) /* Make sure data is writable */ if (unlikely(skb_cow_data(skb, addlen, &lastskb) < 0)) { - PKT_ERROR(pkt, "cfpkt_add_body: cow failed\n"); + PKT_ERROR(pkt, "cow failed\n"); return -EPROTO; } /* @@ -211,8 +214,7 @@ int cfpkt_add_body(struct cfpkt *pkt, const void *data, u16 len) * lengths of the top SKB. */ if (lastskb != skb) { - pr_warning("CAIF: %s(): Packet is non-linear\n", - __func__); + pr_warn("Packet is non-linear\n"); skb->len += len; skb->data_len += len; } @@ -242,14 +244,14 @@ int cfpkt_add_head(struct cfpkt *pkt, const void *data2, u16 len) if (unlikely(is_erronous(pkt))) return -EPROTO; if (unlikely(skb_headroom(skb) < len)) { - PKT_ERROR(pkt, "cfpkt_add_head: no headroom\n"); + PKT_ERROR(pkt, "no headroom\n"); return -EPROTO; } /* Make sure data is writable */ ret = skb_cow_data(skb, 0, &lastskb); if (unlikely(ret < 0)) { - PKT_ERROR(pkt, "cfpkt_add_head: cow failed\n"); + PKT_ERROR(pkt, "cow failed\n"); return ret; } @@ -283,7 +285,7 @@ inline u16 cfpkt_iterate(struct cfpkt *pkt, if (unlikely(is_erronous(pkt))) return -EPROTO; if (unlikely(skb_linearize(&pkt->skb) != 0)) { - PKT_ERROR(pkt, "cfpkt_iterate: linearize failed\n"); + PKT_ERROR(pkt, "linearize failed\n"); return -EPROTO; } return iter_func(data, pkt->skb.data, cfpkt_getlen(pkt)); @@ -309,7 +311,7 @@ int cfpkt_setlen(struct cfpkt *pkt, u16 len) /* Need to expand SKB */ if (unlikely(!cfpkt_pad_trail(pkt, len - skb->len))) - PKT_ERROR(pkt, "cfpkt_setlen: skb_pad_trail failed\n"); + PKT_ERROR(pkt, "skb_pad_trail failed\n"); return cfpkt_getlen(pkt); } @@ -380,8 +382,7 @@ struct cfpkt *cfpkt_split(struct cfpkt *pkt, u16 pos) return NULL; if (skb->data + pos > skb_tail_pointer(skb)) { - PKT_ERROR(pkt, - "cfpkt_split: trying to split beyond end of packet"); + PKT_ERROR(pkt, "trying to split beyond end of packet\n"); return NULL; } @@ -455,17 +456,17 @@ int cfpkt_raw_append(struct cfpkt *pkt, void **buf, unsigned int buflen) return -EPROTO; /* Make sure SKB is writable */ if (unlikely(skb_cow_data(skb, 0, &lastskb) < 0)) { - PKT_ERROR(pkt, "cfpkt_raw_append: skb_cow_data failed\n"); + PKT_ERROR(pkt, "skb_cow_data failed\n"); return -EPROTO; } if (unlikely(skb_linearize(skb) != 0)) { - PKT_ERROR(pkt, "cfpkt_raw_append: linearize failed\n"); + PKT_ERROR(pkt, "linearize failed\n"); return -EPROTO; } if (unlikely(skb_tailroom(skb) < buflen)) { - PKT_ERROR(pkt, "cfpkt_raw_append: buffer too short - failed\n"); + PKT_ERROR(pkt, "buffer too short - failed\n"); return -EPROTO; } @@ -483,14 +484,13 @@ int cfpkt_raw_extract(struct cfpkt *pkt, void **buf, unsigned int buflen) return -EPROTO; if (unlikely(buflen > skb->len)) { - PKT_ERROR(pkt, "cfpkt_raw_extract: buflen too large " - "- failed\n"); + PKT_ERROR(pkt, "buflen too large - failed\n"); return -EPROTO; } if (unlikely(buflen > skb_headlen(skb))) { if (unlikely(skb_linearize(skb) != 0)) { - PKT_ERROR(pkt, "cfpkt_raw_extract: linearize failed\n"); + PKT_ERROR(pkt, "linearize failed\n"); return -EPROTO; } } diff --git a/net/caif/cfrfml.c b/net/caif/cfrfml.c index eb1602022ac..bde8481e8d2 100644 --- a/net/caif/cfrfml.c +++ b/net/caif/cfrfml.c @@ -4,10 +4,12 @@ * License terms: GNU General Public License (GPL) version 2 */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/stddef.h> #include <linux/spinlock.h> #include <linux/slab.h> -#include <linux/unaligned/le_byteshift.h> +#include <asm/unaligned.h> #include <net/caif/caif_layer.h> #include <net/caif/cfsrvl.h> #include <net/caif/cfpkt.h> @@ -48,7 +50,7 @@ struct cflayer *cfrfml_create(u8 channel_id, struct dev_info *dev_info, kzalloc(sizeof(struct cfrfml), GFP_ATOMIC); if (!this) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return NULL; } @@ -178,9 +180,7 @@ out: cfpkt_destroy(rfml->incomplete_frm); rfml->incomplete_frm = NULL; - pr_info("CAIF: %s(): " - "Connection error %d triggered on RFM link\n", - __func__, err); + pr_info("Connection error %d triggered on RFM link\n", err); /* Trigger connection error upon failure.*/ layr->up->ctrlcmd(layr->up, CAIF_CTRLCMD_REMOTE_SHUTDOWN_IND, @@ -280,9 +280,7 @@ static int cfrfml_transmit(struct cflayer *layr, struct cfpkt *pkt) out: if (err != 0) { - pr_info("CAIF: %s(): " - "Connection error %d triggered on RFM link\n", - __func__, err); + pr_info("Connection error %d triggered on RFM link\n", err); /* Trigger connection error upon failure.*/ layr->up->ctrlcmd(layr->up, CAIF_CTRLCMD_REMOTE_SHUTDOWN_IND, diff --git a/net/caif/cfserl.c b/net/caif/cfserl.c index a11fbd68a13..9297f7dea9d 100644 --- a/net/caif/cfserl.c +++ b/net/caif/cfserl.c @@ -4,6 +4,8 @@ * License terms: GNU General Public License (GPL) version 2 */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/stddef.h> #include <linux/spinlock.h> #include <linux/slab.h> @@ -34,7 +36,7 @@ struct cflayer *cfserl_create(int type, int instance, bool use_stx) { struct cfserl *this = kmalloc(sizeof(struct cfserl), GFP_ATOMIC); if (!this) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return NULL; } caif_assert(offsetof(struct cfserl, layer) == 0); diff --git a/net/caif/cfsrvl.c b/net/caif/cfsrvl.c index f40939a9121..ab5e542526b 100644 --- a/net/caif/cfsrvl.c +++ b/net/caif/cfsrvl.c @@ -4,6 +4,8 @@ * License terms: GNU General Public License (GPL) version 2 */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/kernel.h> #include <linux/types.h> #include <linux/errno.h> @@ -79,8 +81,7 @@ static void cfservl_ctrlcmd(struct cflayer *layr, enum caif_ctrlcmd ctrl, layr->up->ctrlcmd(layr->up, ctrl, phyid); break; default: - pr_warning("CAIF: %s(): " - "Unexpected ctrl in cfsrvl (%d)\n", __func__, ctrl); + pr_warn("Unexpected ctrl in cfsrvl (%d)\n", ctrl); /* We have both modem and phy flow on, send flow on */ layr->up->ctrlcmd(layr->up, ctrl, phyid); service->phy_flow_on = true; @@ -107,14 +108,12 @@ static int cfservl_modemcmd(struct cflayer *layr, enum caif_modemcmd ctrl) u8 flow_on = SRVL_FLOW_ON; pkt = cfpkt_create(SRVL_CTRL_PKT_SIZE); if (!pkt) { - pr_warning("CAIF: %s(): Out of memory\n", - __func__); + pr_warn("Out of memory\n"); return -ENOMEM; } if (cfpkt_add_head(pkt, &flow_on, 1) < 0) { - pr_err("CAIF: %s(): Packet is erroneous!\n", - __func__); + pr_err("Packet is erroneous!\n"); cfpkt_destroy(pkt); return -EPROTO; } @@ -131,14 +130,12 @@ static int cfservl_modemcmd(struct cflayer *layr, enum caif_modemcmd ctrl) u8 flow_off = SRVL_FLOW_OFF; pkt = cfpkt_create(SRVL_CTRL_PKT_SIZE); if (!pkt) { - pr_warning("CAIF: %s(): Out of memory\n", - __func__); + pr_warn("Out of memory\n"); return -ENOMEM; } if (cfpkt_add_head(pkt, &flow_off, 1) < 0) { - pr_err("CAIF: %s(): Packet is erroneous!\n", - __func__); + pr_err("Packet is erroneous!\n"); cfpkt_destroy(pkt); return -EPROTO; } diff --git a/net/caif/cfutill.c b/net/caif/cfutill.c index 02795aff57a..efad410e4c8 100644 --- a/net/caif/cfutill.c +++ b/net/caif/cfutill.c @@ -4,6 +4,8 @@ * License terms: GNU General Public License (GPL) version 2 */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/kernel.h> #include <linux/types.h> #include <linux/slab.h> @@ -26,7 +28,7 @@ struct cflayer *cfutill_create(u8 channel_id, struct dev_info *dev_info) { struct cfsrvl *util = kmalloc(sizeof(struct cfsrvl), GFP_ATOMIC); if (!util) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return NULL; } caif_assert(offsetof(struct cfsrvl, layer) == 0); @@ -47,7 +49,7 @@ static int cfutill_receive(struct cflayer *layr, struct cfpkt *pkt) caif_assert(layr->up->receive != NULL); caif_assert(layr->up->ctrlcmd != NULL); if (cfpkt_extr_head(pkt, &cmd, 1) < 0) { - pr_err("CAIF: %s(): Packet is erroneous!\n", __func__); + pr_err("Packet is erroneous!\n"); cfpkt_destroy(pkt); return -EPROTO; } @@ -64,16 +66,14 @@ static int cfutill_receive(struct cflayer *layr, struct cfpkt *pkt) cfpkt_destroy(pkt); return 0; case UTIL_REMOTE_SHUTDOWN: /* Remote Shutdown Request */ - pr_err("CAIF: %s(): REMOTE SHUTDOWN REQUEST RECEIVED\n", - __func__); + pr_err("REMOTE SHUTDOWN REQUEST RECEIVED\n"); layr->ctrlcmd(layr, CAIF_CTRLCMD_REMOTE_SHUTDOWN_IND, 0); service->open = false; cfpkt_destroy(pkt); return 0; default: cfpkt_destroy(pkt); - pr_warning("CAIF: %s(): Unknown service control %d (0x%x)\n", - __func__, cmd, cmd); + pr_warn("Unknown service control %d (0x%x)\n", cmd, cmd); return -EPROTO; } } diff --git a/net/caif/cfveil.c b/net/caif/cfveil.c index 77cc09faac9..3b425b189a9 100644 --- a/net/caif/cfveil.c +++ b/net/caif/cfveil.c @@ -4,6 +4,8 @@ * License terms: GNU General Public License (GPL) version 2 */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/stddef.h> #include <linux/slab.h> #include <net/caif/caif_layer.h> @@ -25,7 +27,7 @@ struct cflayer *cfvei_create(u8 channel_id, struct dev_info *dev_info) { struct cfsrvl *vei = kmalloc(sizeof(struct cfsrvl), GFP_ATOMIC); if (!vei) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return NULL; } caif_assert(offsetof(struct cfsrvl, layer) == 0); @@ -47,7 +49,7 @@ static int cfvei_receive(struct cflayer *layr, struct cfpkt *pkt) if (cfpkt_extr_head(pkt, &cmd, 1) < 0) { - pr_err("CAIF: %s(): Packet is erroneous!\n", __func__); + pr_err("Packet is erroneous!\n"); cfpkt_destroy(pkt); return -EPROTO; } @@ -67,8 +69,7 @@ static int cfvei_receive(struct cflayer *layr, struct cfpkt *pkt) cfpkt_destroy(pkt); return 0; default: /* SET RS232 PIN */ - pr_warning("CAIF: %s():Unknown VEI control packet %d (0x%x)!\n", - __func__, cmd, cmd); + pr_warn("Unknown VEI control packet %d (0x%x)!\n", cmd, cmd); cfpkt_destroy(pkt); return -EPROTO; } @@ -86,7 +87,7 @@ static int cfvei_transmit(struct cflayer *layr, struct cfpkt *pkt) caif_assert(layr->dn->transmit != NULL); if (cfpkt_add_head(pkt, &tmp, 1) < 0) { - pr_err("CAIF: %s(): Packet is erroneous!\n", __func__); + pr_err("Packet is erroneous!\n"); return -EPROTO; } diff --git a/net/caif/cfvidl.c b/net/caif/cfvidl.c index ada6ee2d48f..bf6fef2a0ef 100644 --- a/net/caif/cfvidl.c +++ b/net/caif/cfvidl.c @@ -4,6 +4,8 @@ * License terms: GNU General Public License (GPL) version 2 */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/kernel.h> #include <linux/types.h> #include <linux/slab.h> @@ -21,7 +23,7 @@ struct cflayer *cfvidl_create(u8 channel_id, struct dev_info *dev_info) { struct cfsrvl *vid = kmalloc(sizeof(struct cfsrvl), GFP_ATOMIC); if (!vid) { - pr_warning("CAIF: %s(): Out of memory\n", __func__); + pr_warn("Out of memory\n"); return NULL; } caif_assert(offsetof(struct cfsrvl, layer) == 0); @@ -38,7 +40,7 @@ static int cfvidl_receive(struct cflayer *layr, struct cfpkt *pkt) { u32 videoheader; if (cfpkt_extr_head(pkt, &videoheader, 4) < 0) { - pr_err("CAIF: %s(): Packet is erroneous!\n", __func__); + pr_err("Packet is erroneous!\n"); cfpkt_destroy(pkt); return -EPROTO; } diff --git a/net/caif/chnl_net.c b/net/caif/chnl_net.c index 4293e190ec5..84a422c9894 100644 --- a/net/caif/chnl_net.c +++ b/net/caif/chnl_net.c @@ -5,6 +5,8 @@ * License terms: GNU General Public License (GPL) version 2 */ +#define pr_fmt(fmt) KBUILD_MODNAME ":%s(): " fmt, __func__ + #include <linux/version.h> #include <linux/fs.h> #include <linux/init.h> @@ -28,9 +30,6 @@ #define CONNECT_TIMEOUT (5 * HZ) #define CAIF_NET_DEFAULT_QUEUE_LEN 500 -#undef pr_debug -#define pr_debug pr_warning - /*This list is protected by the rtnl lock. */ static LIST_HEAD(chnl_net_list); @@ -142,8 +141,7 @@ static void chnl_flowctrl_cb(struct cflayer *layr, enum caif_ctrlcmd flow, int phyid) { struct chnl_net *priv = container_of(layr, struct chnl_net, chnl); - pr_debug("CAIF: %s(): NET flowctrl func called flow: %s\n", - __func__, + pr_debug("NET flowctrl func called flow: %s\n", flow == CAIF_CTRLCMD_FLOW_ON_IND ? "ON" : flow == CAIF_CTRLCMD_INIT_RSP ? "INIT" : flow == CAIF_CTRLCMD_FLOW_OFF_IND ? "OFF" : @@ -196,12 +194,12 @@ static int chnl_net_start_xmit(struct sk_buff *skb, struct net_device *dev) priv = netdev_priv(dev); if (skb->len > priv->netdev->mtu) { - pr_warning("CAIF: %s(): Size of skb exceeded MTU\n", __func__); + pr_warn("Size of skb exceeded MTU\n"); return -ENOSPC; } if (!priv->flowenabled) { - pr_debug("CAIF: %s(): dropping packets flow off\n", __func__); + pr_debug("dropping packets flow off\n"); return NETDEV_TX_BUSY; } @@ -237,7 +235,7 @@ static int chnl_net_open(struct net_device *dev) ASSERT_RTNL(); priv = netdev_priv(dev); if (!priv) { - pr_debug("CAIF: %s(): chnl_net_open: no priv\n", __func__); + pr_debug("chnl_net_open: no priv\n"); return -ENODEV; } @@ -246,18 +244,17 @@ static int chnl_net_open(struct net_device *dev) result = caif_connect_client(&priv->conn_req, &priv->chnl, &llifindex, &headroom, &tailroom); if (result != 0) { - pr_debug("CAIF: %s(): err: " - "Unable to register and open device," - " Err:%d\n", - __func__, - result); + pr_debug("err: " + "Unable to register and open device," + " Err:%d\n", + result); goto error; } lldev = dev_get_by_index(dev_net(dev), llifindex); if (lldev == NULL) { - pr_debug("CAIF: %s(): no interface?\n", __func__); + pr_debug("no interface?\n"); result = -ENODEV; goto error; } @@ -279,9 +276,7 @@ static int chnl_net_open(struct net_device *dev) dev_put(lldev); if (mtu < 100) { - pr_warning("CAIF: %s(): " - "CAIF Interface MTU too small (%d)\n", - __func__, mtu); + pr_warn("CAIF Interface MTU too small (%d)\n", mtu); result = -ENODEV; goto error; } @@ -296,33 +291,32 @@ static int chnl_net_open(struct net_device *dev) rtnl_lock(); if (result == -ERESTARTSYS) { - pr_debug("CAIF: %s(): wait_event_interruptible" - " woken by a signal\n", __func__); + pr_debug("wait_event_interruptible woken by a signal\n"); result = -ERESTARTSYS; goto error; } if (result == 0) { - pr_debug("CAIF: %s(): connect timeout\n", __func__); + pr_debug("connect timeout\n"); caif_disconnect_client(&priv->chnl); priv->state = CAIF_DISCONNECTED; - pr_debug("CAIF: %s(): state disconnected\n", __func__); + pr_debug("state disconnected\n"); result = -ETIMEDOUT; goto error; } if (priv->state != CAIF_CONNECTED) { - pr_debug("CAIF: %s(): connect failed\n", __func__); + pr_debug("connect failed\n"); result = -ECONNREFUSED; goto error; } - pr_debug("CAIF: %s(): CAIF Netdevice connected\n", __func__); + pr_debug("CAIF Netdevice connected\n"); return 0; error: caif_disconnect_client(&priv->chnl); priv->state = CAIF_DISCONNECTED; - pr_debug("CAIF: %s(): state disconnected\n", __func__); + pr_debug("state disconnected\n"); return result; } @@ -413,7 +407,7 @@ static void caif_netlink_parms(struct nlattr *data[], struct caif_connect_request *conn_req) { if (!data) { - pr_warning("CAIF: %s: no params data found\n", __func__); + pr_warn("no params data found\n"); return; } if (data[IFLA_CAIF_IPV4_CONNID]) @@ -442,8 +436,7 @@ static int ipcaif_newlink(struct net *src_net, struct net_device *dev, ret = register_netdevice(dev); if (ret) - pr_warning("CAIF: %s(): device rtml registration failed\n", - __func__); + pr_warn("device rtml registration failed\n"); return ret; } diff --git a/net/can/bcm.c b/net/can/bcm.c index 9c65e9deb9c..08ffe9e4be2 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -60,6 +60,13 @@ #include <net/sock.h> #include <net/net_namespace.h> +/* + * To send multiple CAN frame content within TX_SETUP or to filter + * CAN messages with multiplex index within RX_SETUP, the number of + * different filters is limited to 256 due to the one byte index value. + */ +#define MAX_NFRAMES 256 + /* use of last_frames[index].can_dlc */ #define RX_RECV 0x40 /* received data for this element */ #define RX_THR 0x80 /* element not been sent due to throttle feature */ @@ -89,16 +96,16 @@ struct bcm_op { struct list_head list; int ifindex; canid_t can_id; - int flags; + u32 flags; unsigned long frames_abs, frames_filtered; struct timeval ival1, ival2; struct hrtimer timer, thrtimer; struct tasklet_struct tsklet, thrtsklet; ktime_t rx_stamp, kt_ival1, kt_ival2, kt_lastmsg; int rx_ifindex; - int count; - int nframes; - int currframe; + u32 count; + u32 nframes; + u32 currframe; struct can_frame *frames; struct can_frame *last_frames; struct can_frame sframe; @@ -175,7 +182,7 @@ static int bcm_proc_show(struct seq_file *m, void *v) seq_printf(m, "rx_op: %03X %-5s ", op->can_id, bcm_proc_getifname(ifname, op->ifindex)); - seq_printf(m, "[%d]%c ", op->nframes, + seq_printf(m, "[%u]%c ", op->nframes, (op->flags & RX_CHECK_DLC)?'d':' '); if (op->kt_ival1.tv64) seq_printf(m, "timeo=%lld ", @@ -198,7 +205,7 @@ static int bcm_proc_show(struct seq_file *m, void *v) list_for_each_entry(op, &bo->tx_ops, list) { - seq_printf(m, "tx_op: %03X %s [%d] ", + seq_printf(m, "tx_op: %03X %s [%u] ", op->can_id, bcm_proc_getifname(ifname, op->ifindex), op->nframes); @@ -283,7 +290,7 @@ static void bcm_send_to_user(struct bcm_op *op, struct bcm_msg_head *head, struct can_frame *firstframe; struct sockaddr_can *addr; struct sock *sk = op->sk; - int datalen = head->nframes * CFSIZ; + unsigned int datalen = head->nframes * CFSIZ; int err; skb = alloc_skb(sizeof(*head) + datalen, gfp_any()); @@ -468,7 +475,7 @@ rx_changed_settime: * bcm_rx_cmp_to_index - (bit)compares the currently received data to formerly * received data stored in op->last_frames[] */ -static void bcm_rx_cmp_to_index(struct bcm_op *op, int index, +static void bcm_rx_cmp_to_index(struct bcm_op *op, unsigned int index, const struct can_frame *rxdata) { /* @@ -554,7 +561,8 @@ static enum hrtimer_restart bcm_rx_timeout_handler(struct hrtimer *hrtimer) /* * bcm_rx_do_flush - helper for bcm_rx_thr_flush */ -static inline int bcm_rx_do_flush(struct bcm_op *op, int update, int index) +static inline int bcm_rx_do_flush(struct bcm_op *op, int update, + unsigned int index) { if ((op->last_frames) && (op->last_frames[index].can_dlc & RX_THR)) { if (update) @@ -575,7 +583,7 @@ static int bcm_rx_thr_flush(struct bcm_op *op, int update) int updated = 0; if (op->nframes > 1) { - int i; + unsigned int i; /* for MUX filter we start at index 1 */ for (i = 1; i < op->nframes; i++) @@ -624,7 +632,7 @@ static void bcm_rx_handler(struct sk_buff *skb, void *data) { struct bcm_op *op = (struct bcm_op *)data; const struct can_frame *rxframe = (struct can_frame *)skb->data; - int i; + unsigned int i; /* disable timeout */ hrtimer_cancel(&op->timer); @@ -822,14 +830,15 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, { struct bcm_sock *bo = bcm_sk(sk); struct bcm_op *op; - int i, err; + unsigned int i; + int err; /* we need a real device to send frames */ if (!ifindex) return -ENODEV; - /* we need at least one can_frame */ - if (msg_head->nframes < 1) + /* check nframes boundaries - we need at least one can_frame */ + if (msg_head->nframes < 1 || msg_head->nframes > MAX_NFRAMES) return -EINVAL; /* check the given can_id */ @@ -993,6 +1002,10 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, msg_head->nframes = 0; } + /* the first element contains the mux-mask => MAX_NFRAMES + 1 */ + if (msg_head->nframes > MAX_NFRAMES + 1) + return -EINVAL; + if ((msg_head->flags & RX_RTR_FRAME) && ((msg_head->nframes != 1) || (!(msg_head->can_id & CAN_RTR_FLAG)))) diff --git a/net/can/raw.c b/net/can/raw.c index a10e3338f08..e88f610fdb7 100644 --- a/net/can/raw.c +++ b/net/can/raw.c @@ -90,23 +90,39 @@ struct raw_sock { can_err_mask_t err_mask; }; +/* + * Return pointer to store the extra msg flags for raw_recvmsg(). + * We use the space of one unsigned int beyond the 'struct sockaddr_can' + * in skb->cb. + */ +static inline unsigned int *raw_flags(struct sk_buff *skb) +{ + BUILD_BUG_ON(sizeof(skb->cb) <= (sizeof(struct sockaddr_can) + + sizeof(unsigned int))); + + /* return pointer after struct sockaddr_can */ + return (unsigned int *)(&((struct sockaddr_can *)skb->cb)[1]); +} + static inline struct raw_sock *raw_sk(const struct sock *sk) { return (struct raw_sock *)sk; } -static void raw_rcv(struct sk_buff *skb, void *data) +static void raw_rcv(struct sk_buff *oskb, void *data) { struct sock *sk = (struct sock *)data; struct raw_sock *ro = raw_sk(sk); struct sockaddr_can *addr; + struct sk_buff *skb; + unsigned int *pflags; /* check the received tx sock reference */ - if (!ro->recv_own_msgs && skb->sk == sk) + if (!ro->recv_own_msgs && oskb->sk == sk) return; /* clone the given skb to be able to enqueue it into the rcv queue */ - skb = skb_clone(skb, GFP_ATOMIC); + skb = skb_clone(oskb, GFP_ATOMIC); if (!skb) return; @@ -123,6 +139,14 @@ static void raw_rcv(struct sk_buff *skb, void *data) addr->can_family = AF_CAN; addr->can_ifindex = skb->dev->ifindex; + /* add CAN specific message flags for raw_recvmsg() */ + pflags = raw_flags(skb); + *pflags = 0; + if (oskb->sk) + *pflags |= MSG_DONTROUTE; + if (oskb->sk == sk) + *pflags |= MSG_CONFIRM; + if (sock_queue_rcv_skb(sk, skb) < 0) kfree_skb(skb); } @@ -647,12 +671,12 @@ static int raw_sendmsg(struct kiocb *iocb, struct socket *sock, err = memcpy_fromiovec(skb_put(skb, size), msg->msg_iov, size); if (err < 0) goto free_skb; - err = sock_tx_timestamp(msg, sk, skb_tx(skb)); + err = sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags); if (err < 0) goto free_skb; /* to be able to check the received tx sock reference in raw_rcv() */ - skb_tx(skb)->prevent_sk_orphan = 1; + skb_shinfo(skb)->tx_flags |= SKBTX_DRV_NEEDS_SK_REF; skb->dev = dev; skb->sk = sk; @@ -707,6 +731,9 @@ static int raw_recvmsg(struct kiocb *iocb, struct socket *sock, memcpy(msg->msg_name, skb->cb, msg->msg_namelen); } + /* assign the flags that have been recorded in raw_rcv() */ + msg->msg_flags |= *(raw_flags(skb)); + skb_free_datagram(sk, skb); return size; diff --git a/net/ceph/Kconfig b/net/ceph/Kconfig new file mode 100644 index 00000000000..ad424049b0c --- /dev/null +++ b/net/ceph/Kconfig @@ -0,0 +1,28 @@ +config CEPH_LIB + tristate "Ceph core library (EXPERIMENTAL)" + depends on INET && EXPERIMENTAL + select LIBCRC32C + select CRYPTO_AES + select CRYPTO + default n + help + Choose Y or M here to include cephlib, which provides the + common functionality to both the Ceph filesystem and + to the rados block device (rbd). + + More information at http://ceph.newdream.net/. + + If unsure, say N. + +config CEPH_LIB_PRETTYDEBUG + bool "Include file:line in ceph debug output" + depends on CEPH_LIB + default n + help + If you say Y here, debug output will include a filename and + line to aid debugging. This increases kernel size and slows + execution slightly when debug call sites are enabled (e.g., + via CONFIG_DYNAMIC_DEBUG). + + If unsure, say N. + diff --git a/net/ceph/Makefile b/net/ceph/Makefile new file mode 100644 index 00000000000..aab1cabb803 --- /dev/null +++ b/net/ceph/Makefile @@ -0,0 +1,37 @@ +# +# Makefile for CEPH filesystem. +# + +ifneq ($(KERNELRELEASE),) + +obj-$(CONFIG_CEPH_LIB) += libceph.o + +libceph-objs := ceph_common.o messenger.o msgpool.o buffer.o pagelist.o \ + mon_client.o \ + osd_client.o osdmap.o crush/crush.o crush/mapper.o crush/hash.o \ + debugfs.o \ + auth.o auth_none.o \ + crypto.o armor.o \ + auth_x.o \ + ceph_fs.o ceph_strings.o ceph_hash.o \ + pagevec.o + +else +#Otherwise we were called directly from the command +# line; invoke the kernel build system. + +KERNELDIR ?= /lib/modules/$(shell uname -r)/build +PWD := $(shell pwd) + +default: all + +all: + $(MAKE) -C $(KERNELDIR) M=$(PWD) CONFIG_CEPH_LIB=m modules + +modules_install: + $(MAKE) -C $(KERNELDIR) M=$(PWD) CONFIG_CEPH_LIB=m modules_install + +clean: + $(MAKE) -C $(KERNELDIR) M=$(PWD) clean + +endif diff --git a/net/ceph/armor.c b/net/ceph/armor.c new file mode 100644 index 00000000000..eb2a666b0be --- /dev/null +++ b/net/ceph/armor.c @@ -0,0 +1,103 @@ + +#include <linux/errno.h> + +int ceph_armor(char *dst, const char *src, const char *end); +int ceph_unarmor(char *dst, const char *src, const char *end); + +/* + * base64 encode/decode. + */ + +static const char *pem_key = + "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"; + +static int encode_bits(int c) +{ + return pem_key[c]; +} + +static int decode_bits(char c) +{ + if (c >= 'A' && c <= 'Z') + return c - 'A'; + if (c >= 'a' && c <= 'z') + return c - 'a' + 26; + if (c >= '0' && c <= '9') + return c - '0' + 52; + if (c == '+') + return 62; + if (c == '/') + return 63; + if (c == '=') + return 0; /* just non-negative, please */ + return -EINVAL; +} + +int ceph_armor(char *dst, const char *src, const char *end) +{ + int olen = 0; + int line = 0; + + while (src < end) { + unsigned char a, b, c; + + a = *src++; + *dst++ = encode_bits(a >> 2); + if (src < end) { + b = *src++; + *dst++ = encode_bits(((a & 3) << 4) | (b >> 4)); + if (src < end) { + c = *src++; + *dst++ = encode_bits(((b & 15) << 2) | + (c >> 6)); + *dst++ = encode_bits(c & 63); + } else { + *dst++ = encode_bits((b & 15) << 2); + *dst++ = '='; + } + } else { + *dst++ = encode_bits(((a & 3) << 4)); + *dst++ = '='; + *dst++ = '='; + } + olen += 4; + line += 4; + if (line == 64) { + line = 0; + *(dst++) = '\n'; + olen++; + } + } + return olen; +} + +int ceph_unarmor(char *dst, const char *src, const char *end) +{ + int olen = 0; + + while (src < end) { + int a, b, c, d; + + if (src < end && src[0] == '\n') + src++; + if (src + 4 > end) + return -EINVAL; + a = decode_bits(src[0]); + b = decode_bits(src[1]); + c = decode_bits(src[2]); + d = decode_bits(src[3]); + if (a < 0 || b < 0 || c < 0 || d < 0) + return -EINVAL; + + *dst++ = (a << 2) | (b >> 4); + if (src[2] == '=') + return olen + 1; + *dst++ = ((b & 15) << 4) | (c >> 2); + if (src[3] == '=') + return olen + 2; + *dst++ = ((c & 3) << 6) | d; + olen += 3; + src += 4; + } + return olen; +} diff --git a/net/ceph/auth.c b/net/ceph/auth.c new file mode 100644 index 00000000000..549c1f43e1d --- /dev/null +++ b/net/ceph/auth.c @@ -0,0 +1,259 @@ +#include <linux/ceph/ceph_debug.h> + +#include <linux/module.h> +#include <linux/err.h> +#include <linux/slab.h> + +#include <linux/ceph/types.h> +#include <linux/ceph/decode.h> +#include <linux/ceph/libceph.h> +#include <linux/ceph/messenger.h> +#include "auth_none.h" +#include "auth_x.h" + + +/* + * get protocol handler + */ +static u32 supported_protocols[] = { + CEPH_AUTH_NONE, + CEPH_AUTH_CEPHX +}; + +static int ceph_auth_init_protocol(struct ceph_auth_client *ac, int protocol) +{ + switch (protocol) { + case CEPH_AUTH_NONE: + return ceph_auth_none_init(ac); + case CEPH_AUTH_CEPHX: + return ceph_x_init(ac); + default: + return -ENOENT; + } +} + +/* + * setup, teardown. + */ +struct ceph_auth_client *ceph_auth_init(const char *name, const char *secret) +{ + struct ceph_auth_client *ac; + int ret; + + dout("auth_init name '%s' secret '%s'\n", name, secret); + + ret = -ENOMEM; + ac = kzalloc(sizeof(*ac), GFP_NOFS); + if (!ac) + goto out; + + ac->negotiating = true; + if (name) + ac->name = name; + else + ac->name = CEPH_AUTH_NAME_DEFAULT; + dout("auth_init name %s secret %s\n", ac->name, secret); + ac->secret = secret; + return ac; + +out: + return ERR_PTR(ret); +} + +void ceph_auth_destroy(struct ceph_auth_client *ac) +{ + dout("auth_destroy %p\n", ac); + if (ac->ops) + ac->ops->destroy(ac); + kfree(ac); +} + +/* + * Reset occurs when reconnecting to the monitor. + */ +void ceph_auth_reset(struct ceph_auth_client *ac) +{ + dout("auth_reset %p\n", ac); + if (ac->ops && !ac->negotiating) + ac->ops->reset(ac); + ac->negotiating = true; +} + +int ceph_entity_name_encode(const char *name, void **p, void *end) +{ + int len = strlen(name); + + if (*p + 2*sizeof(u32) + len > end) + return -ERANGE; + ceph_encode_32(p, CEPH_ENTITY_TYPE_CLIENT); + ceph_encode_32(p, len); + ceph_encode_copy(p, name, len); + return 0; +} + +/* + * Initiate protocol negotiation with monitor. Include entity name + * and list supported protocols. + */ +int ceph_auth_build_hello(struct ceph_auth_client *ac, void *buf, size_t len) +{ + struct ceph_mon_request_header *monhdr = buf; + void *p = monhdr + 1, *end = buf + len, *lenp; + int i, num; + int ret; + + dout("auth_build_hello\n"); + monhdr->have_version = 0; + monhdr->session_mon = cpu_to_le16(-1); + monhdr->session_mon_tid = 0; + + ceph_encode_32(&p, 0); /* no protocol, yet */ + + lenp = p; + p += sizeof(u32); + + ceph_decode_need(&p, end, 1 + sizeof(u32), bad); + ceph_encode_8(&p, 1); + num = ARRAY_SIZE(supported_protocols); + ceph_encode_32(&p, num); + ceph_decode_need(&p, end, num * sizeof(u32), bad); + for (i = 0; i < num; i++) + ceph_encode_32(&p, supported_protocols[i]); + + ret = ceph_entity_name_encode(ac->name, &p, end); + if (ret < 0) + return ret; + ceph_decode_need(&p, end, sizeof(u64), bad); + ceph_encode_64(&p, ac->global_id); + + ceph_encode_32(&lenp, p - lenp - sizeof(u32)); + return p - buf; + +bad: + return -ERANGE; +} + +static int ceph_build_auth_request(struct ceph_auth_client *ac, + void *msg_buf, size_t msg_len) +{ + struct ceph_mon_request_header *monhdr = msg_buf; + void *p = monhdr + 1; + void *end = msg_buf + msg_len; + int ret; + + monhdr->have_version = 0; + monhdr->session_mon = cpu_to_le16(-1); + monhdr->session_mon_tid = 0; + + ceph_encode_32(&p, ac->protocol); + + ret = ac->ops->build_request(ac, p + sizeof(u32), end); + if (ret < 0) { + pr_err("error %d building auth method %s request\n", ret, + ac->ops->name); + return ret; + } + dout(" built request %d bytes\n", ret); + ceph_encode_32(&p, ret); + return p + ret - msg_buf; +} + +/* + * Handle auth message from monitor. + */ +int ceph_handle_auth_reply(struct ceph_auth_client *ac, + void *buf, size_t len, + void *reply_buf, size_t reply_len) +{ + void *p = buf; + void *end = buf + len; + int protocol; + s32 result; + u64 global_id; + void *payload, *payload_end; + int payload_len; + char *result_msg; + int result_msg_len; + int ret = -EINVAL; + + dout("handle_auth_reply %p %p\n", p, end); + ceph_decode_need(&p, end, sizeof(u32) * 3 + sizeof(u64), bad); + protocol = ceph_decode_32(&p); + result = ceph_decode_32(&p); + global_id = ceph_decode_64(&p); + payload_len = ceph_decode_32(&p); + payload = p; + p += payload_len; + ceph_decode_need(&p, end, sizeof(u32), bad); + result_msg_len = ceph_decode_32(&p); + result_msg = p; + p += result_msg_len; + if (p != end) + goto bad; + + dout(" result %d '%.*s' gid %llu len %d\n", result, result_msg_len, + result_msg, global_id, payload_len); + + payload_end = payload + payload_len; + + if (global_id && ac->global_id != global_id) { + dout(" set global_id %lld -> %lld\n", ac->global_id, global_id); + ac->global_id = global_id; + } + + if (ac->negotiating) { + /* server does not support our protocols? */ + if (!protocol && result < 0) { + ret = result; + goto out; + } + /* set up (new) protocol handler? */ + if (ac->protocol && ac->protocol != protocol) { + ac->ops->destroy(ac); + ac->protocol = 0; + ac->ops = NULL; + } + if (ac->protocol != protocol) { + ret = ceph_auth_init_protocol(ac, protocol); + if (ret) { + pr_err("error %d on auth protocol %d init\n", + ret, protocol); + goto out; + } + } + + ac->negotiating = false; + } + + ret = ac->ops->handle_reply(ac, result, payload, payload_end); + if (ret == -EAGAIN) { + return ceph_build_auth_request(ac, reply_buf, reply_len); + } else if (ret) { + pr_err("auth method '%s' error %d\n", ac->ops->name, ret); + return ret; + } + return 0; + +bad: + pr_err("failed to decode auth msg\n"); +out: + return ret; +} + +int ceph_build_auth(struct ceph_auth_client *ac, + void *msg_buf, size_t msg_len) +{ + if (!ac->protocol) + return ceph_auth_build_hello(ac, msg_buf, msg_len); + BUG_ON(!ac->ops); + if (ac->ops->should_authenticate(ac)) + return ceph_build_auth_request(ac, msg_buf, msg_len); + return 0; +} + +int ceph_auth_is_authenticated(struct ceph_auth_client *ac) +{ + if (!ac->ops) + return 0; + return ac->ops->is_authenticated(ac); +} diff --git a/net/ceph/auth_none.c b/net/ceph/auth_none.c new file mode 100644 index 00000000000..214c2bb43d6 --- /dev/null +++ b/net/ceph/auth_none.c @@ -0,0 +1,132 @@ + +#include <linux/ceph/ceph_debug.h> + +#include <linux/err.h> +#include <linux/module.h> +#include <linux/random.h> +#include <linux/slab.h> + +#include <linux/ceph/decode.h> +#include <linux/ceph/auth.h> + +#include "auth_none.h" + +static void reset(struct ceph_auth_client *ac) +{ + struct ceph_auth_none_info *xi = ac->private; + + xi->starting = true; + xi->built_authorizer = false; +} + +static void destroy(struct ceph_auth_client *ac) +{ + kfree(ac->private); + ac->private = NULL; +} + +static int is_authenticated(struct ceph_auth_client *ac) +{ + struct ceph_auth_none_info *xi = ac->private; + + return !xi->starting; +} + +static int should_authenticate(struct ceph_auth_client *ac) +{ + struct ceph_auth_none_info *xi = ac->private; + + return xi->starting; +} + +/* + * the generic auth code decode the global_id, and we carry no actual + * authenticate state, so nothing happens here. + */ +static int handle_reply(struct ceph_auth_client *ac, int result, + void *buf, void *end) +{ + struct ceph_auth_none_info *xi = ac->private; + + xi->starting = false; + return result; +} + +/* + * build an 'authorizer' with our entity_name and global_id. we can + * reuse a single static copy since it is identical for all services + * we connect to. + */ +static int ceph_auth_none_create_authorizer( + struct ceph_auth_client *ac, int peer_type, + struct ceph_authorizer **a, + void **buf, size_t *len, + void **reply_buf, size_t *reply_len) +{ + struct ceph_auth_none_info *ai = ac->private; + struct ceph_none_authorizer *au = &ai->au; + void *p, *end; + int ret; + + if (!ai->built_authorizer) { + p = au->buf; + end = p + sizeof(au->buf); + ceph_encode_8(&p, 1); + ret = ceph_entity_name_encode(ac->name, &p, end - 8); + if (ret < 0) + goto bad; + ceph_decode_need(&p, end, sizeof(u64), bad2); + ceph_encode_64(&p, ac->global_id); + au->buf_len = p - (void *)au->buf; + ai->built_authorizer = true; + dout("built authorizer len %d\n", au->buf_len); + } + + *a = (struct ceph_authorizer *)au; + *buf = au->buf; + *len = au->buf_len; + *reply_buf = au->reply_buf; + *reply_len = sizeof(au->reply_buf); + return 0; + +bad2: + ret = -ERANGE; +bad: + return ret; +} + +static void ceph_auth_none_destroy_authorizer(struct ceph_auth_client *ac, + struct ceph_authorizer *a) +{ + /* nothing to do */ +} + +static const struct ceph_auth_client_ops ceph_auth_none_ops = { + .name = "none", + .reset = reset, + .destroy = destroy, + .is_authenticated = is_authenticated, + .should_authenticate = should_authenticate, + .handle_reply = handle_reply, + .create_authorizer = ceph_auth_none_create_authorizer, + .destroy_authorizer = ceph_auth_none_destroy_authorizer, +}; + +int ceph_auth_none_init(struct ceph_auth_client *ac) +{ + struct ceph_auth_none_info *xi; + + dout("ceph_auth_none_init %p\n", ac); + xi = kzalloc(sizeof(*xi), GFP_NOFS); + if (!xi) + return -ENOMEM; + + xi->starting = true; + xi->built_authorizer = false; + + ac->protocol = CEPH_AUTH_NONE; + ac->private = xi; + ac->ops = &ceph_auth_none_ops; + return 0; +} + diff --git a/net/ceph/auth_none.h b/net/ceph/auth_none.h new file mode 100644 index 00000000000..ed7d088b1bc --- /dev/null +++ b/net/ceph/auth_none.h @@ -0,0 +1,29 @@ +#ifndef _FS_CEPH_AUTH_NONE_H +#define _FS_CEPH_AUTH_NONE_H + +#include <linux/slab.h> +#include <linux/ceph/auth.h> + +/* + * null security mode. + * + * we use a single static authorizer that simply encodes our entity name + * and global id. + */ + +struct ceph_none_authorizer { + char buf[128]; + int buf_len; + char reply_buf[0]; +}; + +struct ceph_auth_none_info { + bool starting; + bool built_authorizer; + struct ceph_none_authorizer au; /* we only need one; it's static */ +}; + +extern int ceph_auth_none_init(struct ceph_auth_client *ac); + +#endif + diff --git a/net/ceph/auth_x.c b/net/ceph/auth_x.c new file mode 100644 index 00000000000..7fd5dfcf6e1 --- /dev/null +++ b/net/ceph/auth_x.c @@ -0,0 +1,688 @@ + +#include <linux/ceph/ceph_debug.h> + +#include <linux/err.h> +#include <linux/module.h> +#include <linux/random.h> +#include <linux/slab.h> + +#include <linux/ceph/decode.h> +#include <linux/ceph/auth.h> + +#include "crypto.h" +#include "auth_x.h" +#include "auth_x_protocol.h" + +#define TEMP_TICKET_BUF_LEN 256 + +static void ceph_x_validate_tickets(struct ceph_auth_client *ac, int *pneed); + +static int ceph_x_is_authenticated(struct ceph_auth_client *ac) +{ + struct ceph_x_info *xi = ac->private; + int need; + + ceph_x_validate_tickets(ac, &need); + dout("ceph_x_is_authenticated want=%d need=%d have=%d\n", + ac->want_keys, need, xi->have_keys); + return (ac->want_keys & xi->have_keys) == ac->want_keys; +} + +static int ceph_x_should_authenticate(struct ceph_auth_client *ac) +{ + struct ceph_x_info *xi = ac->private; + int need; + + ceph_x_validate_tickets(ac, &need); + dout("ceph_x_should_authenticate want=%d need=%d have=%d\n", + ac->want_keys, need, xi->have_keys); + return need != 0; +} + +static int ceph_x_encrypt_buflen(int ilen) +{ + return sizeof(struct ceph_x_encrypt_header) + ilen + 16 + + sizeof(u32); +} + +static int ceph_x_encrypt(struct ceph_crypto_key *secret, + void *ibuf, int ilen, void *obuf, size_t olen) +{ + struct ceph_x_encrypt_header head = { + .struct_v = 1, + .magic = cpu_to_le64(CEPHX_ENC_MAGIC) + }; + size_t len = olen - sizeof(u32); + int ret; + + ret = ceph_encrypt2(secret, obuf + sizeof(u32), &len, + &head, sizeof(head), ibuf, ilen); + if (ret) + return ret; + ceph_encode_32(&obuf, len); + return len + sizeof(u32); +} + +static int ceph_x_decrypt(struct ceph_crypto_key *secret, + void **p, void *end, void *obuf, size_t olen) +{ + struct ceph_x_encrypt_header head; + size_t head_len = sizeof(head); + int len, ret; + + len = ceph_decode_32(p); + if (*p + len > end) + return -EINVAL; + + dout("ceph_x_decrypt len %d\n", len); + ret = ceph_decrypt2(secret, &head, &head_len, obuf, &olen, + *p, len); + if (ret) + return ret; + if (head.struct_v != 1 || le64_to_cpu(head.magic) != CEPHX_ENC_MAGIC) + return -EPERM; + *p += len; + return olen; +} + +/* + * get existing (or insert new) ticket handler + */ +static struct ceph_x_ticket_handler * +get_ticket_handler(struct ceph_auth_client *ac, int service) +{ + struct ceph_x_ticket_handler *th; + struct ceph_x_info *xi = ac->private; + struct rb_node *parent = NULL, **p = &xi->ticket_handlers.rb_node; + + while (*p) { + parent = *p; + th = rb_entry(parent, struct ceph_x_ticket_handler, node); + if (service < th->service) + p = &(*p)->rb_left; + else if (service > th->service) + p = &(*p)->rb_right; + else + return th; + } + + /* add it */ + th = kzalloc(sizeof(*th), GFP_NOFS); + if (!th) + return ERR_PTR(-ENOMEM); + th->service = service; + rb_link_node(&th->node, parent, p); + rb_insert_color(&th->node, &xi->ticket_handlers); + return th; +} + +static void remove_ticket_handler(struct ceph_auth_client *ac, + struct ceph_x_ticket_handler *th) +{ + struct ceph_x_info *xi = ac->private; + + dout("remove_ticket_handler %p %d\n", th, th->service); + rb_erase(&th->node, &xi->ticket_handlers); + ceph_crypto_key_destroy(&th->session_key); + if (th->ticket_blob) + ceph_buffer_put(th->ticket_blob); + kfree(th); +} + +static int ceph_x_proc_ticket_reply(struct ceph_auth_client *ac, + struct ceph_crypto_key *secret, + void *buf, void *end) +{ + struct ceph_x_info *xi = ac->private; + int num; + void *p = buf; + int ret; + char *dbuf; + char *ticket_buf; + u8 reply_struct_v; + + dbuf = kmalloc(TEMP_TICKET_BUF_LEN, GFP_NOFS); + if (!dbuf) + return -ENOMEM; + + ret = -ENOMEM; + ticket_buf = kmalloc(TEMP_TICKET_BUF_LEN, GFP_NOFS); + if (!ticket_buf) + goto out_dbuf; + + ceph_decode_need(&p, end, 1 + sizeof(u32), bad); + reply_struct_v = ceph_decode_8(&p); + if (reply_struct_v != 1) + goto bad; + num = ceph_decode_32(&p); + dout("%d tickets\n", num); + while (num--) { + int type; + u8 tkt_struct_v, blob_struct_v; + struct ceph_x_ticket_handler *th; + void *dp, *dend; + int dlen; + char is_enc; + struct timespec validity; + struct ceph_crypto_key old_key; + void *tp, *tpend; + struct ceph_timespec new_validity; + struct ceph_crypto_key new_session_key; + struct ceph_buffer *new_ticket_blob; + unsigned long new_expires, new_renew_after; + u64 new_secret_id; + + ceph_decode_need(&p, end, sizeof(u32) + 1, bad); + + type = ceph_decode_32(&p); + dout(" ticket type %d %s\n", type, ceph_entity_type_name(type)); + + tkt_struct_v = ceph_decode_8(&p); + if (tkt_struct_v != 1) + goto bad; + + th = get_ticket_handler(ac, type); + if (IS_ERR(th)) { + ret = PTR_ERR(th); + goto out; + } + + /* blob for me */ + dlen = ceph_x_decrypt(secret, &p, end, dbuf, + TEMP_TICKET_BUF_LEN); + if (dlen <= 0) { + ret = dlen; + goto out; + } + dout(" decrypted %d bytes\n", dlen); + dend = dbuf + dlen; + dp = dbuf; + + tkt_struct_v = ceph_decode_8(&dp); + if (tkt_struct_v != 1) + goto bad; + + memcpy(&old_key, &th->session_key, sizeof(old_key)); + ret = ceph_crypto_key_decode(&new_session_key, &dp, dend); + if (ret) + goto out; + + ceph_decode_copy(&dp, &new_validity, sizeof(new_validity)); + ceph_decode_timespec(&validity, &new_validity); + new_expires = get_seconds() + validity.tv_sec; + new_renew_after = new_expires - (validity.tv_sec / 4); + dout(" expires=%lu renew_after=%lu\n", new_expires, + new_renew_after); + + /* ticket blob for service */ + ceph_decode_8_safe(&p, end, is_enc, bad); + tp = ticket_buf; + if (is_enc) { + /* encrypted */ + dout(" encrypted ticket\n"); + dlen = ceph_x_decrypt(&old_key, &p, end, ticket_buf, + TEMP_TICKET_BUF_LEN); + if (dlen < 0) { + ret = dlen; + goto out; + } + dlen = ceph_decode_32(&tp); + } else { + /* unencrypted */ + ceph_decode_32_safe(&p, end, dlen, bad); + ceph_decode_need(&p, end, dlen, bad); + ceph_decode_copy(&p, ticket_buf, dlen); + } + tpend = tp + dlen; + dout(" ticket blob is %d bytes\n", dlen); + ceph_decode_need(&tp, tpend, 1 + sizeof(u64), bad); + blob_struct_v = ceph_decode_8(&tp); + new_secret_id = ceph_decode_64(&tp); + ret = ceph_decode_buffer(&new_ticket_blob, &tp, tpend); + if (ret) + goto out; + + /* all is well, update our ticket */ + ceph_crypto_key_destroy(&th->session_key); + if (th->ticket_blob) + ceph_buffer_put(th->ticket_blob); + th->session_key = new_session_key; + th->ticket_blob = new_ticket_blob; + th->validity = new_validity; + th->secret_id = new_secret_id; + th->expires = new_expires; + th->renew_after = new_renew_after; + dout(" got ticket service %d (%s) secret_id %lld len %d\n", + type, ceph_entity_type_name(type), th->secret_id, + (int)th->ticket_blob->vec.iov_len); + xi->have_keys |= th->service; + } + + ret = 0; +out: + kfree(ticket_buf); +out_dbuf: + kfree(dbuf); + return ret; + +bad: + ret = -EINVAL; + goto out; +} + +static int ceph_x_build_authorizer(struct ceph_auth_client *ac, + struct ceph_x_ticket_handler *th, + struct ceph_x_authorizer *au) +{ + int maxlen; + struct ceph_x_authorize_a *msg_a; + struct ceph_x_authorize_b msg_b; + void *p, *end; + int ret; + int ticket_blob_len = + (th->ticket_blob ? th->ticket_blob->vec.iov_len : 0); + + dout("build_authorizer for %s %p\n", + ceph_entity_type_name(th->service), au); + + maxlen = sizeof(*msg_a) + sizeof(msg_b) + + ceph_x_encrypt_buflen(ticket_blob_len); + dout(" need len %d\n", maxlen); + if (au->buf && au->buf->alloc_len < maxlen) { + ceph_buffer_put(au->buf); + au->buf = NULL; + } + if (!au->buf) { + au->buf = ceph_buffer_new(maxlen, GFP_NOFS); + if (!au->buf) + return -ENOMEM; + } + au->service = th->service; + + msg_a = au->buf->vec.iov_base; + msg_a->struct_v = 1; + msg_a->global_id = cpu_to_le64(ac->global_id); + msg_a->service_id = cpu_to_le32(th->service); + msg_a->ticket_blob.struct_v = 1; + msg_a->ticket_blob.secret_id = cpu_to_le64(th->secret_id); + msg_a->ticket_blob.blob_len = cpu_to_le32(ticket_blob_len); + if (ticket_blob_len) { + memcpy(msg_a->ticket_blob.blob, th->ticket_blob->vec.iov_base, + th->ticket_blob->vec.iov_len); + } + dout(" th %p secret_id %lld %lld\n", th, th->secret_id, + le64_to_cpu(msg_a->ticket_blob.secret_id)); + + p = msg_a + 1; + p += ticket_blob_len; + end = au->buf->vec.iov_base + au->buf->vec.iov_len; + + get_random_bytes(&au->nonce, sizeof(au->nonce)); + msg_b.struct_v = 1; + msg_b.nonce = cpu_to_le64(au->nonce); + ret = ceph_x_encrypt(&th->session_key, &msg_b, sizeof(msg_b), + p, end - p); + if (ret < 0) + goto out_buf; + p += ret; + au->buf->vec.iov_len = p - au->buf->vec.iov_base; + dout(" built authorizer nonce %llx len %d\n", au->nonce, + (int)au->buf->vec.iov_len); + BUG_ON(au->buf->vec.iov_len > maxlen); + return 0; + +out_buf: + ceph_buffer_put(au->buf); + au->buf = NULL; + return ret; +} + +static int ceph_x_encode_ticket(struct ceph_x_ticket_handler *th, + void **p, void *end) +{ + ceph_decode_need(p, end, 1 + sizeof(u64), bad); + ceph_encode_8(p, 1); + ceph_encode_64(p, th->secret_id); + if (th->ticket_blob) { + const char *buf = th->ticket_blob->vec.iov_base; + u32 len = th->ticket_blob->vec.iov_len; + + ceph_encode_32_safe(p, end, len, bad); + ceph_encode_copy_safe(p, end, buf, len, bad); + } else { + ceph_encode_32_safe(p, end, 0, bad); + } + + return 0; +bad: + return -ERANGE; +} + +static void ceph_x_validate_tickets(struct ceph_auth_client *ac, int *pneed) +{ + int want = ac->want_keys; + struct ceph_x_info *xi = ac->private; + int service; + + *pneed = ac->want_keys & ~(xi->have_keys); + + for (service = 1; service <= want; service <<= 1) { + struct ceph_x_ticket_handler *th; + + if (!(ac->want_keys & service)) + continue; + + if (*pneed & service) + continue; + + th = get_ticket_handler(ac, service); + + if (IS_ERR(th)) { + *pneed |= service; + continue; + } + + if (get_seconds() >= th->renew_after) + *pneed |= service; + if (get_seconds() >= th->expires) + xi->have_keys &= ~service; + } +} + + +static int ceph_x_build_request(struct ceph_auth_client *ac, + void *buf, void *end) +{ + struct ceph_x_info *xi = ac->private; + int need; + struct ceph_x_request_header *head = buf; + int ret; + struct ceph_x_ticket_handler *th = + get_ticket_handler(ac, CEPH_ENTITY_TYPE_AUTH); + + if (IS_ERR(th)) + return PTR_ERR(th); + + ceph_x_validate_tickets(ac, &need); + + dout("build_request want %x have %x need %x\n", + ac->want_keys, xi->have_keys, need); + + if (need & CEPH_ENTITY_TYPE_AUTH) { + struct ceph_x_authenticate *auth = (void *)(head + 1); + void *p = auth + 1; + struct ceph_x_challenge_blob tmp; + char tmp_enc[40]; + u64 *u; + + if (p > end) + return -ERANGE; + + dout(" get_auth_session_key\n"); + head->op = cpu_to_le16(CEPHX_GET_AUTH_SESSION_KEY); + + /* encrypt and hash */ + get_random_bytes(&auth->client_challenge, sizeof(u64)); + tmp.client_challenge = auth->client_challenge; + tmp.server_challenge = cpu_to_le64(xi->server_challenge); + ret = ceph_x_encrypt(&xi->secret, &tmp, sizeof(tmp), + tmp_enc, sizeof(tmp_enc)); + if (ret < 0) + return ret; + + auth->struct_v = 1; + auth->key = 0; + for (u = (u64 *)tmp_enc; u + 1 <= (u64 *)(tmp_enc + ret); u++) + auth->key ^= *(__le64 *)u; + dout(" server_challenge %llx client_challenge %llx key %llx\n", + xi->server_challenge, le64_to_cpu(auth->client_challenge), + le64_to_cpu(auth->key)); + + /* now encode the old ticket if exists */ + ret = ceph_x_encode_ticket(th, &p, end); + if (ret < 0) + return ret; + + return p - buf; + } + + if (need) { + void *p = head + 1; + struct ceph_x_service_ticket_request *req; + + if (p > end) + return -ERANGE; + head->op = cpu_to_le16(CEPHX_GET_PRINCIPAL_SESSION_KEY); + + ret = ceph_x_build_authorizer(ac, th, &xi->auth_authorizer); + if (ret) + return ret; + ceph_encode_copy(&p, xi->auth_authorizer.buf->vec.iov_base, + xi->auth_authorizer.buf->vec.iov_len); + + req = p; + req->keys = cpu_to_le32(need); + p += sizeof(*req); + return p - buf; + } + + return 0; +} + +static int ceph_x_handle_reply(struct ceph_auth_client *ac, int result, + void *buf, void *end) +{ + struct ceph_x_info *xi = ac->private; + struct ceph_x_reply_header *head = buf; + struct ceph_x_ticket_handler *th; + int len = end - buf; + int op; + int ret; + + if (result) + return result; /* XXX hmm? */ + + if (xi->starting) { + /* it's a hello */ + struct ceph_x_server_challenge *sc = buf; + + if (len != sizeof(*sc)) + return -EINVAL; + xi->server_challenge = le64_to_cpu(sc->server_challenge); + dout("handle_reply got server challenge %llx\n", + xi->server_challenge); + xi->starting = false; + xi->have_keys &= ~CEPH_ENTITY_TYPE_AUTH; + return -EAGAIN; + } + + op = le16_to_cpu(head->op); + result = le32_to_cpu(head->result); + dout("handle_reply op %d result %d\n", op, result); + switch (op) { + case CEPHX_GET_AUTH_SESSION_KEY: + /* verify auth key */ + ret = ceph_x_proc_ticket_reply(ac, &xi->secret, + buf + sizeof(*head), end); + break; + + case CEPHX_GET_PRINCIPAL_SESSION_KEY: + th = get_ticket_handler(ac, CEPH_ENTITY_TYPE_AUTH); + if (IS_ERR(th)) + return PTR_ERR(th); + ret = ceph_x_proc_ticket_reply(ac, &th->session_key, + buf + sizeof(*head), end); + break; + + default: + return -EINVAL; + } + if (ret) + return ret; + if (ac->want_keys == xi->have_keys) + return 0; + return -EAGAIN; +} + +static int ceph_x_create_authorizer( + struct ceph_auth_client *ac, int peer_type, + struct ceph_authorizer **a, + void **buf, size_t *len, + void **reply_buf, size_t *reply_len) +{ + struct ceph_x_authorizer *au; + struct ceph_x_ticket_handler *th; + int ret; + + th = get_ticket_handler(ac, peer_type); + if (IS_ERR(th)) + return PTR_ERR(th); + + au = kzalloc(sizeof(*au), GFP_NOFS); + if (!au) + return -ENOMEM; + + ret = ceph_x_build_authorizer(ac, th, au); + if (ret) { + kfree(au); + return ret; + } + + *a = (struct ceph_authorizer *)au; + *buf = au->buf->vec.iov_base; + *len = au->buf->vec.iov_len; + *reply_buf = au->reply_buf; + *reply_len = sizeof(au->reply_buf); + return 0; +} + +static int ceph_x_verify_authorizer_reply(struct ceph_auth_client *ac, + struct ceph_authorizer *a, size_t len) +{ + struct ceph_x_authorizer *au = (void *)a; + struct ceph_x_ticket_handler *th; + int ret = 0; + struct ceph_x_authorize_reply reply; + void *p = au->reply_buf; + void *end = p + sizeof(au->reply_buf); + + th = get_ticket_handler(ac, au->service); + if (IS_ERR(th)) + return PTR_ERR(th); + ret = ceph_x_decrypt(&th->session_key, &p, end, &reply, sizeof(reply)); + if (ret < 0) + return ret; + if (ret != sizeof(reply)) + return -EPERM; + + if (au->nonce + 1 != le64_to_cpu(reply.nonce_plus_one)) + ret = -EPERM; + else + ret = 0; + dout("verify_authorizer_reply nonce %llx got %llx ret %d\n", + au->nonce, le64_to_cpu(reply.nonce_plus_one), ret); + return ret; +} + +static void ceph_x_destroy_authorizer(struct ceph_auth_client *ac, + struct ceph_authorizer *a) +{ + struct ceph_x_authorizer *au = (void *)a; + + ceph_buffer_put(au->buf); + kfree(au); +} + + +static void ceph_x_reset(struct ceph_auth_client *ac) +{ + struct ceph_x_info *xi = ac->private; + + dout("reset\n"); + xi->starting = true; + xi->server_challenge = 0; +} + +static void ceph_x_destroy(struct ceph_auth_client *ac) +{ + struct ceph_x_info *xi = ac->private; + struct rb_node *p; + + dout("ceph_x_destroy %p\n", ac); + ceph_crypto_key_destroy(&xi->secret); + + while ((p = rb_first(&xi->ticket_handlers)) != NULL) { + struct ceph_x_ticket_handler *th = + rb_entry(p, struct ceph_x_ticket_handler, node); + remove_ticket_handler(ac, th); + } + + if (xi->auth_authorizer.buf) + ceph_buffer_put(xi->auth_authorizer.buf); + + kfree(ac->private); + ac->private = NULL; +} + +static void ceph_x_invalidate_authorizer(struct ceph_auth_client *ac, + int peer_type) +{ + struct ceph_x_ticket_handler *th; + + th = get_ticket_handler(ac, peer_type); + if (!IS_ERR(th)) + remove_ticket_handler(ac, th); +} + + +static const struct ceph_auth_client_ops ceph_x_ops = { + .name = "x", + .is_authenticated = ceph_x_is_authenticated, + .should_authenticate = ceph_x_should_authenticate, + .build_request = ceph_x_build_request, + .handle_reply = ceph_x_handle_reply, + .create_authorizer = ceph_x_create_authorizer, + .verify_authorizer_reply = ceph_x_verify_authorizer_reply, + .destroy_authorizer = ceph_x_destroy_authorizer, + .invalidate_authorizer = ceph_x_invalidate_authorizer, + .reset = ceph_x_reset, + .destroy = ceph_x_destroy, +}; + + +int ceph_x_init(struct ceph_auth_client *ac) +{ + struct ceph_x_info *xi; + int ret; + + dout("ceph_x_init %p\n", ac); + ret = -ENOMEM; + xi = kzalloc(sizeof(*xi), GFP_NOFS); + if (!xi) + goto out; + + ret = -EINVAL; + if (!ac->secret) { + pr_err("no secret set (for auth_x protocol)\n"); + goto out_nomem; + } + + ret = ceph_crypto_key_unarmor(&xi->secret, ac->secret); + if (ret) + goto out_nomem; + + xi->starting = true; + xi->ticket_handlers = RB_ROOT; + + ac->protocol = CEPH_AUTH_CEPHX; + ac->private = xi; + ac->ops = &ceph_x_ops; + return 0; + +out_nomem: + kfree(xi); +out: + return ret; +} + + diff --git a/net/ceph/auth_x.h b/net/ceph/auth_x.h new file mode 100644 index 00000000000..e02da7a5c5a --- /dev/null +++ b/net/ceph/auth_x.h @@ -0,0 +1,50 @@ +#ifndef _FS_CEPH_AUTH_X_H +#define _FS_CEPH_AUTH_X_H + +#include <linux/rbtree.h> + +#include <linux/ceph/auth.h> + +#include "crypto.h" +#include "auth_x_protocol.h" + +/* + * Handle ticket for a single service. + */ +struct ceph_x_ticket_handler { + struct rb_node node; + unsigned service; + + struct ceph_crypto_key session_key; + struct ceph_timespec validity; + + u64 secret_id; + struct ceph_buffer *ticket_blob; + + unsigned long renew_after, expires; +}; + + +struct ceph_x_authorizer { + struct ceph_buffer *buf; + unsigned service; + u64 nonce; + char reply_buf[128]; /* big enough for encrypted blob */ +}; + +struct ceph_x_info { + struct ceph_crypto_key secret; + + bool starting; + u64 server_challenge; + + unsigned have_keys; + struct rb_root ticket_handlers; + + struct ceph_x_authorizer auth_authorizer; +}; + +extern int ceph_x_init(struct ceph_auth_client *ac); + +#endif + diff --git a/net/ceph/auth_x_protocol.h b/net/ceph/auth_x_protocol.h new file mode 100644 index 00000000000..671d30576c4 --- /dev/null +++ b/net/ceph/auth_x_protocol.h @@ -0,0 +1,90 @@ +#ifndef __FS_CEPH_AUTH_X_PROTOCOL +#define __FS_CEPH_AUTH_X_PROTOCOL + +#define CEPHX_GET_AUTH_SESSION_KEY 0x0100 +#define CEPHX_GET_PRINCIPAL_SESSION_KEY 0x0200 +#define CEPHX_GET_ROTATING_KEY 0x0400 + +/* common bits */ +struct ceph_x_ticket_blob { + __u8 struct_v; + __le64 secret_id; + __le32 blob_len; + char blob[]; +} __attribute__ ((packed)); + + +/* common request/reply headers */ +struct ceph_x_request_header { + __le16 op; +} __attribute__ ((packed)); + +struct ceph_x_reply_header { + __le16 op; + __le32 result; +} __attribute__ ((packed)); + + +/* authenticate handshake */ + +/* initial hello (no reply header) */ +struct ceph_x_server_challenge { + __u8 struct_v; + __le64 server_challenge; +} __attribute__ ((packed)); + +struct ceph_x_authenticate { + __u8 struct_v; + __le64 client_challenge; + __le64 key; + /* ticket blob */ +} __attribute__ ((packed)); + +struct ceph_x_service_ticket_request { + __u8 struct_v; + __le32 keys; +} __attribute__ ((packed)); + +struct ceph_x_challenge_blob { + __le64 server_challenge; + __le64 client_challenge; +} __attribute__ ((packed)); + + + +/* authorize handshake */ + +/* + * The authorizer consists of two pieces: + * a - service id, ticket blob + * b - encrypted with session key + */ +struct ceph_x_authorize_a { + __u8 struct_v; + __le64 global_id; + __le32 service_id; + struct ceph_x_ticket_blob ticket_blob; +} __attribute__ ((packed)); + +struct ceph_x_authorize_b { + __u8 struct_v; + __le64 nonce; +} __attribute__ ((packed)); + +struct ceph_x_authorize_reply { + __u8 struct_v; + __le64 nonce_plus_one; +} __attribute__ ((packed)); + + +/* + * encyption bundle + */ +#define CEPHX_ENC_MAGIC 0xff009cad8826aa55ull + +struct ceph_x_encrypt_header { + __u8 struct_v; + __le64 magic; +} __attribute__ ((packed)); + +#endif diff --git a/net/ceph/buffer.c b/net/ceph/buffer.c new file mode 100644 index 00000000000..53d8abfa25d --- /dev/null +++ b/net/ceph/buffer.c @@ -0,0 +1,68 @@ + +#include <linux/ceph/ceph_debug.h> + +#include <linux/module.h> +#include <linux/slab.h> + +#include <linux/ceph/buffer.h> +#include <linux/ceph/decode.h> + +struct ceph_buffer *ceph_buffer_new(size_t len, gfp_t gfp) +{ + struct ceph_buffer *b; + + b = kmalloc(sizeof(*b), gfp); + if (!b) + return NULL; + + b->vec.iov_base = kmalloc(len, gfp | __GFP_NOWARN); + if (b->vec.iov_base) { + b->is_vmalloc = false; + } else { + b->vec.iov_base = __vmalloc(len, gfp, PAGE_KERNEL); + if (!b->vec.iov_base) { + kfree(b); + return NULL; + } + b->is_vmalloc = true; + } + + kref_init(&b->kref); + b->alloc_len = len; + b->vec.iov_len = len; + dout("buffer_new %p\n", b); + return b; +} +EXPORT_SYMBOL(ceph_buffer_new); + +void ceph_buffer_release(struct kref *kref) +{ + struct ceph_buffer *b = container_of(kref, struct ceph_buffer, kref); + + dout("buffer_release %p\n", b); + if (b->vec.iov_base) { + if (b->is_vmalloc) + vfree(b->vec.iov_base); + else + kfree(b->vec.iov_base); + } + kfree(b); +} +EXPORT_SYMBOL(ceph_buffer_release); + +int ceph_decode_buffer(struct ceph_buffer **b, void **p, void *end) +{ + size_t len; + + ceph_decode_need(p, end, sizeof(u32), bad); + len = ceph_decode_32(p); + dout("decode_buffer len %d\n", (int)len); + ceph_decode_need(p, end, len, bad); + *b = ceph_buffer_new(len, GFP_NOFS); + if (!*b) + return -ENOMEM; + ceph_decode_copy(p, (*b)->vec.iov_base, len); + return 0; +bad: + return -EINVAL; +} diff --git a/net/ceph/ceph_common.c b/net/ceph/ceph_common.c new file mode 100644 index 00000000000..f3e4a13fea0 --- /dev/null +++ b/net/ceph/ceph_common.c @@ -0,0 +1,529 @@ + +#include <linux/ceph/ceph_debug.h> +#include <linux/backing-dev.h> +#include <linux/ctype.h> +#include <linux/fs.h> +#include <linux/inet.h> +#include <linux/in6.h> +#include <linux/module.h> +#include <linux/mount.h> +#include <linux/parser.h> +#include <linux/sched.h> +#include <linux/seq_file.h> +#include <linux/slab.h> +#include <linux/statfs.h> +#include <linux/string.h> + + +#include <linux/ceph/libceph.h> +#include <linux/ceph/debugfs.h> +#include <linux/ceph/decode.h> +#include <linux/ceph/mon_client.h> +#include <linux/ceph/auth.h> + + + +/* + * find filename portion of a path (/foo/bar/baz -> baz) + */ +const char *ceph_file_part(const char *s, int len) +{ + const char *e = s + len; + + while (e != s && *(e-1) != '/') + e--; + return e; +} +EXPORT_SYMBOL(ceph_file_part); + +const char *ceph_msg_type_name(int type) +{ + switch (type) { + case CEPH_MSG_SHUTDOWN: return "shutdown"; + case CEPH_MSG_PING: return "ping"; + case CEPH_MSG_AUTH: return "auth"; + case CEPH_MSG_AUTH_REPLY: return "auth_reply"; + case CEPH_MSG_MON_MAP: return "mon_map"; + case CEPH_MSG_MON_GET_MAP: return "mon_get_map"; + case CEPH_MSG_MON_SUBSCRIBE: return "mon_subscribe"; + case CEPH_MSG_MON_SUBSCRIBE_ACK: return "mon_subscribe_ack"; + case CEPH_MSG_STATFS: return "statfs"; + case CEPH_MSG_STATFS_REPLY: return "statfs_reply"; + case CEPH_MSG_MDS_MAP: return "mds_map"; + case CEPH_MSG_CLIENT_SESSION: return "client_session"; + case CEPH_MSG_CLIENT_RECONNECT: return "client_reconnect"; + case CEPH_MSG_CLIENT_REQUEST: return "client_request"; + case CEPH_MSG_CLIENT_REQUEST_FORWARD: return "client_request_forward"; + case CEPH_MSG_CLIENT_REPLY: return "client_reply"; + case CEPH_MSG_CLIENT_CAPS: return "client_caps"; + case CEPH_MSG_CLIENT_CAPRELEASE: return "client_cap_release"; + case CEPH_MSG_CLIENT_SNAP: return "client_snap"; + case CEPH_MSG_CLIENT_LEASE: return "client_lease"; + case CEPH_MSG_OSD_MAP: return "osd_map"; + case CEPH_MSG_OSD_OP: return "osd_op"; + case CEPH_MSG_OSD_OPREPLY: return "osd_opreply"; + default: return "unknown"; + } +} +EXPORT_SYMBOL(ceph_msg_type_name); + +/* + * Initially learn our fsid, or verify an fsid matches. + */ +int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid) +{ + if (client->have_fsid) { + if (ceph_fsid_compare(&client->fsid, fsid)) { + pr_err("bad fsid, had %pU got %pU", + &client->fsid, fsid); + return -1; + } + } else { + pr_info("client%lld fsid %pU\n", ceph_client_id(client), fsid); + memcpy(&client->fsid, fsid, sizeof(*fsid)); + ceph_debugfs_client_init(client); + client->have_fsid = true; + } + return 0; +} +EXPORT_SYMBOL(ceph_check_fsid); + +static int strcmp_null(const char *s1, const char *s2) +{ + if (!s1 && !s2) + return 0; + if (s1 && !s2) + return -1; + if (!s1 && s2) + return 1; + return strcmp(s1, s2); +} + +int ceph_compare_options(struct ceph_options *new_opt, + struct ceph_client *client) +{ + struct ceph_options *opt1 = new_opt; + struct ceph_options *opt2 = client->options; + int ofs = offsetof(struct ceph_options, mon_addr); + int i; + int ret; + + ret = memcmp(opt1, opt2, ofs); + if (ret) + return ret; + + ret = strcmp_null(opt1->name, opt2->name); + if (ret) + return ret; + + ret = strcmp_null(opt1->secret, opt2->secret); + if (ret) + return ret; + + /* any matching mon ip implies a match */ + for (i = 0; i < opt1->num_mon; i++) { + if (ceph_monmap_contains(client->monc.monmap, + &opt1->mon_addr[i])) + return 0; + } + return -1; +} +EXPORT_SYMBOL(ceph_compare_options); + + +static int parse_fsid(const char *str, struct ceph_fsid *fsid) +{ + int i = 0; + char tmp[3]; + int err = -EINVAL; + int d; + + dout("parse_fsid '%s'\n", str); + tmp[2] = 0; + while (*str && i < 16) { + if (ispunct(*str)) { + str++; + continue; + } + if (!isxdigit(str[0]) || !isxdigit(str[1])) + break; + tmp[0] = str[0]; + tmp[1] = str[1]; + if (sscanf(tmp, "%x", &d) < 1) + break; + fsid->fsid[i] = d & 0xff; + i++; + str += 2; + } + + if (i == 16) + err = 0; + dout("parse_fsid ret %d got fsid %pU", err, fsid); + return err; +} + +/* + * ceph options + */ +enum { + Opt_osdtimeout, + Opt_osdkeepalivetimeout, + Opt_mount_timeout, + Opt_osd_idle_ttl, + Opt_last_int, + /* int args above */ + Opt_fsid, + Opt_name, + Opt_secret, + Opt_ip, + Opt_last_string, + /* string args above */ + Opt_noshare, + Opt_nocrc, +}; + +static match_table_t opt_tokens = { + {Opt_osdtimeout, "osdtimeout=%d"}, + {Opt_osdkeepalivetimeout, "osdkeepalive=%d"}, + {Opt_mount_timeout, "mount_timeout=%d"}, + {Opt_osd_idle_ttl, "osd_idle_ttl=%d"}, + /* int args above */ + {Opt_fsid, "fsid=%s"}, + {Opt_name, "name=%s"}, + {Opt_secret, "secret=%s"}, + {Opt_ip, "ip=%s"}, + /* string args above */ + {Opt_noshare, "noshare"}, + {Opt_nocrc, "nocrc"}, + {-1, NULL} +}; + +void ceph_destroy_options(struct ceph_options *opt) +{ + dout("destroy_options %p\n", opt); + kfree(opt->name); + kfree(opt->secret); + kfree(opt); +} +EXPORT_SYMBOL(ceph_destroy_options); + +int ceph_parse_options(struct ceph_options **popt, char *options, + const char *dev_name, const char *dev_name_end, + int (*parse_extra_token)(char *c, void *private), + void *private) +{ + struct ceph_options *opt; + const char *c; + int err = -ENOMEM; + substring_t argstr[MAX_OPT_ARGS]; + + opt = kzalloc(sizeof(*opt), GFP_KERNEL); + if (!opt) + return err; + opt->mon_addr = kcalloc(CEPH_MAX_MON, sizeof(*opt->mon_addr), + GFP_KERNEL); + if (!opt->mon_addr) + goto out; + + dout("parse_options %p options '%s' dev_name '%s'\n", opt, options, + dev_name); + + /* start with defaults */ + opt->flags = CEPH_OPT_DEFAULT; + opt->osd_timeout = CEPH_OSD_TIMEOUT_DEFAULT; + opt->osd_keepalive_timeout = CEPH_OSD_KEEPALIVE_DEFAULT; + opt->mount_timeout = CEPH_MOUNT_TIMEOUT_DEFAULT; /* seconds */ + opt->osd_idle_ttl = CEPH_OSD_IDLE_TTL_DEFAULT; /* seconds */ + + /* get mon ip(s) */ + /* ip1[:port1][,ip2[:port2]...] */ + err = ceph_parse_ips(dev_name, dev_name_end, opt->mon_addr, + CEPH_MAX_MON, &opt->num_mon); + if (err < 0) + goto out; + + /* parse mount options */ + while ((c = strsep(&options, ",")) != NULL) { + int token, intval, ret; + if (!*c) + continue; + err = -EINVAL; + token = match_token((char *)c, opt_tokens, argstr); + if (token < 0 && parse_extra_token) { + /* extra? */ + err = parse_extra_token((char *)c, private); + if (err < 0) { + pr_err("bad option at '%s'\n", c); + goto out; + } + continue; + } + if (token < Opt_last_int) { + ret = match_int(&argstr[0], &intval); + if (ret < 0) { + pr_err("bad mount option arg (not int) " + "at '%s'\n", c); + continue; + } + dout("got int token %d val %d\n", token, intval); + } else if (token > Opt_last_int && token < Opt_last_string) { + dout("got string token %d val %s\n", token, + argstr[0].from); + } else { + dout("got token %d\n", token); + } + switch (token) { + case Opt_ip: + err = ceph_parse_ips(argstr[0].from, + argstr[0].to, + &opt->my_addr, + 1, NULL); + if (err < 0) + goto out; + opt->flags |= CEPH_OPT_MYIP; + break; + + case Opt_fsid: + err = parse_fsid(argstr[0].from, &opt->fsid); + if (err == 0) + opt->flags |= CEPH_OPT_FSID; + break; + case Opt_name: + opt->name = kstrndup(argstr[0].from, + argstr[0].to-argstr[0].from, + GFP_KERNEL); + break; + case Opt_secret: + opt->secret = kstrndup(argstr[0].from, + argstr[0].to-argstr[0].from, + GFP_KERNEL); + break; + + /* misc */ + case Opt_osdtimeout: + opt->osd_timeout = intval; + break; + case Opt_osdkeepalivetimeout: + opt->osd_keepalive_timeout = intval; + break; + case Opt_osd_idle_ttl: + opt->osd_idle_ttl = intval; + break; + case Opt_mount_timeout: + opt->mount_timeout = intval; + break; + + case Opt_noshare: + opt->flags |= CEPH_OPT_NOSHARE; + break; + + case Opt_nocrc: + opt->flags |= CEPH_OPT_NOCRC; + break; + + default: + BUG_ON(token); + } + } + + /* success */ + *popt = opt; + return 0; + +out: + ceph_destroy_options(opt); + return err; +} +EXPORT_SYMBOL(ceph_parse_options); + +u64 ceph_client_id(struct ceph_client *client) +{ + return client->monc.auth->global_id; +} +EXPORT_SYMBOL(ceph_client_id); + +/* + * create a fresh client instance + */ +struct ceph_client *ceph_create_client(struct ceph_options *opt, void *private) +{ + struct ceph_client *client; + int err = -ENOMEM; + + client = kzalloc(sizeof(*client), GFP_KERNEL); + if (client == NULL) + return ERR_PTR(-ENOMEM); + + client->private = private; + client->options = opt; + + mutex_init(&client->mount_mutex); + init_waitqueue_head(&client->auth_wq); + client->auth_err = 0; + + client->extra_mon_dispatch = NULL; + client->supported_features = CEPH_FEATURE_SUPPORTED_DEFAULT; + client->required_features = CEPH_FEATURE_REQUIRED_DEFAULT; + + client->msgr = NULL; + + /* subsystems */ + err = ceph_monc_init(&client->monc, client); + if (err < 0) + goto fail; + err = ceph_osdc_init(&client->osdc, client); + if (err < 0) + goto fail_monc; + + return client; + +fail_monc: + ceph_monc_stop(&client->monc); +fail: + kfree(client); + return ERR_PTR(err); +} +EXPORT_SYMBOL(ceph_create_client); + +void ceph_destroy_client(struct ceph_client *client) +{ + dout("destroy_client %p\n", client); + + /* unmount */ + ceph_osdc_stop(&client->osdc); + + /* + * make sure mds and osd connections close out before destroying + * the auth module, which is needed to free those connections' + * ceph_authorizers. + */ + ceph_msgr_flush(); + + ceph_monc_stop(&client->monc); + + ceph_debugfs_client_cleanup(client); + + if (client->msgr) + ceph_messenger_destroy(client->msgr); + + ceph_destroy_options(client->options); + + kfree(client); + dout("destroy_client %p done\n", client); +} +EXPORT_SYMBOL(ceph_destroy_client); + +/* + * true if we have the mon map (and have thus joined the cluster) + */ +static int have_mon_and_osd_map(struct ceph_client *client) +{ + return client->monc.monmap && client->monc.monmap->epoch && + client->osdc.osdmap && client->osdc.osdmap->epoch; +} + +/* + * mount: join the ceph cluster, and open root directory. + */ +int __ceph_open_session(struct ceph_client *client, unsigned long started) +{ + struct ceph_entity_addr *myaddr = NULL; + int err; + unsigned long timeout = client->options->mount_timeout * HZ; + + /* initialize the messenger */ + if (client->msgr == NULL) { + if (ceph_test_opt(client, MYIP)) + myaddr = &client->options->my_addr; + client->msgr = ceph_messenger_create(myaddr, + client->supported_features, + client->required_features); + if (IS_ERR(client->msgr)) { + client->msgr = NULL; + return PTR_ERR(client->msgr); + } + client->msgr->nocrc = ceph_test_opt(client, NOCRC); + } + + /* open session, and wait for mon and osd maps */ + err = ceph_monc_open_session(&client->monc); + if (err < 0) + return err; + + while (!have_mon_and_osd_map(client)) { + err = -EIO; + if (timeout && time_after_eq(jiffies, started + timeout)) + return err; + + /* wait */ + dout("mount waiting for mon_map\n"); + err = wait_event_interruptible_timeout(client->auth_wq, + have_mon_and_osd_map(client) || (client->auth_err < 0), + timeout); + if (err == -EINTR || err == -ERESTARTSYS) + return err; + if (client->auth_err < 0) + return client->auth_err; + } + + return 0; +} +EXPORT_SYMBOL(__ceph_open_session); + + +int ceph_open_session(struct ceph_client *client) +{ + int ret; + unsigned long started = jiffies; /* note the start time */ + + dout("open_session start\n"); + mutex_lock(&client->mount_mutex); + + ret = __ceph_open_session(client, started); + + mutex_unlock(&client->mount_mutex); + return ret; +} +EXPORT_SYMBOL(ceph_open_session); + + +static int __init init_ceph_lib(void) +{ + int ret = 0; + + ret = ceph_debugfs_init(); + if (ret < 0) + goto out; + + ret = ceph_msgr_init(); + if (ret < 0) + goto out_debugfs; + + pr_info("loaded (mon/osd proto %d/%d, osdmap %d/%d %d/%d)\n", + CEPH_MONC_PROTOCOL, CEPH_OSDC_PROTOCOL, + CEPH_OSDMAP_VERSION, CEPH_OSDMAP_VERSION_EXT, + CEPH_OSDMAP_INC_VERSION, CEPH_OSDMAP_INC_VERSION_EXT); + + return 0; + +out_debugfs: + ceph_debugfs_cleanup(); +out: + return ret; +} + +static void __exit exit_ceph_lib(void) +{ + dout("exit_ceph_lib\n"); + ceph_msgr_exit(); + ceph_debugfs_cleanup(); +} + +module_init(init_ceph_lib); +module_exit(exit_ceph_lib); + +MODULE_AUTHOR("Sage Weil <sage@newdream.net>"); +MODULE_AUTHOR("Yehuda Sadeh <yehuda@hq.newdream.net>"); +MODULE_AUTHOR("Patience Warnick <patience@newdream.net>"); +MODULE_DESCRIPTION("Ceph filesystem for Linux"); +MODULE_LICENSE("GPL"); diff --git a/net/ceph/ceph_fs.c b/net/ceph/ceph_fs.c new file mode 100644 index 00000000000..a3a3a31d3c3 --- /dev/null +++ b/net/ceph/ceph_fs.c @@ -0,0 +1,75 @@ +/* + * Some non-inline ceph helpers + */ +#include <linux/module.h> +#include <linux/ceph/types.h> + +/* + * return true if @layout appears to be valid + */ +int ceph_file_layout_is_valid(const struct ceph_file_layout *layout) +{ + __u32 su = le32_to_cpu(layout->fl_stripe_unit); + __u32 sc = le32_to_cpu(layout->fl_stripe_count); + __u32 os = le32_to_cpu(layout->fl_object_size); + + /* stripe unit, object size must be non-zero, 64k increment */ + if (!su || (su & (CEPH_MIN_STRIPE_UNIT-1))) + return 0; + if (!os || (os & (CEPH_MIN_STRIPE_UNIT-1))) + return 0; + /* object size must be a multiple of stripe unit */ + if (os < su || os % su) + return 0; + /* stripe count must be non-zero */ + if (!sc) + return 0; + return 1; +} + + +int ceph_flags_to_mode(int flags) +{ + int mode; + +#ifdef O_DIRECTORY /* fixme */ + if ((flags & O_DIRECTORY) == O_DIRECTORY) + return CEPH_FILE_MODE_PIN; +#endif + if ((flags & O_APPEND) == O_APPEND) + flags |= O_WRONLY; + + if ((flags & O_ACCMODE) == O_RDWR) + mode = CEPH_FILE_MODE_RDWR; + else if ((flags & O_ACCMODE) == O_WRONLY) + mode = CEPH_FILE_MODE_WR; + else + mode = CEPH_FILE_MODE_RD; + +#ifdef O_LAZY + if (flags & O_LAZY) + mode |= CEPH_FILE_MODE_LAZY; +#endif + + return mode; +} +EXPORT_SYMBOL(ceph_flags_to_mode); + +int ceph_caps_for_mode(int mode) +{ + int caps = CEPH_CAP_PIN; + + if (mode & CEPH_FILE_MODE_RD) + caps |= CEPH_CAP_FILE_SHARED | + CEPH_CAP_FILE_RD | CEPH_CAP_FILE_CACHE; + if (mode & CEPH_FILE_MODE_WR) + caps |= CEPH_CAP_FILE_EXCL | + CEPH_CAP_FILE_WR | CEPH_CAP_FILE_BUFFER | + CEPH_CAP_AUTH_SHARED | CEPH_CAP_AUTH_EXCL | + CEPH_CAP_XATTR_SHARED | CEPH_CAP_XATTR_EXCL; + if (mode & CEPH_FILE_MODE_LAZY) + caps |= CEPH_CAP_FILE_LAZYIO; + + return caps; +} +EXPORT_SYMBOL(ceph_caps_for_mode); diff --git a/net/ceph/ceph_hash.c b/net/ceph/ceph_hash.c new file mode 100644 index 00000000000..815ef882679 --- /dev/null +++ b/net/ceph/ceph_hash.c @@ -0,0 +1,118 @@ + +#include <linux/ceph/types.h> + +/* + * Robert Jenkin's hash function. + * http://burtleburtle.net/bob/hash/evahash.html + * This is in the public domain. + */ +#define mix(a, b, c) \ + do { \ + a = a - b; a = a - c; a = a ^ (c >> 13); \ + b = b - c; b = b - a; b = b ^ (a << 8); \ + c = c - a; c = c - b; c = c ^ (b >> 13); \ + a = a - b; a = a - c; a = a ^ (c >> 12); \ + b = b - c; b = b - a; b = b ^ (a << 16); \ + c = c - a; c = c - b; c = c ^ (b >> 5); \ + a = a - b; a = a - c; a = a ^ (c >> 3); \ + b = b - c; b = b - a; b = b ^ (a << 10); \ + c = c - a; c = c - b; c = c ^ (b >> 15); \ + } while (0) + +unsigned ceph_str_hash_rjenkins(const char *str, unsigned length) +{ + const unsigned char *k = (const unsigned char *)str; + __u32 a, b, c; /* the internal state */ + __u32 len; /* how many key bytes still need mixing */ + + /* Set up the internal state */ + len = length; + a = 0x9e3779b9; /* the golden ratio; an arbitrary value */ + b = a; + c = 0; /* variable initialization of internal state */ + + /* handle most of the key */ + while (len >= 12) { + a = a + (k[0] + ((__u32)k[1] << 8) + ((__u32)k[2] << 16) + + ((__u32)k[3] << 24)); + b = b + (k[4] + ((__u32)k[5] << 8) + ((__u32)k[6] << 16) + + ((__u32)k[7] << 24)); + c = c + (k[8] + ((__u32)k[9] << 8) + ((__u32)k[10] << 16) + + ((__u32)k[11] << 24)); + mix(a, b, c); + k = k + 12; + len = len - 12; + } + + /* handle the last 11 bytes */ + c = c + length; + switch (len) { /* all the case statements fall through */ + case 11: + c = c + ((__u32)k[10] << 24); + case 10: + c = c + ((__u32)k[9] << 16); + case 9: + c = c + ((__u32)k[8] << 8); + /* the first byte of c is reserved for the length */ + case 8: + b = b + ((__u32)k[7] << 24); + case 7: + b = b + ((__u32)k[6] << 16); + case 6: + b = b + ((__u32)k[5] << 8); + case 5: + b = b + k[4]; + case 4: + a = a + ((__u32)k[3] << 24); + case 3: + a = a + ((__u32)k[2] << 16); + case 2: + a = a + ((__u32)k[1] << 8); + case 1: + a = a + k[0]; + /* case 0: nothing left to add */ + } + mix(a, b, c); + + return c; +} + +/* + * linux dcache hash + */ +unsigned ceph_str_hash_linux(const char *str, unsigned length) +{ + unsigned long hash = 0; + unsigned char c; + + while (length--) { + c = *str++; + hash = (hash + (c << 4) + (c >> 4)) * 11; + } + return hash; +} + + +unsigned ceph_str_hash(int type, const char *s, unsigned len) +{ + switch (type) { + case CEPH_STR_HASH_LINUX: + return ceph_str_hash_linux(s, len); + case CEPH_STR_HASH_RJENKINS: + return ceph_str_hash_rjenkins(s, len); + default: + return -1; + } +} + +const char *ceph_str_hash_name(int type) +{ + switch (type) { + case CEPH_STR_HASH_LINUX: + return "linux"; + case CEPH_STR_HASH_RJENKINS: + return "rjenkins"; + default: + return "unknown"; + } +} diff --git a/net/ceph/ceph_strings.c b/net/ceph/ceph_strings.c new file mode 100644 index 00000000000..3fbda04de29 --- /dev/null +++ b/net/ceph/ceph_strings.c @@ -0,0 +1,84 @@ +/* + * Ceph string constants + */ +#include <linux/module.h> +#include <linux/ceph/types.h> + +const char *ceph_entity_type_name(int type) +{ + switch (type) { + case CEPH_ENTITY_TYPE_MDS: return "mds"; + case CEPH_ENTITY_TYPE_OSD: return "osd"; + case CEPH_ENTITY_TYPE_MON: return "mon"; + case CEPH_ENTITY_TYPE_CLIENT: return "client"; + case CEPH_ENTITY_TYPE_AUTH: return "auth"; + default: return "unknown"; + } +} + +const char *ceph_osd_op_name(int op) +{ + switch (op) { + case CEPH_OSD_OP_READ: return "read"; + case CEPH_OSD_OP_STAT: return "stat"; + + case CEPH_OSD_OP_MASKTRUNC: return "masktrunc"; + + case CEPH_OSD_OP_WRITE: return "write"; + case CEPH_OSD_OP_DELETE: return "delete"; + case CEPH_OSD_OP_TRUNCATE: return "truncate"; + case CEPH_OSD_OP_ZERO: return "zero"; + case CEPH_OSD_OP_WRITEFULL: return "writefull"; + case CEPH_OSD_OP_ROLLBACK: return "rollback"; + + case CEPH_OSD_OP_APPEND: return "append"; + case CEPH_OSD_OP_STARTSYNC: return "startsync"; + case CEPH_OSD_OP_SETTRUNC: return "settrunc"; + case CEPH_OSD_OP_TRIMTRUNC: return "trimtrunc"; + + case CEPH_OSD_OP_TMAPUP: return "tmapup"; + case CEPH_OSD_OP_TMAPGET: return "tmapget"; + case CEPH_OSD_OP_TMAPPUT: return "tmapput"; + + case CEPH_OSD_OP_GETXATTR: return "getxattr"; + case CEPH_OSD_OP_GETXATTRS: return "getxattrs"; + case CEPH_OSD_OP_SETXATTR: return "setxattr"; + case CEPH_OSD_OP_SETXATTRS: return "setxattrs"; + case CEPH_OSD_OP_RESETXATTRS: return "resetxattrs"; + case CEPH_OSD_OP_RMXATTR: return "rmxattr"; + case CEPH_OSD_OP_CMPXATTR: return "cmpxattr"; + + case CEPH_OSD_OP_PULL: return "pull"; + case CEPH_OSD_OP_PUSH: return "push"; + case CEPH_OSD_OP_BALANCEREADS: return "balance-reads"; + case CEPH_OSD_OP_UNBALANCEREADS: return "unbalance-reads"; + case CEPH_OSD_OP_SCRUB: return "scrub"; + + case CEPH_OSD_OP_WRLOCK: return "wrlock"; + case CEPH_OSD_OP_WRUNLOCK: return "wrunlock"; + case CEPH_OSD_OP_RDLOCK: return "rdlock"; + case CEPH_OSD_OP_RDUNLOCK: return "rdunlock"; + case CEPH_OSD_OP_UPLOCK: return "uplock"; + case CEPH_OSD_OP_DNLOCK: return "dnlock"; + + case CEPH_OSD_OP_CALL: return "call"; + + case CEPH_OSD_OP_PGLS: return "pgls"; + } + return "???"; +} + + +const char *ceph_pool_op_name(int op) +{ + switch (op) { + case POOL_OP_CREATE: return "create"; + case POOL_OP_DELETE: return "delete"; + case POOL_OP_AUID_CHANGE: return "auid change"; + case POOL_OP_CREATE_SNAP: return "create snap"; + case POOL_OP_DELETE_SNAP: return "delete snap"; + case POOL_OP_CREATE_UNMANAGED_SNAP: return "create unmanaged snap"; + case POOL_OP_DELETE_UNMANAGED_SNAP: return "delete unmanaged snap"; + } + return "???"; +} diff --git a/net/ceph/crush/crush.c b/net/ceph/crush/crush.c new file mode 100644 index 00000000000..d6ebb13a18a --- /dev/null +++ b/net/ceph/crush/crush.c @@ -0,0 +1,151 @@ + +#ifdef __KERNEL__ +# include <linux/slab.h> +#else +# include <stdlib.h> +# include <assert.h> +# define kfree(x) do { if (x) free(x); } while (0) +# define BUG_ON(x) assert(!(x)) +#endif + +#include <linux/crush/crush.h> + +const char *crush_bucket_alg_name(int alg) +{ + switch (alg) { + case CRUSH_BUCKET_UNIFORM: return "uniform"; + case CRUSH_BUCKET_LIST: return "list"; + case CRUSH_BUCKET_TREE: return "tree"; + case CRUSH_BUCKET_STRAW: return "straw"; + default: return "unknown"; + } +} + +/** + * crush_get_bucket_item_weight - Get weight of an item in given bucket + * @b: bucket pointer + * @p: item index in bucket + */ +int crush_get_bucket_item_weight(struct crush_bucket *b, int p) +{ + if (p >= b->size) + return 0; + + switch (b->alg) { + case CRUSH_BUCKET_UNIFORM: + return ((struct crush_bucket_uniform *)b)->item_weight; + case CRUSH_BUCKET_LIST: + return ((struct crush_bucket_list *)b)->item_weights[p]; + case CRUSH_BUCKET_TREE: + if (p & 1) + return ((struct crush_bucket_tree *)b)->node_weights[p]; + return 0; + case CRUSH_BUCKET_STRAW: + return ((struct crush_bucket_straw *)b)->item_weights[p]; + } + return 0; +} + +/** + * crush_calc_parents - Calculate parent vectors for the given crush map. + * @map: crush_map pointer + */ +void crush_calc_parents(struct crush_map *map) +{ + int i, b, c; + + for (b = 0; b < map->max_buckets; b++) { + if (map->buckets[b] == NULL) + continue; + for (i = 0; i < map->buckets[b]->size; i++) { + c = map->buckets[b]->items[i]; + BUG_ON(c >= map->max_devices || + c < -map->max_buckets); + if (c >= 0) + map->device_parents[c] = map->buckets[b]->id; + else + map->bucket_parents[-1-c] = map->buckets[b]->id; + } + } +} + +void crush_destroy_bucket_uniform(struct crush_bucket_uniform *b) +{ + kfree(b->h.perm); + kfree(b->h.items); + kfree(b); +} + +void crush_destroy_bucket_list(struct crush_bucket_list *b) +{ + kfree(b->item_weights); + kfree(b->sum_weights); + kfree(b->h.perm); + kfree(b->h.items); + kfree(b); +} + +void crush_destroy_bucket_tree(struct crush_bucket_tree *b) +{ + kfree(b->node_weights); + kfree(b); +} + +void crush_destroy_bucket_straw(struct crush_bucket_straw *b) +{ + kfree(b->straws); + kfree(b->item_weights); + kfree(b->h.perm); + kfree(b->h.items); + kfree(b); +} + +void crush_destroy_bucket(struct crush_bucket *b) +{ + switch (b->alg) { + case CRUSH_BUCKET_UNIFORM: + crush_destroy_bucket_uniform((struct crush_bucket_uniform *)b); + break; + case CRUSH_BUCKET_LIST: + crush_destroy_bucket_list((struct crush_bucket_list *)b); + break; + case CRUSH_BUCKET_TREE: + crush_destroy_bucket_tree((struct crush_bucket_tree *)b); + break; + case CRUSH_BUCKET_STRAW: + crush_destroy_bucket_straw((struct crush_bucket_straw *)b); + break; + } +} + +/** + * crush_destroy - Destroy a crush_map + * @map: crush_map pointer + */ +void crush_destroy(struct crush_map *map) +{ + int b; + + /* buckets */ + if (map->buckets) { + for (b = 0; b < map->max_buckets; b++) { + if (map->buckets[b] == NULL) + continue; + crush_destroy_bucket(map->buckets[b]); + } + kfree(map->buckets); + } + + /* rules */ + if (map->rules) { + for (b = 0; b < map->max_rules; b++) + kfree(map->rules[b]); + kfree(map->rules); + } + + kfree(map->bucket_parents); + kfree(map->device_parents); + kfree(map); +} + + diff --git a/net/ceph/crush/hash.c b/net/ceph/crush/hash.c new file mode 100644 index 00000000000..5bb63e37a8a --- /dev/null +++ b/net/ceph/crush/hash.c @@ -0,0 +1,149 @@ + +#include <linux/types.h> +#include <linux/crush/hash.h> + +/* + * Robert Jenkins' function for mixing 32-bit values + * http://burtleburtle.net/bob/hash/evahash.html + * a, b = random bits, c = input and output + */ +#define crush_hashmix(a, b, c) do { \ + a = a-b; a = a-c; a = a^(c>>13); \ + b = b-c; b = b-a; b = b^(a<<8); \ + c = c-a; c = c-b; c = c^(b>>13); \ + a = a-b; a = a-c; a = a^(c>>12); \ + b = b-c; b = b-a; b = b^(a<<16); \ + c = c-a; c = c-b; c = c^(b>>5); \ + a = a-b; a = a-c; a = a^(c>>3); \ + b = b-c; b = b-a; b = b^(a<<10); \ + c = c-a; c = c-b; c = c^(b>>15); \ + } while (0) + +#define crush_hash_seed 1315423911 + +static __u32 crush_hash32_rjenkins1(__u32 a) +{ + __u32 hash = crush_hash_seed ^ a; + __u32 b = a; + __u32 x = 231232; + __u32 y = 1232; + crush_hashmix(b, x, hash); + crush_hashmix(y, a, hash); + return hash; +} + +static __u32 crush_hash32_rjenkins1_2(__u32 a, __u32 b) +{ + __u32 hash = crush_hash_seed ^ a ^ b; + __u32 x = 231232; + __u32 y = 1232; + crush_hashmix(a, b, hash); + crush_hashmix(x, a, hash); + crush_hashmix(b, y, hash); + return hash; +} + +static __u32 crush_hash32_rjenkins1_3(__u32 a, __u32 b, __u32 c) +{ + __u32 hash = crush_hash_seed ^ a ^ b ^ c; + __u32 x = 231232; + __u32 y = 1232; + crush_hashmix(a, b, hash); + crush_hashmix(c, x, hash); + crush_hashmix(y, a, hash); + crush_hashmix(b, x, hash); + crush_hashmix(y, c, hash); + return hash; +} + +static __u32 crush_hash32_rjenkins1_4(__u32 a, __u32 b, __u32 c, __u32 d) +{ + __u32 hash = crush_hash_seed ^ a ^ b ^ c ^ d; + __u32 x = 231232; + __u32 y = 1232; + crush_hashmix(a, b, hash); + crush_hashmix(c, d, hash); + crush_hashmix(a, x, hash); + crush_hashmix(y, b, hash); + crush_hashmix(c, x, hash); + crush_hashmix(y, d, hash); + return hash; +} + +static __u32 crush_hash32_rjenkins1_5(__u32 a, __u32 b, __u32 c, __u32 d, + __u32 e) +{ + __u32 hash = crush_hash_seed ^ a ^ b ^ c ^ d ^ e; + __u32 x = 231232; + __u32 y = 1232; + crush_hashmix(a, b, hash); + crush_hashmix(c, d, hash); + crush_hashmix(e, x, hash); + crush_hashmix(y, a, hash); + crush_hashmix(b, x, hash); + crush_hashmix(y, c, hash); + crush_hashmix(d, x, hash); + crush_hashmix(y, e, hash); + return hash; +} + + +__u32 crush_hash32(int type, __u32 a) +{ + switch (type) { + case CRUSH_HASH_RJENKINS1: + return crush_hash32_rjenkins1(a); + default: + return 0; + } +} + +__u32 crush_hash32_2(int type, __u32 a, __u32 b) +{ + switch (type) { + case CRUSH_HASH_RJENKINS1: + return crush_hash32_rjenkins1_2(a, b); + default: + return 0; + } +} + +__u32 crush_hash32_3(int type, __u32 a, __u32 b, __u32 c) +{ + switch (type) { + case CRUSH_HASH_RJENKINS1: + return crush_hash32_rjenkins1_3(a, b, c); + default: + return 0; + } +} + +__u32 crush_hash32_4(int type, __u32 a, __u32 b, __u32 c, __u32 d) +{ + switch (type) { + case CRUSH_HASH_RJENKINS1: + return crush_hash32_rjenkins1_4(a, b, c, d); + default: + return 0; + } +} + +__u32 crush_hash32_5(int type, __u32 a, __u32 b, __u32 c, __u32 d, __u32 e) +{ + switch (type) { + case CRUSH_HASH_RJENKINS1: + return crush_hash32_rjenkins1_5(a, b, c, d, e); + default: + return 0; + } +} + +const char *crush_hash_name(int type) +{ + switch (type) { + case CRUSH_HASH_RJENKINS1: + return "rjenkins1"; + default: + return "unknown"; + } +} diff --git a/net/ceph/crush/mapper.c b/net/ceph/crush/mapper.c new file mode 100644 index 00000000000..42599e31dca --- /dev/null +++ b/net/ceph/crush/mapper.c @@ -0,0 +1,609 @@ + +#ifdef __KERNEL__ +# include <linux/string.h> +# include <linux/slab.h> +# include <linux/bug.h> +# include <linux/kernel.h> +# ifndef dprintk +# define dprintk(args...) +# endif +#else +# include <string.h> +# include <stdio.h> +# include <stdlib.h> +# include <assert.h> +# define BUG_ON(x) assert(!(x)) +# define dprintk(args...) /* printf(args) */ +# define kmalloc(x, f) malloc(x) +# define kfree(x) free(x) +#endif + +#include <linux/crush/crush.h> +#include <linux/crush/hash.h> + +/* + * Implement the core CRUSH mapping algorithm. + */ + +/** + * crush_find_rule - find a crush_rule id for a given ruleset, type, and size. + * @map: the crush_map + * @ruleset: the storage ruleset id (user defined) + * @type: storage ruleset type (user defined) + * @size: output set size + */ +int crush_find_rule(struct crush_map *map, int ruleset, int type, int size) +{ + int i; + + for (i = 0; i < map->max_rules; i++) { + if (map->rules[i] && + map->rules[i]->mask.ruleset == ruleset && + map->rules[i]->mask.type == type && + map->rules[i]->mask.min_size <= size && + map->rules[i]->mask.max_size >= size) + return i; + } + return -1; +} + + +/* + * bucket choose methods + * + * For each bucket algorithm, we have a "choose" method that, given a + * crush input @x and replica position (usually, position in output set) @r, + * will produce an item in the bucket. + */ + +/* + * Choose based on a random permutation of the bucket. + * + * We used to use some prime number arithmetic to do this, but it + * wasn't very random, and had some other bad behaviors. Instead, we + * calculate an actual random permutation of the bucket members. + * Since this is expensive, we optimize for the r=0 case, which + * captures the vast majority of calls. + */ +static int bucket_perm_choose(struct crush_bucket *bucket, + int x, int r) +{ + unsigned pr = r % bucket->size; + unsigned i, s; + + /* start a new permutation if @x has changed */ + if (bucket->perm_x != x || bucket->perm_n == 0) { + dprintk("bucket %d new x=%d\n", bucket->id, x); + bucket->perm_x = x; + + /* optimize common r=0 case */ + if (pr == 0) { + s = crush_hash32_3(bucket->hash, x, bucket->id, 0) % + bucket->size; + bucket->perm[0] = s; + bucket->perm_n = 0xffff; /* magic value, see below */ + goto out; + } + + for (i = 0; i < bucket->size; i++) + bucket->perm[i] = i; + bucket->perm_n = 0; + } else if (bucket->perm_n == 0xffff) { + /* clean up after the r=0 case above */ + for (i = 1; i < bucket->size; i++) + bucket->perm[i] = i; + bucket->perm[bucket->perm[0]] = 0; + bucket->perm_n = 1; + } + + /* calculate permutation up to pr */ + for (i = 0; i < bucket->perm_n; i++) + dprintk(" perm_choose have %d: %d\n", i, bucket->perm[i]); + while (bucket->perm_n <= pr) { + unsigned p = bucket->perm_n; + /* no point in swapping the final entry */ + if (p < bucket->size - 1) { + i = crush_hash32_3(bucket->hash, x, bucket->id, p) % + (bucket->size - p); + if (i) { + unsigned t = bucket->perm[p + i]; + bucket->perm[p + i] = bucket->perm[p]; + bucket->perm[p] = t; + } + dprintk(" perm_choose swap %d with %d\n", p, p+i); + } + bucket->perm_n++; + } + for (i = 0; i < bucket->size; i++) + dprintk(" perm_choose %d: %d\n", i, bucket->perm[i]); + + s = bucket->perm[pr]; +out: + dprintk(" perm_choose %d sz=%d x=%d r=%d (%d) s=%d\n", bucket->id, + bucket->size, x, r, pr, s); + return bucket->items[s]; +} + +/* uniform */ +static int bucket_uniform_choose(struct crush_bucket_uniform *bucket, + int x, int r) +{ + return bucket_perm_choose(&bucket->h, x, r); +} + +/* list */ +static int bucket_list_choose(struct crush_bucket_list *bucket, + int x, int r) +{ + int i; + + for (i = bucket->h.size-1; i >= 0; i--) { + __u64 w = crush_hash32_4(bucket->h.hash,x, bucket->h.items[i], + r, bucket->h.id); + w &= 0xffff; + dprintk("list_choose i=%d x=%d r=%d item %d weight %x " + "sw %x rand %llx", + i, x, r, bucket->h.items[i], bucket->item_weights[i], + bucket->sum_weights[i], w); + w *= bucket->sum_weights[i]; + w = w >> 16; + /*dprintk(" scaled %llx\n", w);*/ + if (w < bucket->item_weights[i]) + return bucket->h.items[i]; + } + + BUG_ON(1); + return 0; +} + + +/* (binary) tree */ +static int height(int n) +{ + int h = 0; + while ((n & 1) == 0) { + h++; + n = n >> 1; + } + return h; +} + +static int left(int x) +{ + int h = height(x); + return x - (1 << (h-1)); +} + +static int right(int x) +{ + int h = height(x); + return x + (1 << (h-1)); +} + +static int terminal(int x) +{ + return x & 1; +} + +static int bucket_tree_choose(struct crush_bucket_tree *bucket, + int x, int r) +{ + int n, l; + __u32 w; + __u64 t; + + /* start at root */ + n = bucket->num_nodes >> 1; + + while (!terminal(n)) { + /* pick point in [0, w) */ + w = bucket->node_weights[n]; + t = (__u64)crush_hash32_4(bucket->h.hash, x, n, r, + bucket->h.id) * (__u64)w; + t = t >> 32; + + /* descend to the left or right? */ + l = left(n); + if (t < bucket->node_weights[l]) + n = l; + else + n = right(n); + } + + return bucket->h.items[n >> 1]; +} + + +/* straw */ + +static int bucket_straw_choose(struct crush_bucket_straw *bucket, + int x, int r) +{ + int i; + int high = 0; + __u64 high_draw = 0; + __u64 draw; + + for (i = 0; i < bucket->h.size; i++) { + draw = crush_hash32_3(bucket->h.hash, x, bucket->h.items[i], r); + draw &= 0xffff; + draw *= bucket->straws[i]; + if (i == 0 || draw > high_draw) { + high = i; + high_draw = draw; + } + } + return bucket->h.items[high]; +} + +static int crush_bucket_choose(struct crush_bucket *in, int x, int r) +{ + dprintk(" crush_bucket_choose %d x=%d r=%d\n", in->id, x, r); + switch (in->alg) { + case CRUSH_BUCKET_UNIFORM: + return bucket_uniform_choose((struct crush_bucket_uniform *)in, + x, r); + case CRUSH_BUCKET_LIST: + return bucket_list_choose((struct crush_bucket_list *)in, + x, r); + case CRUSH_BUCKET_TREE: + return bucket_tree_choose((struct crush_bucket_tree *)in, + x, r); + case CRUSH_BUCKET_STRAW: + return bucket_straw_choose((struct crush_bucket_straw *)in, + x, r); + default: + BUG_ON(1); + return in->items[0]; + } +} + +/* + * true if device is marked "out" (failed, fully offloaded) + * of the cluster + */ +static int is_out(struct crush_map *map, __u32 *weight, int item, int x) +{ + if (weight[item] >= 0x10000) + return 0; + if (weight[item] == 0) + return 1; + if ((crush_hash32_2(CRUSH_HASH_RJENKINS1, x, item) & 0xffff) + < weight[item]) + return 0; + return 1; +} + +/** + * crush_choose - choose numrep distinct items of given type + * @map: the crush_map + * @bucket: the bucket we are choose an item from + * @x: crush input value + * @numrep: the number of items to choose + * @type: the type of item to choose + * @out: pointer to output vector + * @outpos: our position in that vector + * @firstn: true if choosing "first n" items, false if choosing "indep" + * @recurse_to_leaf: true if we want one device under each item of given type + * @out2: second output vector for leaf items (if @recurse_to_leaf) + */ +static int crush_choose(struct crush_map *map, + struct crush_bucket *bucket, + __u32 *weight, + int x, int numrep, int type, + int *out, int outpos, + int firstn, int recurse_to_leaf, + int *out2) +{ + int rep; + int ftotal, flocal; + int retry_descent, retry_bucket, skip_rep; + struct crush_bucket *in = bucket; + int r; + int i; + int item = 0; + int itemtype; + int collide, reject; + const int orig_tries = 5; /* attempts before we fall back to search */ + + dprintk("CHOOSE%s bucket %d x %d outpos %d numrep %d\n", recurse_to_leaf ? "_LEAF" : "", + bucket->id, x, outpos, numrep); + + for (rep = outpos; rep < numrep; rep++) { + /* keep trying until we get a non-out, non-colliding item */ + ftotal = 0; + skip_rep = 0; + do { + retry_descent = 0; + in = bucket; /* initial bucket */ + + /* choose through intervening buckets */ + flocal = 0; + do { + collide = 0; + retry_bucket = 0; + r = rep; + if (in->alg == CRUSH_BUCKET_UNIFORM) { + /* be careful */ + if (firstn || numrep >= in->size) + /* r' = r + f_total */ + r += ftotal; + else if (in->size % numrep == 0) + /* r'=r+(n+1)*f_local */ + r += (numrep+1) * + (flocal+ftotal); + else + /* r' = r + n*f_local */ + r += numrep * (flocal+ftotal); + } else { + if (firstn) + /* r' = r + f_total */ + r += ftotal; + else + /* r' = r + n*f_local */ + r += numrep * (flocal+ftotal); + } + + /* bucket choose */ + if (in->size == 0) { + reject = 1; + goto reject; + } + if (flocal >= (in->size>>1) && + flocal > orig_tries) + item = bucket_perm_choose(in, x, r); + else + item = crush_bucket_choose(in, x, r); + BUG_ON(item >= map->max_devices); + + /* desired type? */ + if (item < 0) + itemtype = map->buckets[-1-item]->type; + else + itemtype = 0; + dprintk(" item %d type %d\n", item, itemtype); + + /* keep going? */ + if (itemtype != type) { + BUG_ON(item >= 0 || + (-1-item) >= map->max_buckets); + in = map->buckets[-1-item]; + retry_bucket = 1; + continue; + } + + /* collision? */ + for (i = 0; i < outpos; i++) { + if (out[i] == item) { + collide = 1; + break; + } + } + + reject = 0; + if (recurse_to_leaf) { + if (item < 0) { + if (crush_choose(map, + map->buckets[-1-item], + weight, + x, outpos+1, 0, + out2, outpos, + firstn, 0, + NULL) <= outpos) + /* didn't get leaf */ + reject = 1; + } else { + /* we already have a leaf! */ + out2[outpos] = item; + } + } + + if (!reject) { + /* out? */ + if (itemtype == 0) + reject = is_out(map, weight, + item, x); + else + reject = 0; + } + +reject: + if (reject || collide) { + ftotal++; + flocal++; + + if (collide && flocal < 3) + /* retry locally a few times */ + retry_bucket = 1; + else if (flocal < in->size + orig_tries) + /* exhaustive bucket search */ + retry_bucket = 1; + else if (ftotal < 20) + /* then retry descent */ + retry_descent = 1; + else + /* else give up */ + skip_rep = 1; + dprintk(" reject %d collide %d " + "ftotal %d flocal %d\n", + reject, collide, ftotal, + flocal); + } + } while (retry_bucket); + } while (retry_descent); + + if (skip_rep) { + dprintk("skip rep\n"); + continue; + } + + dprintk("CHOOSE got %d\n", item); + out[outpos] = item; + outpos++; + } + + dprintk("CHOOSE returns %d\n", outpos); + return outpos; +} + + +/** + * crush_do_rule - calculate a mapping with the given input and rule + * @map: the crush_map + * @ruleno: the rule id + * @x: hash input + * @result: pointer to result vector + * @result_max: maximum result size + * @force: force initial replica choice; -1 for none + */ +int crush_do_rule(struct crush_map *map, + int ruleno, int x, int *result, int result_max, + int force, __u32 *weight) +{ + int result_len; + int force_context[CRUSH_MAX_DEPTH]; + int force_pos = -1; + int a[CRUSH_MAX_SET]; + int b[CRUSH_MAX_SET]; + int c[CRUSH_MAX_SET]; + int recurse_to_leaf; + int *w; + int wsize = 0; + int *o; + int osize; + int *tmp; + struct crush_rule *rule; + int step; + int i, j; + int numrep; + int firstn; + int rc = -1; + + BUG_ON(ruleno >= map->max_rules); + + rule = map->rules[ruleno]; + result_len = 0; + w = a; + o = b; + + /* + * determine hierarchical context of force, if any. note + * that this may or may not correspond to the specific types + * referenced by the crush rule. + */ + if (force >= 0) { + if (force >= map->max_devices || + map->device_parents[force] == 0) { + /*dprintk("CRUSH: forcefed device dne\n");*/ + rc = -1; /* force fed device dne */ + goto out; + } + if (!is_out(map, weight, force, x)) { + while (1) { + force_context[++force_pos] = force; + if (force >= 0) + force = map->device_parents[force]; + else + force = map->bucket_parents[-1-force]; + if (force == 0) + break; + } + } + } + + for (step = 0; step < rule->len; step++) { + firstn = 0; + switch (rule->steps[step].op) { + case CRUSH_RULE_TAKE: + w[0] = rule->steps[step].arg1; + if (force_pos >= 0) { + BUG_ON(force_context[force_pos] != w[0]); + force_pos--; + } + wsize = 1; + break; + + case CRUSH_RULE_CHOOSE_LEAF_FIRSTN: + case CRUSH_RULE_CHOOSE_FIRSTN: + firstn = 1; + case CRUSH_RULE_CHOOSE_LEAF_INDEP: + case CRUSH_RULE_CHOOSE_INDEP: + BUG_ON(wsize == 0); + + recurse_to_leaf = + rule->steps[step].op == + CRUSH_RULE_CHOOSE_LEAF_FIRSTN || + rule->steps[step].op == + CRUSH_RULE_CHOOSE_LEAF_INDEP; + + /* reset output */ + osize = 0; + + for (i = 0; i < wsize; i++) { + /* + * see CRUSH_N, CRUSH_N_MINUS macros. + * basically, numrep <= 0 means relative to + * the provided result_max + */ + numrep = rule->steps[step].arg1; + if (numrep <= 0) { + numrep += result_max; + if (numrep <= 0) + continue; + } + j = 0; + if (osize == 0 && force_pos >= 0) { + /* skip any intermediate types */ + while (force_pos && + force_context[force_pos] < 0 && + rule->steps[step].arg2 != + map->buckets[-1 - + force_context[force_pos]]->type) + force_pos--; + o[osize] = force_context[force_pos]; + if (recurse_to_leaf) + c[osize] = force_context[0]; + j++; + force_pos--; + } + osize += crush_choose(map, + map->buckets[-1-w[i]], + weight, + x, numrep, + rule->steps[step].arg2, + o+osize, j, + firstn, + recurse_to_leaf, c+osize); + } + + if (recurse_to_leaf) + /* copy final _leaf_ values to output set */ + memcpy(o, c, osize*sizeof(*o)); + + /* swap t and w arrays */ + tmp = o; + o = w; + w = tmp; + wsize = osize; + break; + + + case CRUSH_RULE_EMIT: + for (i = 0; i < wsize && result_len < result_max; i++) { + result[result_len] = w[i]; + result_len++; + } + wsize = 0; + break; + + default: + BUG_ON(1); + } + } + rc = result_len; + +out: + return rc; +} + + diff --git a/net/ceph/crypto.c b/net/ceph/crypto.c new file mode 100644 index 00000000000..7b505b0c983 --- /dev/null +++ b/net/ceph/crypto.c @@ -0,0 +1,412 @@ + +#include <linux/ceph/ceph_debug.h> + +#include <linux/err.h> +#include <linux/scatterlist.h> +#include <linux/slab.h> +#include <crypto/hash.h> + +#include <linux/ceph/decode.h> +#include "crypto.h" + +int ceph_crypto_key_encode(struct ceph_crypto_key *key, void **p, void *end) +{ + if (*p + sizeof(u16) + sizeof(key->created) + + sizeof(u16) + key->len > end) + return -ERANGE; + ceph_encode_16(p, key->type); + ceph_encode_copy(p, &key->created, sizeof(key->created)); + ceph_encode_16(p, key->len); + ceph_encode_copy(p, key->key, key->len); + return 0; +} + +int ceph_crypto_key_decode(struct ceph_crypto_key *key, void **p, void *end) +{ + ceph_decode_need(p, end, 2*sizeof(u16) + sizeof(key->created), bad); + key->type = ceph_decode_16(p); + ceph_decode_copy(p, &key->created, sizeof(key->created)); + key->len = ceph_decode_16(p); + ceph_decode_need(p, end, key->len, bad); + key->key = kmalloc(key->len, GFP_NOFS); + if (!key->key) + return -ENOMEM; + ceph_decode_copy(p, key->key, key->len); + return 0; + +bad: + dout("failed to decode crypto key\n"); + return -EINVAL; +} + +int ceph_crypto_key_unarmor(struct ceph_crypto_key *key, const char *inkey) +{ + int inlen = strlen(inkey); + int blen = inlen * 3 / 4; + void *buf, *p; + int ret; + + dout("crypto_key_unarmor %s\n", inkey); + buf = kmalloc(blen, GFP_NOFS); + if (!buf) + return -ENOMEM; + blen = ceph_unarmor(buf, inkey, inkey+inlen); + if (blen < 0) { + kfree(buf); + return blen; + } + + p = buf; + ret = ceph_crypto_key_decode(key, &p, p + blen); + kfree(buf); + if (ret) + return ret; + dout("crypto_key_unarmor key %p type %d len %d\n", key, + key->type, key->len); + return 0; +} + + + +#define AES_KEY_SIZE 16 + +static struct crypto_blkcipher *ceph_crypto_alloc_cipher(void) +{ + return crypto_alloc_blkcipher("cbc(aes)", 0, CRYPTO_ALG_ASYNC); +} + +static const u8 *aes_iv = (u8 *)CEPH_AES_IV; + +static int ceph_aes_encrypt(const void *key, int key_len, + void *dst, size_t *dst_len, + const void *src, size_t src_len) +{ + struct scatterlist sg_in[2], sg_out[1]; + struct crypto_blkcipher *tfm = ceph_crypto_alloc_cipher(); + struct blkcipher_desc desc = { .tfm = tfm, .flags = 0 }; + int ret; + void *iv; + int ivsize; + size_t zero_padding = (0x10 - (src_len & 0x0f)); + char pad[16]; + + if (IS_ERR(tfm)) + return PTR_ERR(tfm); + + memset(pad, zero_padding, zero_padding); + + *dst_len = src_len + zero_padding; + + crypto_blkcipher_setkey((void *)tfm, key, key_len); + sg_init_table(sg_in, 2); + sg_set_buf(&sg_in[0], src, src_len); + sg_set_buf(&sg_in[1], pad, zero_padding); + sg_init_table(sg_out, 1); + sg_set_buf(sg_out, dst, *dst_len); + iv = crypto_blkcipher_crt(tfm)->iv; + ivsize = crypto_blkcipher_ivsize(tfm); + + memcpy(iv, aes_iv, ivsize); + /* + print_hex_dump(KERN_ERR, "enc key: ", DUMP_PREFIX_NONE, 16, 1, + key, key_len, 1); + print_hex_dump(KERN_ERR, "enc src: ", DUMP_PREFIX_NONE, 16, 1, + src, src_len, 1); + print_hex_dump(KERN_ERR, "enc pad: ", DUMP_PREFIX_NONE, 16, 1, + pad, zero_padding, 1); + */ + ret = crypto_blkcipher_encrypt(&desc, sg_out, sg_in, + src_len + zero_padding); + crypto_free_blkcipher(tfm); + if (ret < 0) + pr_err("ceph_aes_crypt failed %d\n", ret); + /* + print_hex_dump(KERN_ERR, "enc out: ", DUMP_PREFIX_NONE, 16, 1, + dst, *dst_len, 1); + */ + return 0; +} + +static int ceph_aes_encrypt2(const void *key, int key_len, void *dst, + size_t *dst_len, + const void *src1, size_t src1_len, + const void *src2, size_t src2_len) +{ + struct scatterlist sg_in[3], sg_out[1]; + struct crypto_blkcipher *tfm = ceph_crypto_alloc_cipher(); + struct blkcipher_desc desc = { .tfm = tfm, .flags = 0 }; + int ret; + void *iv; + int ivsize; + size_t zero_padding = (0x10 - ((src1_len + src2_len) & 0x0f)); + char pad[16]; + + if (IS_ERR(tfm)) + return PTR_ERR(tfm); + + memset(pad, zero_padding, zero_padding); + + *dst_len = src1_len + src2_len + zero_padding; + + crypto_blkcipher_setkey((void *)tfm, key, key_len); + sg_init_table(sg_in, 3); + sg_set_buf(&sg_in[0], src1, src1_len); + sg_set_buf(&sg_in[1], src2, src2_len); + sg_set_buf(&sg_in[2], pad, zero_padding); + sg_init_table(sg_out, 1); + sg_set_buf(sg_out, dst, *dst_len); + iv = crypto_blkcipher_crt(tfm)->iv; + ivsize = crypto_blkcipher_ivsize(tfm); + + memcpy(iv, aes_iv, ivsize); + /* + print_hex_dump(KERN_ERR, "enc key: ", DUMP_PREFIX_NONE, 16, 1, + key, key_len, 1); + print_hex_dump(KERN_ERR, "enc src1: ", DUMP_PREFIX_NONE, 16, 1, + src1, src1_len, 1); + print_hex_dump(KERN_ERR, "enc src2: ", DUMP_PREFIX_NONE, 16, 1, + src2, src2_len, 1); + print_hex_dump(KERN_ERR, "enc pad: ", DUMP_PREFIX_NONE, 16, 1, + pad, zero_padding, 1); + */ + ret = crypto_blkcipher_encrypt(&desc, sg_out, sg_in, + src1_len + src2_len + zero_padding); + crypto_free_blkcipher(tfm); + if (ret < 0) + pr_err("ceph_aes_crypt2 failed %d\n", ret); + /* + print_hex_dump(KERN_ERR, "enc out: ", DUMP_PREFIX_NONE, 16, 1, + dst, *dst_len, 1); + */ + return 0; +} + +static int ceph_aes_decrypt(const void *key, int key_len, + void *dst, size_t *dst_len, + const void *src, size_t src_len) +{ + struct scatterlist sg_in[1], sg_out[2]; + struct crypto_blkcipher *tfm = ceph_crypto_alloc_cipher(); + struct blkcipher_desc desc = { .tfm = tfm }; + char pad[16]; + void *iv; + int ivsize; + int ret; + int last_byte; + + if (IS_ERR(tfm)) + return PTR_ERR(tfm); + + crypto_blkcipher_setkey((void *)tfm, key, key_len); + sg_init_table(sg_in, 1); + sg_init_table(sg_out, 2); + sg_set_buf(sg_in, src, src_len); + sg_set_buf(&sg_out[0], dst, *dst_len); + sg_set_buf(&sg_out[1], pad, sizeof(pad)); + + iv = crypto_blkcipher_crt(tfm)->iv; + ivsize = crypto_blkcipher_ivsize(tfm); + + memcpy(iv, aes_iv, ivsize); + + /* + print_hex_dump(KERN_ERR, "dec key: ", DUMP_PREFIX_NONE, 16, 1, + key, key_len, 1); + print_hex_dump(KERN_ERR, "dec in: ", DUMP_PREFIX_NONE, 16, 1, + src, src_len, 1); + */ + + ret = crypto_blkcipher_decrypt(&desc, sg_out, sg_in, src_len); + crypto_free_blkcipher(tfm); + if (ret < 0) { + pr_err("ceph_aes_decrypt failed %d\n", ret); + return ret; + } + + if (src_len <= *dst_len) + last_byte = ((char *)dst)[src_len - 1]; + else + last_byte = pad[src_len - *dst_len - 1]; + if (last_byte <= 16 && src_len >= last_byte) { + *dst_len = src_len - last_byte; + } else { + pr_err("ceph_aes_decrypt got bad padding %d on src len %d\n", + last_byte, (int)src_len); + return -EPERM; /* bad padding */ + } + /* + print_hex_dump(KERN_ERR, "dec out: ", DUMP_PREFIX_NONE, 16, 1, + dst, *dst_len, 1); + */ + return 0; +} + +static int ceph_aes_decrypt2(const void *key, int key_len, + void *dst1, size_t *dst1_len, + void *dst2, size_t *dst2_len, + const void *src, size_t src_len) +{ + struct scatterlist sg_in[1], sg_out[3]; + struct crypto_blkcipher *tfm = ceph_crypto_alloc_cipher(); + struct blkcipher_desc desc = { .tfm = tfm }; + char pad[16]; + void *iv; + int ivsize; + int ret; + int last_byte; + + if (IS_ERR(tfm)) + return PTR_ERR(tfm); + + sg_init_table(sg_in, 1); + sg_set_buf(sg_in, src, src_len); + sg_init_table(sg_out, 3); + sg_set_buf(&sg_out[0], dst1, *dst1_len); + sg_set_buf(&sg_out[1], dst2, *dst2_len); + sg_set_buf(&sg_out[2], pad, sizeof(pad)); + + crypto_blkcipher_setkey((void *)tfm, key, key_len); + iv = crypto_blkcipher_crt(tfm)->iv; + ivsize = crypto_blkcipher_ivsize(tfm); + + memcpy(iv, aes_iv, ivsize); + + /* + print_hex_dump(KERN_ERR, "dec key: ", DUMP_PREFIX_NONE, 16, 1, + key, key_len, 1); + print_hex_dump(KERN_ERR, "dec in: ", DUMP_PREFIX_NONE, 16, 1, + src, src_len, 1); + */ + + ret = crypto_blkcipher_decrypt(&desc, sg_out, sg_in, src_len); + crypto_free_blkcipher(tfm); + if (ret < 0) { + pr_err("ceph_aes_decrypt failed %d\n", ret); + return ret; + } + + if (src_len <= *dst1_len) + last_byte = ((char *)dst1)[src_len - 1]; + else if (src_len <= *dst1_len + *dst2_len) + last_byte = ((char *)dst2)[src_len - *dst1_len - 1]; + else + last_byte = pad[src_len - *dst1_len - *dst2_len - 1]; + if (last_byte <= 16 && src_len >= last_byte) { + src_len -= last_byte; + } else { + pr_err("ceph_aes_decrypt got bad padding %d on src len %d\n", + last_byte, (int)src_len); + return -EPERM; /* bad padding */ + } + + if (src_len < *dst1_len) { + *dst1_len = src_len; + *dst2_len = 0; + } else { + *dst2_len = src_len - *dst1_len; + } + /* + print_hex_dump(KERN_ERR, "dec out1: ", DUMP_PREFIX_NONE, 16, 1, + dst1, *dst1_len, 1); + print_hex_dump(KERN_ERR, "dec out2: ", DUMP_PREFIX_NONE, 16, 1, + dst2, *dst2_len, 1); + */ + + return 0; +} + + +int ceph_decrypt(struct ceph_crypto_key *secret, void *dst, size_t *dst_len, + const void *src, size_t src_len) +{ + switch (secret->type) { + case CEPH_CRYPTO_NONE: + if (*dst_len < src_len) + return -ERANGE; + memcpy(dst, src, src_len); + *dst_len = src_len; + return 0; + + case CEPH_CRYPTO_AES: + return ceph_aes_decrypt(secret->key, secret->len, dst, + dst_len, src, src_len); + + default: + return -EINVAL; + } +} + +int ceph_decrypt2(struct ceph_crypto_key *secret, + void *dst1, size_t *dst1_len, + void *dst2, size_t *dst2_len, + const void *src, size_t src_len) +{ + size_t t; + + switch (secret->type) { + case CEPH_CRYPTO_NONE: + if (*dst1_len + *dst2_len < src_len) + return -ERANGE; + t = min(*dst1_len, src_len); + memcpy(dst1, src, t); + *dst1_len = t; + src += t; + src_len -= t; + if (src_len) { + t = min(*dst2_len, src_len); + memcpy(dst2, src, t); + *dst2_len = t; + } + return 0; + + case CEPH_CRYPTO_AES: + return ceph_aes_decrypt2(secret->key, secret->len, + dst1, dst1_len, dst2, dst2_len, + src, src_len); + + default: + return -EINVAL; + } +} + +int ceph_encrypt(struct ceph_crypto_key *secret, void *dst, size_t *dst_len, + const void *src, size_t src_len) +{ + switch (secret->type) { + case CEPH_CRYPTO_NONE: + if (*dst_len < src_len) + return -ERANGE; + memcpy(dst, src, src_len); + *dst_len = src_len; + return 0; + + case CEPH_CRYPTO_AES: + return ceph_aes_encrypt(secret->key, secret->len, dst, + dst_len, src, src_len); + + default: + return -EINVAL; + } +} + +int ceph_encrypt2(struct ceph_crypto_key *secret, void *dst, size_t *dst_len, + const void *src1, size_t src1_len, + const void *src2, size_t src2_len) +{ + switch (secret->type) { + case CEPH_CRYPTO_NONE: + if (*dst_len < src1_len + src2_len) + return -ERANGE; + memcpy(dst, src1, src1_len); + memcpy(dst + src1_len, src2, src2_len); + *dst_len = src1_len + src2_len; + return 0; + + case CEPH_CRYPTO_AES: + return ceph_aes_encrypt2(secret->key, secret->len, dst, dst_len, + src1, src1_len, src2, src2_len); + + default: + return -EINVAL; + } +} diff --git a/net/ceph/crypto.h b/net/ceph/crypto.h new file mode 100644 index 00000000000..f9eccace592 --- /dev/null +++ b/net/ceph/crypto.h @@ -0,0 +1,48 @@ +#ifndef _FS_CEPH_CRYPTO_H +#define _FS_CEPH_CRYPTO_H + +#include <linux/ceph/types.h> +#include <linux/ceph/buffer.h> + +/* + * cryptographic secret + */ +struct ceph_crypto_key { + int type; + struct ceph_timespec created; + int len; + void *key; +}; + +static inline void ceph_crypto_key_destroy(struct ceph_crypto_key *key) +{ + kfree(key->key); +} + +extern int ceph_crypto_key_encode(struct ceph_crypto_key *key, + void **p, void *end); +extern int ceph_crypto_key_decode(struct ceph_crypto_key *key, + void **p, void *end); +extern int ceph_crypto_key_unarmor(struct ceph_crypto_key *key, const char *in); + +/* crypto.c */ +extern int ceph_decrypt(struct ceph_crypto_key *secret, + void *dst, size_t *dst_len, + const void *src, size_t src_len); +extern int ceph_encrypt(struct ceph_crypto_key *secret, + void *dst, size_t *dst_len, + const void *src, size_t src_len); +extern int ceph_decrypt2(struct ceph_crypto_key *secret, + void *dst1, size_t *dst1_len, + void *dst2, size_t *dst2_len, + const void *src, size_t src_len); +extern int ceph_encrypt2(struct ceph_crypto_key *secret, + void *dst, size_t *dst_len, + const void *src1, size_t src1_len, + const void *src2, size_t src2_len); + +/* armor.c */ +extern int ceph_armor(char *dst, const char *src, const char *end); +extern int ceph_unarmor(char *dst, const char *src, const char *end); + +#endif diff --git a/net/ceph/debugfs.c b/net/ceph/debugfs.c new file mode 100644 index 00000000000..27d4ea315d1 --- /dev/null +++ b/net/ceph/debugfs.c @@ -0,0 +1,267 @@ +#include <linux/ceph/ceph_debug.h> + +#include <linux/device.h> +#include <linux/slab.h> +#include <linux/module.h> +#include <linux/ctype.h> +#include <linux/debugfs.h> +#include <linux/seq_file.h> + +#include <linux/ceph/libceph.h> +#include <linux/ceph/mon_client.h> +#include <linux/ceph/auth.h> +#include <linux/ceph/debugfs.h> + +#ifdef CONFIG_DEBUG_FS + +/* + * Implement /sys/kernel/debug/ceph fun + * + * /sys/kernel/debug/ceph/client* - an instance of the ceph client + * .../osdmap - current osdmap + * .../monmap - current monmap + * .../osdc - active osd requests + * .../monc - mon client state + * .../dentry_lru - dump contents of dentry lru + * .../caps - expose cap (reservation) stats + * .../bdi - symlink to ../../bdi/something + */ + +static struct dentry *ceph_debugfs_dir; + +static int monmap_show(struct seq_file *s, void *p) +{ + int i; + struct ceph_client *client = s->private; + + if (client->monc.monmap == NULL) + return 0; + + seq_printf(s, "epoch %d\n", client->monc.monmap->epoch); + for (i = 0; i < client->monc.monmap->num_mon; i++) { + struct ceph_entity_inst *inst = + &client->monc.monmap->mon_inst[i]; + + seq_printf(s, "\t%s%lld\t%s\n", + ENTITY_NAME(inst->name), + ceph_pr_addr(&inst->addr.in_addr)); + } + return 0; +} + +static int osdmap_show(struct seq_file *s, void *p) +{ + int i; + struct ceph_client *client = s->private; + struct rb_node *n; + + if (client->osdc.osdmap == NULL) + return 0; + seq_printf(s, "epoch %d\n", client->osdc.osdmap->epoch); + seq_printf(s, "flags%s%s\n", + (client->osdc.osdmap->flags & CEPH_OSDMAP_NEARFULL) ? + " NEARFULL" : "", + (client->osdc.osdmap->flags & CEPH_OSDMAP_FULL) ? + " FULL" : ""); + for (n = rb_first(&client->osdc.osdmap->pg_pools); n; n = rb_next(n)) { + struct ceph_pg_pool_info *pool = + rb_entry(n, struct ceph_pg_pool_info, node); + seq_printf(s, "pg_pool %d pg_num %d / %d, lpg_num %d / %d\n", + pool->id, pool->v.pg_num, pool->pg_num_mask, + pool->v.lpg_num, pool->lpg_num_mask); + } + for (i = 0; i < client->osdc.osdmap->max_osd; i++) { + struct ceph_entity_addr *addr = + &client->osdc.osdmap->osd_addr[i]; + int state = client->osdc.osdmap->osd_state[i]; + char sb[64]; + + seq_printf(s, "\tosd%d\t%s\t%3d%%\t(%s)\n", + i, ceph_pr_addr(&addr->in_addr), + ((client->osdc.osdmap->osd_weight[i]*100) >> 16), + ceph_osdmap_state_str(sb, sizeof(sb), state)); + } + return 0; +} + +static int monc_show(struct seq_file *s, void *p) +{ + struct ceph_client *client = s->private; + struct ceph_mon_generic_request *req; + struct ceph_mon_client *monc = &client->monc; + struct rb_node *rp; + + mutex_lock(&monc->mutex); + + if (monc->have_mdsmap) + seq_printf(s, "have mdsmap %u\n", (unsigned)monc->have_mdsmap); + if (monc->have_osdmap) + seq_printf(s, "have osdmap %u\n", (unsigned)monc->have_osdmap); + if (monc->want_next_osdmap) + seq_printf(s, "want next osdmap\n"); + + for (rp = rb_first(&monc->generic_request_tree); rp; rp = rb_next(rp)) { + __u16 op; + req = rb_entry(rp, struct ceph_mon_generic_request, node); + op = le16_to_cpu(req->request->hdr.type); + if (op == CEPH_MSG_STATFS) + seq_printf(s, "%lld statfs\n", req->tid); + else + seq_printf(s, "%lld unknown\n", req->tid); + } + + mutex_unlock(&monc->mutex); + return 0; +} + +static int osdc_show(struct seq_file *s, void *pp) +{ + struct ceph_client *client = s->private; + struct ceph_osd_client *osdc = &client->osdc; + struct rb_node *p; + + mutex_lock(&osdc->request_mutex); + for (p = rb_first(&osdc->requests); p; p = rb_next(p)) { + struct ceph_osd_request *req; + struct ceph_osd_request_head *head; + struct ceph_osd_op *op; + int num_ops; + int opcode, olen; + int i; + + req = rb_entry(p, struct ceph_osd_request, r_node); + + seq_printf(s, "%lld\tosd%d\t%d.%x\t", req->r_tid, + req->r_osd ? req->r_osd->o_osd : -1, + le32_to_cpu(req->r_pgid.pool), + le16_to_cpu(req->r_pgid.ps)); + + head = req->r_request->front.iov_base; + op = (void *)(head + 1); + + num_ops = le16_to_cpu(head->num_ops); + olen = le32_to_cpu(head->object_len); + seq_printf(s, "%.*s", olen, + (const char *)(head->ops + num_ops)); + + if (req->r_reassert_version.epoch) + seq_printf(s, "\t%u'%llu", + (unsigned)le32_to_cpu(req->r_reassert_version.epoch), + le64_to_cpu(req->r_reassert_version.version)); + else + seq_printf(s, "\t"); + + for (i = 0; i < num_ops; i++) { + opcode = le16_to_cpu(op->op); + seq_printf(s, "\t%s", ceph_osd_op_name(opcode)); + op++; + } + + seq_printf(s, "\n"); + } + mutex_unlock(&osdc->request_mutex); + return 0; +} + +CEPH_DEFINE_SHOW_FUNC(monmap_show) +CEPH_DEFINE_SHOW_FUNC(osdmap_show) +CEPH_DEFINE_SHOW_FUNC(monc_show) +CEPH_DEFINE_SHOW_FUNC(osdc_show) + +int ceph_debugfs_init(void) +{ + ceph_debugfs_dir = debugfs_create_dir("ceph", NULL); + if (!ceph_debugfs_dir) + return -ENOMEM; + return 0; +} + +void ceph_debugfs_cleanup(void) +{ + debugfs_remove(ceph_debugfs_dir); +} + +int ceph_debugfs_client_init(struct ceph_client *client) +{ + int ret = -ENOMEM; + char name[80]; + + snprintf(name, sizeof(name), "%pU.client%lld", &client->fsid, + client->monc.auth->global_id); + + client->debugfs_dir = debugfs_create_dir(name, ceph_debugfs_dir); + if (!client->debugfs_dir) + goto out; + + client->monc.debugfs_file = debugfs_create_file("monc", + 0600, + client->debugfs_dir, + client, + &monc_show_fops); + if (!client->monc.debugfs_file) + goto out; + + client->osdc.debugfs_file = debugfs_create_file("osdc", + 0600, + client->debugfs_dir, + client, + &osdc_show_fops); + if (!client->osdc.debugfs_file) + goto out; + + client->debugfs_monmap = debugfs_create_file("monmap", + 0600, + client->debugfs_dir, + client, + &monmap_show_fops); + if (!client->debugfs_monmap) + goto out; + + client->debugfs_osdmap = debugfs_create_file("osdmap", + 0600, + client->debugfs_dir, + client, + &osdmap_show_fops); + if (!client->debugfs_osdmap) + goto out; + + return 0; + +out: + ceph_debugfs_client_cleanup(client); + return ret; +} + +void ceph_debugfs_client_cleanup(struct ceph_client *client) +{ + debugfs_remove(client->debugfs_osdmap); + debugfs_remove(client->debugfs_monmap); + debugfs_remove(client->osdc.debugfs_file); + debugfs_remove(client->monc.debugfs_file); + debugfs_remove(client->debugfs_dir); +} + +#else /* CONFIG_DEBUG_FS */ + +int ceph_debugfs_init(void) +{ + return 0; +} + +void ceph_debugfs_cleanup(void) +{ +} + +int ceph_debugfs_client_init(struct ceph_client *client) +{ + return 0; +} + +void ceph_debugfs_client_cleanup(struct ceph_client *client) +{ +} + +#endif /* CONFIG_DEBUG_FS */ + +EXPORT_SYMBOL(ceph_debugfs_init); +EXPORT_SYMBOL(ceph_debugfs_cleanup); diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c new file mode 100644 index 00000000000..0e8157ee5d4 --- /dev/null +++ b/net/ceph/messenger.c @@ -0,0 +1,2453 @@ +#include <linux/ceph/ceph_debug.h> + +#include <linux/crc32c.h> +#include <linux/ctype.h> +#include <linux/highmem.h> +#include <linux/inet.h> +#include <linux/kthread.h> +#include <linux/net.h> +#include <linux/slab.h> +#include <linux/socket.h> +#include <linux/string.h> +#include <linux/bio.h> +#include <linux/blkdev.h> +#include <net/tcp.h> + +#include <linux/ceph/libceph.h> +#include <linux/ceph/messenger.h> +#include <linux/ceph/decode.h> +#include <linux/ceph/pagelist.h> + +/* + * Ceph uses the messenger to exchange ceph_msg messages with other + * hosts in the system. The messenger provides ordered and reliable + * delivery. We tolerate TCP disconnects by reconnecting (with + * exponential backoff) in the case of a fault (disconnection, bad + * crc, protocol error). Acks allow sent messages to be discarded by + * the sender. + */ + +/* static tag bytes (protocol control messages) */ +static char tag_msg = CEPH_MSGR_TAG_MSG; +static char tag_ack = CEPH_MSGR_TAG_ACK; +static char tag_keepalive = CEPH_MSGR_TAG_KEEPALIVE; + +#ifdef CONFIG_LOCKDEP +static struct lock_class_key socket_class; +#endif + + +static void queue_con(struct ceph_connection *con); +static void con_work(struct work_struct *); +static void ceph_fault(struct ceph_connection *con); + +/* + * nicely render a sockaddr as a string. + */ +#define MAX_ADDR_STR 20 +#define MAX_ADDR_STR_LEN 60 +static char addr_str[MAX_ADDR_STR][MAX_ADDR_STR_LEN]; +static DEFINE_SPINLOCK(addr_str_lock); +static int last_addr_str; + +const char *ceph_pr_addr(const struct sockaddr_storage *ss) +{ + int i; + char *s; + struct sockaddr_in *in4 = (void *)ss; + struct sockaddr_in6 *in6 = (void *)ss; + + spin_lock(&addr_str_lock); + i = last_addr_str++; + if (last_addr_str == MAX_ADDR_STR) + last_addr_str = 0; + spin_unlock(&addr_str_lock); + s = addr_str[i]; + + switch (ss->ss_family) { + case AF_INET: + snprintf(s, MAX_ADDR_STR_LEN, "%pI4:%u", &in4->sin_addr, + (unsigned int)ntohs(in4->sin_port)); + break; + + case AF_INET6: + snprintf(s, MAX_ADDR_STR_LEN, "[%pI6c]:%u", &in6->sin6_addr, + (unsigned int)ntohs(in6->sin6_port)); + break; + + default: + sprintf(s, "(unknown sockaddr family %d)", (int)ss->ss_family); + } + + return s; +} +EXPORT_SYMBOL(ceph_pr_addr); + +static void encode_my_addr(struct ceph_messenger *msgr) +{ + memcpy(&msgr->my_enc_addr, &msgr->inst.addr, sizeof(msgr->my_enc_addr)); + ceph_encode_addr(&msgr->my_enc_addr); +} + +/* + * work queue for all reading and writing to/from the socket. + */ +struct workqueue_struct *ceph_msgr_wq; + +int ceph_msgr_init(void) +{ + ceph_msgr_wq = create_workqueue("ceph-msgr"); + if (IS_ERR(ceph_msgr_wq)) { + int ret = PTR_ERR(ceph_msgr_wq); + pr_err("msgr_init failed to create workqueue: %d\n", ret); + ceph_msgr_wq = NULL; + return ret; + } + return 0; +} +EXPORT_SYMBOL(ceph_msgr_init); + +void ceph_msgr_exit(void) +{ + destroy_workqueue(ceph_msgr_wq); +} +EXPORT_SYMBOL(ceph_msgr_exit); + +void ceph_msgr_flush(void) +{ + flush_workqueue(ceph_msgr_wq); +} +EXPORT_SYMBOL(ceph_msgr_flush); + + +/* + * socket callback functions + */ + +/* data available on socket, or listen socket received a connect */ +static void ceph_data_ready(struct sock *sk, int count_unused) +{ + struct ceph_connection *con = + (struct ceph_connection *)sk->sk_user_data; + if (sk->sk_state != TCP_CLOSE_WAIT) { + dout("ceph_data_ready on %p state = %lu, queueing work\n", + con, con->state); + queue_con(con); + } +} + +/* socket has buffer space for writing */ +static void ceph_write_space(struct sock *sk) +{ + struct ceph_connection *con = + (struct ceph_connection *)sk->sk_user_data; + + /* only queue to workqueue if there is data we want to write. */ + if (test_bit(WRITE_PENDING, &con->state)) { + dout("ceph_write_space %p queueing write work\n", con); + queue_con(con); + } else { + dout("ceph_write_space %p nothing to write\n", con); + } + + /* since we have our own write_space, clear the SOCK_NOSPACE flag */ + clear_bit(SOCK_NOSPACE, &sk->sk_socket->flags); +} + +/* socket's state has changed */ +static void ceph_state_change(struct sock *sk) +{ + struct ceph_connection *con = + (struct ceph_connection *)sk->sk_user_data; + + dout("ceph_state_change %p state = %lu sk_state = %u\n", + con, con->state, sk->sk_state); + + if (test_bit(CLOSED, &con->state)) + return; + + switch (sk->sk_state) { + case TCP_CLOSE: + dout("ceph_state_change TCP_CLOSE\n"); + case TCP_CLOSE_WAIT: + dout("ceph_state_change TCP_CLOSE_WAIT\n"); + if (test_and_set_bit(SOCK_CLOSED, &con->state) == 0) { + if (test_bit(CONNECTING, &con->state)) + con->error_msg = "connection failed"; + else + con->error_msg = "socket closed"; + queue_con(con); + } + break; + case TCP_ESTABLISHED: + dout("ceph_state_change TCP_ESTABLISHED\n"); + queue_con(con); + break; + } +} + +/* + * set up socket callbacks + */ +static void set_sock_callbacks(struct socket *sock, + struct ceph_connection *con) +{ + struct sock *sk = sock->sk; + sk->sk_user_data = (void *)con; + sk->sk_data_ready = ceph_data_ready; + sk->sk_write_space = ceph_write_space; + sk->sk_state_change = ceph_state_change; +} + + +/* + * socket helpers + */ + +/* + * initiate connection to a remote socket. + */ +static struct socket *ceph_tcp_connect(struct ceph_connection *con) +{ + struct sockaddr_storage *paddr = &con->peer_addr.in_addr; + struct socket *sock; + int ret; + + BUG_ON(con->sock); + ret = sock_create_kern(con->peer_addr.in_addr.ss_family, SOCK_STREAM, + IPPROTO_TCP, &sock); + if (ret) + return ERR_PTR(ret); + con->sock = sock; + sock->sk->sk_allocation = GFP_NOFS; + +#ifdef CONFIG_LOCKDEP + lockdep_set_class(&sock->sk->sk_lock, &socket_class); +#endif + + set_sock_callbacks(sock, con); + + dout("connect %s\n", ceph_pr_addr(&con->peer_addr.in_addr)); + + ret = sock->ops->connect(sock, (struct sockaddr *)paddr, sizeof(*paddr), + O_NONBLOCK); + if (ret == -EINPROGRESS) { + dout("connect %s EINPROGRESS sk_state = %u\n", + ceph_pr_addr(&con->peer_addr.in_addr), + sock->sk->sk_state); + ret = 0; + } + if (ret < 0) { + pr_err("connect %s error %d\n", + ceph_pr_addr(&con->peer_addr.in_addr), ret); + sock_release(sock); + con->sock = NULL; + con->error_msg = "connect error"; + } + + if (ret < 0) + return ERR_PTR(ret); + return sock; +} + +static int ceph_tcp_recvmsg(struct socket *sock, void *buf, size_t len) +{ + struct kvec iov = {buf, len}; + struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL }; + + return kernel_recvmsg(sock, &msg, &iov, 1, len, msg.msg_flags); +} + +/* + * write something. @more is true if caller will be sending more data + * shortly. + */ +static int ceph_tcp_sendmsg(struct socket *sock, struct kvec *iov, + size_t kvlen, size_t len, int more) +{ + struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL }; + + if (more) + msg.msg_flags |= MSG_MORE; + else + msg.msg_flags |= MSG_EOR; /* superfluous, but what the hell */ + + return kernel_sendmsg(sock, &msg, iov, kvlen, len); +} + + +/* + * Shutdown/close the socket for the given connection. + */ +static int con_close_socket(struct ceph_connection *con) +{ + int rc; + + dout("con_close_socket on %p sock %p\n", con, con->sock); + if (!con->sock) + return 0; + set_bit(SOCK_CLOSED, &con->state); + rc = con->sock->ops->shutdown(con->sock, SHUT_RDWR); + sock_release(con->sock); + con->sock = NULL; + clear_bit(SOCK_CLOSED, &con->state); + return rc; +} + +/* + * Reset a connection. Discard all incoming and outgoing messages + * and clear *_seq state. + */ +static void ceph_msg_remove(struct ceph_msg *msg) +{ + list_del_init(&msg->list_head); + ceph_msg_put(msg); +} +static void ceph_msg_remove_list(struct list_head *head) +{ + while (!list_empty(head)) { + struct ceph_msg *msg = list_first_entry(head, struct ceph_msg, + list_head); + ceph_msg_remove(msg); + } +} + +static void reset_connection(struct ceph_connection *con) +{ + /* reset connection, out_queue, msg_ and connect_seq */ + /* discard existing out_queue and msg_seq */ + ceph_msg_remove_list(&con->out_queue); + ceph_msg_remove_list(&con->out_sent); + + if (con->in_msg) { + ceph_msg_put(con->in_msg); + con->in_msg = NULL; + } + + con->connect_seq = 0; + con->out_seq = 0; + if (con->out_msg) { + ceph_msg_put(con->out_msg); + con->out_msg = NULL; + } + con->out_keepalive_pending = false; + con->in_seq = 0; + con->in_seq_acked = 0; +} + +/* + * mark a peer down. drop any open connections. + */ +void ceph_con_close(struct ceph_connection *con) +{ + dout("con_close %p peer %s\n", con, + ceph_pr_addr(&con->peer_addr.in_addr)); + set_bit(CLOSED, &con->state); /* in case there's queued work */ + clear_bit(STANDBY, &con->state); /* avoid connect_seq bump */ + clear_bit(LOSSYTX, &con->state); /* so we retry next connect */ + clear_bit(KEEPALIVE_PENDING, &con->state); + clear_bit(WRITE_PENDING, &con->state); + mutex_lock(&con->mutex); + reset_connection(con); + con->peer_global_seq = 0; + cancel_delayed_work(&con->work); + mutex_unlock(&con->mutex); + queue_con(con); +} +EXPORT_SYMBOL(ceph_con_close); + +/* + * Reopen a closed connection, with a new peer address. + */ +void ceph_con_open(struct ceph_connection *con, struct ceph_entity_addr *addr) +{ + dout("con_open %p %s\n", con, ceph_pr_addr(&addr->in_addr)); + set_bit(OPENING, &con->state); + clear_bit(CLOSED, &con->state); + memcpy(&con->peer_addr, addr, sizeof(*addr)); + con->delay = 0; /* reset backoff memory */ + queue_con(con); +} +EXPORT_SYMBOL(ceph_con_open); + +/* + * return true if this connection ever successfully opened + */ +bool ceph_con_opened(struct ceph_connection *con) +{ + return con->connect_seq > 0; +} + +/* + * generic get/put + */ +struct ceph_connection *ceph_con_get(struct ceph_connection *con) +{ + dout("con_get %p nref = %d -> %d\n", con, + atomic_read(&con->nref), atomic_read(&con->nref) + 1); + if (atomic_inc_not_zero(&con->nref)) + return con; + return NULL; +} + +void ceph_con_put(struct ceph_connection *con) +{ + dout("con_put %p nref = %d -> %d\n", con, + atomic_read(&con->nref), atomic_read(&con->nref) - 1); + BUG_ON(atomic_read(&con->nref) == 0); + if (atomic_dec_and_test(&con->nref)) { + BUG_ON(con->sock); + kfree(con); + } +} + +/* + * initialize a new connection. + */ +void ceph_con_init(struct ceph_messenger *msgr, struct ceph_connection *con) +{ + dout("con_init %p\n", con); + memset(con, 0, sizeof(*con)); + atomic_set(&con->nref, 1); + con->msgr = msgr; + mutex_init(&con->mutex); + INIT_LIST_HEAD(&con->out_queue); + INIT_LIST_HEAD(&con->out_sent); + INIT_DELAYED_WORK(&con->work, con_work); +} +EXPORT_SYMBOL(ceph_con_init); + + +/* + * We maintain a global counter to order connection attempts. Get + * a unique seq greater than @gt. + */ +static u32 get_global_seq(struct ceph_messenger *msgr, u32 gt) +{ + u32 ret; + + spin_lock(&msgr->global_seq_lock); + if (msgr->global_seq < gt) + msgr->global_seq = gt; + ret = ++msgr->global_seq; + spin_unlock(&msgr->global_seq_lock); + return ret; +} + + +/* + * Prepare footer for currently outgoing message, and finish things + * off. Assumes out_kvec* are already valid.. we just add on to the end. + */ +static void prepare_write_message_footer(struct ceph_connection *con, int v) +{ + struct ceph_msg *m = con->out_msg; + + dout("prepare_write_message_footer %p\n", con); + con->out_kvec_is_msg = true; + con->out_kvec[v].iov_base = &m->footer; + con->out_kvec[v].iov_len = sizeof(m->footer); + con->out_kvec_bytes += sizeof(m->footer); + con->out_kvec_left++; + con->out_more = m->more_to_follow; + con->out_msg_done = true; +} + +/* + * Prepare headers for the next outgoing message. + */ +static void prepare_write_message(struct ceph_connection *con) +{ + struct ceph_msg *m; + int v = 0; + + con->out_kvec_bytes = 0; + con->out_kvec_is_msg = true; + con->out_msg_done = false; + + /* Sneak an ack in there first? If we can get it into the same + * TCP packet that's a good thing. */ + if (con->in_seq > con->in_seq_acked) { + con->in_seq_acked = con->in_seq; + con->out_kvec[v].iov_base = &tag_ack; + con->out_kvec[v++].iov_len = 1; + con->out_temp_ack = cpu_to_le64(con->in_seq_acked); + con->out_kvec[v].iov_base = &con->out_temp_ack; + con->out_kvec[v++].iov_len = sizeof(con->out_temp_ack); + con->out_kvec_bytes = 1 + sizeof(con->out_temp_ack); + } + + m = list_first_entry(&con->out_queue, + struct ceph_msg, list_head); + con->out_msg = m; + if (test_bit(LOSSYTX, &con->state)) { + list_del_init(&m->list_head); + } else { + /* put message on sent list */ + ceph_msg_get(m); + list_move_tail(&m->list_head, &con->out_sent); + } + + /* + * only assign outgoing seq # if we haven't sent this message + * yet. if it is requeued, resend with it's original seq. + */ + if (m->needs_out_seq) { + m->hdr.seq = cpu_to_le64(++con->out_seq); + m->needs_out_seq = false; + } + + dout("prepare_write_message %p seq %lld type %d len %d+%d+%d %d pgs\n", + m, con->out_seq, le16_to_cpu(m->hdr.type), + le32_to_cpu(m->hdr.front_len), le32_to_cpu(m->hdr.middle_len), + le32_to_cpu(m->hdr.data_len), + m->nr_pages); + BUG_ON(le32_to_cpu(m->hdr.front_len) != m->front.iov_len); + + /* tag + hdr + front + middle */ + con->out_kvec[v].iov_base = &tag_msg; + con->out_kvec[v++].iov_len = 1; + con->out_kvec[v].iov_base = &m->hdr; + con->out_kvec[v++].iov_len = sizeof(m->hdr); + con->out_kvec[v++] = m->front; + if (m->middle) + con->out_kvec[v++] = m->middle->vec; + con->out_kvec_left = v; + con->out_kvec_bytes += 1 + sizeof(m->hdr) + m->front.iov_len + + (m->middle ? m->middle->vec.iov_len : 0); + con->out_kvec_cur = con->out_kvec; + + /* fill in crc (except data pages), footer */ + con->out_msg->hdr.crc = + cpu_to_le32(crc32c(0, (void *)&m->hdr, + sizeof(m->hdr) - sizeof(m->hdr.crc))); + con->out_msg->footer.flags = CEPH_MSG_FOOTER_COMPLETE; + con->out_msg->footer.front_crc = + cpu_to_le32(crc32c(0, m->front.iov_base, m->front.iov_len)); + if (m->middle) + con->out_msg->footer.middle_crc = + cpu_to_le32(crc32c(0, m->middle->vec.iov_base, + m->middle->vec.iov_len)); + else + con->out_msg->footer.middle_crc = 0; + con->out_msg->footer.data_crc = 0; + dout("prepare_write_message front_crc %u data_crc %u\n", + le32_to_cpu(con->out_msg->footer.front_crc), + le32_to_cpu(con->out_msg->footer.middle_crc)); + + /* is there a data payload? */ + if (le32_to_cpu(m->hdr.data_len) > 0) { + /* initialize page iterator */ + con->out_msg_pos.page = 0; + if (m->pages) + con->out_msg_pos.page_pos = + le16_to_cpu(m->hdr.data_off) & ~PAGE_MASK; + else + con->out_msg_pos.page_pos = 0; + con->out_msg_pos.data_pos = 0; + con->out_msg_pos.did_page_crc = 0; + con->out_more = 1; /* data + footer will follow */ + } else { + /* no, queue up footer too and be done */ + prepare_write_message_footer(con, v); + } + + set_bit(WRITE_PENDING, &con->state); +} + +/* + * Prepare an ack. + */ +static void prepare_write_ack(struct ceph_connection *con) +{ + dout("prepare_write_ack %p %llu -> %llu\n", con, + con->in_seq_acked, con->in_seq); + con->in_seq_acked = con->in_seq; + + con->out_kvec[0].iov_base = &tag_ack; + con->out_kvec[0].iov_len = 1; + con->out_temp_ack = cpu_to_le64(con->in_seq_acked); + con->out_kvec[1].iov_base = &con->out_temp_ack; + con->out_kvec[1].iov_len = sizeof(con->out_temp_ack); + con->out_kvec_left = 2; + con->out_kvec_bytes = 1 + sizeof(con->out_temp_ack); + con->out_kvec_cur = con->out_kvec; + con->out_more = 1; /* more will follow.. eventually.. */ + set_bit(WRITE_PENDING, &con->state); +} + +/* + * Prepare to write keepalive byte. + */ +static void prepare_write_keepalive(struct ceph_connection *con) +{ + dout("prepare_write_keepalive %p\n", con); + con->out_kvec[0].iov_base = &tag_keepalive; + con->out_kvec[0].iov_len = 1; + con->out_kvec_left = 1; + con->out_kvec_bytes = 1; + con->out_kvec_cur = con->out_kvec; + set_bit(WRITE_PENDING, &con->state); +} + +/* + * Connection negotiation. + */ + +static void prepare_connect_authorizer(struct ceph_connection *con) +{ + void *auth_buf; + int auth_len = 0; + int auth_protocol = 0; + + mutex_unlock(&con->mutex); + if (con->ops->get_authorizer) + con->ops->get_authorizer(con, &auth_buf, &auth_len, + &auth_protocol, &con->auth_reply_buf, + &con->auth_reply_buf_len, + con->auth_retry); + mutex_lock(&con->mutex); + + con->out_connect.authorizer_protocol = cpu_to_le32(auth_protocol); + con->out_connect.authorizer_len = cpu_to_le32(auth_len); + + con->out_kvec[con->out_kvec_left].iov_base = auth_buf; + con->out_kvec[con->out_kvec_left].iov_len = auth_len; + con->out_kvec_left++; + con->out_kvec_bytes += auth_len; +} + +/* + * We connected to a peer and are saying hello. + */ +static void prepare_write_banner(struct ceph_messenger *msgr, + struct ceph_connection *con) +{ + int len = strlen(CEPH_BANNER); + + con->out_kvec[0].iov_base = CEPH_BANNER; + con->out_kvec[0].iov_len = len; + con->out_kvec[1].iov_base = &msgr->my_enc_addr; + con->out_kvec[1].iov_len = sizeof(msgr->my_enc_addr); + con->out_kvec_left = 2; + con->out_kvec_bytes = len + sizeof(msgr->my_enc_addr); + con->out_kvec_cur = con->out_kvec; + con->out_more = 0; + set_bit(WRITE_PENDING, &con->state); +} + +static void prepare_write_connect(struct ceph_messenger *msgr, + struct ceph_connection *con, + int after_banner) +{ + unsigned global_seq = get_global_seq(con->msgr, 0); + int proto; + + switch (con->peer_name.type) { + case CEPH_ENTITY_TYPE_MON: + proto = CEPH_MONC_PROTOCOL; + break; + case CEPH_ENTITY_TYPE_OSD: + proto = CEPH_OSDC_PROTOCOL; + break; + case CEPH_ENTITY_TYPE_MDS: + proto = CEPH_MDSC_PROTOCOL; + break; + default: + BUG(); + } + + dout("prepare_write_connect %p cseq=%d gseq=%d proto=%d\n", con, + con->connect_seq, global_seq, proto); + + con->out_connect.features = cpu_to_le64(msgr->supported_features); + con->out_connect.host_type = cpu_to_le32(CEPH_ENTITY_TYPE_CLIENT); + con->out_connect.connect_seq = cpu_to_le32(con->connect_seq); + con->out_connect.global_seq = cpu_to_le32(global_seq); + con->out_connect.protocol_version = cpu_to_le32(proto); + con->out_connect.flags = 0; + + if (!after_banner) { + con->out_kvec_left = 0; + con->out_kvec_bytes = 0; + } + con->out_kvec[con->out_kvec_left].iov_base = &con->out_connect; + con->out_kvec[con->out_kvec_left].iov_len = sizeof(con->out_connect); + con->out_kvec_left++; + con->out_kvec_bytes += sizeof(con->out_connect); + con->out_kvec_cur = con->out_kvec; + con->out_more = 0; + set_bit(WRITE_PENDING, &con->state); + + prepare_connect_authorizer(con); +} + + +/* + * write as much of pending kvecs to the socket as we can. + * 1 -> done + * 0 -> socket full, but more to do + * <0 -> error + */ +static int write_partial_kvec(struct ceph_connection *con) +{ + int ret; + + dout("write_partial_kvec %p %d left\n", con, con->out_kvec_bytes); + while (con->out_kvec_bytes > 0) { + ret = ceph_tcp_sendmsg(con->sock, con->out_kvec_cur, + con->out_kvec_left, con->out_kvec_bytes, + con->out_more); + if (ret <= 0) + goto out; + con->out_kvec_bytes -= ret; + if (con->out_kvec_bytes == 0) + break; /* done */ + while (ret > 0) { + if (ret >= con->out_kvec_cur->iov_len) { + ret -= con->out_kvec_cur->iov_len; + con->out_kvec_cur++; + con->out_kvec_left--; + } else { + con->out_kvec_cur->iov_len -= ret; + con->out_kvec_cur->iov_base += ret; + ret = 0; + break; + } + } + } + con->out_kvec_left = 0; + con->out_kvec_is_msg = false; + ret = 1; +out: + dout("write_partial_kvec %p %d left in %d kvecs ret = %d\n", con, + con->out_kvec_bytes, con->out_kvec_left, ret); + return ret; /* done! */ +} + +#ifdef CONFIG_BLOCK +static void init_bio_iter(struct bio *bio, struct bio **iter, int *seg) +{ + if (!bio) { + *iter = NULL; + *seg = 0; + return; + } + *iter = bio; + *seg = bio->bi_idx; +} + +static void iter_bio_next(struct bio **bio_iter, int *seg) +{ + if (*bio_iter == NULL) + return; + + BUG_ON(*seg >= (*bio_iter)->bi_vcnt); + + (*seg)++; + if (*seg == (*bio_iter)->bi_vcnt) + init_bio_iter((*bio_iter)->bi_next, bio_iter, seg); +} +#endif + +/* + * Write as much message data payload as we can. If we finish, queue + * up the footer. + * 1 -> done, footer is now queued in out_kvec[]. + * 0 -> socket full, but more to do + * <0 -> error + */ +static int write_partial_msg_pages(struct ceph_connection *con) +{ + struct ceph_msg *msg = con->out_msg; + unsigned data_len = le32_to_cpu(msg->hdr.data_len); + size_t len; + int crc = con->msgr->nocrc; + int ret; + int total_max_write; + int in_trail = 0; + size_t trail_len = (msg->trail ? msg->trail->length : 0); + + dout("write_partial_msg_pages %p msg %p page %d/%d offset %d\n", + con, con->out_msg, con->out_msg_pos.page, con->out_msg->nr_pages, + con->out_msg_pos.page_pos); + +#ifdef CONFIG_BLOCK + if (msg->bio && !msg->bio_iter) + init_bio_iter(msg->bio, &msg->bio_iter, &msg->bio_seg); +#endif + + while (data_len > con->out_msg_pos.data_pos) { + struct page *page = NULL; + void *kaddr = NULL; + int max_write = PAGE_SIZE; + int page_shift = 0; + + total_max_write = data_len - trail_len - + con->out_msg_pos.data_pos; + + /* + * if we are calculating the data crc (the default), we need + * to map the page. if our pages[] has been revoked, use the + * zero page. + */ + + /* have we reached the trail part of the data? */ + if (con->out_msg_pos.data_pos >= data_len - trail_len) { + in_trail = 1; + + total_max_write = data_len - con->out_msg_pos.data_pos; + + page = list_first_entry(&msg->trail->head, + struct page, lru); + if (crc) + kaddr = kmap(page); + max_write = PAGE_SIZE; + } else if (msg->pages) { + page = msg->pages[con->out_msg_pos.page]; + if (crc) + kaddr = kmap(page); + } else if (msg->pagelist) { + page = list_first_entry(&msg->pagelist->head, + struct page, lru); + if (crc) + kaddr = kmap(page); +#ifdef CONFIG_BLOCK + } else if (msg->bio) { + struct bio_vec *bv; + + bv = bio_iovec_idx(msg->bio_iter, msg->bio_seg); + page = bv->bv_page; + page_shift = bv->bv_offset; + if (crc) + kaddr = kmap(page) + page_shift; + max_write = bv->bv_len; +#endif + } else { + page = con->msgr->zero_page; + if (crc) + kaddr = page_address(con->msgr->zero_page); + } + len = min_t(int, max_write - con->out_msg_pos.page_pos, + total_max_write); + + if (crc && !con->out_msg_pos.did_page_crc) { + void *base = kaddr + con->out_msg_pos.page_pos; + u32 tmpcrc = le32_to_cpu(con->out_msg->footer.data_crc); + + BUG_ON(kaddr == NULL); + con->out_msg->footer.data_crc = + cpu_to_le32(crc32c(tmpcrc, base, len)); + con->out_msg_pos.did_page_crc = 1; + } + ret = kernel_sendpage(con->sock, page, + con->out_msg_pos.page_pos + page_shift, + len, + MSG_DONTWAIT | MSG_NOSIGNAL | + MSG_MORE); + + if (crc && + (msg->pages || msg->pagelist || msg->bio || in_trail)) + kunmap(page); + + if (ret <= 0) + goto out; + + con->out_msg_pos.data_pos += ret; + con->out_msg_pos.page_pos += ret; + if (ret == len) { + con->out_msg_pos.page_pos = 0; + con->out_msg_pos.page++; + con->out_msg_pos.did_page_crc = 0; + if (in_trail) + list_move_tail(&page->lru, + &msg->trail->head); + else if (msg->pagelist) + list_move_tail(&page->lru, + &msg->pagelist->head); +#ifdef CONFIG_BLOCK + else if (msg->bio) + iter_bio_next(&msg->bio_iter, &msg->bio_seg); +#endif + } + } + + dout("write_partial_msg_pages %p msg %p done\n", con, msg); + + /* prepare and queue up footer, too */ + if (!crc) + con->out_msg->footer.flags |= CEPH_MSG_FOOTER_NOCRC; + con->out_kvec_bytes = 0; + con->out_kvec_left = 0; + con->out_kvec_cur = con->out_kvec; + prepare_write_message_footer(con, 0); + ret = 1; +out: + return ret; +} + +/* + * write some zeros + */ +static int write_partial_skip(struct ceph_connection *con) +{ + int ret; + + while (con->out_skip > 0) { + struct kvec iov = { + .iov_base = page_address(con->msgr->zero_page), + .iov_len = min(con->out_skip, (int)PAGE_CACHE_SIZE) + }; + + ret = ceph_tcp_sendmsg(con->sock, &iov, 1, iov.iov_len, 1); + if (ret <= 0) + goto out; + con->out_skip -= ret; + } + ret = 1; +out: + return ret; +} + +/* + * Prepare to read connection handshake, or an ack. + */ +static void prepare_read_banner(struct ceph_connection *con) +{ + dout("prepare_read_banner %p\n", con); + con->in_base_pos = 0; +} + +static void prepare_read_connect(struct ceph_connection *con) +{ + dout("prepare_read_connect %p\n", con); + con->in_base_pos = 0; +} + +static void prepare_read_ack(struct ceph_connection *con) +{ + dout("prepare_read_ack %p\n", con); + con->in_base_pos = 0; +} + +static void prepare_read_tag(struct ceph_connection *con) +{ + dout("prepare_read_tag %p\n", con); + con->in_base_pos = 0; + con->in_tag = CEPH_MSGR_TAG_READY; +} + +/* + * Prepare to read a message. + */ +static int prepare_read_message(struct ceph_connection *con) +{ + dout("prepare_read_message %p\n", con); + BUG_ON(con->in_msg != NULL); + con->in_base_pos = 0; + con->in_front_crc = con->in_middle_crc = con->in_data_crc = 0; + return 0; +} + + +static int read_partial(struct ceph_connection *con, + int *to, int size, void *object) +{ + *to += size; + while (con->in_base_pos < *to) { + int left = *to - con->in_base_pos; + int have = size - left; + int ret = ceph_tcp_recvmsg(con->sock, object + have, left); + if (ret <= 0) + return ret; + con->in_base_pos += ret; + } + return 1; +} + + +/* + * Read all or part of the connect-side handshake on a new connection + */ +static int read_partial_banner(struct ceph_connection *con) +{ + int ret, to = 0; + + dout("read_partial_banner %p at %d\n", con, con->in_base_pos); + + /* peer's banner */ + ret = read_partial(con, &to, strlen(CEPH_BANNER), con->in_banner); + if (ret <= 0) + goto out; + ret = read_partial(con, &to, sizeof(con->actual_peer_addr), + &con->actual_peer_addr); + if (ret <= 0) + goto out; + ret = read_partial(con, &to, sizeof(con->peer_addr_for_me), + &con->peer_addr_for_me); + if (ret <= 0) + goto out; +out: + return ret; +} + +static int read_partial_connect(struct ceph_connection *con) +{ + int ret, to = 0; + + dout("read_partial_connect %p at %d\n", con, con->in_base_pos); + + ret = read_partial(con, &to, sizeof(con->in_reply), &con->in_reply); + if (ret <= 0) + goto out; + ret = read_partial(con, &to, le32_to_cpu(con->in_reply.authorizer_len), + con->auth_reply_buf); + if (ret <= 0) + goto out; + + dout("read_partial_connect %p tag %d, con_seq = %u, g_seq = %u\n", + con, (int)con->in_reply.tag, + le32_to_cpu(con->in_reply.connect_seq), + le32_to_cpu(con->in_reply.global_seq)); +out: + return ret; + +} + +/* + * Verify the hello banner looks okay. + */ +static int verify_hello(struct ceph_connection *con) +{ + if (memcmp(con->in_banner, CEPH_BANNER, strlen(CEPH_BANNER))) { + pr_err("connect to %s got bad banner\n", + ceph_pr_addr(&con->peer_addr.in_addr)); + con->error_msg = "protocol error, bad banner"; + return -1; + } + return 0; +} + +static bool addr_is_blank(struct sockaddr_storage *ss) +{ + switch (ss->ss_family) { + case AF_INET: + return ((struct sockaddr_in *)ss)->sin_addr.s_addr == 0; + case AF_INET6: + return + ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[0] == 0 && + ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[1] == 0 && + ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[2] == 0 && + ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[3] == 0; + } + return false; +} + +static int addr_port(struct sockaddr_storage *ss) +{ + switch (ss->ss_family) { + case AF_INET: + return ntohs(((struct sockaddr_in *)ss)->sin_port); + case AF_INET6: + return ntohs(((struct sockaddr_in6 *)ss)->sin6_port); + } + return 0; +} + +static void addr_set_port(struct sockaddr_storage *ss, int p) +{ + switch (ss->ss_family) { + case AF_INET: + ((struct sockaddr_in *)ss)->sin_port = htons(p); + case AF_INET6: + ((struct sockaddr_in6 *)ss)->sin6_port = htons(p); + } +} + +/* + * Parse an ip[:port] list into an addr array. Use the default + * monitor port if a port isn't specified. + */ +int ceph_parse_ips(const char *c, const char *end, + struct ceph_entity_addr *addr, + int max_count, int *count) +{ + int i; + const char *p = c; + + dout("parse_ips on '%.*s'\n", (int)(end-c), c); + for (i = 0; i < max_count; i++) { + const char *ipend; + struct sockaddr_storage *ss = &addr[i].in_addr; + struct sockaddr_in *in4 = (void *)ss; + struct sockaddr_in6 *in6 = (void *)ss; + int port; + char delim = ','; + + if (*p == '[') { + delim = ']'; + p++; + } + + memset(ss, 0, sizeof(*ss)); + if (in4_pton(p, end - p, (u8 *)&in4->sin_addr.s_addr, + delim, &ipend)) + ss->ss_family = AF_INET; + else if (in6_pton(p, end - p, (u8 *)&in6->sin6_addr.s6_addr, + delim, &ipend)) + ss->ss_family = AF_INET6; + else + goto bad; + p = ipend; + + if (delim == ']') { + if (*p != ']') { + dout("missing matching ']'\n"); + goto bad; + } + p++; + } + + /* port? */ + if (p < end && *p == ':') { + port = 0; + p++; + while (p < end && *p >= '0' && *p <= '9') { + port = (port * 10) + (*p - '0'); + p++; + } + if (port > 65535 || port == 0) + goto bad; + } else { + port = CEPH_MON_PORT; + } + + addr_set_port(ss, port); + + dout("parse_ips got %s\n", ceph_pr_addr(ss)); + + if (p == end) + break; + if (*p != ',') + goto bad; + p++; + } + + if (p != end) + goto bad; + + if (count) + *count = i + 1; + return 0; + +bad: + pr_err("parse_ips bad ip '%.*s'\n", (int)(end - c), c); + return -EINVAL; +} +EXPORT_SYMBOL(ceph_parse_ips); + +static int process_banner(struct ceph_connection *con) +{ + dout("process_banner on %p\n", con); + + if (verify_hello(con) < 0) + return -1; + + ceph_decode_addr(&con->actual_peer_addr); + ceph_decode_addr(&con->peer_addr_for_me); + + /* + * Make sure the other end is who we wanted. note that the other + * end may not yet know their ip address, so if it's 0.0.0.0, give + * them the benefit of the doubt. + */ + if (memcmp(&con->peer_addr, &con->actual_peer_addr, + sizeof(con->peer_addr)) != 0 && + !(addr_is_blank(&con->actual_peer_addr.in_addr) && + con->actual_peer_addr.nonce == con->peer_addr.nonce)) { + pr_warning("wrong peer, want %s/%d, got %s/%d\n", + ceph_pr_addr(&con->peer_addr.in_addr), + (int)le32_to_cpu(con->peer_addr.nonce), + ceph_pr_addr(&con->actual_peer_addr.in_addr), + (int)le32_to_cpu(con->actual_peer_addr.nonce)); + con->error_msg = "wrong peer at address"; + return -1; + } + + /* + * did we learn our address? + */ + if (addr_is_blank(&con->msgr->inst.addr.in_addr)) { + int port = addr_port(&con->msgr->inst.addr.in_addr); + + memcpy(&con->msgr->inst.addr.in_addr, + &con->peer_addr_for_me.in_addr, + sizeof(con->peer_addr_for_me.in_addr)); + addr_set_port(&con->msgr->inst.addr.in_addr, port); + encode_my_addr(con->msgr); + dout("process_banner learned my addr is %s\n", + ceph_pr_addr(&con->msgr->inst.addr.in_addr)); + } + + set_bit(NEGOTIATING, &con->state); + prepare_read_connect(con); + return 0; +} + +static void fail_protocol(struct ceph_connection *con) +{ + reset_connection(con); + set_bit(CLOSED, &con->state); /* in case there's queued work */ + + mutex_unlock(&con->mutex); + if (con->ops->bad_proto) + con->ops->bad_proto(con); + mutex_lock(&con->mutex); +} + +static int process_connect(struct ceph_connection *con) +{ + u64 sup_feat = con->msgr->supported_features; + u64 req_feat = con->msgr->required_features; + u64 server_feat = le64_to_cpu(con->in_reply.features); + + dout("process_connect on %p tag %d\n", con, (int)con->in_tag); + + switch (con->in_reply.tag) { + case CEPH_MSGR_TAG_FEATURES: + pr_err("%s%lld %s feature set mismatch," + " my %llx < server's %llx, missing %llx\n", + ENTITY_NAME(con->peer_name), + ceph_pr_addr(&con->peer_addr.in_addr), + sup_feat, server_feat, server_feat & ~sup_feat); + con->error_msg = "missing required protocol features"; + fail_protocol(con); + return -1; + + case CEPH_MSGR_TAG_BADPROTOVER: + pr_err("%s%lld %s protocol version mismatch," + " my %d != server's %d\n", + ENTITY_NAME(con->peer_name), + ceph_pr_addr(&con->peer_addr.in_addr), + le32_to_cpu(con->out_connect.protocol_version), + le32_to_cpu(con->in_reply.protocol_version)); + con->error_msg = "protocol version mismatch"; + fail_protocol(con); + return -1; + + case CEPH_MSGR_TAG_BADAUTHORIZER: + con->auth_retry++; + dout("process_connect %p got BADAUTHORIZER attempt %d\n", con, + con->auth_retry); + if (con->auth_retry == 2) { + con->error_msg = "connect authorization failure"; + reset_connection(con); + set_bit(CLOSED, &con->state); + return -1; + } + con->auth_retry = 1; + prepare_write_connect(con->msgr, con, 0); + prepare_read_connect(con); + break; + + case CEPH_MSGR_TAG_RESETSESSION: + /* + * If we connected with a large connect_seq but the peer + * has no record of a session with us (no connection, or + * connect_seq == 0), they will send RESETSESION to indicate + * that they must have reset their session, and may have + * dropped messages. + */ + dout("process_connect got RESET peer seq %u\n", + le32_to_cpu(con->in_connect.connect_seq)); + pr_err("%s%lld %s connection reset\n", + ENTITY_NAME(con->peer_name), + ceph_pr_addr(&con->peer_addr.in_addr)); + reset_connection(con); + prepare_write_connect(con->msgr, con, 0); + prepare_read_connect(con); + + /* Tell ceph about it. */ + mutex_unlock(&con->mutex); + pr_info("reset on %s%lld\n", ENTITY_NAME(con->peer_name)); + if (con->ops->peer_reset) + con->ops->peer_reset(con); + mutex_lock(&con->mutex); + break; + + case CEPH_MSGR_TAG_RETRY_SESSION: + /* + * If we sent a smaller connect_seq than the peer has, try + * again with a larger value. + */ + dout("process_connect got RETRY my seq = %u, peer_seq = %u\n", + le32_to_cpu(con->out_connect.connect_seq), + le32_to_cpu(con->in_connect.connect_seq)); + con->connect_seq = le32_to_cpu(con->in_connect.connect_seq); + prepare_write_connect(con->msgr, con, 0); + prepare_read_connect(con); + break; + + case CEPH_MSGR_TAG_RETRY_GLOBAL: + /* + * If we sent a smaller global_seq than the peer has, try + * again with a larger value. + */ + dout("process_connect got RETRY_GLOBAL my %u peer_gseq %u\n", + con->peer_global_seq, + le32_to_cpu(con->in_connect.global_seq)); + get_global_seq(con->msgr, + le32_to_cpu(con->in_connect.global_seq)); + prepare_write_connect(con->msgr, con, 0); + prepare_read_connect(con); + break; + + case CEPH_MSGR_TAG_READY: + if (req_feat & ~server_feat) { + pr_err("%s%lld %s protocol feature mismatch," + " my required %llx > server's %llx, need %llx\n", + ENTITY_NAME(con->peer_name), + ceph_pr_addr(&con->peer_addr.in_addr), + req_feat, server_feat, req_feat & ~server_feat); + con->error_msg = "missing required protocol features"; + fail_protocol(con); + return -1; + } + clear_bit(CONNECTING, &con->state); + con->peer_global_seq = le32_to_cpu(con->in_reply.global_seq); + con->connect_seq++; + con->peer_features = server_feat; + dout("process_connect got READY gseq %d cseq %d (%d)\n", + con->peer_global_seq, + le32_to_cpu(con->in_reply.connect_seq), + con->connect_seq); + WARN_ON(con->connect_seq != + le32_to_cpu(con->in_reply.connect_seq)); + + if (con->in_reply.flags & CEPH_MSG_CONNECT_LOSSY) + set_bit(LOSSYTX, &con->state); + + prepare_read_tag(con); + break; + + case CEPH_MSGR_TAG_WAIT: + /* + * If there is a connection race (we are opening + * connections to each other), one of us may just have + * to WAIT. This shouldn't happen if we are the + * client. + */ + pr_err("process_connect peer connecting WAIT\n"); + + default: + pr_err("connect protocol error, will retry\n"); + con->error_msg = "protocol error, garbage tag during connect"; + return -1; + } + return 0; +} + + +/* + * read (part of) an ack + */ +static int read_partial_ack(struct ceph_connection *con) +{ + int to = 0; + + return read_partial(con, &to, sizeof(con->in_temp_ack), + &con->in_temp_ack); +} + + +/* + * We can finally discard anything that's been acked. + */ +static void process_ack(struct ceph_connection *con) +{ + struct ceph_msg *m; + u64 ack = le64_to_cpu(con->in_temp_ack); + u64 seq; + + while (!list_empty(&con->out_sent)) { + m = list_first_entry(&con->out_sent, struct ceph_msg, + list_head); + seq = le64_to_cpu(m->hdr.seq); + if (seq > ack) + break; + dout("got ack for seq %llu type %d at %p\n", seq, + le16_to_cpu(m->hdr.type), m); + ceph_msg_remove(m); + } + prepare_read_tag(con); +} + + + + +static int read_partial_message_section(struct ceph_connection *con, + struct kvec *section, + unsigned int sec_len, u32 *crc) +{ + int ret, left; + + BUG_ON(!section); + + while (section->iov_len < sec_len) { + BUG_ON(section->iov_base == NULL); + left = sec_len - section->iov_len; + ret = ceph_tcp_recvmsg(con->sock, (char *)section->iov_base + + section->iov_len, left); + if (ret <= 0) + return ret; + section->iov_len += ret; + if (section->iov_len == sec_len) + *crc = crc32c(0, section->iov_base, + section->iov_len); + } + + return 1; +} + +static struct ceph_msg *ceph_alloc_msg(struct ceph_connection *con, + struct ceph_msg_header *hdr, + int *skip); + + +static int read_partial_message_pages(struct ceph_connection *con, + struct page **pages, + unsigned data_len, int datacrc) +{ + void *p; + int ret; + int left; + + left = min((int)(data_len - con->in_msg_pos.data_pos), + (int)(PAGE_SIZE - con->in_msg_pos.page_pos)); + /* (page) data */ + BUG_ON(pages == NULL); + p = kmap(pages[con->in_msg_pos.page]); + ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos, + left); + if (ret > 0 && datacrc) + con->in_data_crc = + crc32c(con->in_data_crc, + p + con->in_msg_pos.page_pos, ret); + kunmap(pages[con->in_msg_pos.page]); + if (ret <= 0) + return ret; + con->in_msg_pos.data_pos += ret; + con->in_msg_pos.page_pos += ret; + if (con->in_msg_pos.page_pos == PAGE_SIZE) { + con->in_msg_pos.page_pos = 0; + con->in_msg_pos.page++; + } + + return ret; +} + +#ifdef CONFIG_BLOCK +static int read_partial_message_bio(struct ceph_connection *con, + struct bio **bio_iter, int *bio_seg, + unsigned data_len, int datacrc) +{ + struct bio_vec *bv = bio_iovec_idx(*bio_iter, *bio_seg); + void *p; + int ret, left; + + if (IS_ERR(bv)) + return PTR_ERR(bv); + + left = min((int)(data_len - con->in_msg_pos.data_pos), + (int)(bv->bv_len - con->in_msg_pos.page_pos)); + + p = kmap(bv->bv_page) + bv->bv_offset; + + ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos, + left); + if (ret > 0 && datacrc) + con->in_data_crc = + crc32c(con->in_data_crc, + p + con->in_msg_pos.page_pos, ret); + kunmap(bv->bv_page); + if (ret <= 0) + return ret; + con->in_msg_pos.data_pos += ret; + con->in_msg_pos.page_pos += ret; + if (con->in_msg_pos.page_pos == bv->bv_len) { + con->in_msg_pos.page_pos = 0; + iter_bio_next(bio_iter, bio_seg); + } + + return ret; +} +#endif + +/* + * read (part of) a message. + */ +static int read_partial_message(struct ceph_connection *con) +{ + struct ceph_msg *m = con->in_msg; + int ret; + int to, left; + unsigned front_len, middle_len, data_len, data_off; + int datacrc = con->msgr->nocrc; + int skip; + u64 seq; + + dout("read_partial_message con %p msg %p\n", con, m); + + /* header */ + while (con->in_base_pos < sizeof(con->in_hdr)) { + left = sizeof(con->in_hdr) - con->in_base_pos; + ret = ceph_tcp_recvmsg(con->sock, + (char *)&con->in_hdr + con->in_base_pos, + left); + if (ret <= 0) + return ret; + con->in_base_pos += ret; + if (con->in_base_pos == sizeof(con->in_hdr)) { + u32 crc = crc32c(0, (void *)&con->in_hdr, + sizeof(con->in_hdr) - sizeof(con->in_hdr.crc)); + if (crc != le32_to_cpu(con->in_hdr.crc)) { + pr_err("read_partial_message bad hdr " + " crc %u != expected %u\n", + crc, con->in_hdr.crc); + return -EBADMSG; + } + } + } + front_len = le32_to_cpu(con->in_hdr.front_len); + if (front_len > CEPH_MSG_MAX_FRONT_LEN) + return -EIO; + middle_len = le32_to_cpu(con->in_hdr.middle_len); + if (middle_len > CEPH_MSG_MAX_DATA_LEN) + return -EIO; + data_len = le32_to_cpu(con->in_hdr.data_len); + if (data_len > CEPH_MSG_MAX_DATA_LEN) + return -EIO; + data_off = le16_to_cpu(con->in_hdr.data_off); + + /* verify seq# */ + seq = le64_to_cpu(con->in_hdr.seq); + if ((s64)seq - (s64)con->in_seq < 1) { + pr_info("skipping %s%lld %s seq %lld, expected %lld\n", + ENTITY_NAME(con->peer_name), + ceph_pr_addr(&con->peer_addr.in_addr), + seq, con->in_seq + 1); + con->in_base_pos = -front_len - middle_len - data_len - + sizeof(m->footer); + con->in_tag = CEPH_MSGR_TAG_READY; + con->in_seq++; + return 0; + } else if ((s64)seq - (s64)con->in_seq > 1) { + pr_err("read_partial_message bad seq %lld expected %lld\n", + seq, con->in_seq + 1); + con->error_msg = "bad message sequence # for incoming message"; + return -EBADMSG; + } + + /* allocate message? */ + if (!con->in_msg) { + dout("got hdr type %d front %d data %d\n", con->in_hdr.type, + con->in_hdr.front_len, con->in_hdr.data_len); + skip = 0; + con->in_msg = ceph_alloc_msg(con, &con->in_hdr, &skip); + if (skip) { + /* skip this message */ + dout("alloc_msg said skip message\n"); + BUG_ON(con->in_msg); + con->in_base_pos = -front_len - middle_len - data_len - + sizeof(m->footer); + con->in_tag = CEPH_MSGR_TAG_READY; + con->in_seq++; + return 0; + } + if (!con->in_msg) { + con->error_msg = + "error allocating memory for incoming message"; + return -ENOMEM; + } + m = con->in_msg; + m->front.iov_len = 0; /* haven't read it yet */ + if (m->middle) + m->middle->vec.iov_len = 0; + + con->in_msg_pos.page = 0; + if (m->pages) + con->in_msg_pos.page_pos = data_off & ~PAGE_MASK; + else + con->in_msg_pos.page_pos = 0; + con->in_msg_pos.data_pos = 0; + } + + /* front */ + ret = read_partial_message_section(con, &m->front, front_len, + &con->in_front_crc); + if (ret <= 0) + return ret; + + /* middle */ + if (m->middle) { + ret = read_partial_message_section(con, &m->middle->vec, + middle_len, + &con->in_middle_crc); + if (ret <= 0) + return ret; + } +#ifdef CONFIG_BLOCK + if (m->bio && !m->bio_iter) + init_bio_iter(m->bio, &m->bio_iter, &m->bio_seg); +#endif + + /* (page) data */ + while (con->in_msg_pos.data_pos < data_len) { + if (m->pages) { + ret = read_partial_message_pages(con, m->pages, + data_len, datacrc); + if (ret <= 0) + return ret; +#ifdef CONFIG_BLOCK + } else if (m->bio) { + + ret = read_partial_message_bio(con, + &m->bio_iter, &m->bio_seg, + data_len, datacrc); + if (ret <= 0) + return ret; +#endif + } else { + BUG_ON(1); + } + } + + /* footer */ + to = sizeof(m->hdr) + sizeof(m->footer); + while (con->in_base_pos < to) { + left = to - con->in_base_pos; + ret = ceph_tcp_recvmsg(con->sock, (char *)&m->footer + + (con->in_base_pos - sizeof(m->hdr)), + left); + if (ret <= 0) + return ret; + con->in_base_pos += ret; + } + dout("read_partial_message got msg %p %d (%u) + %d (%u) + %d (%u)\n", + m, front_len, m->footer.front_crc, middle_len, + m->footer.middle_crc, data_len, m->footer.data_crc); + + /* crc ok? */ + if (con->in_front_crc != le32_to_cpu(m->footer.front_crc)) { + pr_err("read_partial_message %p front crc %u != exp. %u\n", + m, con->in_front_crc, m->footer.front_crc); + return -EBADMSG; + } + if (con->in_middle_crc != le32_to_cpu(m->footer.middle_crc)) { + pr_err("read_partial_message %p middle crc %u != exp %u\n", + m, con->in_middle_crc, m->footer.middle_crc); + return -EBADMSG; + } + if (datacrc && + (m->footer.flags & CEPH_MSG_FOOTER_NOCRC) == 0 && + con->in_data_crc != le32_to_cpu(m->footer.data_crc)) { + pr_err("read_partial_message %p data crc %u != exp. %u\n", m, + con->in_data_crc, le32_to_cpu(m->footer.data_crc)); + return -EBADMSG; + } + + return 1; /* done! */ +} + +/* + * Process message. This happens in the worker thread. The callback should + * be careful not to do anything that waits on other incoming messages or it + * may deadlock. + */ +static void process_message(struct ceph_connection *con) +{ + struct ceph_msg *msg; + + msg = con->in_msg; + con->in_msg = NULL; + + /* if first message, set peer_name */ + if (con->peer_name.type == 0) + con->peer_name = msg->hdr.src; + + con->in_seq++; + mutex_unlock(&con->mutex); + + dout("===== %p %llu from %s%lld %d=%s len %d+%d (%u %u %u) =====\n", + msg, le64_to_cpu(msg->hdr.seq), + ENTITY_NAME(msg->hdr.src), + le16_to_cpu(msg->hdr.type), + ceph_msg_type_name(le16_to_cpu(msg->hdr.type)), + le32_to_cpu(msg->hdr.front_len), + le32_to_cpu(msg->hdr.data_len), + con->in_front_crc, con->in_middle_crc, con->in_data_crc); + con->ops->dispatch(con, msg); + + mutex_lock(&con->mutex); + prepare_read_tag(con); +} + + +/* + * Write something to the socket. Called in a worker thread when the + * socket appears to be writeable and we have something ready to send. + */ +static int try_write(struct ceph_connection *con) +{ + struct ceph_messenger *msgr = con->msgr; + int ret = 1; + + dout("try_write start %p state %lu nref %d\n", con, con->state, + atomic_read(&con->nref)); + +more: + dout("try_write out_kvec_bytes %d\n", con->out_kvec_bytes); + + /* open the socket first? */ + if (con->sock == NULL) { + /* + * if we were STANDBY and are reconnecting _this_ + * connection, bump connect_seq now. Always bump + * global_seq. + */ + if (test_and_clear_bit(STANDBY, &con->state)) + con->connect_seq++; + + prepare_write_banner(msgr, con); + prepare_write_connect(msgr, con, 1); + prepare_read_banner(con); + set_bit(CONNECTING, &con->state); + clear_bit(NEGOTIATING, &con->state); + + BUG_ON(con->in_msg); + con->in_tag = CEPH_MSGR_TAG_READY; + dout("try_write initiating connect on %p new state %lu\n", + con, con->state); + con->sock = ceph_tcp_connect(con); + if (IS_ERR(con->sock)) { + con->sock = NULL; + con->error_msg = "connect error"; + ret = -1; + goto out; + } + } + +more_kvec: + /* kvec data queued? */ + if (con->out_skip) { + ret = write_partial_skip(con); + if (ret <= 0) + goto done; + if (ret < 0) { + dout("try_write write_partial_skip err %d\n", ret); + goto done; + } + } + if (con->out_kvec_left) { + ret = write_partial_kvec(con); + if (ret <= 0) + goto done; + } + + /* msg pages? */ + if (con->out_msg) { + if (con->out_msg_done) { + ceph_msg_put(con->out_msg); + con->out_msg = NULL; /* we're done with this one */ + goto do_next; + } + + ret = write_partial_msg_pages(con); + if (ret == 1) + goto more_kvec; /* we need to send the footer, too! */ + if (ret == 0) + goto done; + if (ret < 0) { + dout("try_write write_partial_msg_pages err %d\n", + ret); + goto done; + } + } + +do_next: + if (!test_bit(CONNECTING, &con->state)) { + /* is anything else pending? */ + if (!list_empty(&con->out_queue)) { + prepare_write_message(con); + goto more; + } + if (con->in_seq > con->in_seq_acked) { + prepare_write_ack(con); + goto more; + } + if (test_and_clear_bit(KEEPALIVE_PENDING, &con->state)) { + prepare_write_keepalive(con); + goto more; + } + } + + /* Nothing to do! */ + clear_bit(WRITE_PENDING, &con->state); + dout("try_write nothing else to write.\n"); +done: + ret = 0; +out: + dout("try_write done on %p\n", con); + return ret; +} + + + +/* + * Read what we can from the socket. + */ +static int try_read(struct ceph_connection *con) +{ + int ret = -1; + + if (!con->sock) + return 0; + + if (test_bit(STANDBY, &con->state)) + return 0; + + dout("try_read start on %p\n", con); + +more: + dout("try_read tag %d in_base_pos %d\n", (int)con->in_tag, + con->in_base_pos); + if (test_bit(CONNECTING, &con->state)) { + if (!test_bit(NEGOTIATING, &con->state)) { + dout("try_read connecting\n"); + ret = read_partial_banner(con); + if (ret <= 0) + goto done; + if (process_banner(con) < 0) { + ret = -1; + goto out; + } + } + ret = read_partial_connect(con); + if (ret <= 0) + goto done; + if (process_connect(con) < 0) { + ret = -1; + goto out; + } + goto more; + } + + if (con->in_base_pos < 0) { + /* + * skipping + discarding content. + * + * FIXME: there must be a better way to do this! + */ + static char buf[1024]; + int skip = min(1024, -con->in_base_pos); + dout("skipping %d / %d bytes\n", skip, -con->in_base_pos); + ret = ceph_tcp_recvmsg(con->sock, buf, skip); + if (ret <= 0) + goto done; + con->in_base_pos += ret; + if (con->in_base_pos) + goto more; + } + if (con->in_tag == CEPH_MSGR_TAG_READY) { + /* + * what's next? + */ + ret = ceph_tcp_recvmsg(con->sock, &con->in_tag, 1); + if (ret <= 0) + goto done; + dout("try_read got tag %d\n", (int)con->in_tag); + switch (con->in_tag) { + case CEPH_MSGR_TAG_MSG: + prepare_read_message(con); + break; + case CEPH_MSGR_TAG_ACK: + prepare_read_ack(con); + break; + case CEPH_MSGR_TAG_CLOSE: + set_bit(CLOSED, &con->state); /* fixme */ + goto done; + default: + goto bad_tag; + } + } + if (con->in_tag == CEPH_MSGR_TAG_MSG) { + ret = read_partial_message(con); + if (ret <= 0) { + switch (ret) { + case -EBADMSG: + con->error_msg = "bad crc"; + ret = -EIO; + goto out; + case -EIO: + con->error_msg = "io error"; + goto out; + default: + goto done; + } + } + if (con->in_tag == CEPH_MSGR_TAG_READY) + goto more; + process_message(con); + goto more; + } + if (con->in_tag == CEPH_MSGR_TAG_ACK) { + ret = read_partial_ack(con); + if (ret <= 0) + goto done; + process_ack(con); + goto more; + } + +done: + ret = 0; +out: + dout("try_read done on %p\n", con); + return ret; + +bad_tag: + pr_err("try_read bad con->in_tag = %d\n", (int)con->in_tag); + con->error_msg = "protocol error, garbage tag"; + ret = -1; + goto out; +} + + +/* + * Atomically queue work on a connection. Bump @con reference to + * avoid races with connection teardown. + * + * There is some trickery going on with QUEUED and BUSY because we + * only want a _single_ thread operating on each connection at any + * point in time, but we want to use all available CPUs. + * + * The worker thread only proceeds if it can atomically set BUSY. It + * clears QUEUED and does it's thing. When it thinks it's done, it + * clears BUSY, then rechecks QUEUED.. if it's set again, it loops + * (tries again to set BUSY). + * + * To queue work, we first set QUEUED, _then_ if BUSY isn't set, we + * try to queue work. If that fails (work is already queued, or BUSY) + * we give up (work also already being done or is queued) but leave QUEUED + * set so that the worker thread will loop if necessary. + */ +static void queue_con(struct ceph_connection *con) +{ + if (test_bit(DEAD, &con->state)) { + dout("queue_con %p ignoring: DEAD\n", + con); + return; + } + + if (!con->ops->get(con)) { + dout("queue_con %p ref count 0\n", con); + return; + } + + set_bit(QUEUED, &con->state); + if (test_bit(BUSY, &con->state)) { + dout("queue_con %p - already BUSY\n", con); + con->ops->put(con); + } else if (!queue_work(ceph_msgr_wq, &con->work.work)) { + dout("queue_con %p - already queued\n", con); + con->ops->put(con); + } else { + dout("queue_con %p\n", con); + } +} + +/* + * Do some work on a connection. Drop a connection ref when we're done. + */ +static void con_work(struct work_struct *work) +{ + struct ceph_connection *con = container_of(work, struct ceph_connection, + work.work); + int backoff = 0; + +more: + if (test_and_set_bit(BUSY, &con->state) != 0) { + dout("con_work %p BUSY already set\n", con); + goto out; + } + dout("con_work %p start, clearing QUEUED\n", con); + clear_bit(QUEUED, &con->state); + + mutex_lock(&con->mutex); + + if (test_bit(CLOSED, &con->state)) { /* e.g. if we are replaced */ + dout("con_work CLOSED\n"); + con_close_socket(con); + goto done; + } + if (test_and_clear_bit(OPENING, &con->state)) { + /* reopen w/ new peer */ + dout("con_work OPENING\n"); + con_close_socket(con); + } + + if (test_and_clear_bit(SOCK_CLOSED, &con->state) || + try_read(con) < 0 || + try_write(con) < 0) { + mutex_unlock(&con->mutex); + backoff = 1; + ceph_fault(con); /* error/fault path */ + goto done_unlocked; + } + +done: + mutex_unlock(&con->mutex); + +done_unlocked: + clear_bit(BUSY, &con->state); + dout("con->state=%lu\n", con->state); + if (test_bit(QUEUED, &con->state)) { + if (!backoff || test_bit(OPENING, &con->state)) { + dout("con_work %p QUEUED reset, looping\n", con); + goto more; + } + dout("con_work %p QUEUED reset, but just faulted\n", con); + clear_bit(QUEUED, &con->state); + } + dout("con_work %p done\n", con); + +out: + con->ops->put(con); +} + + +/* + * Generic error/fault handler. A retry mechanism is used with + * exponential backoff + */ +static void ceph_fault(struct ceph_connection *con) +{ + pr_err("%s%lld %s %s\n", ENTITY_NAME(con->peer_name), + ceph_pr_addr(&con->peer_addr.in_addr), con->error_msg); + dout("fault %p state %lu to peer %s\n", + con, con->state, ceph_pr_addr(&con->peer_addr.in_addr)); + + if (test_bit(LOSSYTX, &con->state)) { + dout("fault on LOSSYTX channel\n"); + goto out; + } + + mutex_lock(&con->mutex); + if (test_bit(CLOSED, &con->state)) + goto out_unlock; + + con_close_socket(con); + + if (con->in_msg) { + ceph_msg_put(con->in_msg); + con->in_msg = NULL; + } + + /* Requeue anything that hasn't been acked */ + list_splice_init(&con->out_sent, &con->out_queue); + + /* If there are no messages in the queue, place the connection + * in a STANDBY state (i.e., don't try to reconnect just yet). */ + if (list_empty(&con->out_queue) && !con->out_keepalive_pending) { + dout("fault setting STANDBY\n"); + set_bit(STANDBY, &con->state); + } else { + /* retry after a delay. */ + if (con->delay == 0) + con->delay = BASE_DELAY_INTERVAL; + else if (con->delay < MAX_DELAY_INTERVAL) + con->delay *= 2; + dout("fault queueing %p delay %lu\n", con, con->delay); + con->ops->get(con); + if (queue_delayed_work(ceph_msgr_wq, &con->work, + round_jiffies_relative(con->delay)) == 0) + con->ops->put(con); + } + +out_unlock: + mutex_unlock(&con->mutex); +out: + /* + * in case we faulted due to authentication, invalidate our + * current tickets so that we can get new ones. + */ + if (con->auth_retry && con->ops->invalidate_authorizer) { + dout("calling invalidate_authorizer()\n"); + con->ops->invalidate_authorizer(con); + } + + if (con->ops->fault) + con->ops->fault(con); +} + + + +/* + * create a new messenger instance + */ +struct ceph_messenger *ceph_messenger_create(struct ceph_entity_addr *myaddr, + u32 supported_features, + u32 required_features) +{ + struct ceph_messenger *msgr; + + msgr = kzalloc(sizeof(*msgr), GFP_KERNEL); + if (msgr == NULL) + return ERR_PTR(-ENOMEM); + + msgr->supported_features = supported_features; + msgr->required_features = required_features; + + spin_lock_init(&msgr->global_seq_lock); + + /* the zero page is needed if a request is "canceled" while the message + * is being written over the socket */ + msgr->zero_page = __page_cache_alloc(GFP_KERNEL | __GFP_ZERO); + if (!msgr->zero_page) { + kfree(msgr); + return ERR_PTR(-ENOMEM); + } + kmap(msgr->zero_page); + + if (myaddr) + msgr->inst.addr = *myaddr; + + /* select a random nonce */ + msgr->inst.addr.type = 0; + get_random_bytes(&msgr->inst.addr.nonce, sizeof(msgr->inst.addr.nonce)); + encode_my_addr(msgr); + + dout("messenger_create %p\n", msgr); + return msgr; +} +EXPORT_SYMBOL(ceph_messenger_create); + +void ceph_messenger_destroy(struct ceph_messenger *msgr) +{ + dout("destroy %p\n", msgr); + kunmap(msgr->zero_page); + __free_page(msgr->zero_page); + kfree(msgr); + dout("destroyed messenger %p\n", msgr); +} +EXPORT_SYMBOL(ceph_messenger_destroy); + +/* + * Queue up an outgoing message on the given connection. + */ +void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg) +{ + if (test_bit(CLOSED, &con->state)) { + dout("con_send %p closed, dropping %p\n", con, msg); + ceph_msg_put(msg); + return; + } + + /* set src+dst */ + msg->hdr.src = con->msgr->inst.name; + + BUG_ON(msg->front.iov_len != le32_to_cpu(msg->hdr.front_len)); + + msg->needs_out_seq = true; + + /* queue */ + mutex_lock(&con->mutex); + BUG_ON(!list_empty(&msg->list_head)); + list_add_tail(&msg->list_head, &con->out_queue); + dout("----- %p to %s%lld %d=%s len %d+%d+%d -----\n", msg, + ENTITY_NAME(con->peer_name), le16_to_cpu(msg->hdr.type), + ceph_msg_type_name(le16_to_cpu(msg->hdr.type)), + le32_to_cpu(msg->hdr.front_len), + le32_to_cpu(msg->hdr.middle_len), + le32_to_cpu(msg->hdr.data_len)); + mutex_unlock(&con->mutex); + + /* if there wasn't anything waiting to send before, queue + * new work */ + if (test_and_set_bit(WRITE_PENDING, &con->state) == 0) + queue_con(con); +} +EXPORT_SYMBOL(ceph_con_send); + +/* + * Revoke a message that was previously queued for send + */ +void ceph_con_revoke(struct ceph_connection *con, struct ceph_msg *msg) +{ + mutex_lock(&con->mutex); + if (!list_empty(&msg->list_head)) { + dout("con_revoke %p msg %p - was on queue\n", con, msg); + list_del_init(&msg->list_head); + ceph_msg_put(msg); + msg->hdr.seq = 0; + } + if (con->out_msg == msg) { + dout("con_revoke %p msg %p - was sending\n", con, msg); + con->out_msg = NULL; + if (con->out_kvec_is_msg) { + con->out_skip = con->out_kvec_bytes; + con->out_kvec_is_msg = false; + } + ceph_msg_put(msg); + msg->hdr.seq = 0; + } + mutex_unlock(&con->mutex); +} + +/* + * Revoke a message that we may be reading data into + */ +void ceph_con_revoke_message(struct ceph_connection *con, struct ceph_msg *msg) +{ + mutex_lock(&con->mutex); + if (con->in_msg && con->in_msg == msg) { + unsigned front_len = le32_to_cpu(con->in_hdr.front_len); + unsigned middle_len = le32_to_cpu(con->in_hdr.middle_len); + unsigned data_len = le32_to_cpu(con->in_hdr.data_len); + + /* skip rest of message */ + dout("con_revoke_pages %p msg %p revoked\n", con, msg); + con->in_base_pos = con->in_base_pos - + sizeof(struct ceph_msg_header) - + front_len - + middle_len - + data_len - + sizeof(struct ceph_msg_footer); + ceph_msg_put(con->in_msg); + con->in_msg = NULL; + con->in_tag = CEPH_MSGR_TAG_READY; + con->in_seq++; + } else { + dout("con_revoke_pages %p msg %p pages %p no-op\n", + con, con->in_msg, msg); + } + mutex_unlock(&con->mutex); +} + +/* + * Queue a keepalive byte to ensure the tcp connection is alive. + */ +void ceph_con_keepalive(struct ceph_connection *con) +{ + if (test_and_set_bit(KEEPALIVE_PENDING, &con->state) == 0 && + test_and_set_bit(WRITE_PENDING, &con->state) == 0) + queue_con(con); +} +EXPORT_SYMBOL(ceph_con_keepalive); + + +/* + * construct a new message with given type, size + * the new msg has a ref count of 1. + */ +struct ceph_msg *ceph_msg_new(int type, int front_len, gfp_t flags) +{ + struct ceph_msg *m; + + m = kmalloc(sizeof(*m), flags); + if (m == NULL) + goto out; + kref_init(&m->kref); + INIT_LIST_HEAD(&m->list_head); + + m->hdr.tid = 0; + m->hdr.type = cpu_to_le16(type); + m->hdr.priority = cpu_to_le16(CEPH_MSG_PRIO_DEFAULT); + m->hdr.version = 0; + m->hdr.front_len = cpu_to_le32(front_len); + m->hdr.middle_len = 0; + m->hdr.data_len = 0; + m->hdr.data_off = 0; + m->hdr.reserved = 0; + m->footer.front_crc = 0; + m->footer.middle_crc = 0; + m->footer.data_crc = 0; + m->footer.flags = 0; + m->front_max = front_len; + m->front_is_vmalloc = false; + m->more_to_follow = false; + m->pool = NULL; + + /* front */ + if (front_len) { + if (front_len > PAGE_CACHE_SIZE) { + m->front.iov_base = __vmalloc(front_len, flags, + PAGE_KERNEL); + m->front_is_vmalloc = true; + } else { + m->front.iov_base = kmalloc(front_len, flags); + } + if (m->front.iov_base == NULL) { + pr_err("msg_new can't allocate %d bytes\n", + front_len); + goto out2; + } + } else { + m->front.iov_base = NULL; + } + m->front.iov_len = front_len; + + /* middle */ + m->middle = NULL; + + /* data */ + m->nr_pages = 0; + m->pages = NULL; + m->pagelist = NULL; + m->bio = NULL; + m->bio_iter = NULL; + m->bio_seg = 0; + m->trail = NULL; + + dout("ceph_msg_new %p front %d\n", m, front_len); + return m; + +out2: + ceph_msg_put(m); +out: + pr_err("msg_new can't create type %d front %d\n", type, front_len); + return NULL; +} +EXPORT_SYMBOL(ceph_msg_new); + +/* + * Allocate "middle" portion of a message, if it is needed and wasn't + * allocated by alloc_msg. This allows us to read a small fixed-size + * per-type header in the front and then gracefully fail (i.e., + * propagate the error to the caller based on info in the front) when + * the middle is too large. + */ +static int ceph_alloc_middle(struct ceph_connection *con, struct ceph_msg *msg) +{ + int type = le16_to_cpu(msg->hdr.type); + int middle_len = le32_to_cpu(msg->hdr.middle_len); + + dout("alloc_middle %p type %d %s middle_len %d\n", msg, type, + ceph_msg_type_name(type), middle_len); + BUG_ON(!middle_len); + BUG_ON(msg->middle); + + msg->middle = ceph_buffer_new(middle_len, GFP_NOFS); + if (!msg->middle) + return -ENOMEM; + return 0; +} + +/* + * Generic message allocator, for incoming messages. + */ +static struct ceph_msg *ceph_alloc_msg(struct ceph_connection *con, + struct ceph_msg_header *hdr, + int *skip) +{ + int type = le16_to_cpu(hdr->type); + int front_len = le32_to_cpu(hdr->front_len); + int middle_len = le32_to_cpu(hdr->middle_len); + struct ceph_msg *msg = NULL; + int ret; + + if (con->ops->alloc_msg) { + mutex_unlock(&con->mutex); + msg = con->ops->alloc_msg(con, hdr, skip); + mutex_lock(&con->mutex); + if (!msg || *skip) + return NULL; + } + if (!msg) { + *skip = 0; + msg = ceph_msg_new(type, front_len, GFP_NOFS); + if (!msg) { + pr_err("unable to allocate msg type %d len %d\n", + type, front_len); + return NULL; + } + } + memcpy(&msg->hdr, &con->in_hdr, sizeof(con->in_hdr)); + + if (middle_len && !msg->middle) { + ret = ceph_alloc_middle(con, msg); + if (ret < 0) { + ceph_msg_put(msg); + return NULL; + } + } + + return msg; +} + + +/* + * Free a generically kmalloc'd message. + */ +void ceph_msg_kfree(struct ceph_msg *m) +{ + dout("msg_kfree %p\n", m); + if (m->front_is_vmalloc) + vfree(m->front.iov_base); + else + kfree(m->front.iov_base); + kfree(m); +} + +/* + * Drop a msg ref. Destroy as needed. + */ +void ceph_msg_last_put(struct kref *kref) +{ + struct ceph_msg *m = container_of(kref, struct ceph_msg, kref); + + dout("ceph_msg_put last one on %p\n", m); + WARN_ON(!list_empty(&m->list_head)); + + /* drop middle, data, if any */ + if (m->middle) { + ceph_buffer_put(m->middle); + m->middle = NULL; + } + m->nr_pages = 0; + m->pages = NULL; + + if (m->pagelist) { + ceph_pagelist_release(m->pagelist); + kfree(m->pagelist); + m->pagelist = NULL; + } + + m->trail = NULL; + + if (m->pool) + ceph_msgpool_put(m->pool, m); + else + ceph_msg_kfree(m); +} +EXPORT_SYMBOL(ceph_msg_last_put); + +void ceph_msg_dump(struct ceph_msg *msg) +{ + pr_debug("msg_dump %p (front_max %d nr_pages %d)\n", msg, + msg->front_max, msg->nr_pages); + print_hex_dump(KERN_DEBUG, "header: ", + DUMP_PREFIX_OFFSET, 16, 1, + &msg->hdr, sizeof(msg->hdr), true); + print_hex_dump(KERN_DEBUG, " front: ", + DUMP_PREFIX_OFFSET, 16, 1, + msg->front.iov_base, msg->front.iov_len, true); + if (msg->middle) + print_hex_dump(KERN_DEBUG, "middle: ", + DUMP_PREFIX_OFFSET, 16, 1, + msg->middle->vec.iov_base, + msg->middle->vec.iov_len, true); + print_hex_dump(KERN_DEBUG, "footer: ", + DUMP_PREFIX_OFFSET, 16, 1, + &msg->footer, sizeof(msg->footer), true); +} +EXPORT_SYMBOL(ceph_msg_dump); diff --git a/net/ceph/mon_client.c b/net/ceph/mon_client.c new file mode 100644 index 00000000000..8a079399174 --- /dev/null +++ b/net/ceph/mon_client.c @@ -0,0 +1,1027 @@ +#include <linux/ceph/ceph_debug.h> + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/slab.h> +#include <linux/random.h> +#include <linux/sched.h> + +#include <linux/ceph/mon_client.h> +#include <linux/ceph/libceph.h> +#include <linux/ceph/decode.h> + +#include <linux/ceph/auth.h> + +/* + * Interact with Ceph monitor cluster. Handle requests for new map + * versions, and periodically resend as needed. Also implement + * statfs() and umount(). + * + * A small cluster of Ceph "monitors" are responsible for managing critical + * cluster configuration and state information. An odd number (e.g., 3, 5) + * of cmon daemons use a modified version of the Paxos part-time parliament + * algorithm to manage the MDS map (mds cluster membership), OSD map, and + * list of clients who have mounted the file system. + * + * We maintain an open, active session with a monitor at all times in order to + * receive timely MDSMap updates. We periodically send a keepalive byte on the + * TCP socket to ensure we detect a failure. If the connection does break, we + * randomly hunt for a new monitor. Once the connection is reestablished, we + * resend any outstanding requests. + */ + +static const struct ceph_connection_operations mon_con_ops; + +static int __validate_auth(struct ceph_mon_client *monc); + +/* + * Decode a monmap blob (e.g., during mount). + */ +struct ceph_monmap *ceph_monmap_decode(void *p, void *end) +{ + struct ceph_monmap *m = NULL; + int i, err = -EINVAL; + struct ceph_fsid fsid; + u32 epoch, num_mon; + u16 version; + u32 len; + + ceph_decode_32_safe(&p, end, len, bad); + ceph_decode_need(&p, end, len, bad); + + dout("monmap_decode %p %p len %d\n", p, end, (int)(end-p)); + + ceph_decode_16_safe(&p, end, version, bad); + + ceph_decode_need(&p, end, sizeof(fsid) + 2*sizeof(u32), bad); + ceph_decode_copy(&p, &fsid, sizeof(fsid)); + epoch = ceph_decode_32(&p); + + num_mon = ceph_decode_32(&p); + ceph_decode_need(&p, end, num_mon*sizeof(m->mon_inst[0]), bad); + + if (num_mon >= CEPH_MAX_MON) + goto bad; + m = kmalloc(sizeof(*m) + sizeof(m->mon_inst[0])*num_mon, GFP_NOFS); + if (m == NULL) + return ERR_PTR(-ENOMEM); + m->fsid = fsid; + m->epoch = epoch; + m->num_mon = num_mon; + ceph_decode_copy(&p, m->mon_inst, num_mon*sizeof(m->mon_inst[0])); + for (i = 0; i < num_mon; i++) + ceph_decode_addr(&m->mon_inst[i].addr); + + dout("monmap_decode epoch %d, num_mon %d\n", m->epoch, + m->num_mon); + for (i = 0; i < m->num_mon; i++) + dout("monmap_decode mon%d is %s\n", i, + ceph_pr_addr(&m->mon_inst[i].addr.in_addr)); + return m; + +bad: + dout("monmap_decode failed with %d\n", err); + kfree(m); + return ERR_PTR(err); +} + +/* + * return true if *addr is included in the monmap. + */ +int ceph_monmap_contains(struct ceph_monmap *m, struct ceph_entity_addr *addr) +{ + int i; + + for (i = 0; i < m->num_mon; i++) + if (memcmp(addr, &m->mon_inst[i].addr, sizeof(*addr)) == 0) + return 1; + return 0; +} + +/* + * Send an auth request. + */ +static void __send_prepared_auth_request(struct ceph_mon_client *monc, int len) +{ + monc->pending_auth = 1; + monc->m_auth->front.iov_len = len; + monc->m_auth->hdr.front_len = cpu_to_le32(len); + ceph_con_revoke(monc->con, monc->m_auth); + ceph_msg_get(monc->m_auth); /* keep our ref */ + ceph_con_send(monc->con, monc->m_auth); +} + +/* + * Close monitor session, if any. + */ +static void __close_session(struct ceph_mon_client *monc) +{ + if (monc->con) { + dout("__close_session closing mon%d\n", monc->cur_mon); + ceph_con_revoke(monc->con, monc->m_auth); + ceph_con_close(monc->con); + monc->cur_mon = -1; + monc->pending_auth = 0; + ceph_auth_reset(monc->auth); + } +} + +/* + * Open a session with a (new) monitor. + */ +static int __open_session(struct ceph_mon_client *monc) +{ + char r; + int ret; + + if (monc->cur_mon < 0) { + get_random_bytes(&r, 1); + monc->cur_mon = r % monc->monmap->num_mon; + dout("open_session num=%d r=%d -> mon%d\n", + monc->monmap->num_mon, r, monc->cur_mon); + monc->sub_sent = 0; + monc->sub_renew_after = jiffies; /* i.e., expired */ + monc->want_next_osdmap = !!monc->want_next_osdmap; + + dout("open_session mon%d opening\n", monc->cur_mon); + monc->con->peer_name.type = CEPH_ENTITY_TYPE_MON; + monc->con->peer_name.num = cpu_to_le64(monc->cur_mon); + ceph_con_open(monc->con, + &monc->monmap->mon_inst[monc->cur_mon].addr); + + /* initiatiate authentication handshake */ + ret = ceph_auth_build_hello(monc->auth, + monc->m_auth->front.iov_base, + monc->m_auth->front_max); + __send_prepared_auth_request(monc, ret); + } else { + dout("open_session mon%d already open\n", monc->cur_mon); + } + return 0; +} + +static bool __sub_expired(struct ceph_mon_client *monc) +{ + return time_after_eq(jiffies, monc->sub_renew_after); +} + +/* + * Reschedule delayed work timer. + */ +static void __schedule_delayed(struct ceph_mon_client *monc) +{ + unsigned delay; + + if (monc->cur_mon < 0 || __sub_expired(monc)) + delay = 10 * HZ; + else + delay = 20 * HZ; + dout("__schedule_delayed after %u\n", delay); + schedule_delayed_work(&monc->delayed_work, delay); +} + +/* + * Send subscribe request for mdsmap and/or osdmap. + */ +static void __send_subscribe(struct ceph_mon_client *monc) +{ + dout("__send_subscribe sub_sent=%u exp=%u want_osd=%d\n", + (unsigned)monc->sub_sent, __sub_expired(monc), + monc->want_next_osdmap); + if ((__sub_expired(monc) && !monc->sub_sent) || + monc->want_next_osdmap == 1) { + struct ceph_msg *msg = monc->m_subscribe; + struct ceph_mon_subscribe_item *i; + void *p, *end; + int num; + + p = msg->front.iov_base; + end = p + msg->front_max; + + num = 1 + !!monc->want_next_osdmap + !!monc->want_mdsmap; + ceph_encode_32(&p, num); + + if (monc->want_next_osdmap) { + dout("__send_subscribe to 'osdmap' %u\n", + (unsigned)monc->have_osdmap); + ceph_encode_string(&p, end, "osdmap", 6); + i = p; + i->have = cpu_to_le64(monc->have_osdmap); + i->onetime = 1; + p += sizeof(*i); + monc->want_next_osdmap = 2; /* requested */ + } + if (monc->want_mdsmap) { + dout("__send_subscribe to 'mdsmap' %u+\n", + (unsigned)monc->have_mdsmap); + ceph_encode_string(&p, end, "mdsmap", 6); + i = p; + i->have = cpu_to_le64(monc->have_mdsmap); + i->onetime = 0; + p += sizeof(*i); + } + ceph_encode_string(&p, end, "monmap", 6); + i = p; + i->have = 0; + i->onetime = 0; + p += sizeof(*i); + + msg->front.iov_len = p - msg->front.iov_base; + msg->hdr.front_len = cpu_to_le32(msg->front.iov_len); + ceph_con_revoke(monc->con, msg); + ceph_con_send(monc->con, ceph_msg_get(msg)); + + monc->sub_sent = jiffies | 1; /* never 0 */ + } +} + +static void handle_subscribe_ack(struct ceph_mon_client *monc, + struct ceph_msg *msg) +{ + unsigned seconds; + struct ceph_mon_subscribe_ack *h = msg->front.iov_base; + + if (msg->front.iov_len < sizeof(*h)) + goto bad; + seconds = le32_to_cpu(h->duration); + + mutex_lock(&monc->mutex); + if (monc->hunting) { + pr_info("mon%d %s session established\n", + monc->cur_mon, + ceph_pr_addr(&monc->con->peer_addr.in_addr)); + monc->hunting = false; + } + dout("handle_subscribe_ack after %d seconds\n", seconds); + monc->sub_renew_after = monc->sub_sent + (seconds >> 1)*HZ - 1; + monc->sub_sent = 0; + mutex_unlock(&monc->mutex); + return; +bad: + pr_err("got corrupt subscribe-ack msg\n"); + ceph_msg_dump(msg); +} + +/* + * Keep track of which maps we have + */ +int ceph_monc_got_mdsmap(struct ceph_mon_client *monc, u32 got) +{ + mutex_lock(&monc->mutex); + monc->have_mdsmap = got; + mutex_unlock(&monc->mutex); + return 0; +} +EXPORT_SYMBOL(ceph_monc_got_mdsmap); + +int ceph_monc_got_osdmap(struct ceph_mon_client *monc, u32 got) +{ + mutex_lock(&monc->mutex); + monc->have_osdmap = got; + monc->want_next_osdmap = 0; + mutex_unlock(&monc->mutex); + return 0; +} + +/* + * Register interest in the next osdmap + */ +void ceph_monc_request_next_osdmap(struct ceph_mon_client *monc) +{ + dout("request_next_osdmap have %u\n", monc->have_osdmap); + mutex_lock(&monc->mutex); + if (!monc->want_next_osdmap) + monc->want_next_osdmap = 1; + if (monc->want_next_osdmap < 2) + __send_subscribe(monc); + mutex_unlock(&monc->mutex); +} + +/* + * + */ +int ceph_monc_open_session(struct ceph_mon_client *monc) +{ + if (!monc->con) { + monc->con = kmalloc(sizeof(*monc->con), GFP_KERNEL); + if (!monc->con) + return -ENOMEM; + ceph_con_init(monc->client->msgr, monc->con); + monc->con->private = monc; + monc->con->ops = &mon_con_ops; + } + + mutex_lock(&monc->mutex); + __open_session(monc); + __schedule_delayed(monc); + mutex_unlock(&monc->mutex); + return 0; +} +EXPORT_SYMBOL(ceph_monc_open_session); + +/* + * The monitor responds with mount ack indicate mount success. The + * included client ticket allows the client to talk to MDSs and OSDs. + */ +static void ceph_monc_handle_map(struct ceph_mon_client *monc, + struct ceph_msg *msg) +{ + struct ceph_client *client = monc->client; + struct ceph_monmap *monmap = NULL, *old = monc->monmap; + void *p, *end; + + mutex_lock(&monc->mutex); + + dout("handle_monmap\n"); + p = msg->front.iov_base; + end = p + msg->front.iov_len; + + monmap = ceph_monmap_decode(p, end); + if (IS_ERR(monmap)) { + pr_err("problem decoding monmap, %d\n", + (int)PTR_ERR(monmap)); + goto out; + } + + if (ceph_check_fsid(monc->client, &monmap->fsid) < 0) { + kfree(monmap); + goto out; + } + + client->monc.monmap = monmap; + kfree(old); + +out: + mutex_unlock(&monc->mutex); + wake_up_all(&client->auth_wq); +} + +/* + * generic requests (e.g., statfs, poolop) + */ +static struct ceph_mon_generic_request *__lookup_generic_req( + struct ceph_mon_client *monc, u64 tid) +{ + struct ceph_mon_generic_request *req; + struct rb_node *n = monc->generic_request_tree.rb_node; + + while (n) { + req = rb_entry(n, struct ceph_mon_generic_request, node); + if (tid < req->tid) + n = n->rb_left; + else if (tid > req->tid) + n = n->rb_right; + else + return req; + } + return NULL; +} + +static void __insert_generic_request(struct ceph_mon_client *monc, + struct ceph_mon_generic_request *new) +{ + struct rb_node **p = &monc->generic_request_tree.rb_node; + struct rb_node *parent = NULL; + struct ceph_mon_generic_request *req = NULL; + + while (*p) { + parent = *p; + req = rb_entry(parent, struct ceph_mon_generic_request, node); + if (new->tid < req->tid) + p = &(*p)->rb_left; + else if (new->tid > req->tid) + p = &(*p)->rb_right; + else + BUG(); + } + + rb_link_node(&new->node, parent, p); + rb_insert_color(&new->node, &monc->generic_request_tree); +} + +static void release_generic_request(struct kref *kref) +{ + struct ceph_mon_generic_request *req = + container_of(kref, struct ceph_mon_generic_request, kref); + + if (req->reply) + ceph_msg_put(req->reply); + if (req->request) + ceph_msg_put(req->request); + + kfree(req); +} + +static void put_generic_request(struct ceph_mon_generic_request *req) +{ + kref_put(&req->kref, release_generic_request); +} + +static void get_generic_request(struct ceph_mon_generic_request *req) +{ + kref_get(&req->kref); +} + +static struct ceph_msg *get_generic_reply(struct ceph_connection *con, + struct ceph_msg_header *hdr, + int *skip) +{ + struct ceph_mon_client *monc = con->private; + struct ceph_mon_generic_request *req; + u64 tid = le64_to_cpu(hdr->tid); + struct ceph_msg *m; + + mutex_lock(&monc->mutex); + req = __lookup_generic_req(monc, tid); + if (!req) { + dout("get_generic_reply %lld dne\n", tid); + *skip = 1; + m = NULL; + } else { + dout("get_generic_reply %lld got %p\n", tid, req->reply); + m = ceph_msg_get(req->reply); + /* + * we don't need to track the connection reading into + * this reply because we only have one open connection + * at a time, ever. + */ + } + mutex_unlock(&monc->mutex); + return m; +} + +static int do_generic_request(struct ceph_mon_client *monc, + struct ceph_mon_generic_request *req) +{ + int err; + + /* register request */ + mutex_lock(&monc->mutex); + req->tid = ++monc->last_tid; + req->request->hdr.tid = cpu_to_le64(req->tid); + __insert_generic_request(monc, req); + monc->num_generic_requests++; + ceph_con_send(monc->con, ceph_msg_get(req->request)); + mutex_unlock(&monc->mutex); + + err = wait_for_completion_interruptible(&req->completion); + + mutex_lock(&monc->mutex); + rb_erase(&req->node, &monc->generic_request_tree); + monc->num_generic_requests--; + mutex_unlock(&monc->mutex); + + if (!err) + err = req->result; + return err; +} + +/* + * statfs + */ +static void handle_statfs_reply(struct ceph_mon_client *monc, + struct ceph_msg *msg) +{ + struct ceph_mon_generic_request *req; + struct ceph_mon_statfs_reply *reply = msg->front.iov_base; + u64 tid = le64_to_cpu(msg->hdr.tid); + + if (msg->front.iov_len != sizeof(*reply)) + goto bad; + dout("handle_statfs_reply %p tid %llu\n", msg, tid); + + mutex_lock(&monc->mutex); + req = __lookup_generic_req(monc, tid); + if (req) { + *(struct ceph_statfs *)req->buf = reply->st; + req->result = 0; + get_generic_request(req); + } + mutex_unlock(&monc->mutex); + if (req) { + complete_all(&req->completion); + put_generic_request(req); + } + return; + +bad: + pr_err("corrupt generic reply, tid %llu\n", tid); + ceph_msg_dump(msg); +} + +/* + * Do a synchronous statfs(). + */ +int ceph_monc_do_statfs(struct ceph_mon_client *monc, struct ceph_statfs *buf) +{ + struct ceph_mon_generic_request *req; + struct ceph_mon_statfs *h; + int err; + + req = kzalloc(sizeof(*req), GFP_NOFS); + if (!req) + return -ENOMEM; + + kref_init(&req->kref); + req->buf = buf; + req->buf_len = sizeof(*buf); + init_completion(&req->completion); + + err = -ENOMEM; + req->request = ceph_msg_new(CEPH_MSG_STATFS, sizeof(*h), GFP_NOFS); + if (!req->request) + goto out; + req->reply = ceph_msg_new(CEPH_MSG_STATFS_REPLY, 1024, GFP_NOFS); + if (!req->reply) + goto out; + + /* fill out request */ + h = req->request->front.iov_base; + h->monhdr.have_version = 0; + h->monhdr.session_mon = cpu_to_le16(-1); + h->monhdr.session_mon_tid = 0; + h->fsid = monc->monmap->fsid; + + err = do_generic_request(monc, req); + +out: + kref_put(&req->kref, release_generic_request); + return err; +} +EXPORT_SYMBOL(ceph_monc_do_statfs); + +/* + * pool ops + */ +static int get_poolop_reply_buf(const char *src, size_t src_len, + char *dst, size_t dst_len) +{ + u32 buf_len; + + if (src_len != sizeof(u32) + dst_len) + return -EINVAL; + + buf_len = le32_to_cpu(*(u32 *)src); + if (buf_len != dst_len) + return -EINVAL; + + memcpy(dst, src + sizeof(u32), dst_len); + return 0; +} + +static void handle_poolop_reply(struct ceph_mon_client *monc, + struct ceph_msg *msg) +{ + struct ceph_mon_generic_request *req; + struct ceph_mon_poolop_reply *reply = msg->front.iov_base; + u64 tid = le64_to_cpu(msg->hdr.tid); + + if (msg->front.iov_len < sizeof(*reply)) + goto bad; + dout("handle_poolop_reply %p tid %llu\n", msg, tid); + + mutex_lock(&monc->mutex); + req = __lookup_generic_req(monc, tid); + if (req) { + if (req->buf_len && + get_poolop_reply_buf(msg->front.iov_base + sizeof(*reply), + msg->front.iov_len - sizeof(*reply), + req->buf, req->buf_len) < 0) { + mutex_unlock(&monc->mutex); + goto bad; + } + req->result = le32_to_cpu(reply->reply_code); + get_generic_request(req); + } + mutex_unlock(&monc->mutex); + if (req) { + complete(&req->completion); + put_generic_request(req); + } + return; + +bad: + pr_err("corrupt generic reply, tid %llu\n", tid); + ceph_msg_dump(msg); +} + +/* + * Do a synchronous pool op. + */ +int ceph_monc_do_poolop(struct ceph_mon_client *monc, u32 op, + u32 pool, u64 snapid, + char *buf, int len) +{ + struct ceph_mon_generic_request *req; + struct ceph_mon_poolop *h; + int err; + + req = kzalloc(sizeof(*req), GFP_NOFS); + if (!req) + return -ENOMEM; + + kref_init(&req->kref); + req->buf = buf; + req->buf_len = len; + init_completion(&req->completion); + + err = -ENOMEM; + req->request = ceph_msg_new(CEPH_MSG_POOLOP, sizeof(*h), GFP_NOFS); + if (!req->request) + goto out; + req->reply = ceph_msg_new(CEPH_MSG_POOLOP_REPLY, 1024, GFP_NOFS); + if (!req->reply) + goto out; + + /* fill out request */ + req->request->hdr.version = cpu_to_le16(2); + h = req->request->front.iov_base; + h->monhdr.have_version = 0; + h->monhdr.session_mon = cpu_to_le16(-1); + h->monhdr.session_mon_tid = 0; + h->fsid = monc->monmap->fsid; + h->pool = cpu_to_le32(pool); + h->op = cpu_to_le32(op); + h->auid = 0; + h->snapid = cpu_to_le64(snapid); + h->name_len = 0; + + err = do_generic_request(monc, req); + +out: + kref_put(&req->kref, release_generic_request); + return err; +} + +int ceph_monc_create_snapid(struct ceph_mon_client *monc, + u32 pool, u64 *snapid) +{ + return ceph_monc_do_poolop(monc, POOL_OP_CREATE_UNMANAGED_SNAP, + pool, 0, (char *)snapid, sizeof(*snapid)); + +} +EXPORT_SYMBOL(ceph_monc_create_snapid); + +int ceph_monc_delete_snapid(struct ceph_mon_client *monc, + u32 pool, u64 snapid) +{ + return ceph_monc_do_poolop(monc, POOL_OP_CREATE_UNMANAGED_SNAP, + pool, snapid, 0, 0); + +} + +/* + * Resend pending generic requests. + */ +static void __resend_generic_request(struct ceph_mon_client *monc) +{ + struct ceph_mon_generic_request *req; + struct rb_node *p; + + for (p = rb_first(&monc->generic_request_tree); p; p = rb_next(p)) { + req = rb_entry(p, struct ceph_mon_generic_request, node); + ceph_con_revoke(monc->con, req->request); + ceph_con_send(monc->con, ceph_msg_get(req->request)); + } +} + +/* + * Delayed work. If we haven't mounted yet, retry. Otherwise, + * renew/retry subscription as needed (in case it is timing out, or we + * got an ENOMEM). And keep the monitor connection alive. + */ +static void delayed_work(struct work_struct *work) +{ + struct ceph_mon_client *monc = + container_of(work, struct ceph_mon_client, delayed_work.work); + + dout("monc delayed_work\n"); + mutex_lock(&monc->mutex); + if (monc->hunting) { + __close_session(monc); + __open_session(monc); /* continue hunting */ + } else { + ceph_con_keepalive(monc->con); + + __validate_auth(monc); + + if (monc->auth->ops->is_authenticated(monc->auth)) + __send_subscribe(monc); + } + __schedule_delayed(monc); + mutex_unlock(&monc->mutex); +} + +/* + * On startup, we build a temporary monmap populated with the IPs + * provided by mount(2). + */ +static int build_initial_monmap(struct ceph_mon_client *monc) +{ + struct ceph_options *opt = monc->client->options; + struct ceph_entity_addr *mon_addr = opt->mon_addr; + int num_mon = opt->num_mon; + int i; + + /* build initial monmap */ + monc->monmap = kzalloc(sizeof(*monc->monmap) + + num_mon*sizeof(monc->monmap->mon_inst[0]), + GFP_KERNEL); + if (!monc->monmap) + return -ENOMEM; + for (i = 0; i < num_mon; i++) { + monc->monmap->mon_inst[i].addr = mon_addr[i]; + monc->monmap->mon_inst[i].addr.nonce = 0; + monc->monmap->mon_inst[i].name.type = + CEPH_ENTITY_TYPE_MON; + monc->monmap->mon_inst[i].name.num = cpu_to_le64(i); + } + monc->monmap->num_mon = num_mon; + monc->have_fsid = false; + return 0; +} + +int ceph_monc_init(struct ceph_mon_client *monc, struct ceph_client *cl) +{ + int err = 0; + + dout("init\n"); + memset(monc, 0, sizeof(*monc)); + monc->client = cl; + monc->monmap = NULL; + mutex_init(&monc->mutex); + + err = build_initial_monmap(monc); + if (err) + goto out; + + monc->con = NULL; + + /* authentication */ + monc->auth = ceph_auth_init(cl->options->name, + cl->options->secret); + if (IS_ERR(monc->auth)) + return PTR_ERR(monc->auth); + monc->auth->want_keys = + CEPH_ENTITY_TYPE_AUTH | CEPH_ENTITY_TYPE_MON | + CEPH_ENTITY_TYPE_OSD | CEPH_ENTITY_TYPE_MDS; + + /* msgs */ + err = -ENOMEM; + monc->m_subscribe_ack = ceph_msg_new(CEPH_MSG_MON_SUBSCRIBE_ACK, + sizeof(struct ceph_mon_subscribe_ack), + GFP_NOFS); + if (!monc->m_subscribe_ack) + goto out_monmap; + + monc->m_subscribe = ceph_msg_new(CEPH_MSG_MON_SUBSCRIBE, 96, GFP_NOFS); + if (!monc->m_subscribe) + goto out_subscribe_ack; + + monc->m_auth_reply = ceph_msg_new(CEPH_MSG_AUTH_REPLY, 4096, GFP_NOFS); + if (!monc->m_auth_reply) + goto out_subscribe; + + monc->m_auth = ceph_msg_new(CEPH_MSG_AUTH, 4096, GFP_NOFS); + monc->pending_auth = 0; + if (!monc->m_auth) + goto out_auth_reply; + + monc->cur_mon = -1; + monc->hunting = true; + monc->sub_renew_after = jiffies; + monc->sub_sent = 0; + + INIT_DELAYED_WORK(&monc->delayed_work, delayed_work); + monc->generic_request_tree = RB_ROOT; + monc->num_generic_requests = 0; + monc->last_tid = 0; + + monc->have_mdsmap = 0; + monc->have_osdmap = 0; + monc->want_next_osdmap = 1; + return 0; + +out_auth_reply: + ceph_msg_put(monc->m_auth_reply); +out_subscribe: + ceph_msg_put(monc->m_subscribe); +out_subscribe_ack: + ceph_msg_put(monc->m_subscribe_ack); +out_monmap: + kfree(monc->monmap); +out: + return err; +} +EXPORT_SYMBOL(ceph_monc_init); + +void ceph_monc_stop(struct ceph_mon_client *monc) +{ + dout("stop\n"); + cancel_delayed_work_sync(&monc->delayed_work); + + mutex_lock(&monc->mutex); + __close_session(monc); + if (monc->con) { + monc->con->private = NULL; + monc->con->ops->put(monc->con); + monc->con = NULL; + } + mutex_unlock(&monc->mutex); + + ceph_auth_destroy(monc->auth); + + ceph_msg_put(monc->m_auth); + ceph_msg_put(monc->m_auth_reply); + ceph_msg_put(monc->m_subscribe); + ceph_msg_put(monc->m_subscribe_ack); + + kfree(monc->monmap); +} +EXPORT_SYMBOL(ceph_monc_stop); + +static void handle_auth_reply(struct ceph_mon_client *monc, + struct ceph_msg *msg) +{ + int ret; + int was_auth = 0; + + mutex_lock(&monc->mutex); + if (monc->auth->ops) + was_auth = monc->auth->ops->is_authenticated(monc->auth); + monc->pending_auth = 0; + ret = ceph_handle_auth_reply(monc->auth, msg->front.iov_base, + msg->front.iov_len, + monc->m_auth->front.iov_base, + monc->m_auth->front_max); + if (ret < 0) { + monc->client->auth_err = ret; + wake_up_all(&monc->client->auth_wq); + } else if (ret > 0) { + __send_prepared_auth_request(monc, ret); + } else if (!was_auth && monc->auth->ops->is_authenticated(monc->auth)) { + dout("authenticated, starting session\n"); + + monc->client->msgr->inst.name.type = CEPH_ENTITY_TYPE_CLIENT; + monc->client->msgr->inst.name.num = + cpu_to_le64(monc->auth->global_id); + + __send_subscribe(monc); + __resend_generic_request(monc); + } + mutex_unlock(&monc->mutex); +} + +static int __validate_auth(struct ceph_mon_client *monc) +{ + int ret; + + if (monc->pending_auth) + return 0; + + ret = ceph_build_auth(monc->auth, monc->m_auth->front.iov_base, + monc->m_auth->front_max); + if (ret <= 0) + return ret; /* either an error, or no need to authenticate */ + __send_prepared_auth_request(monc, ret); + return 0; +} + +int ceph_monc_validate_auth(struct ceph_mon_client *monc) +{ + int ret; + + mutex_lock(&monc->mutex); + ret = __validate_auth(monc); + mutex_unlock(&monc->mutex); + return ret; +} +EXPORT_SYMBOL(ceph_monc_validate_auth); + +/* + * handle incoming message + */ +static void dispatch(struct ceph_connection *con, struct ceph_msg *msg) +{ + struct ceph_mon_client *monc = con->private; + int type = le16_to_cpu(msg->hdr.type); + + if (!monc) + return; + + switch (type) { + case CEPH_MSG_AUTH_REPLY: + handle_auth_reply(monc, msg); + break; + + case CEPH_MSG_MON_SUBSCRIBE_ACK: + handle_subscribe_ack(monc, msg); + break; + + case CEPH_MSG_STATFS_REPLY: + handle_statfs_reply(monc, msg); + break; + + case CEPH_MSG_POOLOP_REPLY: + handle_poolop_reply(monc, msg); + break; + + case CEPH_MSG_MON_MAP: + ceph_monc_handle_map(monc, msg); + break; + + case CEPH_MSG_OSD_MAP: + ceph_osdc_handle_map(&monc->client->osdc, msg); + break; + + default: + /* can the chained handler handle it? */ + if (monc->client->extra_mon_dispatch && + monc->client->extra_mon_dispatch(monc->client, msg) == 0) + break; + + pr_err("received unknown message type %d %s\n", type, + ceph_msg_type_name(type)); + } + ceph_msg_put(msg); +} + +/* + * Allocate memory for incoming message + */ +static struct ceph_msg *mon_alloc_msg(struct ceph_connection *con, + struct ceph_msg_header *hdr, + int *skip) +{ + struct ceph_mon_client *monc = con->private; + int type = le16_to_cpu(hdr->type); + int front_len = le32_to_cpu(hdr->front_len); + struct ceph_msg *m = NULL; + + *skip = 0; + + switch (type) { + case CEPH_MSG_MON_SUBSCRIBE_ACK: + m = ceph_msg_get(monc->m_subscribe_ack); + break; + case CEPH_MSG_POOLOP_REPLY: + case CEPH_MSG_STATFS_REPLY: + return get_generic_reply(con, hdr, skip); + case CEPH_MSG_AUTH_REPLY: + m = ceph_msg_get(monc->m_auth_reply); + break; + case CEPH_MSG_MON_MAP: + case CEPH_MSG_MDS_MAP: + case CEPH_MSG_OSD_MAP: + m = ceph_msg_new(type, front_len, GFP_NOFS); + break; + } + + if (!m) { + pr_info("alloc_msg unknown type %d\n", type); + *skip = 1; + } + return m; +} + +/* + * If the monitor connection resets, pick a new monitor and resubmit + * any pending requests. + */ +static void mon_fault(struct ceph_connection *con) +{ + struct ceph_mon_client *monc = con->private; + + if (!monc) + return; + + dout("mon_fault\n"); + mutex_lock(&monc->mutex); + if (!con->private) + goto out; + + if (monc->con && !monc->hunting) + pr_info("mon%d %s session lost, " + "hunting for new mon\n", monc->cur_mon, + ceph_pr_addr(&monc->con->peer_addr.in_addr)); + + __close_session(monc); + if (!monc->hunting) { + /* start hunting */ + monc->hunting = true; + __open_session(monc); + } else { + /* already hunting, let's wait a bit */ + __schedule_delayed(monc); + } +out: + mutex_unlock(&monc->mutex); +} + +static const struct ceph_connection_operations mon_con_ops = { + .get = ceph_con_get, + .put = ceph_con_put, + .dispatch = dispatch, + .fault = mon_fault, + .alloc_msg = mon_alloc_msg, +}; diff --git a/net/ceph/msgpool.c b/net/ceph/msgpool.c new file mode 100644 index 00000000000..d5f2d97ac05 --- /dev/null +++ b/net/ceph/msgpool.c @@ -0,0 +1,64 @@ +#include <linux/ceph/ceph_debug.h> + +#include <linux/err.h> +#include <linux/sched.h> +#include <linux/types.h> +#include <linux/vmalloc.h> + +#include <linux/ceph/msgpool.h> + +static void *alloc_fn(gfp_t gfp_mask, void *arg) +{ + struct ceph_msgpool *pool = arg; + void *p; + + p = ceph_msg_new(0, pool->front_len, gfp_mask); + if (!p) + pr_err("msgpool %s alloc failed\n", pool->name); + return p; +} + +static void free_fn(void *element, void *arg) +{ + ceph_msg_put(element); +} + +int ceph_msgpool_init(struct ceph_msgpool *pool, + int front_len, int size, bool blocking, const char *name) +{ + pool->front_len = front_len; + pool->pool = mempool_create(size, alloc_fn, free_fn, pool); + if (!pool->pool) + return -ENOMEM; + pool->name = name; + return 0; +} + +void ceph_msgpool_destroy(struct ceph_msgpool *pool) +{ + mempool_destroy(pool->pool); +} + +struct ceph_msg *ceph_msgpool_get(struct ceph_msgpool *pool, + int front_len) +{ + if (front_len > pool->front_len) { + pr_err("msgpool_get pool %s need front %d, pool size is %d\n", + pool->name, front_len, pool->front_len); + WARN_ON(1); + + /* try to alloc a fresh message */ + return ceph_msg_new(0, front_len, GFP_NOFS); + } + + return mempool_alloc(pool->pool, GFP_NOFS); +} + +void ceph_msgpool_put(struct ceph_msgpool *pool, struct ceph_msg *msg) +{ + /* reset msg front_len; user may have changed it */ + msg->front.iov_len = pool->front_len; + msg->hdr.front_len = cpu_to_le32(pool->front_len); + + kref_init(&msg->kref); /* retake single ref */ +} diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c new file mode 100644 index 00000000000..79391994b3e --- /dev/null +++ b/net/ceph/osd_client.c @@ -0,0 +1,1773 @@ +#include <linux/ceph/ceph_debug.h> + +#include <linux/module.h> +#include <linux/err.h> +#include <linux/highmem.h> +#include <linux/mm.h> +#include <linux/pagemap.h> +#include <linux/slab.h> +#include <linux/uaccess.h> +#ifdef CONFIG_BLOCK +#include <linux/bio.h> +#endif + +#include <linux/ceph/libceph.h> +#include <linux/ceph/osd_client.h> +#include <linux/ceph/messenger.h> +#include <linux/ceph/decode.h> +#include <linux/ceph/auth.h> +#include <linux/ceph/pagelist.h> + +#define OSD_OP_FRONT_LEN 4096 +#define OSD_OPREPLY_FRONT_LEN 512 + +static const struct ceph_connection_operations osd_con_ops; +static int __kick_requests(struct ceph_osd_client *osdc, + struct ceph_osd *kickosd); + +static void kick_requests(struct ceph_osd_client *osdc, struct ceph_osd *osd); + +static int op_needs_trail(int op) +{ + switch (op) { + case CEPH_OSD_OP_GETXATTR: + case CEPH_OSD_OP_SETXATTR: + case CEPH_OSD_OP_CMPXATTR: + case CEPH_OSD_OP_CALL: + return 1; + default: + return 0; + } +} + +static int op_has_extent(int op) +{ + return (op == CEPH_OSD_OP_READ || + op == CEPH_OSD_OP_WRITE); +} + +void ceph_calc_raw_layout(struct ceph_osd_client *osdc, + struct ceph_file_layout *layout, + u64 snapid, + u64 off, u64 *plen, u64 *bno, + struct ceph_osd_request *req, + struct ceph_osd_req_op *op) +{ + struct ceph_osd_request_head *reqhead = req->r_request->front.iov_base; + u64 orig_len = *plen; + u64 objoff, objlen; /* extent in object */ + + reqhead->snapid = cpu_to_le64(snapid); + + /* object extent? */ + ceph_calc_file_object_mapping(layout, off, plen, bno, + &objoff, &objlen); + if (*plen < orig_len) + dout(" skipping last %llu, final file extent %llu~%llu\n", + orig_len - *plen, off, *plen); + + if (op_has_extent(op->op)) { + op->extent.offset = objoff; + op->extent.length = objlen; + } + req->r_num_pages = calc_pages_for(off, *plen); + if (op->op == CEPH_OSD_OP_WRITE) + op->payload_len = *plen; + + dout("calc_layout bno=%llx %llu~%llu (%d pages)\n", + *bno, objoff, objlen, req->r_num_pages); + +} +EXPORT_SYMBOL(ceph_calc_raw_layout); + +/* + * Implement client access to distributed object storage cluster. + * + * All data objects are stored within a cluster/cloud of OSDs, or + * "object storage devices." (Note that Ceph OSDs have _nothing_ to + * do with the T10 OSD extensions to SCSI.) Ceph OSDs are simply + * remote daemons serving up and coordinating consistent and safe + * access to storage. + * + * Cluster membership and the mapping of data objects onto storage devices + * are described by the osd map. + * + * We keep track of pending OSD requests (read, write), resubmit + * requests to different OSDs when the cluster topology/data layout + * change, or retry the affected requests when the communications + * channel with an OSD is reset. + */ + +/* + * calculate the mapping of a file extent onto an object, and fill out the + * request accordingly. shorten extent as necessary if it crosses an + * object boundary. + * + * fill osd op in request message. + */ +static void calc_layout(struct ceph_osd_client *osdc, + struct ceph_vino vino, + struct ceph_file_layout *layout, + u64 off, u64 *plen, + struct ceph_osd_request *req, + struct ceph_osd_req_op *op) +{ + u64 bno; + + ceph_calc_raw_layout(osdc, layout, vino.snap, off, + plen, &bno, req, op); + + sprintf(req->r_oid, "%llx.%08llx", vino.ino, bno); + req->r_oid_len = strlen(req->r_oid); +} + +/* + * requests + */ +void ceph_osdc_release_request(struct kref *kref) +{ + struct ceph_osd_request *req = container_of(kref, + struct ceph_osd_request, + r_kref); + + if (req->r_request) + ceph_msg_put(req->r_request); + if (req->r_reply) + ceph_msg_put(req->r_reply); + if (req->r_con_filling_msg) { + dout("release_request revoking pages %p from con %p\n", + req->r_pages, req->r_con_filling_msg); + ceph_con_revoke_message(req->r_con_filling_msg, + req->r_reply); + ceph_con_put(req->r_con_filling_msg); + } + if (req->r_own_pages) + ceph_release_page_vector(req->r_pages, + req->r_num_pages); +#ifdef CONFIG_BLOCK + if (req->r_bio) + bio_put(req->r_bio); +#endif + ceph_put_snap_context(req->r_snapc); + if (req->r_trail) { + ceph_pagelist_release(req->r_trail); + kfree(req->r_trail); + } + if (req->r_mempool) + mempool_free(req, req->r_osdc->req_mempool); + else + kfree(req); +} +EXPORT_SYMBOL(ceph_osdc_release_request); + +static int get_num_ops(struct ceph_osd_req_op *ops, int *needs_trail) +{ + int i = 0; + + if (needs_trail) + *needs_trail = 0; + while (ops[i].op) { + if (needs_trail && op_needs_trail(ops[i].op)) + *needs_trail = 1; + i++; + } + + return i; +} + +struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc, + int flags, + struct ceph_snap_context *snapc, + struct ceph_osd_req_op *ops, + bool use_mempool, + gfp_t gfp_flags, + struct page **pages, + struct bio *bio) +{ + struct ceph_osd_request *req; + struct ceph_msg *msg; + int needs_trail; + int num_op = get_num_ops(ops, &needs_trail); + size_t msg_size = sizeof(struct ceph_osd_request_head); + + msg_size += num_op*sizeof(struct ceph_osd_op); + + if (use_mempool) { + req = mempool_alloc(osdc->req_mempool, gfp_flags); + memset(req, 0, sizeof(*req)); + } else { + req = kzalloc(sizeof(*req), gfp_flags); + } + if (req == NULL) + return NULL; + + req->r_osdc = osdc; + req->r_mempool = use_mempool; + + kref_init(&req->r_kref); + init_completion(&req->r_completion); + init_completion(&req->r_safe_completion); + INIT_LIST_HEAD(&req->r_unsafe_item); + req->r_flags = flags; + + WARN_ON((flags & (CEPH_OSD_FLAG_READ|CEPH_OSD_FLAG_WRITE)) == 0); + + /* create reply message */ + if (use_mempool) + msg = ceph_msgpool_get(&osdc->msgpool_op_reply, 0); + else + msg = ceph_msg_new(CEPH_MSG_OSD_OPREPLY, + OSD_OPREPLY_FRONT_LEN, gfp_flags); + if (!msg) { + ceph_osdc_put_request(req); + return NULL; + } + req->r_reply = msg; + + /* allocate space for the trailing data */ + if (needs_trail) { + req->r_trail = kmalloc(sizeof(struct ceph_pagelist), gfp_flags); + if (!req->r_trail) { + ceph_osdc_put_request(req); + return NULL; + } + ceph_pagelist_init(req->r_trail); + } + /* create request message; allow space for oid */ + msg_size += 40; + if (snapc) + msg_size += sizeof(u64) * snapc->num_snaps; + if (use_mempool) + msg = ceph_msgpool_get(&osdc->msgpool_op, 0); + else + msg = ceph_msg_new(CEPH_MSG_OSD_OP, msg_size, gfp_flags); + if (!msg) { + ceph_osdc_put_request(req); + return NULL; + } + + msg->hdr.type = cpu_to_le16(CEPH_MSG_OSD_OP); + memset(msg->front.iov_base, 0, msg->front.iov_len); + + req->r_request = msg; + req->r_pages = pages; +#ifdef CONFIG_BLOCK + if (bio) { + req->r_bio = bio; + bio_get(req->r_bio); + } +#endif + + return req; +} +EXPORT_SYMBOL(ceph_osdc_alloc_request); + +static void osd_req_encode_op(struct ceph_osd_request *req, + struct ceph_osd_op *dst, + struct ceph_osd_req_op *src) +{ + dst->op = cpu_to_le16(src->op); + + switch (dst->op) { + case CEPH_OSD_OP_READ: + case CEPH_OSD_OP_WRITE: + dst->extent.offset = + cpu_to_le64(src->extent.offset); + dst->extent.length = + cpu_to_le64(src->extent.length); + dst->extent.truncate_size = + cpu_to_le64(src->extent.truncate_size); + dst->extent.truncate_seq = + cpu_to_le32(src->extent.truncate_seq); + break; + + case CEPH_OSD_OP_GETXATTR: + case CEPH_OSD_OP_SETXATTR: + case CEPH_OSD_OP_CMPXATTR: + BUG_ON(!req->r_trail); + + dst->xattr.name_len = cpu_to_le32(src->xattr.name_len); + dst->xattr.value_len = cpu_to_le32(src->xattr.value_len); + dst->xattr.cmp_op = src->xattr.cmp_op; + dst->xattr.cmp_mode = src->xattr.cmp_mode; + ceph_pagelist_append(req->r_trail, src->xattr.name, + src->xattr.name_len); + ceph_pagelist_append(req->r_trail, src->xattr.val, + src->xattr.value_len); + break; + case CEPH_OSD_OP_CALL: + BUG_ON(!req->r_trail); + + dst->cls.class_len = src->cls.class_len; + dst->cls.method_len = src->cls.method_len; + dst->cls.indata_len = cpu_to_le32(src->cls.indata_len); + + ceph_pagelist_append(req->r_trail, src->cls.class_name, + src->cls.class_len); + ceph_pagelist_append(req->r_trail, src->cls.method_name, + src->cls.method_len); + ceph_pagelist_append(req->r_trail, src->cls.indata, + src->cls.indata_len); + break; + case CEPH_OSD_OP_ROLLBACK: + dst->snap.snapid = cpu_to_le64(src->snap.snapid); + break; + case CEPH_OSD_OP_STARTSYNC: + break; + default: + pr_err("unrecognized osd opcode %d\n", dst->op); + WARN_ON(1); + break; + } + dst->payload_len = cpu_to_le32(src->payload_len); +} + +/* + * build new request AND message + * + */ +void ceph_osdc_build_request(struct ceph_osd_request *req, + u64 off, u64 *plen, + struct ceph_osd_req_op *src_ops, + struct ceph_snap_context *snapc, + struct timespec *mtime, + const char *oid, + int oid_len) +{ + struct ceph_msg *msg = req->r_request; + struct ceph_osd_request_head *head; + struct ceph_osd_req_op *src_op; + struct ceph_osd_op *op; + void *p; + int num_op = get_num_ops(src_ops, NULL); + size_t msg_size = sizeof(*head) + num_op*sizeof(*op); + int flags = req->r_flags; + u64 data_len = 0; + int i; + + head = msg->front.iov_base; + op = (void *)(head + 1); + p = (void *)(op + num_op); + + req->r_snapc = ceph_get_snap_context(snapc); + + head->client_inc = cpu_to_le32(1); /* always, for now. */ + head->flags = cpu_to_le32(flags); + if (flags & CEPH_OSD_FLAG_WRITE) + ceph_encode_timespec(&head->mtime, mtime); + head->num_ops = cpu_to_le16(num_op); + + + /* fill in oid */ + head->object_len = cpu_to_le32(oid_len); + memcpy(p, oid, oid_len); + p += oid_len; + + src_op = src_ops; + while (src_op->op) { + osd_req_encode_op(req, op, src_op); + src_op++; + op++; + } + + if (req->r_trail) + data_len += req->r_trail->length; + + if (snapc) { + head->snap_seq = cpu_to_le64(snapc->seq); + head->num_snaps = cpu_to_le32(snapc->num_snaps); + for (i = 0; i < snapc->num_snaps; i++) { + put_unaligned_le64(snapc->snaps[i], p); + p += sizeof(u64); + } + } + + if (flags & CEPH_OSD_FLAG_WRITE) { + req->r_request->hdr.data_off = cpu_to_le16(off); + req->r_request->hdr.data_len = cpu_to_le32(*plen + data_len); + } else if (data_len) { + req->r_request->hdr.data_off = 0; + req->r_request->hdr.data_len = cpu_to_le32(data_len); + } + + BUG_ON(p > msg->front.iov_base + msg->front.iov_len); + msg_size = p - msg->front.iov_base; + msg->front.iov_len = msg_size; + msg->hdr.front_len = cpu_to_le32(msg_size); + return; +} +EXPORT_SYMBOL(ceph_osdc_build_request); + +/* + * build new request AND message, calculate layout, and adjust file + * extent as needed. + * + * if the file was recently truncated, we include information about its + * old and new size so that the object can be updated appropriately. (we + * avoid synchronously deleting truncated objects because it's slow.) + * + * if @do_sync, include a 'startsync' command so that the osd will flush + * data quickly. + */ +struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *osdc, + struct ceph_file_layout *layout, + struct ceph_vino vino, + u64 off, u64 *plen, + int opcode, int flags, + struct ceph_snap_context *snapc, + int do_sync, + u32 truncate_seq, + u64 truncate_size, + struct timespec *mtime, + bool use_mempool, int num_reply) +{ + struct ceph_osd_req_op ops[3]; + struct ceph_osd_request *req; + + ops[0].op = opcode; + ops[0].extent.truncate_seq = truncate_seq; + ops[0].extent.truncate_size = truncate_size; + ops[0].payload_len = 0; + + if (do_sync) { + ops[1].op = CEPH_OSD_OP_STARTSYNC; + ops[1].payload_len = 0; + ops[2].op = 0; + } else + ops[1].op = 0; + + req = ceph_osdc_alloc_request(osdc, flags, + snapc, ops, + use_mempool, + GFP_NOFS, NULL, NULL); + if (IS_ERR(req)) + return req; + + /* calculate max write size */ + calc_layout(osdc, vino, layout, off, plen, req, ops); + req->r_file_layout = *layout; /* keep a copy */ + + ceph_osdc_build_request(req, off, plen, ops, + snapc, + mtime, + req->r_oid, req->r_oid_len); + + return req; +} +EXPORT_SYMBOL(ceph_osdc_new_request); + +/* + * We keep osd requests in an rbtree, sorted by ->r_tid. + */ +static void __insert_request(struct ceph_osd_client *osdc, + struct ceph_osd_request *new) +{ + struct rb_node **p = &osdc->requests.rb_node; + struct rb_node *parent = NULL; + struct ceph_osd_request *req = NULL; + + while (*p) { + parent = *p; + req = rb_entry(parent, struct ceph_osd_request, r_node); + if (new->r_tid < req->r_tid) + p = &(*p)->rb_left; + else if (new->r_tid > req->r_tid) + p = &(*p)->rb_right; + else + BUG(); + } + + rb_link_node(&new->r_node, parent, p); + rb_insert_color(&new->r_node, &osdc->requests); +} + +static struct ceph_osd_request *__lookup_request(struct ceph_osd_client *osdc, + u64 tid) +{ + struct ceph_osd_request *req; + struct rb_node *n = osdc->requests.rb_node; + + while (n) { + req = rb_entry(n, struct ceph_osd_request, r_node); + if (tid < req->r_tid) + n = n->rb_left; + else if (tid > req->r_tid) + n = n->rb_right; + else + return req; + } + return NULL; +} + +static struct ceph_osd_request * +__lookup_request_ge(struct ceph_osd_client *osdc, + u64 tid) +{ + struct ceph_osd_request *req; + struct rb_node *n = osdc->requests.rb_node; + + while (n) { + req = rb_entry(n, struct ceph_osd_request, r_node); + if (tid < req->r_tid) { + if (!n->rb_left) + return req; + n = n->rb_left; + } else if (tid > req->r_tid) { + n = n->rb_right; + } else { + return req; + } + } + return NULL; +} + + +/* + * If the osd connection drops, we need to resubmit all requests. + */ +static void osd_reset(struct ceph_connection *con) +{ + struct ceph_osd *osd = con->private; + struct ceph_osd_client *osdc; + + if (!osd) + return; + dout("osd_reset osd%d\n", osd->o_osd); + osdc = osd->o_osdc; + down_read(&osdc->map_sem); + kick_requests(osdc, osd); + up_read(&osdc->map_sem); +} + +/* + * Track open sessions with osds. + */ +static struct ceph_osd *create_osd(struct ceph_osd_client *osdc) +{ + struct ceph_osd *osd; + + osd = kzalloc(sizeof(*osd), GFP_NOFS); + if (!osd) + return NULL; + + atomic_set(&osd->o_ref, 1); + osd->o_osdc = osdc; + INIT_LIST_HEAD(&osd->o_requests); + INIT_LIST_HEAD(&osd->o_osd_lru); + osd->o_incarnation = 1; + + ceph_con_init(osdc->client->msgr, &osd->o_con); + osd->o_con.private = osd; + osd->o_con.ops = &osd_con_ops; + osd->o_con.peer_name.type = CEPH_ENTITY_TYPE_OSD; + + INIT_LIST_HEAD(&osd->o_keepalive_item); + return osd; +} + +static struct ceph_osd *get_osd(struct ceph_osd *osd) +{ + if (atomic_inc_not_zero(&osd->o_ref)) { + dout("get_osd %p %d -> %d\n", osd, atomic_read(&osd->o_ref)-1, + atomic_read(&osd->o_ref)); + return osd; + } else { + dout("get_osd %p FAIL\n", osd); + return NULL; + } +} + +static void put_osd(struct ceph_osd *osd) +{ + dout("put_osd %p %d -> %d\n", osd, atomic_read(&osd->o_ref), + atomic_read(&osd->o_ref) - 1); + if (atomic_dec_and_test(&osd->o_ref)) { + struct ceph_auth_client *ac = osd->o_osdc->client->monc.auth; + + if (osd->o_authorizer) + ac->ops->destroy_authorizer(ac, osd->o_authorizer); + kfree(osd); + } +} + +/* + * remove an osd from our map + */ +static void __remove_osd(struct ceph_osd_client *osdc, struct ceph_osd *osd) +{ + dout("__remove_osd %p\n", osd); + BUG_ON(!list_empty(&osd->o_requests)); + rb_erase(&osd->o_node, &osdc->osds); + list_del_init(&osd->o_osd_lru); + ceph_con_close(&osd->o_con); + put_osd(osd); +} + +static void __move_osd_to_lru(struct ceph_osd_client *osdc, + struct ceph_osd *osd) +{ + dout("__move_osd_to_lru %p\n", osd); + BUG_ON(!list_empty(&osd->o_osd_lru)); + list_add_tail(&osd->o_osd_lru, &osdc->osd_lru); + osd->lru_ttl = jiffies + osdc->client->options->osd_idle_ttl * HZ; +} + +static void __remove_osd_from_lru(struct ceph_osd *osd) +{ + dout("__remove_osd_from_lru %p\n", osd); + if (!list_empty(&osd->o_osd_lru)) + list_del_init(&osd->o_osd_lru); +} + +static void remove_old_osds(struct ceph_osd_client *osdc, int remove_all) +{ + struct ceph_osd *osd, *nosd; + + dout("__remove_old_osds %p\n", osdc); + mutex_lock(&osdc->request_mutex); + list_for_each_entry_safe(osd, nosd, &osdc->osd_lru, o_osd_lru) { + if (!remove_all && time_before(jiffies, osd->lru_ttl)) + break; + __remove_osd(osdc, osd); + } + mutex_unlock(&osdc->request_mutex); +} + +/* + * reset osd connect + */ +static int __reset_osd(struct ceph_osd_client *osdc, struct ceph_osd *osd) +{ + struct ceph_osd_request *req; + int ret = 0; + + dout("__reset_osd %p osd%d\n", osd, osd->o_osd); + if (list_empty(&osd->o_requests)) { + __remove_osd(osdc, osd); + } else if (memcmp(&osdc->osdmap->osd_addr[osd->o_osd], + &osd->o_con.peer_addr, + sizeof(osd->o_con.peer_addr)) == 0 && + !ceph_con_opened(&osd->o_con)) { + dout(" osd addr hasn't changed and connection never opened," + " letting msgr retry"); + /* touch each r_stamp for handle_timeout()'s benfit */ + list_for_each_entry(req, &osd->o_requests, r_osd_item) + req->r_stamp = jiffies; + ret = -EAGAIN; + } else { + ceph_con_close(&osd->o_con); + ceph_con_open(&osd->o_con, &osdc->osdmap->osd_addr[osd->o_osd]); + osd->o_incarnation++; + } + return ret; +} + +static void __insert_osd(struct ceph_osd_client *osdc, struct ceph_osd *new) +{ + struct rb_node **p = &osdc->osds.rb_node; + struct rb_node *parent = NULL; + struct ceph_osd *osd = NULL; + + while (*p) { + parent = *p; + osd = rb_entry(parent, struct ceph_osd, o_node); + if (new->o_osd < osd->o_osd) + p = &(*p)->rb_left; + else if (new->o_osd > osd->o_osd) + p = &(*p)->rb_right; + else + BUG(); + } + + rb_link_node(&new->o_node, parent, p); + rb_insert_color(&new->o_node, &osdc->osds); +} + +static struct ceph_osd *__lookup_osd(struct ceph_osd_client *osdc, int o) +{ + struct ceph_osd *osd; + struct rb_node *n = osdc->osds.rb_node; + + while (n) { + osd = rb_entry(n, struct ceph_osd, o_node); + if (o < osd->o_osd) + n = n->rb_left; + else if (o > osd->o_osd) + n = n->rb_right; + else + return osd; + } + return NULL; +} + +static void __schedule_osd_timeout(struct ceph_osd_client *osdc) +{ + schedule_delayed_work(&osdc->timeout_work, + osdc->client->options->osd_keepalive_timeout * HZ); +} + +static void __cancel_osd_timeout(struct ceph_osd_client *osdc) +{ + cancel_delayed_work(&osdc->timeout_work); +} + +/* + * Register request, assign tid. If this is the first request, set up + * the timeout event. + */ +static void register_request(struct ceph_osd_client *osdc, + struct ceph_osd_request *req) +{ + mutex_lock(&osdc->request_mutex); + req->r_tid = ++osdc->last_tid; + req->r_request->hdr.tid = cpu_to_le64(req->r_tid); + INIT_LIST_HEAD(&req->r_req_lru_item); + + dout("register_request %p tid %lld\n", req, req->r_tid); + __insert_request(osdc, req); + ceph_osdc_get_request(req); + osdc->num_requests++; + + if (osdc->num_requests == 1) { + dout(" first request, scheduling timeout\n"); + __schedule_osd_timeout(osdc); + } + mutex_unlock(&osdc->request_mutex); +} + +/* + * called under osdc->request_mutex + */ +static void __unregister_request(struct ceph_osd_client *osdc, + struct ceph_osd_request *req) +{ + dout("__unregister_request %p tid %lld\n", req, req->r_tid); + rb_erase(&req->r_node, &osdc->requests); + osdc->num_requests--; + + if (req->r_osd) { + /* make sure the original request isn't in flight. */ + ceph_con_revoke(&req->r_osd->o_con, req->r_request); + + list_del_init(&req->r_osd_item); + if (list_empty(&req->r_osd->o_requests)) + __move_osd_to_lru(osdc, req->r_osd); + req->r_osd = NULL; + } + + ceph_osdc_put_request(req); + + list_del_init(&req->r_req_lru_item); + if (osdc->num_requests == 0) { + dout(" no requests, canceling timeout\n"); + __cancel_osd_timeout(osdc); + } +} + +/* + * Cancel a previously queued request message + */ +static void __cancel_request(struct ceph_osd_request *req) +{ + if (req->r_sent && req->r_osd) { + ceph_con_revoke(&req->r_osd->o_con, req->r_request); + req->r_sent = 0; + } + list_del_init(&req->r_req_lru_item); +} + +/* + * Pick an osd (the first 'up' osd in the pg), allocate the osd struct + * (as needed), and set the request r_osd appropriately. If there is + * no up osd, set r_osd to NULL. + * + * Return 0 if unchanged, 1 if changed, or negative on error. + * + * Caller should hold map_sem for read and request_mutex. + */ +static int __map_osds(struct ceph_osd_client *osdc, + struct ceph_osd_request *req) +{ + struct ceph_osd_request_head *reqhead = req->r_request->front.iov_base; + struct ceph_pg pgid; + int acting[CEPH_PG_MAX_SIZE]; + int o = -1, num = 0; + int err; + + dout("map_osds %p tid %lld\n", req, req->r_tid); + err = ceph_calc_object_layout(&reqhead->layout, req->r_oid, + &req->r_file_layout, osdc->osdmap); + if (err) + return err; + pgid = reqhead->layout.ol_pgid; + req->r_pgid = pgid; + + err = ceph_calc_pg_acting(osdc->osdmap, pgid, acting); + if (err > 0) { + o = acting[0]; + num = err; + } + + if ((req->r_osd && req->r_osd->o_osd == o && + req->r_sent >= req->r_osd->o_incarnation && + req->r_num_pg_osds == num && + memcmp(req->r_pg_osds, acting, sizeof(acting[0])*num) == 0) || + (req->r_osd == NULL && o == -1)) + return 0; /* no change */ + + dout("map_osds tid %llu pgid %d.%x osd%d (was osd%d)\n", + req->r_tid, le32_to_cpu(pgid.pool), le16_to_cpu(pgid.ps), o, + req->r_osd ? req->r_osd->o_osd : -1); + + /* record full pg acting set */ + memcpy(req->r_pg_osds, acting, sizeof(acting[0]) * num); + req->r_num_pg_osds = num; + + if (req->r_osd) { + __cancel_request(req); + list_del_init(&req->r_osd_item); + req->r_osd = NULL; + } + + req->r_osd = __lookup_osd(osdc, o); + if (!req->r_osd && o >= 0) { + err = -ENOMEM; + req->r_osd = create_osd(osdc); + if (!req->r_osd) + goto out; + + dout("map_osds osd %p is osd%d\n", req->r_osd, o); + req->r_osd->o_osd = o; + req->r_osd->o_con.peer_name.num = cpu_to_le64(o); + __insert_osd(osdc, req->r_osd); + + ceph_con_open(&req->r_osd->o_con, &osdc->osdmap->osd_addr[o]); + } + + if (req->r_osd) { + __remove_osd_from_lru(req->r_osd); + list_add(&req->r_osd_item, &req->r_osd->o_requests); + } + err = 1; /* osd or pg changed */ + +out: + return err; +} + +/* + * caller should hold map_sem (for read) and request_mutex + */ +static int __send_request(struct ceph_osd_client *osdc, + struct ceph_osd_request *req) +{ + struct ceph_osd_request_head *reqhead; + int err; + + err = __map_osds(osdc, req); + if (err < 0) + return err; + if (req->r_osd == NULL) { + dout("send_request %p no up osds in pg\n", req); + ceph_monc_request_next_osdmap(&osdc->client->monc); + return 0; + } + + dout("send_request %p tid %llu to osd%d flags %d\n", + req, req->r_tid, req->r_osd->o_osd, req->r_flags); + + reqhead = req->r_request->front.iov_base; + reqhead->osdmap_epoch = cpu_to_le32(osdc->osdmap->epoch); + reqhead->flags |= cpu_to_le32(req->r_flags); /* e.g., RETRY */ + reqhead->reassert_version = req->r_reassert_version; + + req->r_stamp = jiffies; + list_move_tail(&req->r_req_lru_item, &osdc->req_lru); + + ceph_msg_get(req->r_request); /* send consumes a ref */ + ceph_con_send(&req->r_osd->o_con, req->r_request); + req->r_sent = req->r_osd->o_incarnation; + return 0; +} + +/* + * Timeout callback, called every N seconds when 1 or more osd + * requests has been active for more than N seconds. When this + * happens, we ping all OSDs with requests who have timed out to + * ensure any communications channel reset is detected. Reset the + * request timeouts another N seconds in the future as we go. + * Reschedule the timeout event another N seconds in future (unless + * there are no open requests). + */ +static void handle_timeout(struct work_struct *work) +{ + struct ceph_osd_client *osdc = + container_of(work, struct ceph_osd_client, timeout_work.work); + struct ceph_osd_request *req, *last_req = NULL; + struct ceph_osd *osd; + unsigned long timeout = osdc->client->options->osd_timeout * HZ; + unsigned long keepalive = + osdc->client->options->osd_keepalive_timeout * HZ; + unsigned long last_stamp = 0; + struct rb_node *p; + struct list_head slow_osds; + + dout("timeout\n"); + down_read(&osdc->map_sem); + + ceph_monc_request_next_osdmap(&osdc->client->monc); + + mutex_lock(&osdc->request_mutex); + for (p = rb_first(&osdc->requests); p; p = rb_next(p)) { + req = rb_entry(p, struct ceph_osd_request, r_node); + + if (req->r_resend) { + int err; + + dout("osdc resending prev failed %lld\n", req->r_tid); + err = __send_request(osdc, req); + if (err) + dout("osdc failed again on %lld\n", req->r_tid); + else + req->r_resend = false; + continue; + } + } + + /* + * reset osds that appear to be _really_ unresponsive. this + * is a failsafe measure.. we really shouldn't be getting to + * this point if the system is working properly. the monitors + * should mark the osd as failed and we should find out about + * it from an updated osd map. + */ + while (timeout && !list_empty(&osdc->req_lru)) { + req = list_entry(osdc->req_lru.next, struct ceph_osd_request, + r_req_lru_item); + + if (time_before(jiffies, req->r_stamp + timeout)) + break; + + BUG_ON(req == last_req && req->r_stamp == last_stamp); + last_req = req; + last_stamp = req->r_stamp; + + osd = req->r_osd; + BUG_ON(!osd); + pr_warning(" tid %llu timed out on osd%d, will reset osd\n", + req->r_tid, osd->o_osd); + __kick_requests(osdc, osd); + } + + /* + * ping osds that are a bit slow. this ensures that if there + * is a break in the TCP connection we will notice, and reopen + * a connection with that osd (from the fault callback). + */ + INIT_LIST_HEAD(&slow_osds); + list_for_each_entry(req, &osdc->req_lru, r_req_lru_item) { + if (time_before(jiffies, req->r_stamp + keepalive)) + break; + + osd = req->r_osd; + BUG_ON(!osd); + dout(" tid %llu is slow, will send keepalive on osd%d\n", + req->r_tid, osd->o_osd); + list_move_tail(&osd->o_keepalive_item, &slow_osds); + } + while (!list_empty(&slow_osds)) { + osd = list_entry(slow_osds.next, struct ceph_osd, + o_keepalive_item); + list_del_init(&osd->o_keepalive_item); + ceph_con_keepalive(&osd->o_con); + } + + __schedule_osd_timeout(osdc); + mutex_unlock(&osdc->request_mutex); + + up_read(&osdc->map_sem); +} + +static void handle_osds_timeout(struct work_struct *work) +{ + struct ceph_osd_client *osdc = + container_of(work, struct ceph_osd_client, + osds_timeout_work.work); + unsigned long delay = + osdc->client->options->osd_idle_ttl * HZ >> 2; + + dout("osds timeout\n"); + down_read(&osdc->map_sem); + remove_old_osds(osdc, 0); + up_read(&osdc->map_sem); + + schedule_delayed_work(&osdc->osds_timeout_work, + round_jiffies_relative(delay)); +} + +/* + * handle osd op reply. either call the callback if it is specified, + * or do the completion to wake up the waiting thread. + */ +static void handle_reply(struct ceph_osd_client *osdc, struct ceph_msg *msg, + struct ceph_connection *con) +{ + struct ceph_osd_reply_head *rhead = msg->front.iov_base; + struct ceph_osd_request *req; + u64 tid; + int numops, object_len, flags; + s32 result; + + tid = le64_to_cpu(msg->hdr.tid); + if (msg->front.iov_len < sizeof(*rhead)) + goto bad; + numops = le32_to_cpu(rhead->num_ops); + object_len = le32_to_cpu(rhead->object_len); + result = le32_to_cpu(rhead->result); + if (msg->front.iov_len != sizeof(*rhead) + object_len + + numops * sizeof(struct ceph_osd_op)) + goto bad; + dout("handle_reply %p tid %llu result %d\n", msg, tid, (int)result); + + /* lookup */ + mutex_lock(&osdc->request_mutex); + req = __lookup_request(osdc, tid); + if (req == NULL) { + dout("handle_reply tid %llu dne\n", tid); + mutex_unlock(&osdc->request_mutex); + return; + } + ceph_osdc_get_request(req); + flags = le32_to_cpu(rhead->flags); + + /* + * if this connection filled our message, drop our reference now, to + * avoid a (safe but slower) revoke later. + */ + if (req->r_con_filling_msg == con && req->r_reply == msg) { + dout(" dropping con_filling_msg ref %p\n", con); + req->r_con_filling_msg = NULL; + ceph_con_put(con); + } + + if (!req->r_got_reply) { + unsigned bytes; + + req->r_result = le32_to_cpu(rhead->result); + bytes = le32_to_cpu(msg->hdr.data_len); + dout("handle_reply result %d bytes %d\n", req->r_result, + bytes); + if (req->r_result == 0) + req->r_result = bytes; + + /* in case this is a write and we need to replay, */ + req->r_reassert_version = rhead->reassert_version; + + req->r_got_reply = 1; + } else if ((flags & CEPH_OSD_FLAG_ONDISK) == 0) { + dout("handle_reply tid %llu dup ack\n", tid); + mutex_unlock(&osdc->request_mutex); + goto done; + } + + dout("handle_reply tid %llu flags %d\n", tid, flags); + + /* either this is a read, or we got the safe response */ + if (result < 0 || + (flags & CEPH_OSD_FLAG_ONDISK) || + ((flags & CEPH_OSD_FLAG_WRITE) == 0)) + __unregister_request(osdc, req); + + mutex_unlock(&osdc->request_mutex); + + if (req->r_callback) + req->r_callback(req, msg); + else + complete_all(&req->r_completion); + + if (flags & CEPH_OSD_FLAG_ONDISK) { + if (req->r_safe_callback) + req->r_safe_callback(req, msg); + complete_all(&req->r_safe_completion); /* fsync waiter */ + } + +done: + ceph_osdc_put_request(req); + return; + +bad: + pr_err("corrupt osd_op_reply got %d %d expected %d\n", + (int)msg->front.iov_len, le32_to_cpu(msg->hdr.front_len), + (int)sizeof(*rhead)); + ceph_msg_dump(msg); +} + + +static int __kick_requests(struct ceph_osd_client *osdc, + struct ceph_osd *kickosd) +{ + struct ceph_osd_request *req; + struct rb_node *p, *n; + int needmap = 0; + int err; + + dout("kick_requests osd%d\n", kickosd ? kickosd->o_osd : -1); + if (kickosd) { + err = __reset_osd(osdc, kickosd); + if (err == -EAGAIN) + return 1; + } else { + for (p = rb_first(&osdc->osds); p; p = n) { + struct ceph_osd *osd = + rb_entry(p, struct ceph_osd, o_node); + + n = rb_next(p); + if (!ceph_osd_is_up(osdc->osdmap, osd->o_osd) || + memcmp(&osd->o_con.peer_addr, + ceph_osd_addr(osdc->osdmap, + osd->o_osd), + sizeof(struct ceph_entity_addr)) != 0) + __reset_osd(osdc, osd); + } + } + + for (p = rb_first(&osdc->requests); p; p = rb_next(p)) { + req = rb_entry(p, struct ceph_osd_request, r_node); + + if (req->r_resend) { + dout(" r_resend set on tid %llu\n", req->r_tid); + __cancel_request(req); + goto kick; + } + if (req->r_osd && kickosd == req->r_osd) { + __cancel_request(req); + goto kick; + } + + err = __map_osds(osdc, req); + if (err == 0) + continue; /* no change */ + if (err < 0) { + /* + * FIXME: really, we should set the request + * error and fail if this isn't a 'nofail' + * request, but that's a fair bit more + * complicated to do. So retry! + */ + dout(" setting r_resend on %llu\n", req->r_tid); + req->r_resend = true; + continue; + } + if (req->r_osd == NULL) { + dout("tid %llu maps to no valid osd\n", req->r_tid); + needmap++; /* request a newer map */ + continue; + } + +kick: + dout("kicking %p tid %llu osd%d\n", req, req->r_tid, + req->r_osd ? req->r_osd->o_osd : -1); + req->r_flags |= CEPH_OSD_FLAG_RETRY; + err = __send_request(osdc, req); + if (err) { + dout(" setting r_resend on %llu\n", req->r_tid); + req->r_resend = true; + } + } + + return needmap; +} + +/* + * Resubmit osd requests whose osd or osd address has changed. Request + * a new osd map if osds are down, or we are otherwise unable to determine + * how to direct a request. + * + * Close connections to down osds. + * + * If @who is specified, resubmit requests for that specific osd. + * + * Caller should hold map_sem for read and request_mutex. + */ +static void kick_requests(struct ceph_osd_client *osdc, + struct ceph_osd *kickosd) +{ + int needmap; + + mutex_lock(&osdc->request_mutex); + needmap = __kick_requests(osdc, kickosd); + mutex_unlock(&osdc->request_mutex); + + if (needmap) { + dout("%d requests for down osds, need new map\n", needmap); + ceph_monc_request_next_osdmap(&osdc->client->monc); + } + +} +/* + * Process updated osd map. + * + * The message contains any number of incremental and full maps, normally + * indicating some sort of topology change in the cluster. Kick requests + * off to different OSDs as needed. + */ +void ceph_osdc_handle_map(struct ceph_osd_client *osdc, struct ceph_msg *msg) +{ + void *p, *end, *next; + u32 nr_maps, maplen; + u32 epoch; + struct ceph_osdmap *newmap = NULL, *oldmap; + int err; + struct ceph_fsid fsid; + + dout("handle_map have %u\n", osdc->osdmap ? osdc->osdmap->epoch : 0); + p = msg->front.iov_base; + end = p + msg->front.iov_len; + + /* verify fsid */ + ceph_decode_need(&p, end, sizeof(fsid), bad); + ceph_decode_copy(&p, &fsid, sizeof(fsid)); + if (ceph_check_fsid(osdc->client, &fsid) < 0) + return; + + down_write(&osdc->map_sem); + + /* incremental maps */ + ceph_decode_32_safe(&p, end, nr_maps, bad); + dout(" %d inc maps\n", nr_maps); + while (nr_maps > 0) { + ceph_decode_need(&p, end, 2*sizeof(u32), bad); + epoch = ceph_decode_32(&p); + maplen = ceph_decode_32(&p); + ceph_decode_need(&p, end, maplen, bad); + next = p + maplen; + if (osdc->osdmap && osdc->osdmap->epoch+1 == epoch) { + dout("applying incremental map %u len %d\n", + epoch, maplen); + newmap = osdmap_apply_incremental(&p, next, + osdc->osdmap, + osdc->client->msgr); + if (IS_ERR(newmap)) { + err = PTR_ERR(newmap); + goto bad; + } + BUG_ON(!newmap); + if (newmap != osdc->osdmap) { + ceph_osdmap_destroy(osdc->osdmap); + osdc->osdmap = newmap; + } + } else { + dout("ignoring incremental map %u len %d\n", + epoch, maplen); + } + p = next; + nr_maps--; + } + if (newmap) + goto done; + + /* full maps */ + ceph_decode_32_safe(&p, end, nr_maps, bad); + dout(" %d full maps\n", nr_maps); + while (nr_maps) { + ceph_decode_need(&p, end, 2*sizeof(u32), bad); + epoch = ceph_decode_32(&p); + maplen = ceph_decode_32(&p); + ceph_decode_need(&p, end, maplen, bad); + if (nr_maps > 1) { + dout("skipping non-latest full map %u len %d\n", + epoch, maplen); + } else if (osdc->osdmap && osdc->osdmap->epoch >= epoch) { + dout("skipping full map %u len %d, " + "older than our %u\n", epoch, maplen, + osdc->osdmap->epoch); + } else { + dout("taking full map %u len %d\n", epoch, maplen); + newmap = osdmap_decode(&p, p+maplen); + if (IS_ERR(newmap)) { + err = PTR_ERR(newmap); + goto bad; + } + BUG_ON(!newmap); + oldmap = osdc->osdmap; + osdc->osdmap = newmap; + if (oldmap) + ceph_osdmap_destroy(oldmap); + } + p += maplen; + nr_maps--; + } + +done: + downgrade_write(&osdc->map_sem); + ceph_monc_got_osdmap(&osdc->client->monc, osdc->osdmap->epoch); + if (newmap) + kick_requests(osdc, NULL); + up_read(&osdc->map_sem); + wake_up_all(&osdc->client->auth_wq); + return; + +bad: + pr_err("osdc handle_map corrupt msg\n"); + ceph_msg_dump(msg); + up_write(&osdc->map_sem); + return; +} + +/* + * Register request, send initial attempt. + */ +int ceph_osdc_start_request(struct ceph_osd_client *osdc, + struct ceph_osd_request *req, + bool nofail) +{ + int rc = 0; + + req->r_request->pages = req->r_pages; + req->r_request->nr_pages = req->r_num_pages; +#ifdef CONFIG_BLOCK + req->r_request->bio = req->r_bio; +#endif + req->r_request->trail = req->r_trail; + + register_request(osdc, req); + + down_read(&osdc->map_sem); + mutex_lock(&osdc->request_mutex); + /* + * a racing kick_requests() may have sent the message for us + * while we dropped request_mutex above, so only send now if + * the request still han't been touched yet. + */ + if (req->r_sent == 0) { + rc = __send_request(osdc, req); + if (rc) { + if (nofail) { + dout("osdc_start_request failed send, " + " marking %lld\n", req->r_tid); + req->r_resend = true; + rc = 0; + } else { + __unregister_request(osdc, req); + } + } + } + mutex_unlock(&osdc->request_mutex); + up_read(&osdc->map_sem); + return rc; +} +EXPORT_SYMBOL(ceph_osdc_start_request); + +/* + * wait for a request to complete + */ +int ceph_osdc_wait_request(struct ceph_osd_client *osdc, + struct ceph_osd_request *req) +{ + int rc; + + rc = wait_for_completion_interruptible(&req->r_completion); + if (rc < 0) { + mutex_lock(&osdc->request_mutex); + __cancel_request(req); + __unregister_request(osdc, req); + mutex_unlock(&osdc->request_mutex); + dout("wait_request tid %llu canceled/timed out\n", req->r_tid); + return rc; + } + + dout("wait_request tid %llu result %d\n", req->r_tid, req->r_result); + return req->r_result; +} +EXPORT_SYMBOL(ceph_osdc_wait_request); + +/* + * sync - wait for all in-flight requests to flush. avoid starvation. + */ +void ceph_osdc_sync(struct ceph_osd_client *osdc) +{ + struct ceph_osd_request *req; + u64 last_tid, next_tid = 0; + + mutex_lock(&osdc->request_mutex); + last_tid = osdc->last_tid; + while (1) { + req = __lookup_request_ge(osdc, next_tid); + if (!req) + break; + if (req->r_tid > last_tid) + break; + + next_tid = req->r_tid + 1; + if ((req->r_flags & CEPH_OSD_FLAG_WRITE) == 0) + continue; + + ceph_osdc_get_request(req); + mutex_unlock(&osdc->request_mutex); + dout("sync waiting on tid %llu (last is %llu)\n", + req->r_tid, last_tid); + wait_for_completion(&req->r_safe_completion); + mutex_lock(&osdc->request_mutex); + ceph_osdc_put_request(req); + } + mutex_unlock(&osdc->request_mutex); + dout("sync done (thru tid %llu)\n", last_tid); +} +EXPORT_SYMBOL(ceph_osdc_sync); + +/* + * init, shutdown + */ +int ceph_osdc_init(struct ceph_osd_client *osdc, struct ceph_client *client) +{ + int err; + + dout("init\n"); + osdc->client = client; + osdc->osdmap = NULL; + init_rwsem(&osdc->map_sem); + init_completion(&osdc->map_waiters); + osdc->last_requested_map = 0; + mutex_init(&osdc->request_mutex); + osdc->last_tid = 0; + osdc->osds = RB_ROOT; + INIT_LIST_HEAD(&osdc->osd_lru); + osdc->requests = RB_ROOT; + INIT_LIST_HEAD(&osdc->req_lru); + osdc->num_requests = 0; + INIT_DELAYED_WORK(&osdc->timeout_work, handle_timeout); + INIT_DELAYED_WORK(&osdc->osds_timeout_work, handle_osds_timeout); + + schedule_delayed_work(&osdc->osds_timeout_work, + round_jiffies_relative(osdc->client->options->osd_idle_ttl * HZ)); + + err = -ENOMEM; + osdc->req_mempool = mempool_create_kmalloc_pool(10, + sizeof(struct ceph_osd_request)); + if (!osdc->req_mempool) + goto out; + + err = ceph_msgpool_init(&osdc->msgpool_op, OSD_OP_FRONT_LEN, 10, true, + "osd_op"); + if (err < 0) + goto out_mempool; + err = ceph_msgpool_init(&osdc->msgpool_op_reply, + OSD_OPREPLY_FRONT_LEN, 10, true, + "osd_op_reply"); + if (err < 0) + goto out_msgpool; + return 0; + +out_msgpool: + ceph_msgpool_destroy(&osdc->msgpool_op); +out_mempool: + mempool_destroy(osdc->req_mempool); +out: + return err; +} +EXPORT_SYMBOL(ceph_osdc_init); + +void ceph_osdc_stop(struct ceph_osd_client *osdc) +{ + cancel_delayed_work_sync(&osdc->timeout_work); + cancel_delayed_work_sync(&osdc->osds_timeout_work); + if (osdc->osdmap) { + ceph_osdmap_destroy(osdc->osdmap); + osdc->osdmap = NULL; + } + remove_old_osds(osdc, 1); + mempool_destroy(osdc->req_mempool); + ceph_msgpool_destroy(&osdc->msgpool_op); + ceph_msgpool_destroy(&osdc->msgpool_op_reply); +} +EXPORT_SYMBOL(ceph_osdc_stop); + +/* + * Read some contiguous pages. If we cross a stripe boundary, shorten + * *plen. Return number of bytes read, or error. + */ +int ceph_osdc_readpages(struct ceph_osd_client *osdc, + struct ceph_vino vino, struct ceph_file_layout *layout, + u64 off, u64 *plen, + u32 truncate_seq, u64 truncate_size, + struct page **pages, int num_pages) +{ + struct ceph_osd_request *req; + int rc = 0; + + dout("readpages on ino %llx.%llx on %llu~%llu\n", vino.ino, + vino.snap, off, *plen); + req = ceph_osdc_new_request(osdc, layout, vino, off, plen, + CEPH_OSD_OP_READ, CEPH_OSD_FLAG_READ, + NULL, 0, truncate_seq, truncate_size, NULL, + false, 1); + if (!req) + return -ENOMEM; + + /* it may be a short read due to an object boundary */ + req->r_pages = pages; + + dout("readpages final extent is %llu~%llu (%d pages)\n", + off, *plen, req->r_num_pages); + + rc = ceph_osdc_start_request(osdc, req, false); + if (!rc) + rc = ceph_osdc_wait_request(osdc, req); + + ceph_osdc_put_request(req); + dout("readpages result %d\n", rc); + return rc; +} +EXPORT_SYMBOL(ceph_osdc_readpages); + +/* + * do a synchronous write on N pages + */ +int ceph_osdc_writepages(struct ceph_osd_client *osdc, struct ceph_vino vino, + struct ceph_file_layout *layout, + struct ceph_snap_context *snapc, + u64 off, u64 len, + u32 truncate_seq, u64 truncate_size, + struct timespec *mtime, + struct page **pages, int num_pages, + int flags, int do_sync, bool nofail) +{ + struct ceph_osd_request *req; + int rc = 0; + + BUG_ON(vino.snap != CEPH_NOSNAP); + req = ceph_osdc_new_request(osdc, layout, vino, off, &len, + CEPH_OSD_OP_WRITE, + flags | CEPH_OSD_FLAG_ONDISK | + CEPH_OSD_FLAG_WRITE, + snapc, do_sync, + truncate_seq, truncate_size, mtime, + nofail, 1); + if (!req) + return -ENOMEM; + + /* it may be a short write due to an object boundary */ + req->r_pages = pages; + dout("writepages %llu~%llu (%d pages)\n", off, len, + req->r_num_pages); + + rc = ceph_osdc_start_request(osdc, req, nofail); + if (!rc) + rc = ceph_osdc_wait_request(osdc, req); + + ceph_osdc_put_request(req); + if (rc == 0) + rc = len; + dout("writepages result %d\n", rc); + return rc; +} +EXPORT_SYMBOL(ceph_osdc_writepages); + +/* + * handle incoming message + */ +static void dispatch(struct ceph_connection *con, struct ceph_msg *msg) +{ + struct ceph_osd *osd = con->private; + struct ceph_osd_client *osdc; + int type = le16_to_cpu(msg->hdr.type); + + if (!osd) + goto out; + osdc = osd->o_osdc; + + switch (type) { + case CEPH_MSG_OSD_MAP: + ceph_osdc_handle_map(osdc, msg); + break; + case CEPH_MSG_OSD_OPREPLY: + handle_reply(osdc, msg, con); + break; + + default: + pr_err("received unknown message type %d %s\n", type, + ceph_msg_type_name(type)); + } +out: + ceph_msg_put(msg); +} + +/* + * lookup and return message for incoming reply. set up reply message + * pages. + */ +static struct ceph_msg *get_reply(struct ceph_connection *con, + struct ceph_msg_header *hdr, + int *skip) +{ + struct ceph_osd *osd = con->private; + struct ceph_osd_client *osdc = osd->o_osdc; + struct ceph_msg *m; + struct ceph_osd_request *req; + int front = le32_to_cpu(hdr->front_len); + int data_len = le32_to_cpu(hdr->data_len); + u64 tid; + + tid = le64_to_cpu(hdr->tid); + mutex_lock(&osdc->request_mutex); + req = __lookup_request(osdc, tid); + if (!req) { + *skip = 1; + m = NULL; + pr_info("get_reply unknown tid %llu from osd%d\n", tid, + osd->o_osd); + goto out; + } + + if (req->r_con_filling_msg) { + dout("get_reply revoking msg %p from old con %p\n", + req->r_reply, req->r_con_filling_msg); + ceph_con_revoke_message(req->r_con_filling_msg, req->r_reply); + ceph_con_put(req->r_con_filling_msg); + req->r_con_filling_msg = NULL; + } + + if (front > req->r_reply->front.iov_len) { + pr_warning("get_reply front %d > preallocated %d\n", + front, (int)req->r_reply->front.iov_len); + m = ceph_msg_new(CEPH_MSG_OSD_OPREPLY, front, GFP_NOFS); + if (!m) + goto out; + ceph_msg_put(req->r_reply); + req->r_reply = m; + } + m = ceph_msg_get(req->r_reply); + + if (data_len > 0) { + unsigned data_off = le16_to_cpu(hdr->data_off); + int want = calc_pages_for(data_off & ~PAGE_MASK, data_len); + + if (unlikely(req->r_num_pages < want)) { + pr_warning("tid %lld reply %d > expected %d pages\n", + tid, want, m->nr_pages); + *skip = 1; + ceph_msg_put(m); + m = NULL; + goto out; + } + m->pages = req->r_pages; + m->nr_pages = req->r_num_pages; +#ifdef CONFIG_BLOCK + m->bio = req->r_bio; +#endif + } + *skip = 0; + req->r_con_filling_msg = ceph_con_get(con); + dout("get_reply tid %lld %p\n", tid, m); + +out: + mutex_unlock(&osdc->request_mutex); + return m; + +} + +static struct ceph_msg *alloc_msg(struct ceph_connection *con, + struct ceph_msg_header *hdr, + int *skip) +{ + struct ceph_osd *osd = con->private; + int type = le16_to_cpu(hdr->type); + int front = le32_to_cpu(hdr->front_len); + + switch (type) { + case CEPH_MSG_OSD_MAP: + return ceph_msg_new(type, front, GFP_NOFS); + case CEPH_MSG_OSD_OPREPLY: + return get_reply(con, hdr, skip); + default: + pr_info("alloc_msg unexpected msg type %d from osd%d\n", type, + osd->o_osd); + *skip = 1; + return NULL; + } +} + +/* + * Wrappers to refcount containing ceph_osd struct + */ +static struct ceph_connection *get_osd_con(struct ceph_connection *con) +{ + struct ceph_osd *osd = con->private; + if (get_osd(osd)) + return con; + return NULL; +} + +static void put_osd_con(struct ceph_connection *con) +{ + struct ceph_osd *osd = con->private; + put_osd(osd); +} + +/* + * authentication + */ +static int get_authorizer(struct ceph_connection *con, + void **buf, int *len, int *proto, + void **reply_buf, int *reply_len, int force_new) +{ + struct ceph_osd *o = con->private; + struct ceph_osd_client *osdc = o->o_osdc; + struct ceph_auth_client *ac = osdc->client->monc.auth; + int ret = 0; + + if (force_new && o->o_authorizer) { + ac->ops->destroy_authorizer(ac, o->o_authorizer); + o->o_authorizer = NULL; + } + if (o->o_authorizer == NULL) { + ret = ac->ops->create_authorizer( + ac, CEPH_ENTITY_TYPE_OSD, + &o->o_authorizer, + &o->o_authorizer_buf, + &o->o_authorizer_buf_len, + &o->o_authorizer_reply_buf, + &o->o_authorizer_reply_buf_len); + if (ret) + return ret; + } + + *proto = ac->protocol; + *buf = o->o_authorizer_buf; + *len = o->o_authorizer_buf_len; + *reply_buf = o->o_authorizer_reply_buf; + *reply_len = o->o_authorizer_reply_buf_len; + return 0; +} + + +static int verify_authorizer_reply(struct ceph_connection *con, int len) +{ + struct ceph_osd *o = con->private; + struct ceph_osd_client *osdc = o->o_osdc; + struct ceph_auth_client *ac = osdc->client->monc.auth; + + return ac->ops->verify_authorizer_reply(ac, o->o_authorizer, len); +} + +static int invalidate_authorizer(struct ceph_connection *con) +{ + struct ceph_osd *o = con->private; + struct ceph_osd_client *osdc = o->o_osdc; + struct ceph_auth_client *ac = osdc->client->monc.auth; + + if (ac->ops->invalidate_authorizer) + ac->ops->invalidate_authorizer(ac, CEPH_ENTITY_TYPE_OSD); + + return ceph_monc_validate_auth(&osdc->client->monc); +} + +static const struct ceph_connection_operations osd_con_ops = { + .get = get_osd_con, + .put = put_osd_con, + .dispatch = dispatch, + .get_authorizer = get_authorizer, + .verify_authorizer_reply = verify_authorizer_reply, + .invalidate_authorizer = invalidate_authorizer, + .alloc_msg = alloc_msg, + .fault = osd_reset, +}; diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c new file mode 100644 index 00000000000..d73f3f6efa3 --- /dev/null +++ b/net/ceph/osdmap.c @@ -0,0 +1,1128 @@ + +#include <linux/ceph/ceph_debug.h> + +#include <linux/module.h> +#include <linux/slab.h> +#include <asm/div64.h> + +#include <linux/ceph/libceph.h> +#include <linux/ceph/osdmap.h> +#include <linux/ceph/decode.h> +#include <linux/crush/hash.h> +#include <linux/crush/mapper.h> + +char *ceph_osdmap_state_str(char *str, int len, int state) +{ + int flag = 0; + + if (!len) + goto done; + + *str = '\0'; + if (state) { + if (state & CEPH_OSD_EXISTS) { + snprintf(str, len, "exists"); + flag = 1; + } + if (state & CEPH_OSD_UP) { + snprintf(str, len, "%s%s%s", str, (flag ? ", " : ""), + "up"); + flag = 1; + } + } else { + snprintf(str, len, "doesn't exist"); + } +done: + return str; +} + +/* maps */ + +static int calc_bits_of(unsigned t) +{ + int b = 0; + while (t) { + t = t >> 1; + b++; + } + return b; +} + +/* + * the foo_mask is the smallest value 2^n-1 that is >= foo. + */ +static void calc_pg_masks(struct ceph_pg_pool_info *pi) +{ + pi->pg_num_mask = (1 << calc_bits_of(le32_to_cpu(pi->v.pg_num)-1)) - 1; + pi->pgp_num_mask = + (1 << calc_bits_of(le32_to_cpu(pi->v.pgp_num)-1)) - 1; + pi->lpg_num_mask = + (1 << calc_bits_of(le32_to_cpu(pi->v.lpg_num)-1)) - 1; + pi->lpgp_num_mask = + (1 << calc_bits_of(le32_to_cpu(pi->v.lpgp_num)-1)) - 1; +} + +/* + * decode crush map + */ +static int crush_decode_uniform_bucket(void **p, void *end, + struct crush_bucket_uniform *b) +{ + dout("crush_decode_uniform_bucket %p to %p\n", *p, end); + ceph_decode_need(p, end, (1+b->h.size) * sizeof(u32), bad); + b->item_weight = ceph_decode_32(p); + return 0; +bad: + return -EINVAL; +} + +static int crush_decode_list_bucket(void **p, void *end, + struct crush_bucket_list *b) +{ + int j; + dout("crush_decode_list_bucket %p to %p\n", *p, end); + b->item_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS); + if (b->item_weights == NULL) + return -ENOMEM; + b->sum_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS); + if (b->sum_weights == NULL) + return -ENOMEM; + ceph_decode_need(p, end, 2 * b->h.size * sizeof(u32), bad); + for (j = 0; j < b->h.size; j++) { + b->item_weights[j] = ceph_decode_32(p); + b->sum_weights[j] = ceph_decode_32(p); + } + return 0; +bad: + return -EINVAL; +} + +static int crush_decode_tree_bucket(void **p, void *end, + struct crush_bucket_tree *b) +{ + int j; + dout("crush_decode_tree_bucket %p to %p\n", *p, end); + ceph_decode_32_safe(p, end, b->num_nodes, bad); + b->node_weights = kcalloc(b->num_nodes, sizeof(u32), GFP_NOFS); + if (b->node_weights == NULL) + return -ENOMEM; + ceph_decode_need(p, end, b->num_nodes * sizeof(u32), bad); + for (j = 0; j < b->num_nodes; j++) + b->node_weights[j] = ceph_decode_32(p); + return 0; +bad: + return -EINVAL; +} + +static int crush_decode_straw_bucket(void **p, void *end, + struct crush_bucket_straw *b) +{ + int j; + dout("crush_decode_straw_bucket %p to %p\n", *p, end); + b->item_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS); + if (b->item_weights == NULL) + return -ENOMEM; + b->straws = kcalloc(b->h.size, sizeof(u32), GFP_NOFS); + if (b->straws == NULL) + return -ENOMEM; + ceph_decode_need(p, end, 2 * b->h.size * sizeof(u32), bad); + for (j = 0; j < b->h.size; j++) { + b->item_weights[j] = ceph_decode_32(p); + b->straws[j] = ceph_decode_32(p); + } + return 0; +bad: + return -EINVAL; +} + +static struct crush_map *crush_decode(void *pbyval, void *end) +{ + struct crush_map *c; + int err = -EINVAL; + int i, j; + void **p = &pbyval; + void *start = pbyval; + u32 magic; + + dout("crush_decode %p to %p len %d\n", *p, end, (int)(end - *p)); + + c = kzalloc(sizeof(*c), GFP_NOFS); + if (c == NULL) + return ERR_PTR(-ENOMEM); + + ceph_decode_need(p, end, 4*sizeof(u32), bad); + magic = ceph_decode_32(p); + if (magic != CRUSH_MAGIC) { + pr_err("crush_decode magic %x != current %x\n", + (unsigned)magic, (unsigned)CRUSH_MAGIC); + goto bad; + } + c->max_buckets = ceph_decode_32(p); + c->max_rules = ceph_decode_32(p); + c->max_devices = ceph_decode_32(p); + + c->device_parents = kcalloc(c->max_devices, sizeof(u32), GFP_NOFS); + if (c->device_parents == NULL) + goto badmem; + c->bucket_parents = kcalloc(c->max_buckets, sizeof(u32), GFP_NOFS); + if (c->bucket_parents == NULL) + goto badmem; + + c->buckets = kcalloc(c->max_buckets, sizeof(*c->buckets), GFP_NOFS); + if (c->buckets == NULL) + goto badmem; + c->rules = kcalloc(c->max_rules, sizeof(*c->rules), GFP_NOFS); + if (c->rules == NULL) + goto badmem; + + /* buckets */ + for (i = 0; i < c->max_buckets; i++) { + int size = 0; + u32 alg; + struct crush_bucket *b; + + ceph_decode_32_safe(p, end, alg, bad); + if (alg == 0) { + c->buckets[i] = NULL; + continue; + } + dout("crush_decode bucket %d off %x %p to %p\n", + i, (int)(*p-start), *p, end); + + switch (alg) { + case CRUSH_BUCKET_UNIFORM: + size = sizeof(struct crush_bucket_uniform); + break; + case CRUSH_BUCKET_LIST: + size = sizeof(struct crush_bucket_list); + break; + case CRUSH_BUCKET_TREE: + size = sizeof(struct crush_bucket_tree); + break; + case CRUSH_BUCKET_STRAW: + size = sizeof(struct crush_bucket_straw); + break; + default: + err = -EINVAL; + goto bad; + } + BUG_ON(size == 0); + b = c->buckets[i] = kzalloc(size, GFP_NOFS); + if (b == NULL) + goto badmem; + + ceph_decode_need(p, end, 4*sizeof(u32), bad); + b->id = ceph_decode_32(p); + b->type = ceph_decode_16(p); + b->alg = ceph_decode_8(p); + b->hash = ceph_decode_8(p); + b->weight = ceph_decode_32(p); + b->size = ceph_decode_32(p); + + dout("crush_decode bucket size %d off %x %p to %p\n", + b->size, (int)(*p-start), *p, end); + + b->items = kcalloc(b->size, sizeof(__s32), GFP_NOFS); + if (b->items == NULL) + goto badmem; + b->perm = kcalloc(b->size, sizeof(u32), GFP_NOFS); + if (b->perm == NULL) + goto badmem; + b->perm_n = 0; + + ceph_decode_need(p, end, b->size*sizeof(u32), bad); + for (j = 0; j < b->size; j++) + b->items[j] = ceph_decode_32(p); + + switch (b->alg) { + case CRUSH_BUCKET_UNIFORM: + err = crush_decode_uniform_bucket(p, end, + (struct crush_bucket_uniform *)b); + if (err < 0) + goto bad; + break; + case CRUSH_BUCKET_LIST: + err = crush_decode_list_bucket(p, end, + (struct crush_bucket_list *)b); + if (err < 0) + goto bad; + break; + case CRUSH_BUCKET_TREE: + err = crush_decode_tree_bucket(p, end, + (struct crush_bucket_tree *)b); + if (err < 0) + goto bad; + break; + case CRUSH_BUCKET_STRAW: + err = crush_decode_straw_bucket(p, end, + (struct crush_bucket_straw *)b); + if (err < 0) + goto bad; + break; + } + } + + /* rules */ + dout("rule vec is %p\n", c->rules); + for (i = 0; i < c->max_rules; i++) { + u32 yes; + struct crush_rule *r; + + ceph_decode_32_safe(p, end, yes, bad); + if (!yes) { + dout("crush_decode NO rule %d off %x %p to %p\n", + i, (int)(*p-start), *p, end); + c->rules[i] = NULL; + continue; + } + + dout("crush_decode rule %d off %x %p to %p\n", + i, (int)(*p-start), *p, end); + + /* len */ + ceph_decode_32_safe(p, end, yes, bad); +#if BITS_PER_LONG == 32 + err = -EINVAL; + if (yes > ULONG_MAX / sizeof(struct crush_rule_step)) + goto bad; +#endif + r = c->rules[i] = kmalloc(sizeof(*r) + + yes*sizeof(struct crush_rule_step), + GFP_NOFS); + if (r == NULL) + goto badmem; + dout(" rule %d is at %p\n", i, r); + r->len = yes; + ceph_decode_copy_safe(p, end, &r->mask, 4, bad); /* 4 u8's */ + ceph_decode_need(p, end, r->len*3*sizeof(u32), bad); + for (j = 0; j < r->len; j++) { + r->steps[j].op = ceph_decode_32(p); + r->steps[j].arg1 = ceph_decode_32(p); + r->steps[j].arg2 = ceph_decode_32(p); + } + } + + /* ignore trailing name maps. */ + + dout("crush_decode success\n"); + return c; + +badmem: + err = -ENOMEM; +bad: + dout("crush_decode fail %d\n", err); + crush_destroy(c); + return ERR_PTR(err); +} + +/* + * rbtree of pg_mapping for handling pg_temp (explicit mapping of pgid + * to a set of osds) + */ +static int pgid_cmp(struct ceph_pg l, struct ceph_pg r) +{ + u64 a = *(u64 *)&l; + u64 b = *(u64 *)&r; + + if (a < b) + return -1; + if (a > b) + return 1; + return 0; +} + +static int __insert_pg_mapping(struct ceph_pg_mapping *new, + struct rb_root *root) +{ + struct rb_node **p = &root->rb_node; + struct rb_node *parent = NULL; + struct ceph_pg_mapping *pg = NULL; + int c; + + while (*p) { + parent = *p; + pg = rb_entry(parent, struct ceph_pg_mapping, node); + c = pgid_cmp(new->pgid, pg->pgid); + if (c < 0) + p = &(*p)->rb_left; + else if (c > 0) + p = &(*p)->rb_right; + else + return -EEXIST; + } + + rb_link_node(&new->node, parent, p); + rb_insert_color(&new->node, root); + return 0; +} + +static struct ceph_pg_mapping *__lookup_pg_mapping(struct rb_root *root, + struct ceph_pg pgid) +{ + struct rb_node *n = root->rb_node; + struct ceph_pg_mapping *pg; + int c; + + while (n) { + pg = rb_entry(n, struct ceph_pg_mapping, node); + c = pgid_cmp(pgid, pg->pgid); + if (c < 0) + n = n->rb_left; + else if (c > 0) + n = n->rb_right; + else + return pg; + } + return NULL; +} + +/* + * rbtree of pg pool info + */ +static int __insert_pg_pool(struct rb_root *root, struct ceph_pg_pool_info *new) +{ + struct rb_node **p = &root->rb_node; + struct rb_node *parent = NULL; + struct ceph_pg_pool_info *pi = NULL; + + while (*p) { + parent = *p; + pi = rb_entry(parent, struct ceph_pg_pool_info, node); + if (new->id < pi->id) + p = &(*p)->rb_left; + else if (new->id > pi->id) + p = &(*p)->rb_right; + else + return -EEXIST; + } + + rb_link_node(&new->node, parent, p); + rb_insert_color(&new->node, root); + return 0; +} + +static struct ceph_pg_pool_info *__lookup_pg_pool(struct rb_root *root, int id) +{ + struct ceph_pg_pool_info *pi; + struct rb_node *n = root->rb_node; + + while (n) { + pi = rb_entry(n, struct ceph_pg_pool_info, node); + if (id < pi->id) + n = n->rb_left; + else if (id > pi->id) + n = n->rb_right; + else + return pi; + } + return NULL; +} + +int ceph_pg_poolid_by_name(struct ceph_osdmap *map, const char *name) +{ + struct rb_node *rbp; + + for (rbp = rb_first(&map->pg_pools); rbp; rbp = rb_next(rbp)) { + struct ceph_pg_pool_info *pi = + rb_entry(rbp, struct ceph_pg_pool_info, node); + if (pi->name && strcmp(pi->name, name) == 0) + return pi->id; + } + return -ENOENT; +} +EXPORT_SYMBOL(ceph_pg_poolid_by_name); + +static void __remove_pg_pool(struct rb_root *root, struct ceph_pg_pool_info *pi) +{ + rb_erase(&pi->node, root); + kfree(pi->name); + kfree(pi); +} + +static int __decode_pool(void **p, void *end, struct ceph_pg_pool_info *pi) +{ + unsigned n, m; + + ceph_decode_copy(p, &pi->v, sizeof(pi->v)); + calc_pg_masks(pi); + + /* num_snaps * snap_info_t */ + n = le32_to_cpu(pi->v.num_snaps); + while (n--) { + ceph_decode_need(p, end, sizeof(u64) + 1 + sizeof(u64) + + sizeof(struct ceph_timespec), bad); + *p += sizeof(u64) + /* key */ + 1 + sizeof(u64) + /* u8, snapid */ + sizeof(struct ceph_timespec); + m = ceph_decode_32(p); /* snap name */ + *p += m; + } + + *p += le32_to_cpu(pi->v.num_removed_snap_intervals) * sizeof(u64) * 2; + return 0; + +bad: + return -EINVAL; +} + +static int __decode_pool_names(void **p, void *end, struct ceph_osdmap *map) +{ + struct ceph_pg_pool_info *pi; + u32 num, len, pool; + + ceph_decode_32_safe(p, end, num, bad); + dout(" %d pool names\n", num); + while (num--) { + ceph_decode_32_safe(p, end, pool, bad); + ceph_decode_32_safe(p, end, len, bad); + dout(" pool %d len %d\n", pool, len); + pi = __lookup_pg_pool(&map->pg_pools, pool); + if (pi) { + kfree(pi->name); + pi->name = kmalloc(len + 1, GFP_NOFS); + if (pi->name) { + memcpy(pi->name, *p, len); + pi->name[len] = '\0'; + dout(" name is %s\n", pi->name); + } + } + *p += len; + } + return 0; + +bad: + return -EINVAL; +} + +/* + * osd map + */ +void ceph_osdmap_destroy(struct ceph_osdmap *map) +{ + dout("osdmap_destroy %p\n", map); + if (map->crush) + crush_destroy(map->crush); + while (!RB_EMPTY_ROOT(&map->pg_temp)) { + struct ceph_pg_mapping *pg = + rb_entry(rb_first(&map->pg_temp), + struct ceph_pg_mapping, node); + rb_erase(&pg->node, &map->pg_temp); + kfree(pg); + } + while (!RB_EMPTY_ROOT(&map->pg_pools)) { + struct ceph_pg_pool_info *pi = + rb_entry(rb_first(&map->pg_pools), + struct ceph_pg_pool_info, node); + __remove_pg_pool(&map->pg_pools, pi); + } + kfree(map->osd_state); + kfree(map->osd_weight); + kfree(map->osd_addr); + kfree(map); +} + +/* + * adjust max osd value. reallocate arrays. + */ +static int osdmap_set_max_osd(struct ceph_osdmap *map, int max) +{ + u8 *state; + struct ceph_entity_addr *addr; + u32 *weight; + + state = kcalloc(max, sizeof(*state), GFP_NOFS); + addr = kcalloc(max, sizeof(*addr), GFP_NOFS); + weight = kcalloc(max, sizeof(*weight), GFP_NOFS); + if (state == NULL || addr == NULL || weight == NULL) { + kfree(state); + kfree(addr); + kfree(weight); + return -ENOMEM; + } + + /* copy old? */ + if (map->osd_state) { + memcpy(state, map->osd_state, map->max_osd*sizeof(*state)); + memcpy(addr, map->osd_addr, map->max_osd*sizeof(*addr)); + memcpy(weight, map->osd_weight, map->max_osd*sizeof(*weight)); + kfree(map->osd_state); + kfree(map->osd_addr); + kfree(map->osd_weight); + } + + map->osd_state = state; + map->osd_weight = weight; + map->osd_addr = addr; + map->max_osd = max; + return 0; +} + +/* + * decode a full map. + */ +struct ceph_osdmap *osdmap_decode(void **p, void *end) +{ + struct ceph_osdmap *map; + u16 version; + u32 len, max, i; + u8 ev; + int err = -EINVAL; + void *start = *p; + struct ceph_pg_pool_info *pi; + + dout("osdmap_decode %p to %p len %d\n", *p, end, (int)(end - *p)); + + map = kzalloc(sizeof(*map), GFP_NOFS); + if (map == NULL) + return ERR_PTR(-ENOMEM); + map->pg_temp = RB_ROOT; + + ceph_decode_16_safe(p, end, version, bad); + if (version > CEPH_OSDMAP_VERSION) { + pr_warning("got unknown v %d > %d of osdmap\n", version, + CEPH_OSDMAP_VERSION); + goto bad; + } + + ceph_decode_need(p, end, 2*sizeof(u64)+6*sizeof(u32), bad); + ceph_decode_copy(p, &map->fsid, sizeof(map->fsid)); + map->epoch = ceph_decode_32(p); + ceph_decode_copy(p, &map->created, sizeof(map->created)); + ceph_decode_copy(p, &map->modified, sizeof(map->modified)); + + ceph_decode_32_safe(p, end, max, bad); + while (max--) { + ceph_decode_need(p, end, 4 + 1 + sizeof(pi->v), bad); + pi = kzalloc(sizeof(*pi), GFP_NOFS); + if (!pi) + goto bad; + pi->id = ceph_decode_32(p); + ev = ceph_decode_8(p); /* encoding version */ + if (ev > CEPH_PG_POOL_VERSION) { + pr_warning("got unknown v %d > %d of ceph_pg_pool\n", + ev, CEPH_PG_POOL_VERSION); + kfree(pi); + goto bad; + } + err = __decode_pool(p, end, pi); + if (err < 0) + goto bad; + __insert_pg_pool(&map->pg_pools, pi); + } + + if (version >= 5 && __decode_pool_names(p, end, map) < 0) + goto bad; + + ceph_decode_32_safe(p, end, map->pool_max, bad); + + ceph_decode_32_safe(p, end, map->flags, bad); + + max = ceph_decode_32(p); + + /* (re)alloc osd arrays */ + err = osdmap_set_max_osd(map, max); + if (err < 0) + goto bad; + dout("osdmap_decode max_osd = %d\n", map->max_osd); + + /* osds */ + err = -EINVAL; + ceph_decode_need(p, end, 3*sizeof(u32) + + map->max_osd*(1 + sizeof(*map->osd_weight) + + sizeof(*map->osd_addr)), bad); + *p += 4; /* skip length field (should match max) */ + ceph_decode_copy(p, map->osd_state, map->max_osd); + + *p += 4; /* skip length field (should match max) */ + for (i = 0; i < map->max_osd; i++) + map->osd_weight[i] = ceph_decode_32(p); + + *p += 4; /* skip length field (should match max) */ + ceph_decode_copy(p, map->osd_addr, map->max_osd*sizeof(*map->osd_addr)); + for (i = 0; i < map->max_osd; i++) + ceph_decode_addr(&map->osd_addr[i]); + + /* pg_temp */ + ceph_decode_32_safe(p, end, len, bad); + for (i = 0; i < len; i++) { + int n, j; + struct ceph_pg pgid; + struct ceph_pg_mapping *pg; + + ceph_decode_need(p, end, sizeof(u32) + sizeof(u64), bad); + ceph_decode_copy(p, &pgid, sizeof(pgid)); + n = ceph_decode_32(p); + ceph_decode_need(p, end, n * sizeof(u32), bad); + err = -ENOMEM; + pg = kmalloc(sizeof(*pg) + n*sizeof(u32), GFP_NOFS); + if (!pg) + goto bad; + pg->pgid = pgid; + pg->len = n; + for (j = 0; j < n; j++) + pg->osds[j] = ceph_decode_32(p); + + err = __insert_pg_mapping(pg, &map->pg_temp); + if (err) + goto bad; + dout(" added pg_temp %llx len %d\n", *(u64 *)&pgid, len); + } + + /* crush */ + ceph_decode_32_safe(p, end, len, bad); + dout("osdmap_decode crush len %d from off 0x%x\n", len, + (int)(*p - start)); + ceph_decode_need(p, end, len, bad); + map->crush = crush_decode(*p, end); + *p += len; + if (IS_ERR(map->crush)) { + err = PTR_ERR(map->crush); + map->crush = NULL; + goto bad; + } + + /* ignore the rest of the map */ + *p = end; + + dout("osdmap_decode done %p %p\n", *p, end); + return map; + +bad: + dout("osdmap_decode fail\n"); + ceph_osdmap_destroy(map); + return ERR_PTR(err); +} + +/* + * decode and apply an incremental map update. + */ +struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end, + struct ceph_osdmap *map, + struct ceph_messenger *msgr) +{ + struct crush_map *newcrush = NULL; + struct ceph_fsid fsid; + u32 epoch = 0; + struct ceph_timespec modified; + u32 len, pool; + __s32 new_pool_max, new_flags, max; + void *start = *p; + int err = -EINVAL; + u16 version; + struct rb_node *rbp; + + ceph_decode_16_safe(p, end, version, bad); + if (version > CEPH_OSDMAP_INC_VERSION) { + pr_warning("got unknown v %d > %d of inc osdmap\n", version, + CEPH_OSDMAP_INC_VERSION); + goto bad; + } + + ceph_decode_need(p, end, sizeof(fsid)+sizeof(modified)+2*sizeof(u32), + bad); + ceph_decode_copy(p, &fsid, sizeof(fsid)); + epoch = ceph_decode_32(p); + BUG_ON(epoch != map->epoch+1); + ceph_decode_copy(p, &modified, sizeof(modified)); + new_pool_max = ceph_decode_32(p); + new_flags = ceph_decode_32(p); + + /* full map? */ + ceph_decode_32_safe(p, end, len, bad); + if (len > 0) { + dout("apply_incremental full map len %d, %p to %p\n", + len, *p, end); + return osdmap_decode(p, min(*p+len, end)); + } + + /* new crush? */ + ceph_decode_32_safe(p, end, len, bad); + if (len > 0) { + dout("apply_incremental new crush map len %d, %p to %p\n", + len, *p, end); + newcrush = crush_decode(*p, min(*p+len, end)); + if (IS_ERR(newcrush)) + return ERR_CAST(newcrush); + *p += len; + } + + /* new flags? */ + if (new_flags >= 0) + map->flags = new_flags; + if (new_pool_max >= 0) + map->pool_max = new_pool_max; + + ceph_decode_need(p, end, 5*sizeof(u32), bad); + + /* new max? */ + max = ceph_decode_32(p); + if (max >= 0) { + err = osdmap_set_max_osd(map, max); + if (err < 0) + goto bad; + } + + map->epoch++; + map->modified = map->modified; + if (newcrush) { + if (map->crush) + crush_destroy(map->crush); + map->crush = newcrush; + newcrush = NULL; + } + + /* new_pool */ + ceph_decode_32_safe(p, end, len, bad); + while (len--) { + __u8 ev; + struct ceph_pg_pool_info *pi; + + ceph_decode_32_safe(p, end, pool, bad); + ceph_decode_need(p, end, 1 + sizeof(pi->v), bad); + ev = ceph_decode_8(p); /* encoding version */ + if (ev > CEPH_PG_POOL_VERSION) { + pr_warning("got unknown v %d > %d of ceph_pg_pool\n", + ev, CEPH_PG_POOL_VERSION); + goto bad; + } + pi = __lookup_pg_pool(&map->pg_pools, pool); + if (!pi) { + pi = kzalloc(sizeof(*pi), GFP_NOFS); + if (!pi) { + err = -ENOMEM; + goto bad; + } + pi->id = pool; + __insert_pg_pool(&map->pg_pools, pi); + } + err = __decode_pool(p, end, pi); + if (err < 0) + goto bad; + } + if (version >= 5 && __decode_pool_names(p, end, map) < 0) + goto bad; + + /* old_pool */ + ceph_decode_32_safe(p, end, len, bad); + while (len--) { + struct ceph_pg_pool_info *pi; + + ceph_decode_32_safe(p, end, pool, bad); + pi = __lookup_pg_pool(&map->pg_pools, pool); + if (pi) + __remove_pg_pool(&map->pg_pools, pi); + } + + /* new_up */ + err = -EINVAL; + ceph_decode_32_safe(p, end, len, bad); + while (len--) { + u32 osd; + struct ceph_entity_addr addr; + ceph_decode_32_safe(p, end, osd, bad); + ceph_decode_copy_safe(p, end, &addr, sizeof(addr), bad); + ceph_decode_addr(&addr); + pr_info("osd%d up\n", osd); + BUG_ON(osd >= map->max_osd); + map->osd_state[osd] |= CEPH_OSD_UP; + map->osd_addr[osd] = addr; + } + + /* new_down */ + ceph_decode_32_safe(p, end, len, bad); + while (len--) { + u32 osd; + ceph_decode_32_safe(p, end, osd, bad); + (*p)++; /* clean flag */ + pr_info("osd%d down\n", osd); + if (osd < map->max_osd) + map->osd_state[osd] &= ~CEPH_OSD_UP; + } + + /* new_weight */ + ceph_decode_32_safe(p, end, len, bad); + while (len--) { + u32 osd, off; + ceph_decode_need(p, end, sizeof(u32)*2, bad); + osd = ceph_decode_32(p); + off = ceph_decode_32(p); + pr_info("osd%d weight 0x%x %s\n", osd, off, + off == CEPH_OSD_IN ? "(in)" : + (off == CEPH_OSD_OUT ? "(out)" : "")); + if (osd < map->max_osd) + map->osd_weight[osd] = off; + } + + /* new_pg_temp */ + rbp = rb_first(&map->pg_temp); + ceph_decode_32_safe(p, end, len, bad); + while (len--) { + struct ceph_pg_mapping *pg; + int j; + struct ceph_pg pgid; + u32 pglen; + ceph_decode_need(p, end, sizeof(u64) + sizeof(u32), bad); + ceph_decode_copy(p, &pgid, sizeof(pgid)); + pglen = ceph_decode_32(p); + + /* remove any? */ + while (rbp && pgid_cmp(rb_entry(rbp, struct ceph_pg_mapping, + node)->pgid, pgid) <= 0) { + struct ceph_pg_mapping *cur = + rb_entry(rbp, struct ceph_pg_mapping, node); + + rbp = rb_next(rbp); + dout(" removed pg_temp %llx\n", *(u64 *)&cur->pgid); + rb_erase(&cur->node, &map->pg_temp); + kfree(cur); + } + + if (pglen) { + /* insert */ + ceph_decode_need(p, end, pglen*sizeof(u32), bad); + pg = kmalloc(sizeof(*pg) + sizeof(u32)*pglen, GFP_NOFS); + if (!pg) { + err = -ENOMEM; + goto bad; + } + pg->pgid = pgid; + pg->len = pglen; + for (j = 0; j < pglen; j++) + pg->osds[j] = ceph_decode_32(p); + err = __insert_pg_mapping(pg, &map->pg_temp); + if (err) { + kfree(pg); + goto bad; + } + dout(" added pg_temp %llx len %d\n", *(u64 *)&pgid, + pglen); + } + } + while (rbp) { + struct ceph_pg_mapping *cur = + rb_entry(rbp, struct ceph_pg_mapping, node); + + rbp = rb_next(rbp); + dout(" removed pg_temp %llx\n", *(u64 *)&cur->pgid); + rb_erase(&cur->node, &map->pg_temp); + kfree(cur); + } + + /* ignore the rest */ + *p = end; + return map; + +bad: + pr_err("corrupt inc osdmap epoch %d off %d (%p of %p-%p)\n", + epoch, (int)(*p - start), *p, start, end); + print_hex_dump(KERN_DEBUG, "osdmap: ", + DUMP_PREFIX_OFFSET, 16, 1, + start, end - start, true); + if (newcrush) + crush_destroy(newcrush); + return ERR_PTR(err); +} + + + + +/* + * calculate file layout from given offset, length. + * fill in correct oid, logical length, and object extent + * offset, length. + * + * for now, we write only a single su, until we can + * pass a stride back to the caller. + */ +void ceph_calc_file_object_mapping(struct ceph_file_layout *layout, + u64 off, u64 *plen, + u64 *ono, + u64 *oxoff, u64 *oxlen) +{ + u32 osize = le32_to_cpu(layout->fl_object_size); + u32 su = le32_to_cpu(layout->fl_stripe_unit); + u32 sc = le32_to_cpu(layout->fl_stripe_count); + u32 bl, stripeno, stripepos, objsetno; + u32 su_per_object; + u64 t, su_offset; + + dout("mapping %llu~%llu osize %u fl_su %u\n", off, *plen, + osize, su); + su_per_object = osize / su; + dout("osize %u / su %u = su_per_object %u\n", osize, su, + su_per_object); + + BUG_ON((su & ~PAGE_MASK) != 0); + /* bl = *off / su; */ + t = off; + do_div(t, su); + bl = t; + dout("off %llu / su %u = bl %u\n", off, su, bl); + + stripeno = bl / sc; + stripepos = bl % sc; + objsetno = stripeno / su_per_object; + + *ono = objsetno * sc + stripepos; + dout("objset %u * sc %u = ono %u\n", objsetno, sc, (unsigned)*ono); + + /* *oxoff = *off % layout->fl_stripe_unit; # offset in su */ + t = off; + su_offset = do_div(t, su); + *oxoff = su_offset + (stripeno % su_per_object) * su; + + /* + * Calculate the length of the extent being written to the selected + * object. This is the minimum of the full length requested (plen) or + * the remainder of the current stripe being written to. + */ + *oxlen = min_t(u64, *plen, su - su_offset); + *plen = *oxlen; + + dout(" obj extent %llu~%llu\n", *oxoff, *oxlen); +} +EXPORT_SYMBOL(ceph_calc_file_object_mapping); + +/* + * calculate an object layout (i.e. pgid) from an oid, + * file_layout, and osdmap + */ +int ceph_calc_object_layout(struct ceph_object_layout *ol, + const char *oid, + struct ceph_file_layout *fl, + struct ceph_osdmap *osdmap) +{ + unsigned num, num_mask; + struct ceph_pg pgid; + s32 preferred = (s32)le32_to_cpu(fl->fl_pg_preferred); + int poolid = le32_to_cpu(fl->fl_pg_pool); + struct ceph_pg_pool_info *pool; + unsigned ps; + + BUG_ON(!osdmap); + + pool = __lookup_pg_pool(&osdmap->pg_pools, poolid); + if (!pool) + return -EIO; + ps = ceph_str_hash(pool->v.object_hash, oid, strlen(oid)); + if (preferred >= 0) { + ps += preferred; + num = le32_to_cpu(pool->v.lpg_num); + num_mask = pool->lpg_num_mask; + } else { + num = le32_to_cpu(pool->v.pg_num); + num_mask = pool->pg_num_mask; + } + + pgid.ps = cpu_to_le16(ps); + pgid.preferred = cpu_to_le16(preferred); + pgid.pool = fl->fl_pg_pool; + if (preferred >= 0) + dout("calc_object_layout '%s' pgid %d.%xp%d\n", oid, poolid, ps, + (int)preferred); + else + dout("calc_object_layout '%s' pgid %d.%x\n", oid, poolid, ps); + + ol->ol_pgid = pgid; + ol->ol_stripe_unit = fl->fl_object_stripe_unit; + return 0; +} +EXPORT_SYMBOL(ceph_calc_object_layout); + +/* + * Calculate raw osd vector for the given pgid. Return pointer to osd + * array, or NULL on failure. + */ +static int *calc_pg_raw(struct ceph_osdmap *osdmap, struct ceph_pg pgid, + int *osds, int *num) +{ + struct ceph_pg_mapping *pg; + struct ceph_pg_pool_info *pool; + int ruleno; + unsigned poolid, ps, pps; + int preferred; + + /* pg_temp? */ + pg = __lookup_pg_mapping(&osdmap->pg_temp, pgid); + if (pg) { + *num = pg->len; + return pg->osds; + } + + /* crush */ + poolid = le32_to_cpu(pgid.pool); + ps = le16_to_cpu(pgid.ps); + preferred = (s16)le16_to_cpu(pgid.preferred); + + /* don't forcefeed bad device ids to crush */ + if (preferred >= osdmap->max_osd || + preferred >= osdmap->crush->max_devices) + preferred = -1; + + pool = __lookup_pg_pool(&osdmap->pg_pools, poolid); + if (!pool) + return NULL; + ruleno = crush_find_rule(osdmap->crush, pool->v.crush_ruleset, + pool->v.type, pool->v.size); + if (ruleno < 0) { + pr_err("no crush rule pool %d ruleset %d type %d size %d\n", + poolid, pool->v.crush_ruleset, pool->v.type, + pool->v.size); + return NULL; + } + + if (preferred >= 0) + pps = ceph_stable_mod(ps, + le32_to_cpu(pool->v.lpgp_num), + pool->lpgp_num_mask); + else + pps = ceph_stable_mod(ps, + le32_to_cpu(pool->v.pgp_num), + pool->pgp_num_mask); + pps += poolid; + *num = crush_do_rule(osdmap->crush, ruleno, pps, osds, + min_t(int, pool->v.size, *num), + preferred, osdmap->osd_weight); + return osds; +} + +/* + * Return acting set for given pgid. + */ +int ceph_calc_pg_acting(struct ceph_osdmap *osdmap, struct ceph_pg pgid, + int *acting) +{ + int rawosds[CEPH_PG_MAX_SIZE], *osds; + int i, o, num = CEPH_PG_MAX_SIZE; + + osds = calc_pg_raw(osdmap, pgid, rawosds, &num); + if (!osds) + return -1; + + /* primary is first up osd */ + o = 0; + for (i = 0; i < num; i++) + if (ceph_osd_is_up(osdmap, osds[i])) + acting[o++] = osds[i]; + return o; +} + +/* + * Return primary osd for given pgid, or -1 if none. + */ +int ceph_calc_pg_primary(struct ceph_osdmap *osdmap, struct ceph_pg pgid) +{ + int rawosds[CEPH_PG_MAX_SIZE], *osds; + int i, num = CEPH_PG_MAX_SIZE; + + osds = calc_pg_raw(osdmap, pgid, rawosds, &num); + if (!osds) + return -1; + + /* primary is first up osd */ + for (i = 0; i < num; i++) + if (ceph_osd_is_up(osdmap, osds[i])) + return osds[i]; + return -1; +} +EXPORT_SYMBOL(ceph_calc_pg_primary); diff --git a/net/ceph/pagelist.c b/net/ceph/pagelist.c new file mode 100644 index 00000000000..13cb409a7bb --- /dev/null +++ b/net/ceph/pagelist.c @@ -0,0 +1,154 @@ + +#include <linux/module.h> +#include <linux/gfp.h> +#include <linux/pagemap.h> +#include <linux/highmem.h> +#include <linux/ceph/pagelist.h> + +static void ceph_pagelist_unmap_tail(struct ceph_pagelist *pl) +{ + if (pl->mapped_tail) { + struct page *page = list_entry(pl->head.prev, struct page, lru); + kunmap(page); + pl->mapped_tail = NULL; + } +} + +int ceph_pagelist_release(struct ceph_pagelist *pl) +{ + ceph_pagelist_unmap_tail(pl); + while (!list_empty(&pl->head)) { + struct page *page = list_first_entry(&pl->head, struct page, + lru); + list_del(&page->lru); + __free_page(page); + } + ceph_pagelist_free_reserve(pl); + return 0; +} +EXPORT_SYMBOL(ceph_pagelist_release); + +static int ceph_pagelist_addpage(struct ceph_pagelist *pl) +{ + struct page *page; + + if (!pl->num_pages_free) { + page = __page_cache_alloc(GFP_NOFS); + } else { + page = list_first_entry(&pl->free_list, struct page, lru); + list_del(&page->lru); + --pl->num_pages_free; + } + if (!page) + return -ENOMEM; + pl->room += PAGE_SIZE; + ceph_pagelist_unmap_tail(pl); + list_add_tail(&page->lru, &pl->head); + pl->mapped_tail = kmap(page); + return 0; +} + +int ceph_pagelist_append(struct ceph_pagelist *pl, const void *buf, size_t len) +{ + while (pl->room < len) { + size_t bit = pl->room; + int ret; + + memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK), + buf, bit); + pl->length += bit; + pl->room -= bit; + buf += bit; + len -= bit; + ret = ceph_pagelist_addpage(pl); + if (ret) + return ret; + } + + memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK), buf, len); + pl->length += len; + pl->room -= len; + return 0; +} +EXPORT_SYMBOL(ceph_pagelist_append); + +/** + * Allocate enough pages for a pagelist to append the given amount + * of data without without allocating. + * Returns: 0 on success, -ENOMEM on error. + */ +int ceph_pagelist_reserve(struct ceph_pagelist *pl, size_t space) +{ + if (space <= pl->room) + return 0; + space -= pl->room; + space = (space + PAGE_SIZE - 1) >> PAGE_SHIFT; /* conv to num pages */ + + while (space > pl->num_pages_free) { + struct page *page = __page_cache_alloc(GFP_NOFS); + if (!page) + return -ENOMEM; + list_add_tail(&page->lru, &pl->free_list); + ++pl->num_pages_free; + } + return 0; +} +EXPORT_SYMBOL(ceph_pagelist_reserve); + +/** + * Free any pages that have been preallocated. + */ +int ceph_pagelist_free_reserve(struct ceph_pagelist *pl) +{ + while (!list_empty(&pl->free_list)) { + struct page *page = list_first_entry(&pl->free_list, + struct page, lru); + list_del(&page->lru); + __free_page(page); + --pl->num_pages_free; + } + BUG_ON(pl->num_pages_free); + return 0; +} +EXPORT_SYMBOL(ceph_pagelist_free_reserve); + +/** + * Create a truncation point. + */ +void ceph_pagelist_set_cursor(struct ceph_pagelist *pl, + struct ceph_pagelist_cursor *c) +{ + c->pl = pl; + c->page_lru = pl->head.prev; + c->room = pl->room; +} +EXPORT_SYMBOL(ceph_pagelist_set_cursor); + +/** + * Truncate a pagelist to the given point. Move extra pages to reserve. + * This won't sleep. + * Returns: 0 on success, + * -EINVAL if the pagelist doesn't match the trunc point pagelist + */ +int ceph_pagelist_truncate(struct ceph_pagelist *pl, + struct ceph_pagelist_cursor *c) +{ + struct page *page; + + if (pl != c->pl) + return -EINVAL; + ceph_pagelist_unmap_tail(pl); + while (pl->head.prev != c->page_lru) { + page = list_entry(pl->head.prev, struct page, lru); + list_del(&page->lru); /* remove from pagelist */ + list_add_tail(&page->lru, &pl->free_list); /* add to reserve */ + ++pl->num_pages_free; + } + pl->room = c->room; + if (!list_empty(&pl->head)) { + page = list_entry(pl->head.prev, struct page, lru); + pl->mapped_tail = kmap(page); + } + return 0; +} +EXPORT_SYMBOL(ceph_pagelist_truncate); diff --git a/net/ceph/pagevec.c b/net/ceph/pagevec.c new file mode 100644 index 00000000000..54caf068715 --- /dev/null +++ b/net/ceph/pagevec.c @@ -0,0 +1,223 @@ +#include <linux/ceph/ceph_debug.h> + +#include <linux/module.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/file.h> +#include <linux/namei.h> +#include <linux/writeback.h> + +#include <linux/ceph/libceph.h> + +/* + * build a vector of user pages + */ +struct page **ceph_get_direct_page_vector(const char __user *data, + int num_pages, + loff_t off, size_t len) +{ + struct page **pages; + int rc; + + pages = kmalloc(sizeof(*pages) * num_pages, GFP_NOFS); + if (!pages) + return ERR_PTR(-ENOMEM); + + down_read(¤t->mm->mmap_sem); + rc = get_user_pages(current, current->mm, (unsigned long)data, + num_pages, 0, 0, pages, NULL); + up_read(¤t->mm->mmap_sem); + if (rc < 0) + goto fail; + return pages; + +fail: + kfree(pages); + return ERR_PTR(rc); +} +EXPORT_SYMBOL(ceph_get_direct_page_vector); + +void ceph_put_page_vector(struct page **pages, int num_pages) +{ + int i; + + for (i = 0; i < num_pages; i++) + put_page(pages[i]); + kfree(pages); +} +EXPORT_SYMBOL(ceph_put_page_vector); + +void ceph_release_page_vector(struct page **pages, int num_pages) +{ + int i; + + for (i = 0; i < num_pages; i++) + __free_pages(pages[i], 0); + kfree(pages); +} +EXPORT_SYMBOL(ceph_release_page_vector); + +/* + * allocate a vector new pages + */ +struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags) +{ + struct page **pages; + int i; + + pages = kmalloc(sizeof(*pages) * num_pages, flags); + if (!pages) + return ERR_PTR(-ENOMEM); + for (i = 0; i < num_pages; i++) { + pages[i] = __page_cache_alloc(flags); + if (pages[i] == NULL) { + ceph_release_page_vector(pages, i); + return ERR_PTR(-ENOMEM); + } + } + return pages; +} +EXPORT_SYMBOL(ceph_alloc_page_vector); + +/* + * copy user data into a page vector + */ +int ceph_copy_user_to_page_vector(struct page **pages, + const char __user *data, + loff_t off, size_t len) +{ + int i = 0; + int po = off & ~PAGE_CACHE_MASK; + int left = len; + int l, bad; + + while (left > 0) { + l = min_t(int, PAGE_CACHE_SIZE-po, left); + bad = copy_from_user(page_address(pages[i]) + po, data, l); + if (bad == l) + return -EFAULT; + data += l - bad; + left -= l - bad; + po += l - bad; + if (po == PAGE_CACHE_SIZE) { + po = 0; + i++; + } + } + return len; +} +EXPORT_SYMBOL(ceph_copy_user_to_page_vector); + +int ceph_copy_to_page_vector(struct page **pages, + const char *data, + loff_t off, size_t len) +{ + int i = 0; + size_t po = off & ~PAGE_CACHE_MASK; + size_t left = len; + size_t l; + + while (left > 0) { + l = min_t(size_t, PAGE_CACHE_SIZE-po, left); + memcpy(page_address(pages[i]) + po, data, l); + data += l; + left -= l; + po += l; + if (po == PAGE_CACHE_SIZE) { + po = 0; + i++; + } + } + return len; +} +EXPORT_SYMBOL(ceph_copy_to_page_vector); + +int ceph_copy_from_page_vector(struct page **pages, + char *data, + loff_t off, size_t len) +{ + int i = 0; + size_t po = off & ~PAGE_CACHE_MASK; + size_t left = len; + size_t l; + + while (left > 0) { + l = min_t(size_t, PAGE_CACHE_SIZE-po, left); + memcpy(data, page_address(pages[i]) + po, l); + data += l; + left -= l; + po += l; + if (po == PAGE_CACHE_SIZE) { + po = 0; + i++; + } + } + return len; +} +EXPORT_SYMBOL(ceph_copy_from_page_vector); + +/* + * copy user data from a page vector into a user pointer + */ +int ceph_copy_page_vector_to_user(struct page **pages, + char __user *data, + loff_t off, size_t len) +{ + int i = 0; + int po = off & ~PAGE_CACHE_MASK; + int left = len; + int l, bad; + + while (left > 0) { + l = min_t(int, left, PAGE_CACHE_SIZE-po); + bad = copy_to_user(data, page_address(pages[i]) + po, l); + if (bad == l) + return -EFAULT; + data += l - bad; + left -= l - bad; + if (po) { + po += l - bad; + if (po == PAGE_CACHE_SIZE) + po = 0; + } + i++; + } + return len; +} +EXPORT_SYMBOL(ceph_copy_page_vector_to_user); + +/* + * Zero an extent within a page vector. Offset is relative to the + * start of the first page. + */ +void ceph_zero_page_vector_range(int off, int len, struct page **pages) +{ + int i = off >> PAGE_CACHE_SHIFT; + + off &= ~PAGE_CACHE_MASK; + + dout("zero_page_vector_page %u~%u\n", off, len); + + /* leading partial page? */ + if (off) { + int end = min((int)PAGE_CACHE_SIZE, off + len); + dout("zeroing %d %p head from %d\n", i, pages[i], + (int)off); + zero_user_segment(pages[i], off, end); + len -= (end - off); + i++; + } + while (len >= PAGE_CACHE_SIZE) { + dout("zeroing %d %p len=%d\n", i, pages[i], len); + zero_user_segment(pages[i], 0, PAGE_CACHE_SIZE); + len -= PAGE_CACHE_SIZE; + i++; + } + /* trailing partial page? */ + if (len) { + dout("zeroing %d %p tail to %d\n", i, pages[i], (int)len); + zero_user_segment(pages[i], 0, len); + } +} +EXPORT_SYMBOL(ceph_zero_page_vector_range); + diff --git a/net/core/datagram.c b/net/core/datagram.c index 251997a9548..cd1e039c875 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -243,6 +243,7 @@ void skb_free_datagram_locked(struct sock *sk, struct sk_buff *skb) unlock_sock_fast(sk, slow); /* skb is now orphaned, can be freed outside of locked section */ + trace_kfree_skb(skb, skb_free_datagram_locked); __kfree_skb(skb); } EXPORT_SYMBOL(skb_free_datagram_locked); @@ -746,13 +747,12 @@ unsigned int datagram_poll(struct file *file, struct socket *sock, if (sk->sk_err || !skb_queue_empty(&sk->sk_error_queue)) mask |= POLLERR; if (sk->sk_shutdown & RCV_SHUTDOWN) - mask |= POLLRDHUP; + mask |= POLLRDHUP | POLLIN | POLLRDNORM; if (sk->sk_shutdown == SHUTDOWN_MASK) mask |= POLLHUP; /* readable? */ - if (!skb_queue_empty(&sk->sk_receive_queue) || - (sk->sk_shutdown & RCV_SHUTDOWN)) + if (!skb_queue_empty(&sk->sk_receive_queue)) mask |= POLLIN | POLLRDNORM; /* Connection-based need to check for termination and startup */ diff --git a/net/core/dev.c b/net/core/dev.c index 1ae65439144..78b5a89b0f4 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -128,7 +128,10 @@ #include <linux/jhash.h> #include <linux/random.h> #include <trace/events/napi.h> +#include <trace/events/net.h> +#include <trace/events/skb.h> #include <linux/pci.h> +#include <linux/inetdevice.h> #include "net-sysfs.h" @@ -371,6 +374,14 @@ static inline void netdev_set_addr_lockdep_class(struct net_device *dev) * --ANK (980803) */ +static inline struct list_head *ptype_head(const struct packet_type *pt) +{ + if (pt->type == htons(ETH_P_ALL)) + return &ptype_all; + else + return &ptype_base[ntohs(pt->type) & PTYPE_HASH_MASK]; +} + /** * dev_add_pack - add packet handler * @pt: packet type declaration @@ -386,16 +397,11 @@ static inline void netdev_set_addr_lockdep_class(struct net_device *dev) void dev_add_pack(struct packet_type *pt) { - int hash; + struct list_head *head = ptype_head(pt); - spin_lock_bh(&ptype_lock); - if (pt->type == htons(ETH_P_ALL)) - list_add_rcu(&pt->list, &ptype_all); - else { - hash = ntohs(pt->type) & PTYPE_HASH_MASK; - list_add_rcu(&pt->list, &ptype_base[hash]); - } - spin_unlock_bh(&ptype_lock); + spin_lock(&ptype_lock); + list_add_rcu(&pt->list, head); + spin_unlock(&ptype_lock); } EXPORT_SYMBOL(dev_add_pack); @@ -414,15 +420,10 @@ EXPORT_SYMBOL(dev_add_pack); */ void __dev_remove_pack(struct packet_type *pt) { - struct list_head *head; + struct list_head *head = ptype_head(pt); struct packet_type *pt1; - spin_lock_bh(&ptype_lock); - - if (pt->type == htons(ETH_P_ALL)) - head = &ptype_all; - else - head = &ptype_base[ntohs(pt->type) & PTYPE_HASH_MASK]; + spin_lock(&ptype_lock); list_for_each_entry(pt1, head, list) { if (pt == pt1) { @@ -433,7 +434,7 @@ void __dev_remove_pack(struct packet_type *pt) printk(KERN_WARNING "dev_remove_pack: %p not found.\n", pt); out: - spin_unlock_bh(&ptype_lock); + spin_unlock(&ptype_lock); } EXPORT_SYMBOL(__dev_remove_pack); @@ -1484,8 +1485,9 @@ int dev_forward_skb(struct net_device *dev, struct sk_buff *skb) skb_orphan(skb); nf_reset(skb); - if (!(dev->flags & IFF_UP) || - (skb->len > (dev->mtu + dev->hard_header_len))) { + if (unlikely(!(dev->flags & IFF_UP) || + (skb->len > (dev->mtu + dev->hard_header_len + VLAN_HLEN)))) { + atomic_long_inc(&dev->rx_dropped); kfree_skb(skb); return NET_RX_DROP; } @@ -1553,21 +1555,56 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev) * Routine to help set real_num_tx_queues. To avoid skbs mapped to queues * greater then real_num_tx_queues stale skbs on the qdisc must be flushed. */ -void netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq) +int netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq) { - unsigned int real_num = dev->real_num_tx_queues; + if (txq < 1 || txq > dev->num_tx_queues) + return -EINVAL; - if (unlikely(txq > dev->num_tx_queues)) - ; - else if (txq > real_num) - dev->real_num_tx_queues = txq; - else if (txq < real_num) { - dev->real_num_tx_queues = txq; - qdisc_reset_all_tx_gt(dev, txq); + if (dev->reg_state == NETREG_REGISTERED) { + ASSERT_RTNL(); + + if (txq < dev->real_num_tx_queues) + qdisc_reset_all_tx_gt(dev, txq); } + + dev->real_num_tx_queues = txq; + return 0; } EXPORT_SYMBOL(netif_set_real_num_tx_queues); +#ifdef CONFIG_RPS +/** + * netif_set_real_num_rx_queues - set actual number of RX queues used + * @dev: Network device + * @rxq: Actual number of RX queues + * + * This must be called either with the rtnl_lock held or before + * registration of the net device. Returns 0 on success, or a + * negative error code. If called before registration, it always + * succeeds. + */ +int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq) +{ + int rc; + + if (rxq < 1 || rxq > dev->num_rx_queues) + return -EINVAL; + + if (dev->reg_state == NETREG_REGISTERED) { + ASSERT_RTNL(); + + rc = net_rx_queue_update_kobjects(dev, dev->real_num_rx_queues, + rxq); + if (rc) + return rc; + } + + dev->real_num_rx_queues = rxq; + return 0; +} +EXPORT_SYMBOL(netif_set_real_num_rx_queues); +#endif + static inline void __netif_reschedule(struct Qdisc *q) { struct softnet_data *sd; @@ -1659,7 +1696,12 @@ static bool can_checksum_protocol(unsigned long features, __be16 protocol) static bool dev_can_checksum(struct net_device *dev, struct sk_buff *skb) { - if (can_checksum_protocol(dev->features, skb->protocol)) + int features = dev->features; + + if (vlan_tx_tag_present(skb)) + features &= dev->vlan_features; + + if (can_checksum_protocol(features, skb->protocol)) return true; if (skb->protocol == htons(ETH_P_8021Q)) { @@ -1758,6 +1800,16 @@ struct sk_buff *skb_gso_segment(struct sk_buff *skb, int features) __be16 type = skb->protocol; int err; + if (type == htons(ETH_P_8021Q)) { + struct vlan_ethhdr *veh; + + if (unlikely(!pskb_may_pull(skb, VLAN_ETH_HLEN))) + return ERR_PTR(-EINVAL); + + veh = (struct vlan_ethhdr *)skb->data; + type = veh->h_vlan_encapsulated_proto; + } + skb_reset_mac_header(skb); skb->mac_len = skb->network_header - skb->mac_header; __skb_pull(skb, skb->mac_len); @@ -1902,14 +1954,14 @@ static int dev_gso_segment(struct sk_buff *skb) /* * Try to orphan skb early, right before transmission by the device. - * We cannot orphan skb if tx timestamp is requested, since - * drivers need to call skb_tstamp_tx() to send the timestamp. + * We cannot orphan skb if tx timestamp is requested or the sk-reference + * is needed on driver level for other reasons, e.g. see net/can/raw.c */ static inline void skb_orphan_try(struct sk_buff *skb) { struct sock *sk = skb->sk; - if (sk && !skb_tx(skb)->flags) { + if (sk && !skb_shinfo(skb)->tx_flags) { /* skb_tx_hash() wont be able to get sk. * We copy sk_hash into skb->rxhash */ @@ -1929,9 +1981,14 @@ static inline void skb_orphan_try(struct sk_buff *skb) static inline int skb_needs_linearize(struct sk_buff *skb, struct net_device *dev) { + int features = dev->features; + + if (skb->protocol == htons(ETH_P_8021Q) || vlan_tx_tag_present(skb)) + features &= dev->vlan_features; + return skb_is_nonlinear(skb) && - ((skb_has_frags(skb) && !(dev->features & NETIF_F_FRAGLIST)) || - (skb_shinfo(skb)->nr_frags && (!(dev->features & NETIF_F_SG) || + ((skb_has_frag_list(skb) && !(features & NETIF_F_FRAGLIST)) || + (skb_shinfo(skb)->nr_frags && (!(features & NETIF_F_SG) || illegal_highdma(dev, skb)))); } @@ -1954,6 +2011,15 @@ int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, skb_orphan_try(skb); + if (vlan_tx_tag_present(skb) && + !(dev->features & NETIF_F_HW_VLAN_TX)) { + skb = __vlan_put_tag(skb, vlan_tx_tag_get(skb)); + if (unlikely(!skb)) + goto out; + + skb->vlan_tci = 0; + } + if (netif_needs_gso(dev, skb)) { if (unlikely(dev_gso_segment(skb))) goto out_kfree_skb; @@ -1978,6 +2044,7 @@ int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, } rc = ops->ndo_start_xmit(skb, dev); + trace_net_dev_xmit(skb, rc); if (rc == NETDEV_TX_OK) txq_trans_update(txq); return rc; @@ -1998,6 +2065,7 @@ gso: skb_dst_drop(nskb); rc = ops->ndo_start_xmit(nskb, dev); + trace_net_dev_xmit(nskb, rc); if (unlikely(rc != NETDEV_TX_OK)) { if (rc & ~NETDEV_TX_MASK) goto out_kfree_gso_skb; @@ -2015,6 +2083,7 @@ out_kfree_gso_skb: skb->destructor = DEV_GSO_CB(skb)->destructor; out_kfree_skb: kfree_skb(skb); +out: return rc; } @@ -2058,16 +2127,16 @@ static struct netdev_queue *dev_pick_tx(struct net_device *dev, struct sk_buff *skb) { int queue_index; - struct sock *sk = skb->sk; + const struct net_device_ops *ops = dev->netdev_ops; - queue_index = sk_tx_queue_get(sk); - if (queue_index < 0) { - const struct net_device_ops *ops = dev->netdev_ops; + if (ops->ndo_select_queue) { + queue_index = ops->ndo_select_queue(dev, skb); + queue_index = dev_cap_txqueue(dev, queue_index); + } else { + struct sock *sk = skb->sk; + queue_index = sk_tx_queue_get(sk); + if (queue_index < 0) { - if (ops->ndo_select_queue) { - queue_index = ops->ndo_select_queue(dev, skb); - queue_index = dev_cap_txqueue(dev, queue_index); - } else { queue_index = 0; if (dev->real_num_tx_queues > 1) queue_index = skb_tx_hash(dev, skb); @@ -2143,6 +2212,9 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, return rc; } +static DEFINE_PER_CPU(int, xmit_recursion); +#define RECURSION_LIMIT 3 + /** * dev_queue_xmit - transmit a buffer * @skb: buffer to transmit @@ -2186,6 +2258,7 @@ int dev_queue_xmit(struct sk_buff *skb) #ifdef CONFIG_NET_CLS_ACT skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_EGRESS); #endif + trace_net_dev_queue(skb); if (q->enqueue) { rc = __dev_xmit_skb(skb, q, dev, txq); goto out; @@ -2208,10 +2281,15 @@ int dev_queue_xmit(struct sk_buff *skb) if (txq->xmit_lock_owner != cpu) { + if (__this_cpu_read(xmit_recursion) > RECURSION_LIMIT) + goto recursion_alert; + HARD_TX_LOCK(dev, txq, cpu); if (!netif_tx_queue_stopped(txq)) { + __this_cpu_inc(xmit_recursion); rc = dev_hard_start_xmit(skb, dev, txq); + __this_cpu_dec(xmit_recursion); if (dev_xmit_complete(rc)) { HARD_TX_UNLOCK(dev, txq); goto out; @@ -2223,7 +2301,9 @@ int dev_queue_xmit(struct sk_buff *skb) "queue packet!\n", dev->name); } else { /* Recursion is detected! It is possible, - * unfortunately */ + * unfortunately + */ +recursion_alert: if (net_ratelimit()) printk(KERN_CRIT "Dead loop on virtual device " "%s, fix it urgently!\n", dev->name); @@ -2259,69 +2339,44 @@ static inline void ____napi_schedule(struct softnet_data *sd, __raise_softirq_irqoff(NET_RX_SOFTIRQ); } -#ifdef CONFIG_RPS - -/* One global table that all flow-based protocols share. */ -struct rps_sock_flow_table *rps_sock_flow_table __read_mostly; -EXPORT_SYMBOL(rps_sock_flow_table); - /* - * get_rps_cpu is called from netif_receive_skb and returns the target - * CPU from the RPS map of the receiving queue for a given skb. - * rcu_read_lock must be held on entry. + * __skb_get_rxhash: calculate a flow hash based on src/dst addresses + * and src/dst port numbers. Returns a non-zero hash number on success + * and 0 on failure. */ -static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, - struct rps_dev_flow **rflowp) +__u32 __skb_get_rxhash(struct sk_buff *skb) { + int nhoff, hash = 0, poff; struct ipv6hdr *ip6; struct iphdr *ip; - struct netdev_rx_queue *rxqueue; - struct rps_map *map; - struct rps_dev_flow_table *flow_table; - struct rps_sock_flow_table *sock_flow_table; - int cpu = -1; u8 ip_proto; - u16 tcpu; u32 addr1, addr2, ihl; union { u32 v32; u16 v16[2]; } ports; - if (skb_rx_queue_recorded(skb)) { - u16 index = skb_get_rx_queue(skb); - if (unlikely(index >= dev->num_rx_queues)) { - WARN_ONCE(dev->num_rx_queues > 1, "%s received packet " - "on queue %u, but number of RX queues is %u\n", - dev->name, index, dev->num_rx_queues); - goto done; - } - rxqueue = dev->_rx + index; - } else - rxqueue = dev->_rx; - - if (!rxqueue->rps_map && !rxqueue->rps_flow_table) - goto done; - - if (skb->rxhash) - goto got_hash; /* Skip hash computation on packet header */ + nhoff = skb_network_offset(skb); switch (skb->protocol) { case __constant_htons(ETH_P_IP): - if (!pskb_may_pull(skb, sizeof(*ip))) + if (!pskb_may_pull(skb, sizeof(*ip) + nhoff)) goto done; - ip = (struct iphdr *) skb->data; - ip_proto = ip->protocol; + ip = (struct iphdr *) (skb->data + nhoff); + if (ip->frag_off & htons(IP_MF | IP_OFFSET)) + ip_proto = 0; + else + ip_proto = ip->protocol; addr1 = (__force u32) ip->saddr; addr2 = (__force u32) ip->daddr; ihl = ip->ihl; break; case __constant_htons(ETH_P_IPV6): - if (!pskb_may_pull(skb, sizeof(*ip6))) + if (!pskb_may_pull(skb, sizeof(*ip6) + nhoff)) goto done; - ip6 = (struct ipv6hdr *) skb->data; + ip6 = (struct ipv6hdr *) (skb->data + nhoff); ip_proto = ip6->nexthdr; addr1 = (__force u32) ip6->saddr.s6_addr32[3]; addr2 = (__force u32) ip6->daddr.s6_addr32[3]; @@ -2330,33 +2385,81 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, default: goto done; } - switch (ip_proto) { - case IPPROTO_TCP: - case IPPROTO_UDP: - case IPPROTO_DCCP: - case IPPROTO_ESP: - case IPPROTO_AH: - case IPPROTO_SCTP: - case IPPROTO_UDPLITE: - if (pskb_may_pull(skb, (ihl * 4) + 4)) { - ports.v32 = * (__force u32 *) (skb->data + (ihl * 4)); + + ports.v32 = 0; + poff = proto_ports_offset(ip_proto); + if (poff >= 0) { + nhoff += ihl * 4 + poff; + if (pskb_may_pull(skb, nhoff + 4)) { + ports.v32 = * (__force u32 *) (skb->data + nhoff); if (ports.v16[1] < ports.v16[0]) swap(ports.v16[0], ports.v16[1]); - break; } - default: - ports.v32 = 0; - break; } /* get a consistent hash (same value on both flow directions) */ if (addr2 < addr1) swap(addr1, addr2); - skb->rxhash = jhash_3words(addr1, addr2, ports.v32, hashrnd); - if (!skb->rxhash) - skb->rxhash = 1; -got_hash: + hash = jhash_3words(addr1, addr2, ports.v32, hashrnd); + if (!hash) + hash = 1; + +done: + return hash; +} +EXPORT_SYMBOL(__skb_get_rxhash); + +#ifdef CONFIG_RPS + +/* One global table that all flow-based protocols share. */ +struct rps_sock_flow_table *rps_sock_flow_table __read_mostly; +EXPORT_SYMBOL(rps_sock_flow_table); + +/* + * get_rps_cpu is called from netif_receive_skb and returns the target + * CPU from the RPS map of the receiving queue for a given skb. + * rcu_read_lock must be held on entry. + */ +static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, + struct rps_dev_flow **rflowp) +{ + struct netdev_rx_queue *rxqueue; + struct rps_map *map = NULL; + struct rps_dev_flow_table *flow_table; + struct rps_sock_flow_table *sock_flow_table; + int cpu = -1; + u16 tcpu; + + if (skb_rx_queue_recorded(skb)) { + u16 index = skb_get_rx_queue(skb); + if (unlikely(index >= dev->real_num_rx_queues)) { + WARN_ONCE(dev->real_num_rx_queues > 1, + "%s received packet on queue %u, but number " + "of RX queues is %u\n", + dev->name, index, dev->real_num_rx_queues); + goto done; + } + rxqueue = dev->_rx + index; + } else + rxqueue = dev->_rx; + + if (rxqueue->rps_map) { + map = rcu_dereference(rxqueue->rps_map); + if (map && map->len == 1) { + tcpu = map->cpus[0]; + if (cpu_online(tcpu)) + cpu = tcpu; + goto done; + } + } else if (!rxqueue->rps_flow_table) { + goto done; + } + + skb_reset_network_header(skb); + if (!skb_get_rxhash(skb)) + goto done; + flow_table = rcu_dereference(rxqueue->rps_flow_table); sock_flow_table = rcu_dereference(rps_sock_flow_table); if (flow_table && sock_flow_table) { @@ -2396,7 +2499,6 @@ got_hash: } } - map = rcu_dereference(rxqueue->rps_map); if (map) { tcpu = map->cpus[((u64) skb->rxhash * map->len) >> 32]; @@ -2482,6 +2584,7 @@ enqueue: local_irq_restore(flags); + atomic_long_inc(&skb->dev->rx_dropped); kfree_skb(skb); return NET_RX_DROP; } @@ -2512,6 +2615,7 @@ int netif_rx(struct sk_buff *skb) if (netdev_tstamp_prequeue) net_timestamp_check(skb); + trace_netif_rx(skb); #ifdef CONFIG_RPS { struct rps_dev_flow voidflow, *rflow = &voidflow; @@ -2571,6 +2675,7 @@ static void net_tx_action(struct softirq_action *h) clist = clist->next; WARN_ON(atomic_read(&skb->users)); + trace_kfree_skb(skb, net_tx_action); __kfree_skb(skb); } } @@ -2636,11 +2741,10 @@ EXPORT_SYMBOL_GPL(br_fdb_test_addr_hook); * the ingress scheduler, you just cant add policies on ingress. * */ -static int ing_filter(struct sk_buff *skb) +static int ing_filter(struct sk_buff *skb, struct netdev_queue *rxq) { struct net_device *dev = skb->dev; u32 ttl = G_TC_RTTL(skb->tc_verd); - struct netdev_queue *rxq; int result = TC_ACT_OK; struct Qdisc *q; @@ -2654,8 +2758,6 @@ static int ing_filter(struct sk_buff *skb) skb->tc_verd = SET_TC_RTTL(skb->tc_verd, ttl); skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_INGRESS); - rxq = &dev->rx_queue; - q = rxq->qdisc; if (q != &noop_qdisc) { spin_lock(qdisc_lock(q)); @@ -2671,7 +2773,9 @@ static inline struct sk_buff *handle_ing(struct sk_buff *skb, struct packet_type **pt_prev, int *ret, struct net_device *orig_dev) { - if (skb->dev->rx_queue.qdisc == &noop_qdisc) + struct netdev_queue *rxq = rcu_dereference(skb->dev->ingress_queue); + + if (!rxq || rxq->qdisc == &noop_qdisc) goto out; if (*pt_prev) { @@ -2679,7 +2783,7 @@ static inline struct sk_buff *handle_ing(struct sk_buff *skb, *pt_prev = NULL; } - switch (ing_filter(skb)) { + switch (ing_filter(skb, rxq)) { case TC_ACT_SHOT: case TC_ACT_STOLEN: kfree_skb(skb); @@ -2692,33 +2796,6 @@ out: } #endif -/* - * netif_nit_deliver - deliver received packets to network taps - * @skb: buffer - * - * This function is used to deliver incoming packets to network - * taps. It should be used when the normal netif_receive_skb path - * is bypassed, for example because of VLAN acceleration. - */ -void netif_nit_deliver(struct sk_buff *skb) -{ - struct packet_type *ptype; - - if (list_empty(&ptype_all)) - return; - - skb_reset_network_header(skb); - skb_reset_transport_header(skb); - skb->mac_len = skb->network_header - skb->mac_header; - - rcu_read_lock(); - list_for_each_entry_rcu(ptype, &ptype_all, list) { - if (!ptype->dev || ptype->dev == skb->dev) - deliver_skb(skb, ptype, skb->dev); - } - rcu_read_unlock(); -} - /** * netdev_rx_handler_register - register receive handler * @dev: device to register a handler for @@ -2828,8 +2905,7 @@ static int __netif_receive_skb(struct sk_buff *skb) if (!netdev_tstamp_prequeue) net_timestamp_check(skb); - if (vlan_tx_tag_present(skb) && vlan_hwaccel_do_receive(skb)) - return NET_RX_SUCCESS; + trace_netif_receive_skb(skb); /* if we've gotten here through NAPI, check netpoll */ if (netpoll_receive_skb(skb)) @@ -2843,8 +2919,7 @@ static int __netif_receive_skb(struct sk_buff *skb) * be delivered to pkt handlers that are exact matches. Also * the deliver_no_wcard flag will be set. If packet handlers * are sensitive to duplicate packets these skbs will need to - * be dropped at the handler. The vlan accel path may have - * already set the deliver_no_wcard flag. + * be dropped at the handler. */ null_or_orig = NULL; orig_dev = skb->dev; @@ -2903,6 +2978,18 @@ ncls: goto out; } + if (vlan_tx_tag_present(skb)) { + if (pt_prev) { + ret = deliver_skb(skb, pt_prev, orig_dev); + pt_prev = NULL; + } + if (vlan_hwaccel_do_receive(&skb)) { + ret = __netif_receive_skb(skb); + goto out; + } else if (unlikely(!skb)) + goto out; + } + /* * Make sure frames received on VLAN interfaces stacked on * bonding interfaces still make their way to any base bonding @@ -2930,6 +3017,7 @@ ncls: if (pt_prev) { ret = pt_prev->func(skb, skb->dev, pt_prev, orig_dev); } else { + atomic_long_inc(&skb->dev->rx_dropped); kfree_skb(skb); /* Jamal, now you will not able to escape explaining * me how you were going to use this. :-) @@ -3050,7 +3138,7 @@ out: return netif_receive_skb(skb); } -static void napi_gro_flush(struct napi_struct *napi) +inline void napi_gro_flush(struct napi_struct *napi) { struct sk_buff *skb, *next; @@ -3063,6 +3151,7 @@ static void napi_gro_flush(struct napi_struct *napi) napi->gro_count = 0; napi->gro_list = NULL; } +EXPORT_SYMBOL(napi_gro_flush); enum gro_result dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb) { @@ -3077,7 +3166,7 @@ enum gro_result dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb) if (!(skb->dev->features & NETIF_F_GRO) || netpoll_rx_on(skb)) goto normal; - if (skb_is_gso(skb) || skb_has_frags(skb)) + if (skb_is_gso(skb) || skb_has_frag_list(skb)) goto normal; rcu_read_lock(); @@ -3143,7 +3232,7 @@ pull: put_page(skb_shinfo(skb)->frags[0].page); memmove(skb_shinfo(skb)->frags, skb_shinfo(skb)->frags + 1, - --skb_shinfo(skb)->nr_frags); + --skb_shinfo(skb)->nr_frags * sizeof(skb_frag_t)); } } @@ -3156,16 +3245,19 @@ normal: } EXPORT_SYMBOL(dev_gro_receive); -static gro_result_t +static inline gro_result_t __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) { struct sk_buff *p; for (p = napi->gro_list; p; p = p->next) { - NAPI_GRO_CB(p)->same_flow = - (p->dev == skb->dev) && - !compare_ether_header(skb_mac_header(p), + unsigned long diffs; + + diffs = (unsigned long)p->dev ^ (unsigned long)skb->dev; + diffs |= p->vlan_tci ^ skb->vlan_tci; + diffs |= compare_ether_header(skb_mac_header(p), skb_gro_mac_header(skb)); + NAPI_GRO_CB(p)->same_flow = !diffs; NAPI_GRO_CB(p)->flush = 0; } @@ -3218,14 +3310,14 @@ gro_result_t napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) } EXPORT_SYMBOL(napi_gro_receive); -void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb) +static void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb) { __skb_pull(skb, skb_headlen(skb)); skb_reserve(skb, NET_IP_ALIGN - skb_headroom(skb)); + skb->vlan_tci = 0; napi->skb = skb; } -EXPORT_SYMBOL(napi_reuse_skb); struct sk_buff *napi_get_frags(struct napi_struct *napi) { @@ -4845,7 +4937,7 @@ static void rollback_registered_many(struct list_head *head) dev = list_first_entry(head, struct net_device, unreg_list); call_netdevice_notifiers(NETDEV_UNREGISTER_BATCH, dev); - synchronize_net(); + rcu_barrier(); list_for_each_entry(dev, head, unreg_list) dev_put(dev); @@ -4859,21 +4951,6 @@ static void rollback_registered(struct net_device *dev) rollback_registered_many(&single); } -static void __netdev_init_queue_locks_one(struct net_device *dev, - struct netdev_queue *dev_queue, - void *_unused) -{ - spin_lock_init(&dev_queue->_xmit_lock); - netdev_set_xmit_lockdep_class(&dev_queue->_xmit_lock, dev->type); - dev_queue->xmit_lock_owner = -1; -} - -static void netdev_init_queue_locks(struct net_device *dev) -{ - netdev_for_each_tx_queue(dev, __netdev_init_queue_locks_one, NULL); - __netdev_init_queue_locks_one(dev, &dev->rx_queue, NULL); -} - unsigned long netdev_fix_features(unsigned long features, const char *name) { /* Fix illegal SG+CSUM combinations. */ @@ -4941,6 +5018,66 @@ void netif_stacked_transfer_operstate(const struct net_device *rootdev, } EXPORT_SYMBOL(netif_stacked_transfer_operstate); +static int netif_alloc_rx_queues(struct net_device *dev) +{ +#ifdef CONFIG_RPS + unsigned int i, count = dev->num_rx_queues; + struct netdev_rx_queue *rx; + + BUG_ON(count < 1); + + rx = kcalloc(count, sizeof(struct netdev_rx_queue), GFP_KERNEL); + if (!rx) { + pr_err("netdev: Unable to allocate %u rx queues.\n", count); + return -ENOMEM; + } + dev->_rx = rx; + + /* + * Set a pointer to first element in the array which holds the + * reference count. + */ + for (i = 0; i < count; i++) + rx[i].first = rx; +#endif + return 0; +} + +static int netif_alloc_netdev_queues(struct net_device *dev) +{ + unsigned int count = dev->num_tx_queues; + struct netdev_queue *tx; + + BUG_ON(count < 1); + + tx = kcalloc(count, sizeof(struct netdev_queue), GFP_KERNEL); + if (!tx) { + pr_err("netdev: Unable to allocate %u tx queues.\n", + count); + return -ENOMEM; + } + dev->_tx = tx; + return 0; +} + +static void netdev_init_one_queue(struct net_device *dev, + struct netdev_queue *queue, + void *_unused) +{ + queue->dev = dev; + + /* Initialize queue lock */ + spin_lock_init(&queue->_xmit_lock); + netdev_set_xmit_lockdep_class(&queue->_xmit_lock, dev->type); + queue->xmit_lock_owner = -1; +} + +static void netdev_init_queues(struct net_device *dev) +{ + netdev_for_each_tx_queue(dev, netdev_init_one_queue, NULL); + spin_lock_init(&dev->tx_global_lock); +} + /** * register_netdevice - register a network device * @dev: device to register @@ -4974,28 +5111,19 @@ int register_netdevice(struct net_device *dev) spin_lock_init(&dev->addr_list_lock); netdev_set_addr_lockdep_class(dev); - netdev_init_queue_locks(dev); dev->iflink = -1; -#ifdef CONFIG_RPS - if (!dev->num_rx_queues) { - /* - * Allocate a single RX queue if driver never called - * alloc_netdev_mq - */ + ret = netif_alloc_rx_queues(dev); + if (ret) + goto out; - dev->_rx = kzalloc(sizeof(struct netdev_rx_queue), GFP_KERNEL); - if (!dev->_rx) { - ret = -ENOMEM; - goto out; - } + ret = netif_alloc_netdev_queues(dev); + if (ret) + goto out; + + netdev_init_queues(dev); - dev->_rx->first = dev->_rx; - atomic_set(&dev->_rx->count, 1); - dev->num_rx_queues = 1; - } -#endif /* Init, if this function is available */ if (dev->netdev_ops->ndo_init) { ret = dev->netdev_ops->ndo_init(dev); @@ -5035,6 +5163,12 @@ int register_netdevice(struct net_device *dev) if (dev->features & NETIF_F_SG) dev->features |= NETIF_F_GSO; + /* Enable GRO and NETIF_F_HIGHDMA for vlans by default, + * vlan_dev_init() will do the dev->features check, so these features + * are enabled only if supported by underlying device. + */ + dev->vlan_features |= (NETIF_F_GRO | NETIF_F_HIGHDMA); + ret = call_netdevice_notifiers(NETDEV_POST_INIT, dev); ret = notifier_to_errno(ret); if (ret) @@ -5105,9 +5239,6 @@ int init_dummy_netdev(struct net_device *dev) */ dev->reg_state = NETREG_DUMMY; - /* initialize the ref count */ - atomic_set(&dev->refcnt, 1); - /* NAPI wants this */ INIT_LIST_HEAD(&dev->napi_list); @@ -5115,6 +5246,11 @@ int init_dummy_netdev(struct net_device *dev) set_bit(__LINK_STATE_PRESENT, &dev->state); set_bit(__LINK_STATE_START, &dev->state); + /* Note : We dont allocate pcpu_refcnt for dummy devices, + * because users of this 'device' dont need to change + * its refcount. + */ + return 0; } EXPORT_SYMBOL_GPL(init_dummy_netdev); @@ -5156,6 +5292,16 @@ out: } EXPORT_SYMBOL(register_netdev); +int netdev_refcnt_read(const struct net_device *dev) +{ + int i, refcnt = 0; + + for_each_possible_cpu(i) + refcnt += *per_cpu_ptr(dev->pcpu_refcnt, i); + return refcnt; +} +EXPORT_SYMBOL(netdev_refcnt_read); + /* * netdev_wait_allrefs - wait until all references are gone. * @@ -5170,11 +5316,14 @@ EXPORT_SYMBOL(register_netdev); static void netdev_wait_allrefs(struct net_device *dev) { unsigned long rebroadcast_time, warning_time; + int refcnt; linkwatch_forget_dev(dev); rebroadcast_time = warning_time = jiffies; - while (atomic_read(&dev->refcnt) != 0) { + refcnt = netdev_refcnt_read(dev); + + while (refcnt != 0) { if (time_after(jiffies, rebroadcast_time + 1 * HZ)) { rtnl_lock(); @@ -5201,11 +5350,13 @@ static void netdev_wait_allrefs(struct net_device *dev) msleep(250); + refcnt = netdev_refcnt_read(dev); + if (time_after(jiffies, warning_time + 10 * HZ)) { printk(KERN_EMERG "unregister_netdevice: " "waiting for %s to become free. Usage " "count = %d\n", - dev->name, atomic_read(&dev->refcnt)); + dev->name, refcnt); warning_time = jiffies; } } @@ -5263,8 +5414,8 @@ void netdev_run_todo(void) netdev_wait_allrefs(dev); /* paranoia */ - BUG_ON(atomic_read(&dev->refcnt)); - WARN_ON(dev->ip_ptr); + BUG_ON(netdev_refcnt_read(dev)); + WARN_ON(rcu_dereference_raw(dev->ip_ptr)); WARN_ON(dev->ip6_ptr); WARN_ON(dev->dn_ptr); @@ -5342,30 +5493,34 @@ struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev, if (ops->ndo_get_stats64) { memset(storage, 0, sizeof(*storage)); - return ops->ndo_get_stats64(dev, storage); - } - if (ops->ndo_get_stats) { + ops->ndo_get_stats64(dev, storage); + } else if (ops->ndo_get_stats) { netdev_stats_to_stats64(storage, ops->ndo_get_stats(dev)); - return storage; + } else { + netdev_stats_to_stats64(storage, &dev->stats); + dev_txq_stats_fold(dev, storage); } - netdev_stats_to_stats64(storage, &dev->stats); - dev_txq_stats_fold(dev, storage); + storage->rx_dropped += atomic_long_read(&dev->rx_dropped); return storage; } EXPORT_SYMBOL(dev_get_stats); -static void netdev_init_one_queue(struct net_device *dev, - struct netdev_queue *queue, - void *_unused) +struct netdev_queue *dev_ingress_queue_create(struct net_device *dev) { - queue->dev = dev; -} + struct netdev_queue *queue = dev_ingress_queue(dev); -static void netdev_init_queues(struct net_device *dev) -{ - netdev_init_one_queue(dev, &dev->rx_queue, NULL); - netdev_for_each_tx_queue(dev, netdev_init_one_queue, NULL); - spin_lock_init(&dev->tx_global_lock); +#ifdef CONFIG_NET_CLS_ACT + if (queue) + return queue; + queue = kzalloc(sizeof(*queue), GFP_KERNEL); + if (!queue) + return NULL; + netdev_init_one_queue(dev, queue, NULL); + queue->qdisc = &noop_qdisc; + queue->qdisc_sleeping = &noop_qdisc; + rcu_assign_pointer(dev->ingress_queue, queue); +#endif + return queue; } /** @@ -5382,17 +5537,18 @@ static void netdev_init_queues(struct net_device *dev) struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, void (*setup)(struct net_device *), unsigned int queue_count) { - struct netdev_queue *tx; struct net_device *dev; size_t alloc_size; struct net_device *p; -#ifdef CONFIG_RPS - struct netdev_rx_queue *rx; - int i; -#endif BUG_ON(strlen(name) >= sizeof(dev->name)); + if (queue_count < 1) { + pr_err("alloc_netdev: Unable to allocate device " + "with zero queues.\n"); + return NULL; + } + alloc_size = sizeof(struct net_device); if (sizeof_priv) { /* ensure 32-byte alignment of private area */ @@ -5408,55 +5564,31 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, return NULL; } - tx = kcalloc(queue_count, sizeof(struct netdev_queue), GFP_KERNEL); - if (!tx) { - printk(KERN_ERR "alloc_netdev: Unable to allocate " - "tx qdiscs.\n"); - goto free_p; - } - -#ifdef CONFIG_RPS - rx = kcalloc(queue_count, sizeof(struct netdev_rx_queue), GFP_KERNEL); - if (!rx) { - printk(KERN_ERR "alloc_netdev: Unable to allocate " - "rx queues.\n"); - goto free_tx; - } - - atomic_set(&rx->count, queue_count); - - /* - * Set a pointer to first element in the array which holds the - * reference count. - */ - for (i = 0; i < queue_count; i++) - rx[i].first = rx; -#endif - dev = PTR_ALIGN(p, NETDEV_ALIGN); dev->padded = (char *)dev - (char *)p; + dev->pcpu_refcnt = alloc_percpu(int); + if (!dev->pcpu_refcnt) + goto free_p; + if (dev_addr_init(dev)) - goto free_rx; + goto free_pcpu; dev_mc_init(dev); dev_uc_init(dev); dev_net_set(dev, &init_net); - dev->_tx = tx; dev->num_tx_queues = queue_count; dev->real_num_tx_queues = queue_count; #ifdef CONFIG_RPS - dev->_rx = rx; dev->num_rx_queues = queue_count; + dev->real_num_rx_queues = queue_count; #endif dev->gso_max_size = GSO_MAX_SIZE; - netdev_init_queues(dev); - INIT_LIST_HEAD(&dev->ethtool_ntuple_list.list); dev->ethtool_ntuple_list.count = 0; INIT_LIST_HEAD(&dev->napi_list); @@ -5467,12 +5599,8 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, strcpy(dev->name, name); return dev; -free_rx: -#ifdef CONFIG_RPS - kfree(rx); -free_tx: -#endif - kfree(tx); +free_pcpu: + free_percpu(dev->pcpu_refcnt); free_p: kfree(p); return NULL; @@ -5495,6 +5623,8 @@ void free_netdev(struct net_device *dev) kfree(dev->_tx); + kfree(rcu_dereference_raw(dev->ingress_queue)); + /* Flush device addresses */ dev_addr_flush(dev); @@ -5504,6 +5634,9 @@ void free_netdev(struct net_device *dev) list_for_each_entry_safe(p, n, &dev->napi_list, dev_list) netif_napi_del(p); + free_percpu(dev->pcpu_refcnt); + dev->pcpu_refcnt = NULL; + /* Compatibility with error handling in drivers */ if (dev->reg_state == NETREG_UNINITIALIZED) { kfree((char *)dev - dev->padded); @@ -5658,6 +5791,10 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char /* Notify protocols, that we are about to destroy this device. They should clean all the things. + + Note that dev->reg_state stays at NETREG_REGISTERED. + This is wanted because this way 8021q and macvlan know + the device is just moving and can keep their slaves up. */ call_netdevice_notifiers(NETDEV_UNREGISTER, dev); call_netdevice_notifiers(NETDEV_UNREGISTER_BATCH, dev); diff --git a/net/core/dst.c b/net/core/dst.c index 6c41b1fac3d..8abe628b79f 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -168,7 +168,7 @@ void *dst_alloc(struct dst_ops *ops) { struct dst_entry *dst; - if (ops->gc && atomic_read(&ops->entries) > ops->gc_thresh) { + if (ops->gc && dst_entries_get_fast(ops) > ops->gc_thresh) { if (ops->gc(ops)) return NULL; } @@ -183,7 +183,7 @@ void *dst_alloc(struct dst_ops *ops) #if RT_CACHE_DEBUG >= 2 atomic_inc(&dst_total); #endif - atomic_inc(&ops->entries); + dst_entries_add(ops, 1); return dst; } EXPORT_SYMBOL(dst_alloc); @@ -228,15 +228,15 @@ again: child = dst->child; dst->hh = NULL; - if (hh && atomic_dec_and_test(&hh->hh_refcnt)) - kfree(hh); + if (hh) + hh_cache_put(hh); if (neigh) { dst->neighbour = NULL; neigh_release(neigh); } - atomic_dec(&dst->ops->entries); + dst_entries_add(dst->ops, -1); if (dst->ops->destroy) dst->ops->destroy(dst); @@ -271,13 +271,40 @@ void dst_release(struct dst_entry *dst) if (dst) { int newrefcnt; - smp_mb__before_atomic_dec(); newrefcnt = atomic_dec_return(&dst->__refcnt); WARN_ON(newrefcnt < 0); + if (unlikely(dst->flags & DST_NOCACHE) && !newrefcnt) { + dst = dst_destroy(dst); + if (dst) + __dst_free(dst); + } } } EXPORT_SYMBOL(dst_release); +/** + * skb_dst_set_noref - sets skb dst, without a reference + * @skb: buffer + * @dst: dst entry + * + * Sets skb dst, assuming a reference was not taken on dst + * skb_dst_drop() should not dst_release() this dst + */ +void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst) +{ + WARN_ON(!rcu_read_lock_held() && !rcu_read_lock_bh_held()); + /* If dst not in cache, we must take a reference, because + * dst_release() will destroy dst as soon as its refcount becomes zero + */ + if (unlikely(dst->flags & DST_NOCACHE)) { + dst_hold(dst); + skb_dst_set(skb, dst); + } else { + skb->_skb_refdst = (unsigned long)dst | SKB_DST_NOREF; + } +} +EXPORT_SYMBOL(skb_dst_set_noref); + /* Dirty hack. We did it in 2.2 (in __dst_free), * we have _very_ good reasons not to repeat * this mistake in 2.3, but we have no choice diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 7a85367b3c2..956a9f4971c 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -19,6 +19,7 @@ #include <linux/netdevice.h> #include <linux/bitops.h> #include <linux/uaccess.h> +#include <linux/vmalloc.h> #include <linux/slab.h> /* @@ -131,7 +132,8 @@ EXPORT_SYMBOL(ethtool_op_set_ufo); * NETIF_F_xxx values in include/linux/netdevice.h */ static const u32 flags_dup_features = - (ETH_FLAG_LRO | ETH_FLAG_NTUPLE | ETH_FLAG_RXHASH); + (ETH_FLAG_LRO | ETH_FLAG_RXVLAN | ETH_FLAG_TXVLAN | ETH_FLAG_NTUPLE | + ETH_FLAG_RXHASH); u32 ethtool_op_get_flags(struct net_device *dev) { @@ -205,18 +207,24 @@ static noinline_for_stack int ethtool_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo info; const struct ethtool_ops *ops = dev->ethtool_ops; - if (!ops->get_drvinfo) - return -EOPNOTSUPP; - memset(&info, 0, sizeof(info)); info.cmd = ETHTOOL_GDRVINFO; - ops->get_drvinfo(dev, &info); + if (ops && ops->get_drvinfo) { + ops->get_drvinfo(dev, &info); + } else if (dev->dev.parent && dev->dev.parent->driver) { + strlcpy(info.bus_info, dev_name(dev->dev.parent), + sizeof(info.bus_info)); + strlcpy(info.driver, dev->dev.parent->driver->name, + sizeof(info.driver)); + } else { + return -EOPNOTSUPP; + } /* * this method of obtaining string set info is deprecated; * Use ETHTOOL_GSSET_INFO instead. */ - if (ops->get_sset_count) { + if (ops && ops->get_sset_count) { int rc; rc = ops->get_sset_count(dev, ETH_SS_TEST); @@ -229,9 +237,9 @@ static noinline_for_stack int ethtool_get_drvinfo(struct net_device *dev, if (rc >= 0) info.n_priv_flags = rc; } - if (ops->get_regs_len) + if (ops && ops->get_regs_len) info.regdump_len = ops->get_regs_len(dev); - if (ops->get_eeprom_len) + if (ops && ops->get_eeprom_len) info.eedump_len = ops->get_eeprom_len(dev); if (copy_to_user(useraddr, &info, sizeof(info))) @@ -348,7 +356,7 @@ static noinline_for_stack int ethtool_get_rxnfc(struct net_device *dev, if (info.cmd == ETHTOOL_GRXCLSRLALL) { if (info.rule_cnt > 0) { if (info.rule_cnt <= KMALLOC_MAX_SIZE / sizeof(u32)) - rule_buf = kmalloc(info.rule_cnt * sizeof(u32), + rule_buf = kzalloc(info.rule_cnt * sizeof(u32), GFP_USER); if (!rule_buf) return -ENOMEM; @@ -397,7 +405,7 @@ static noinline_for_stack int ethtool_get_rxfh_indir(struct net_device *dev, (KMALLOC_MAX_SIZE - sizeof(*indir)) / sizeof(*indir->ring_index)) return -ENOMEM; full_size = sizeof(*indir) + sizeof(*indir->ring_index) * table_size; - indir = kmalloc(full_size, GFP_USER); + indir = kzalloc(full_size, GFP_USER); if (!indir) return -ENOMEM; @@ -479,6 +487,38 @@ static void __rx_ntuple_filter_add(struct ethtool_rx_ntuple_list *list, list->count++; } +/* + * ethtool does not (or did not) set masks for flow parameters that are + * not specified, so if both value and mask are 0 then this must be + * treated as equivalent to a mask with all bits set. Implement that + * here rather than in drivers. + */ +static void rx_ntuple_fix_masks(struct ethtool_rx_ntuple_flow_spec *fs) +{ + struct ethtool_tcpip4_spec *entry = &fs->h_u.tcp_ip4_spec; + struct ethtool_tcpip4_spec *mask = &fs->m_u.tcp_ip4_spec; + + if (fs->flow_type != TCP_V4_FLOW && + fs->flow_type != UDP_V4_FLOW && + fs->flow_type != SCTP_V4_FLOW) + return; + + if (!(entry->ip4src | mask->ip4src)) + mask->ip4src = htonl(0xffffffff); + if (!(entry->ip4dst | mask->ip4dst)) + mask->ip4dst = htonl(0xffffffff); + if (!(entry->psrc | mask->psrc)) + mask->psrc = htons(0xffff); + if (!(entry->pdst | mask->pdst)) + mask->pdst = htons(0xffff); + if (!(entry->tos | mask->tos)) + mask->tos = 0xff; + if (!(fs->vlan_tag | fs->vlan_tag_mask)) + fs->vlan_tag_mask = 0xffff; + if (!(fs->data | fs->data_mask)) + fs->data_mask = 0xffffffffffffffffULL; +} + static noinline_for_stack int ethtool_set_rx_ntuple(struct net_device *dev, void __user *useraddr) { @@ -493,6 +533,8 @@ static noinline_for_stack int ethtool_set_rx_ntuple(struct net_device *dev, if (copy_from_user(&cmd, useraddr, sizeof(cmd))) return -EFAULT; + rx_ntuple_fix_masks(&cmd.fs); + /* * Cache filter in dev struct for GET operation only if * the underlying driver doesn't have its own GET operation, and @@ -538,7 +580,7 @@ static int ethtool_get_rx_ntuple(struct net_device *dev, void __user *useraddr) gstrings.len = ret; - data = kmalloc(gstrings.len * ETH_GSTRING_LEN, GFP_USER); + data = kzalloc(gstrings.len * ETH_GSTRING_LEN, GFP_USER); if (!data) return -ENOMEM; @@ -667,19 +709,19 @@ static int ethtool_get_rx_ntuple(struct net_device *dev, void __user *useraddr) break; case IP_USER_FLOW: sprintf(p, "\tSrc IP addr: 0x%x\n", - fsc->fs.h_u.raw_ip4_spec.ip4src); + fsc->fs.h_u.usr_ip4_spec.ip4src); p += ETH_GSTRING_LEN; num_strings++; sprintf(p, "\tSrc IP mask: 0x%x\n", - fsc->fs.m_u.raw_ip4_spec.ip4src); + fsc->fs.m_u.usr_ip4_spec.ip4src); p += ETH_GSTRING_LEN; num_strings++; sprintf(p, "\tDest IP addr: 0x%x\n", - fsc->fs.h_u.raw_ip4_spec.ip4dst); + fsc->fs.h_u.usr_ip4_spec.ip4dst); p += ETH_GSTRING_LEN; num_strings++; sprintf(p, "\tDest IP mask: 0x%x\n", - fsc->fs.m_u.raw_ip4_spec.ip4dst); + fsc->fs.m_u.usr_ip4_spec.ip4dst); p += ETH_GSTRING_LEN; num_strings++; break; @@ -775,7 +817,7 @@ static int ethtool_get_regs(struct net_device *dev, char __user *useraddr) if (regs.len > reglen) regs.len = reglen; - regbuf = kmalloc(reglen, GFP_USER); + regbuf = vmalloc(reglen); if (!regbuf) return -ENOMEM; @@ -790,7 +832,7 @@ static int ethtool_get_regs(struct net_device *dev, char __user *useraddr) ret = 0; out: - kfree(regbuf); + vfree(regbuf); return ret; } @@ -1175,8 +1217,11 @@ static int ethtool_set_gro(struct net_device *dev, char __user *useraddr) return -EFAULT; if (edata.data) { - if (!dev->ethtool_ops->get_rx_csum || - !dev->ethtool_ops->get_rx_csum(dev)) + u32 rxcsum = dev->ethtool_ops->get_rx_csum ? + dev->ethtool_ops->get_rx_csum(dev) : + ethtool_op_get_rx_csum(dev); + + if (!rxcsum) return -EINVAL; dev->features |= NETIF_F_GRO; } else @@ -1402,14 +1447,22 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) if (!dev || !netif_device_present(dev)) return -ENODEV; - if (!dev->ethtool_ops) - return -EOPNOTSUPP; - if (copy_from_user(ðcmd, useraddr, sizeof(ethcmd))) return -EFAULT; + if (!dev->ethtool_ops) { + /* ETHTOOL_GDRVINFO does not require any driver support. + * It is also unprivileged and does not change anything, + * so we can take a shortcut to it. */ + if (ethcmd == ETHTOOL_GDRVINFO) + return ethtool_get_drvinfo(dev, useraddr); + else + return -EOPNOTSUPP; + } + /* Allow some commands to be done by anyone */ switch (ethcmd) { + case ETHTOOL_GSET: case ETHTOOL_GDRVINFO: case ETHTOOL_GMSGLVL: case ETHTOOL_GCOALESCE: diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 42e84e08a1b..1bc3f253ba6 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -144,7 +144,7 @@ fib_rules_register(const struct fib_rules_ops *tmpl, struct net *net) } EXPORT_SYMBOL_GPL(fib_rules_register); -void fib_rules_cleanup_ops(struct fib_rules_ops *ops) +static void fib_rules_cleanup_ops(struct fib_rules_ops *ops) { struct fib_rule *rule, *tmp; @@ -153,7 +153,6 @@ void fib_rules_cleanup_ops(struct fib_rules_ops *ops) fib_rule_put(rule); } } -EXPORT_SYMBOL_GPL(fib_rules_cleanup_ops); static void fib_rules_put_rcu(struct rcu_head *head) { @@ -182,7 +181,8 @@ static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops, { int ret = 0; - if (rule->iifindex && (rule->iifindex != fl->iif)) + if (rule->iifindex && (rule->iifindex != fl->iif) && + !(fl->flags & FLOWI_FLAG_MATCH_ANY_IIF)) goto out; if (rule->oifindex && (rule->oifindex != fl->oif)) @@ -225,9 +225,12 @@ jumped: err = ops->action(rule, fl, flags, arg); if (err != -EAGAIN) { - fib_rule_get(rule); - arg->rule = rule; - goto out; + if ((arg->flags & FIB_LOOKUP_NOREF) || + likely(atomic_inc_not_zero(&rule->refcnt))) { + arg->rule = rule; + goto out; + } + break; } } @@ -491,7 +494,6 @@ static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) } } - synchronize_rcu(); notify_rule_change(RTM_DELRULE, rule, ops, nlh, NETLINK_CB(skb).pid); fib_rule_put(rule); diff --git a/net/core/filter.c b/net/core/filter.c index 52b051f82a0..7adf5035291 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -638,10 +638,9 @@ int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk) return err; } - rcu_read_lock_bh(); - old_fp = rcu_dereference_bh(sk->sk_filter); + old_fp = rcu_dereference_protected(sk->sk_filter, + sock_owned_by_user(sk)); rcu_assign_pointer(sk->sk_filter, fp); - rcu_read_unlock_bh(); if (old_fp) sk_filter_delayed_uncharge(sk, old_fp); @@ -654,14 +653,13 @@ int sk_detach_filter(struct sock *sk) int ret = -ENOENT; struct sk_filter *filter; - rcu_read_lock_bh(); - filter = rcu_dereference_bh(sk->sk_filter); + filter = rcu_dereference_protected(sk->sk_filter, + sock_owned_by_user(sk)); if (filter) { rcu_assign_pointer(sk->sk_filter, NULL); sk_filter_delayed_uncharge(sk, filter); ret = 0; } - rcu_read_unlock_bh(); return ret; } EXPORT_SYMBOL_GPL(sk_detach_filter); diff --git a/net/core/flow.c b/net/core/flow.c index f67dcbfe54e..127c8a7ffd6 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -53,8 +53,7 @@ struct flow_flush_info { struct flow_cache { u32 hash_shift; - unsigned long order; - struct flow_cache_percpu *percpu; + struct flow_cache_percpu __percpu *percpu; struct notifier_block hotcpu_notifier; int low_watermark; int high_watermark; @@ -64,7 +63,7 @@ struct flow_cache { atomic_t flow_cache_genid = ATOMIC_INIT(0); EXPORT_SYMBOL(flow_cache_genid); static struct flow_cache flow_cache_global; -static struct kmem_cache *flow_cachep; +static struct kmem_cache *flow_cachep __read_mostly; static DEFINE_SPINLOCK(flow_cache_gc_lock); static LIST_HEAD(flow_cache_gc_list); @@ -177,15 +176,11 @@ static u32 flow_hash_code(struct flow_cache *fc, { u32 *k = (u32 *) key; - return (jhash2(k, (sizeof(*key) / sizeof(u32)), fcp->hash_rnd) - & (flow_cache_hash_size(fc) - 1)); + return jhash2(k, (sizeof(*key) / sizeof(u32)), fcp->hash_rnd) + & (flow_cache_hash_size(fc) - 1); } -#if (BITS_PER_LONG == 64) -typedef u64 flow_compare_t; -#else -typedef u32 flow_compare_t; -#endif +typedef unsigned long flow_compare_t; /* I hear what you're saying, use memcmp. But memcmp cannot make * important assumptions that we can here, such as alignment and @@ -357,62 +352,73 @@ void flow_cache_flush(void) put_online_cpus(); } -static void __init flow_cache_cpu_prepare(struct flow_cache *fc, - struct flow_cache_percpu *fcp) +static int __cpuinit flow_cache_cpu_prepare(struct flow_cache *fc, int cpu) { - fcp->hash_table = (struct hlist_head *) - __get_free_pages(GFP_KERNEL|__GFP_ZERO, fc->order); - if (!fcp->hash_table) - panic("NET: failed to allocate flow cache order %lu\n", fc->order); + struct flow_cache_percpu *fcp = per_cpu_ptr(fc->percpu, cpu); + size_t sz = sizeof(struct hlist_head) * flow_cache_hash_size(fc); - fcp->hash_rnd_recalc = 1; - fcp->hash_count = 0; - tasklet_init(&fcp->flush_tasklet, flow_cache_flush_tasklet, 0); + if (!fcp->hash_table) { + fcp->hash_table = kzalloc_node(sz, GFP_KERNEL, cpu_to_node(cpu)); + if (!fcp->hash_table) { + pr_err("NET: failed to allocate flow cache sz %zu\n", sz); + return -ENOMEM; + } + fcp->hash_rnd_recalc = 1; + fcp->hash_count = 0; + tasklet_init(&fcp->flush_tasklet, flow_cache_flush_tasklet, 0); + } + return 0; } -static int flow_cache_cpu(struct notifier_block *nfb, +static int __cpuinit flow_cache_cpu(struct notifier_block *nfb, unsigned long action, void *hcpu) { struct flow_cache *fc = container_of(nfb, struct flow_cache, hotcpu_notifier); - int cpu = (unsigned long) hcpu; + int res, cpu = (unsigned long) hcpu; struct flow_cache_percpu *fcp = per_cpu_ptr(fc->percpu, cpu); - if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) + switch (action) { + case CPU_UP_PREPARE: + case CPU_UP_PREPARE_FROZEN: + res = flow_cache_cpu_prepare(fc, cpu); + if (res) + return notifier_from_errno(res); + break; + case CPU_DEAD: + case CPU_DEAD_FROZEN: __flow_cache_shrink(fc, fcp, 0); + break; + } return NOTIFY_OK; } -static int flow_cache_init(struct flow_cache *fc) +static int __init flow_cache_init(struct flow_cache *fc) { - unsigned long order; int i; fc->hash_shift = 10; fc->low_watermark = 2 * flow_cache_hash_size(fc); fc->high_watermark = 4 * flow_cache_hash_size(fc); - for (order = 0; - (PAGE_SIZE << order) < - (sizeof(struct hlist_head)*flow_cache_hash_size(fc)); - order++) - /* NOTHING */; - fc->order = order; fc->percpu = alloc_percpu(struct flow_cache_percpu); + if (!fc->percpu) + return -ENOMEM; - setup_timer(&fc->rnd_timer, flow_cache_new_hashrnd, - (unsigned long) fc); - fc->rnd_timer.expires = jiffies + FLOW_HASH_RND_PERIOD; - add_timer(&fc->rnd_timer); - - for_each_possible_cpu(i) - flow_cache_cpu_prepare(fc, per_cpu_ptr(fc->percpu, i)); - + for_each_online_cpu(i) { + if (flow_cache_cpu_prepare(fc, i)) + return -ENOMEM; + } fc->hotcpu_notifier = (struct notifier_block){ .notifier_call = flow_cache_cpu, }; register_hotcpu_notifier(&fc->hotcpu_notifier); + setup_timer(&fc->rnd_timer, flow_cache_new_hashrnd, + (unsigned long) fc); + fc->rnd_timer.expires = jiffies + FLOW_HASH_RND_PERIOD; + add_timer(&fc->rnd_timer); + return 0; } diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c index 9fbe7f7429b..7c2373321b7 100644 --- a/net/core/gen_estimator.c +++ b/net/core/gen_estimator.c @@ -232,7 +232,7 @@ int gen_new_estimator(struct gnet_stats_basic_packed *bstats, est->last_packets = bstats->packets; est->avpps = rate_est->pps<<10; - spin_lock(&est_tree_lock); + spin_lock_bh(&est_tree_lock); if (!elist[idx].timer.function) { INIT_LIST_HEAD(&elist[idx].list); setup_timer(&elist[idx].timer, est_timer, idx); @@ -243,7 +243,7 @@ int gen_new_estimator(struct gnet_stats_basic_packed *bstats, list_add_rcu(&est->list, &elist[idx].list); gen_add_node(est); - spin_unlock(&est_tree_lock); + spin_unlock_bh(&est_tree_lock); return 0; } @@ -270,18 +270,18 @@ void gen_kill_estimator(struct gnet_stats_basic_packed *bstats, { struct gen_estimator *e; - spin_lock(&est_tree_lock); + spin_lock_bh(&est_tree_lock); while ((e = gen_find_node(bstats, rate_est))) { rb_erase(&e->node, &est_root); - write_lock_bh(&est_lock); + write_lock(&est_lock); e->bstats = NULL; - write_unlock_bh(&est_lock); + write_unlock(&est_lock); list_del_rcu(&e->list); call_rcu(&e->e_rcu, __gen_kill_estimator); } - spin_unlock(&est_tree_lock); + spin_unlock_bh(&est_tree_lock); } EXPORT_SYMBOL(gen_kill_estimator); @@ -320,9 +320,9 @@ bool gen_estimator_active(const struct gnet_stats_basic_packed *bstats, ASSERT_RTNL(); - spin_lock(&est_tree_lock); + spin_lock_bh(&est_tree_lock); res = gen_find_node(bstats, rate_est) != NULL; - spin_unlock(&est_tree_lock); + spin_unlock_bh(&est_tree_lock); return res; } diff --git a/net/core/iovec.c b/net/core/iovec.c index 1cd98df412d..72aceb1fe4f 100644 --- a/net/core/iovec.c +++ b/net/core/iovec.c @@ -35,13 +35,16 @@ * in any case. */ -int verify_iovec(struct msghdr *m, struct iovec *iov, struct sockaddr *address, int mode) +long verify_iovec(struct msghdr *m, struct iovec *iov, struct sockaddr *address, int mode) { - int size, err, ct; + int size, ct; + long err; if (m->msg_namelen) { if (mode == VERIFY_READ) { - err = move_addr_to_kernel(m->msg_name, m->msg_namelen, + void __user *namep; + namep = (void __user __force *) m->msg_name; + err = move_addr_to_kernel(namep, m->msg_namelen, address); if (err < 0) return err; @@ -52,7 +55,7 @@ int verify_iovec(struct msghdr *m, struct iovec *iov, struct sockaddr *address, } size = m->msg_iovlen * sizeof(struct iovec); - if (copy_from_user(iov, m->msg_iov, size)) + if (copy_from_user(iov, (void __user __force *) m->msg_iov, size)) return -EFAULT; m->msg_iov = iov; diff --git a/net/core/neighbour.c b/net/core/neighbour.c index a4e0a7482c2..8cc8f9a79db 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -122,7 +122,7 @@ static void neigh_cleanup_and_release(struct neighbour *neigh) unsigned long neigh_rand_reach_time(unsigned long base) { - return (base ? (net_random() % base) + (base >> 1) : 0); + return base ? (net_random() % base) + (base >> 1) : 0; } EXPORT_SYMBOL(neigh_rand_reach_time); @@ -131,15 +131,20 @@ static int neigh_forced_gc(struct neigh_table *tbl) { int shrunk = 0; int i; + struct neigh_hash_table *nht; NEIGH_CACHE_STAT_INC(tbl, forced_gc_runs); write_lock_bh(&tbl->lock); - for (i = 0; i <= tbl->hash_mask; i++) { - struct neighbour *n, **np; + nht = rcu_dereference_protected(tbl->nht, + lockdep_is_held(&tbl->lock)); + for (i = 0; i <= nht->hash_mask; i++) { + struct neighbour *n; + struct neighbour __rcu **np; - np = &tbl->hash_buckets[i]; - while ((n = *np) != NULL) { + np = &nht->hash_buckets[i]; + while ((n = rcu_dereference_protected(*np, + lockdep_is_held(&tbl->lock))) != NULL) { /* Neighbour record may be discarded if: * - nobody refers to it. * - it is not permanent @@ -147,7 +152,9 @@ static int neigh_forced_gc(struct neigh_table *tbl) write_lock(&n->lock); if (atomic_read(&n->refcnt) == 1 && !(n->nud_state & NUD_PERMANENT)) { - *np = n->next; + rcu_assign_pointer(*np, + rcu_dereference_protected(n->next, + lockdep_is_held(&tbl->lock))); n->dead = 1; shrunk = 1; write_unlock(&n->lock); @@ -199,16 +206,24 @@ static void pneigh_queue_purge(struct sk_buff_head *list) static void neigh_flush_dev(struct neigh_table *tbl, struct net_device *dev) { int i; + struct neigh_hash_table *nht; - for (i = 0; i <= tbl->hash_mask; i++) { - struct neighbour *n, **np = &tbl->hash_buckets[i]; + nht = rcu_dereference_protected(tbl->nht, + lockdep_is_held(&tbl->lock)); - while ((n = *np) != NULL) { + for (i = 0; i <= nht->hash_mask; i++) { + struct neighbour *n; + struct neighbour __rcu **np = &nht->hash_buckets[i]; + + while ((n = rcu_dereference_protected(*np, + lockdep_is_held(&tbl->lock))) != NULL) { if (dev && n->dev != dev) { np = &n->next; continue; } - *np = n->next; + rcu_assign_pointer(*np, + rcu_dereference_protected(n->next, + lockdep_is_held(&tbl->lock))); write_lock(&n->lock); neigh_del_timer(n); n->dead = 1; @@ -279,6 +294,7 @@ static struct neighbour *neigh_alloc(struct neigh_table *tbl) skb_queue_head_init(&n->arp_queue); rwlock_init(&n->lock); + seqlock_init(&n->ha_lock); n->updated = n->used = now; n->nud_state = NUD_NONE; n->output = neigh_blackhole; @@ -297,64 +313,86 @@ out_entries: goto out; } -static struct neighbour **neigh_hash_alloc(unsigned int entries) +static struct neigh_hash_table *neigh_hash_alloc(unsigned int entries) { - unsigned long size = entries * sizeof(struct neighbour *); - struct neighbour **ret; + size_t size = entries * sizeof(struct neighbour *); + struct neigh_hash_table *ret; + struct neighbour **buckets; - if (size <= PAGE_SIZE) { - ret = kzalloc(size, GFP_ATOMIC); - } else { - ret = (struct neighbour **) - __get_free_pages(GFP_ATOMIC|__GFP_ZERO, get_order(size)); + ret = kmalloc(sizeof(*ret), GFP_ATOMIC); + if (!ret) + return NULL; + if (size <= PAGE_SIZE) + buckets = kzalloc(size, GFP_ATOMIC); + else + buckets = (struct neighbour **) + __get_free_pages(GFP_ATOMIC | __GFP_ZERO, + get_order(size)); + if (!buckets) { + kfree(ret); + return NULL; } + rcu_assign_pointer(ret->hash_buckets, buckets); + ret->hash_mask = entries - 1; + get_random_bytes(&ret->hash_rnd, sizeof(ret->hash_rnd)); return ret; } -static void neigh_hash_free(struct neighbour **hash, unsigned int entries) +static void neigh_hash_free_rcu(struct rcu_head *head) { - unsigned long size = entries * sizeof(struct neighbour *); + struct neigh_hash_table *nht = container_of(head, + struct neigh_hash_table, + rcu); + size_t size = (nht->hash_mask + 1) * sizeof(struct neighbour *); + struct neighbour **buckets = nht->hash_buckets; if (size <= PAGE_SIZE) - kfree(hash); + kfree(buckets); else - free_pages((unsigned long)hash, get_order(size)); + free_pages((unsigned long)buckets, get_order(size)); + kfree(nht); } -static void neigh_hash_grow(struct neigh_table *tbl, unsigned long new_entries) +static struct neigh_hash_table *neigh_hash_grow(struct neigh_table *tbl, + unsigned long new_entries) { - struct neighbour **new_hash, **old_hash; - unsigned int i, new_hash_mask, old_entries; + unsigned int i, hash; + struct neigh_hash_table *new_nht, *old_nht; NEIGH_CACHE_STAT_INC(tbl, hash_grows); BUG_ON(!is_power_of_2(new_entries)); - new_hash = neigh_hash_alloc(new_entries); - if (!new_hash) - return; - - old_entries = tbl->hash_mask + 1; - new_hash_mask = new_entries - 1; - old_hash = tbl->hash_buckets; + old_nht = rcu_dereference_protected(tbl->nht, + lockdep_is_held(&tbl->lock)); + new_nht = neigh_hash_alloc(new_entries); + if (!new_nht) + return old_nht; - get_random_bytes(&tbl->hash_rnd, sizeof(tbl->hash_rnd)); - for (i = 0; i < old_entries; i++) { + for (i = 0; i <= old_nht->hash_mask; i++) { struct neighbour *n, *next; - for (n = old_hash[i]; n; n = next) { - unsigned int hash_val = tbl->hash(n->primary_key, n->dev); - - hash_val &= new_hash_mask; - next = n->next; - - n->next = new_hash[hash_val]; - new_hash[hash_val] = n; + for (n = rcu_dereference_protected(old_nht->hash_buckets[i], + lockdep_is_held(&tbl->lock)); + n != NULL; + n = next) { + hash = tbl->hash(n->primary_key, n->dev, + new_nht->hash_rnd); + + hash &= new_nht->hash_mask; + next = rcu_dereference_protected(n->next, + lockdep_is_held(&tbl->lock)); + + rcu_assign_pointer(n->next, + rcu_dereference_protected( + new_nht->hash_buckets[hash], + lockdep_is_held(&tbl->lock))); + rcu_assign_pointer(new_nht->hash_buckets[hash], n); } } - tbl->hash_buckets = new_hash; - tbl->hash_mask = new_hash_mask; - neigh_hash_free(old_hash, old_entries); + rcu_assign_pointer(tbl->nht, new_nht); + call_rcu(&old_nht->rcu, neigh_hash_free_rcu); + return new_nht; } struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey, @@ -363,19 +401,26 @@ struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey, struct neighbour *n; int key_len = tbl->key_len; u32 hash_val; + struct neigh_hash_table *nht; NEIGH_CACHE_STAT_INC(tbl, lookups); - read_lock_bh(&tbl->lock); - hash_val = tbl->hash(pkey, dev); - for (n = tbl->hash_buckets[hash_val & tbl->hash_mask]; n; n = n->next) { + rcu_read_lock_bh(); + nht = rcu_dereference_bh(tbl->nht); + hash_val = tbl->hash(pkey, dev, nht->hash_rnd) & nht->hash_mask; + + for (n = rcu_dereference_bh(nht->hash_buckets[hash_val]); + n != NULL; + n = rcu_dereference_bh(n->next)) { if (dev == n->dev && !memcmp(n->primary_key, pkey, key_len)) { - neigh_hold(n); + if (!atomic_inc_not_zero(&n->refcnt)) + n = NULL; NEIGH_CACHE_STAT_INC(tbl, hits); break; } } - read_unlock_bh(&tbl->lock); + + rcu_read_unlock_bh(); return n; } EXPORT_SYMBOL(neigh_lookup); @@ -386,20 +431,27 @@ struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net, struct neighbour *n; int key_len = tbl->key_len; u32 hash_val; + struct neigh_hash_table *nht; NEIGH_CACHE_STAT_INC(tbl, lookups); - read_lock_bh(&tbl->lock); - hash_val = tbl->hash(pkey, NULL); - for (n = tbl->hash_buckets[hash_val & tbl->hash_mask]; n; n = n->next) { + rcu_read_lock_bh(); + nht = rcu_dereference_bh(tbl->nht); + hash_val = tbl->hash(pkey, NULL, nht->hash_rnd) & nht->hash_mask; + + for (n = rcu_dereference_bh(nht->hash_buckets[hash_val]); + n != NULL; + n = rcu_dereference_bh(n->next)) { if (!memcmp(n->primary_key, pkey, key_len) && net_eq(dev_net(n->dev), net)) { - neigh_hold(n); + if (!atomic_inc_not_zero(&n->refcnt)) + n = NULL; NEIGH_CACHE_STAT_INC(tbl, hits); break; } } - read_unlock_bh(&tbl->lock); + + rcu_read_unlock_bh(); return n; } EXPORT_SYMBOL(neigh_lookup_nodev); @@ -411,6 +463,7 @@ struct neighbour *neigh_create(struct neigh_table *tbl, const void *pkey, int key_len = tbl->key_len; int error; struct neighbour *n1, *rc, *n = neigh_alloc(tbl); + struct neigh_hash_table *nht; if (!n) { rc = ERR_PTR(-ENOBUFS); @@ -437,18 +490,24 @@ struct neighbour *neigh_create(struct neigh_table *tbl, const void *pkey, n->confirmed = jiffies - (n->parms->base_reachable_time << 1); write_lock_bh(&tbl->lock); + nht = rcu_dereference_protected(tbl->nht, + lockdep_is_held(&tbl->lock)); - if (atomic_read(&tbl->entries) > (tbl->hash_mask + 1)) - neigh_hash_grow(tbl, (tbl->hash_mask + 1) << 1); + if (atomic_read(&tbl->entries) > (nht->hash_mask + 1)) + nht = neigh_hash_grow(tbl, (nht->hash_mask + 1) << 1); - hash_val = tbl->hash(pkey, dev) & tbl->hash_mask; + hash_val = tbl->hash(pkey, dev, nht->hash_rnd) & nht->hash_mask; if (n->parms->dead) { rc = ERR_PTR(-EINVAL); goto out_tbl_unlock; } - for (n1 = tbl->hash_buckets[hash_val]; n1; n1 = n1->next) { + for (n1 = rcu_dereference_protected(nht->hash_buckets[hash_val], + lockdep_is_held(&tbl->lock)); + n1 != NULL; + n1 = rcu_dereference_protected(n1->next, + lockdep_is_held(&tbl->lock))) { if (dev == n1->dev && !memcmp(n1->primary_key, pkey, key_len)) { neigh_hold(n1); rc = n1; @@ -456,10 +515,12 @@ struct neighbour *neigh_create(struct neigh_table *tbl, const void *pkey, } } - n->next = tbl->hash_buckets[hash_val]; - tbl->hash_buckets[hash_val] = n; n->dead = 0; neigh_hold(n); + rcu_assign_pointer(n->next, + rcu_dereference_protected(nht->hash_buckets[hash_val], + lockdep_is_held(&tbl->lock))); + rcu_assign_pointer(nht->hash_buckets[hash_val], n); write_unlock_bh(&tbl->lock); NEIGH_PRINTK2("neigh %p is created.\n", n); rc = n; @@ -616,6 +677,12 @@ static inline void neigh_parms_put(struct neigh_parms *parms) neigh_parms_destroy(parms); } +static void neigh_destroy_rcu(struct rcu_head *head) +{ + struct neighbour *neigh = container_of(head, struct neighbour, rcu); + + kmem_cache_free(neigh->tbl->kmem_cachep, neigh); +} /* * neighbour must already be out of the table; * @@ -643,8 +710,7 @@ void neigh_destroy(struct neighbour *neigh) write_seqlock_bh(&hh->hh_lock); hh->hh_output = neigh_blackhole; write_sequnlock_bh(&hh->hh_lock); - if (atomic_dec_and_test(&hh->hh_refcnt)) - kfree(hh); + hh_cache_put(hh); } skb_queue_purge(&neigh->arp_queue); @@ -655,7 +721,7 @@ void neigh_destroy(struct neighbour *neigh) NEIGH_PRINTK2("neigh %p is destroyed.\n", neigh); atomic_dec(&neigh->tbl->entries); - kmem_cache_free(neigh->tbl->kmem_cachep, neigh); + call_rcu(&neigh->rcu, neigh_destroy_rcu); } EXPORT_SYMBOL(neigh_destroy); @@ -696,12 +762,16 @@ static void neigh_connect(struct neighbour *neigh) static void neigh_periodic_work(struct work_struct *work) { struct neigh_table *tbl = container_of(work, struct neigh_table, gc_work.work); - struct neighbour *n, **np; + struct neighbour *n; + struct neighbour __rcu **np; unsigned int i; + struct neigh_hash_table *nht; NEIGH_CACHE_STAT_INC(tbl, periodic_gc_runs); write_lock_bh(&tbl->lock); + nht = rcu_dereference_protected(tbl->nht, + lockdep_is_held(&tbl->lock)); /* * periodically recompute ReachableTime from random function @@ -715,10 +785,11 @@ static void neigh_periodic_work(struct work_struct *work) neigh_rand_reach_time(p->base_reachable_time); } - for (i = 0 ; i <= tbl->hash_mask; i++) { - np = &tbl->hash_buckets[i]; + for (i = 0 ; i <= nht->hash_mask; i++) { + np = &nht->hash_buckets[i]; - while ((n = *np) != NULL) { + while ((n = rcu_dereference_protected(*np, + lockdep_is_held(&tbl->lock))) != NULL) { unsigned int state; write_lock(&n->lock); @@ -766,9 +837,9 @@ next_elt: static __inline__ int neigh_max_probes(struct neighbour *n) { struct neigh_parms *p = n->parms; - return (n->nud_state & NUD_PROBE ? + return (n->nud_state & NUD_PROBE) ? p->ucast_probes : - p->ucast_probes + p->app_probes + p->mcast_probes); + p->ucast_probes + p->app_probes + p->mcast_probes; } static void neigh_invalidate(struct neighbour *neigh) @@ -945,7 +1016,7 @@ out_unlock_bh: } EXPORT_SYMBOL(__neigh_event_send); -static void neigh_update_hhs(struct neighbour *neigh) +static void neigh_update_hhs(const struct neighbour *neigh) { struct hh_cache *hh; void (*update)(struct hh_cache*, const struct net_device*, const unsigned char *) @@ -1081,7 +1152,9 @@ int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new, } if (lladdr != neigh->ha) { + write_seqlock(&neigh->ha_lock); memcpy(&neigh->ha, lladdr, dev->addr_len); + write_sequnlock(&neigh->ha_lock); neigh_update_hhs(neigh); if (!(new & NUD_CONNECTED)) neigh->confirmed = jiffies - @@ -1139,44 +1212,73 @@ struct neighbour *neigh_event_ns(struct neigh_table *tbl, } EXPORT_SYMBOL(neigh_event_ns); +static inline bool neigh_hh_lookup(struct neighbour *n, struct dst_entry *dst, + __be16 protocol) +{ + struct hh_cache *hh; + + smp_rmb(); /* paired with smp_wmb() in neigh_hh_init() */ + for (hh = n->hh; hh; hh = hh->hh_next) { + if (hh->hh_type == protocol) { + atomic_inc(&hh->hh_refcnt); + if (unlikely(cmpxchg(&dst->hh, NULL, hh) != NULL)) + hh_cache_put(hh); + return true; + } + } + return false; +} + +/* called with read_lock_bh(&n->lock); */ static void neigh_hh_init(struct neighbour *n, struct dst_entry *dst, __be16 protocol) { struct hh_cache *hh; struct net_device *dev = dst->dev; - for (hh = n->hh; hh; hh = hh->hh_next) - if (hh->hh_type == protocol) - break; + if (likely(neigh_hh_lookup(n, dst, protocol))) + return; - if (!hh && (hh = kzalloc(sizeof(*hh), GFP_ATOMIC)) != NULL) { - seqlock_init(&hh->hh_lock); - hh->hh_type = protocol; - atomic_set(&hh->hh_refcnt, 0); - hh->hh_next = NULL; + /* slow path */ + hh = kzalloc(sizeof(*hh), GFP_ATOMIC); + if (!hh) + return; - if (dev->header_ops->cache(n, hh)) { - kfree(hh); - hh = NULL; - } else { - atomic_inc(&hh->hh_refcnt); - hh->hh_next = n->hh; - n->hh = hh; - if (n->nud_state & NUD_CONNECTED) - hh->hh_output = n->ops->hh_output; - else - hh->hh_output = n->ops->output; - } + seqlock_init(&hh->hh_lock); + hh->hh_type = protocol; + atomic_set(&hh->hh_refcnt, 2); + + if (dev->header_ops->cache(n, hh)) { + kfree(hh); + return; } - if (hh) { - atomic_inc(&hh->hh_refcnt); - dst->hh = hh; + + write_lock_bh(&n->lock); + + /* must check if another thread already did the insert */ + if (neigh_hh_lookup(n, dst, protocol)) { + kfree(hh); + goto end; } + + if (n->nud_state & NUD_CONNECTED) + hh->hh_output = n->ops->hh_output; + else + hh->hh_output = n->ops->output; + + hh->hh_next = n->hh; + smp_wmb(); /* paired with smp_rmb() in neigh_hh_lookup() */ + n->hh = hh; + + if (unlikely(cmpxchg(&dst->hh, NULL, hh) != NULL)) + hh_cache_put(hh); +end: + write_unlock_bh(&n->lock); } /* This function can be used in contexts, where only old dev_queue_xmit - worked, f.e. if you want to override normal output path (eql, shaper), - but resolution is not made yet. + * worked, f.e. if you want to override normal output path (eql, shaper), + * but resolution is not made yet. */ int neigh_compat_output(struct sk_buff *skb) @@ -1210,19 +1312,19 @@ int neigh_resolve_output(struct sk_buff *skb) if (!neigh_event_send(neigh, skb)) { int err; struct net_device *dev = neigh->dev; - if (dev->header_ops->cache && !dst->hh) { - write_lock_bh(&neigh->lock); - if (!dst->hh) - neigh_hh_init(neigh, dst, dst->ops->protocol); - err = dev_hard_header(skb, dev, ntohs(skb->protocol), - neigh->ha, NULL, skb->len); - write_unlock_bh(&neigh->lock); - } else { - read_lock_bh(&neigh->lock); + unsigned int seq; + + if (dev->header_ops->cache && + !dst->hh && + !(dst->flags & DST_NOCACHE)) + neigh_hh_init(neigh, dst, dst->ops->protocol); + + do { + seq = read_seqbegin(&neigh->ha_lock); err = dev_hard_header(skb, dev, ntohs(skb->protocol), neigh->ha, NULL, skb->len); - read_unlock_bh(&neigh->lock); - } + } while (read_seqretry(&neigh->ha_lock, seq)); + if (err >= 0) rc = neigh->ops->queue_xmit(skb); else @@ -1248,13 +1350,16 @@ int neigh_connected_output(struct sk_buff *skb) struct dst_entry *dst = skb_dst(skb); struct neighbour *neigh = dst->neighbour; struct net_device *dev = neigh->dev; + unsigned int seq; __skb_pull(skb, skb_network_offset(skb)); - read_lock_bh(&neigh->lock); - err = dev_hard_header(skb, dev, ntohs(skb->protocol), - neigh->ha, NULL, skb->len); - read_unlock_bh(&neigh->lock); + do { + seq = read_seqbegin(&neigh->ha_lock); + err = dev_hard_header(skb, dev, ntohs(skb->protocol), + neigh->ha, NULL, skb->len); + } while (read_seqretry(&neigh->ha_lock, seq)); + if (err >= 0) err = neigh->ops->queue_xmit(skb); else { @@ -1436,17 +1541,14 @@ void neigh_table_init_no_netlink(struct neigh_table *tbl) panic("cannot create neighbour proc dir entry"); #endif - tbl->hash_mask = 1; - tbl->hash_buckets = neigh_hash_alloc(tbl->hash_mask + 1); + tbl->nht = neigh_hash_alloc(8); phsize = (PNEIGH_HASHMASK + 1) * sizeof(struct pneigh_entry *); tbl->phash_buckets = kzalloc(phsize, GFP_KERNEL); - if (!tbl->hash_buckets || !tbl->phash_buckets) + if (!tbl->nht || !tbl->phash_buckets) panic("cannot allocate neighbour cache hashes"); - get_random_bytes(&tbl->hash_rnd, sizeof(tbl->hash_rnd)); - rwlock_init(&tbl->lock); INIT_DELAYED_WORK_DEFERRABLE(&tbl->gc_work, neigh_periodic_work); schedule_delayed_work(&tbl->gc_work, tbl->parms.reachable_time); @@ -1486,8 +1588,7 @@ int neigh_table_clear(struct neigh_table *tbl) struct neigh_table **tp; /* It is not clean... Fix it to unload IPv6 module safely */ - cancel_delayed_work(&tbl->gc_work); - flush_scheduled_work(); + cancel_delayed_work_sync(&tbl->gc_work); del_timer_sync(&tbl->proxy_timer); pneigh_queue_purge(&tbl->proxy_queue); neigh_ifdown(tbl, NULL); @@ -1502,8 +1603,8 @@ int neigh_table_clear(struct neigh_table *tbl) } write_unlock(&neigh_tbl_lock); - neigh_hash_free(tbl->hash_buckets, tbl->hash_mask + 1); - tbl->hash_buckets = NULL; + call_rcu(&tbl->nht->rcu, neigh_hash_free_rcu); + tbl->nht = NULL; kfree(tbl->phash_buckets); tbl->phash_buckets = NULL; @@ -1529,6 +1630,7 @@ static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) struct net_device *dev = NULL; int err = -EINVAL; + ASSERT_RTNL(); if (nlmsg_len(nlh) < sizeof(*ndm)) goto out; @@ -1538,7 +1640,7 @@ static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) ndm = nlmsg_data(nlh); if (ndm->ndm_ifindex) { - dev = dev_get_by_index(net, ndm->ndm_ifindex); + dev = __dev_get_by_index(net, ndm->ndm_ifindex); if (dev == NULL) { err = -ENODEV; goto out; @@ -1554,34 +1656,31 @@ static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) read_unlock(&neigh_tbl_lock); if (nla_len(dst_attr) < tbl->key_len) - goto out_dev_put; + goto out; if (ndm->ndm_flags & NTF_PROXY) { err = pneigh_delete(tbl, net, nla_data(dst_attr), dev); - goto out_dev_put; + goto out; } if (dev == NULL) - goto out_dev_put; + goto out; neigh = neigh_lookup(tbl, nla_data(dst_attr), dev); if (neigh == NULL) { err = -ENOENT; - goto out_dev_put; + goto out; } err = neigh_update(neigh, NULL, NUD_FAILED, NEIGH_UPDATE_F_OVERRIDE | NEIGH_UPDATE_F_ADMIN); neigh_release(neigh); - goto out_dev_put; + goto out; } read_unlock(&neigh_tbl_lock); err = -EAFNOSUPPORT; -out_dev_put: - if (dev) - dev_put(dev); out: return err; } @@ -1595,6 +1694,7 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) struct net_device *dev = NULL; int err; + ASSERT_RTNL(); err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL); if (err < 0) goto out; @@ -1605,14 +1705,14 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) ndm = nlmsg_data(nlh); if (ndm->ndm_ifindex) { - dev = dev_get_by_index(net, ndm->ndm_ifindex); + dev = __dev_get_by_index(net, ndm->ndm_ifindex); if (dev == NULL) { err = -ENODEV; goto out; } if (tb[NDA_LLADDR] && nla_len(tb[NDA_LLADDR]) < dev->addr_len) - goto out_dev_put; + goto out; } read_lock(&neigh_tbl_lock); @@ -1626,7 +1726,7 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) read_unlock(&neigh_tbl_lock); if (nla_len(tb[NDA_DST]) < tbl->key_len) - goto out_dev_put; + goto out; dst = nla_data(tb[NDA_DST]); lladdr = tb[NDA_LLADDR] ? nla_data(tb[NDA_LLADDR]) : NULL; @@ -1639,29 +1739,29 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) pn->flags = ndm->ndm_flags; err = 0; } - goto out_dev_put; + goto out; } if (dev == NULL) - goto out_dev_put; + goto out; neigh = neigh_lookup(tbl, dst, dev); if (neigh == NULL) { if (!(nlh->nlmsg_flags & NLM_F_CREATE)) { err = -ENOENT; - goto out_dev_put; + goto out; } neigh = __neigh_lookup_errno(tbl, dst, dev); if (IS_ERR(neigh)) { err = PTR_ERR(neigh); - goto out_dev_put; + goto out; } } else { if (nlh->nlmsg_flags & NLM_F_EXCL) { err = -EEXIST; neigh_release(neigh); - goto out_dev_put; + goto out; } if (!(nlh->nlmsg_flags & NLM_F_REPLACE)) @@ -1674,15 +1774,11 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) } else err = neigh_update(neigh, lladdr, ndm->ndm_state, flags); neigh_release(neigh); - goto out_dev_put; + goto out; } read_unlock(&neigh_tbl_lock); err = -EAFNOSUPPORT; - -out_dev_put: - if (dev) - dev_put(dev); out: return err; } @@ -1748,18 +1844,22 @@ static int neightbl_fill_info(struct sk_buff *skb, struct neigh_table *tbl, unsigned long now = jiffies; unsigned int flush_delta = now - tbl->last_flush; unsigned int rand_delta = now - tbl->last_rand; - + struct neigh_hash_table *nht; struct ndt_config ndc = { .ndtc_key_len = tbl->key_len, .ndtc_entry_size = tbl->entry_size, .ndtc_entries = atomic_read(&tbl->entries), .ndtc_last_flush = jiffies_to_msecs(flush_delta), .ndtc_last_rand = jiffies_to_msecs(rand_delta), - .ndtc_hash_rnd = tbl->hash_rnd, - .ndtc_hash_mask = tbl->hash_mask, .ndtc_proxy_qlen = tbl->proxy_queue.qlen, }; + rcu_read_lock_bh(); + nht = rcu_dereference_bh(tbl->nht); + ndc.ndtc_hash_rnd = nht->hash_rnd; + ndc.ndtc_hash_mask = nht->hash_mask; + rcu_read_unlock_bh(); + NLA_PUT(skb, NDTA_CONFIG, sizeof(ndc), &ndc); } @@ -2056,10 +2156,14 @@ static int neigh_fill_info(struct sk_buff *skb, struct neighbour *neigh, read_lock_bh(&neigh->lock); ndm->ndm_state = neigh->nud_state; - if ((neigh->nud_state & NUD_VALID) && - nla_put(skb, NDA_LLADDR, neigh->dev->addr_len, neigh->ha) < 0) { - read_unlock_bh(&neigh->lock); - goto nla_put_failure; + if (neigh->nud_state & NUD_VALID) { + char haddr[MAX_ADDR_LEN]; + + neigh_ha_snapshot(haddr, neigh, neigh->dev); + if (nla_put(skb, NDA_LLADDR, neigh->dev->addr_len, haddr) < 0) { + read_unlock_bh(&neigh->lock); + goto nla_put_failure; + } } ci.ndm_used = jiffies_to_clock_t(now - neigh->used); @@ -2087,18 +2191,23 @@ static void neigh_update_notify(struct neighbour *neigh) static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, struct netlink_callback *cb) { - struct net * net = sock_net(skb->sk); + struct net *net = sock_net(skb->sk); struct neighbour *n; int rc, h, s_h = cb->args[1]; int idx, s_idx = idx = cb->args[2]; + struct neigh_hash_table *nht; - read_lock_bh(&tbl->lock); - for (h = 0; h <= tbl->hash_mask; h++) { + rcu_read_lock_bh(); + nht = rcu_dereference_bh(tbl->nht); + + for (h = 0; h <= nht->hash_mask; h++) { if (h < s_h) continue; if (h > s_h) s_idx = 0; - for (n = tbl->hash_buckets[h], idx = 0; n; n = n->next) { + for (n = rcu_dereference_bh(nht->hash_buckets[h]), idx = 0; + n != NULL; + n = rcu_dereference_bh(n->next)) { if (!net_eq(dev_net(n->dev), net)) continue; if (idx < s_idx) @@ -2107,17 +2216,16 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, cb->nlh->nlmsg_seq, RTM_NEWNEIGH, NLM_F_MULTI) <= 0) { - read_unlock_bh(&tbl->lock); rc = -1; goto out; } - next: +next: idx++; } } - read_unlock_bh(&tbl->lock); rc = skb->len; out: + rcu_read_unlock_bh(); cb->args[1] = h; cb->args[2] = idx; return rc; @@ -2150,15 +2258,22 @@ static int neigh_dump_info(struct sk_buff *skb, struct netlink_callback *cb) void neigh_for_each(struct neigh_table *tbl, void (*cb)(struct neighbour *, void *), void *cookie) { int chain; + struct neigh_hash_table *nht; - read_lock_bh(&tbl->lock); - for (chain = 0; chain <= tbl->hash_mask; chain++) { + rcu_read_lock_bh(); + nht = rcu_dereference_bh(tbl->nht); + + read_lock(&tbl->lock); /* avoid resizes */ + for (chain = 0; chain <= nht->hash_mask; chain++) { struct neighbour *n; - for (n = tbl->hash_buckets[chain]; n; n = n->next) + for (n = rcu_dereference_bh(nht->hash_buckets[chain]); + n != NULL; + n = rcu_dereference_bh(n->next)) cb(n, cookie); } - read_unlock_bh(&tbl->lock); + read_unlock(&tbl->lock); + rcu_read_unlock_bh(); } EXPORT_SYMBOL(neigh_for_each); @@ -2167,18 +2282,25 @@ void __neigh_for_each_release(struct neigh_table *tbl, int (*cb)(struct neighbour *)) { int chain; + struct neigh_hash_table *nht; - for (chain = 0; chain <= tbl->hash_mask; chain++) { - struct neighbour *n, **np; + nht = rcu_dereference_protected(tbl->nht, + lockdep_is_held(&tbl->lock)); + for (chain = 0; chain <= nht->hash_mask; chain++) { + struct neighbour *n; + struct neighbour __rcu **np; - np = &tbl->hash_buckets[chain]; - while ((n = *np) != NULL) { + np = &nht->hash_buckets[chain]; + while ((n = rcu_dereference_protected(*np, + lockdep_is_held(&tbl->lock))) != NULL) { int release; write_lock(&n->lock); release = cb(n); if (release) { - *np = n->next; + rcu_assign_pointer(*np, + rcu_dereference_protected(n->next, + lockdep_is_held(&tbl->lock))); n->dead = 1; } else np = &n->next; @@ -2196,13 +2318,13 @@ static struct neighbour *neigh_get_first(struct seq_file *seq) { struct neigh_seq_state *state = seq->private; struct net *net = seq_file_net(seq); - struct neigh_table *tbl = state->tbl; + struct neigh_hash_table *nht = state->nht; struct neighbour *n = NULL; int bucket = state->bucket; state->flags &= ~NEIGH_SEQ_IS_PNEIGH; - for (bucket = 0; bucket <= tbl->hash_mask; bucket++) { - n = tbl->hash_buckets[bucket]; + for (bucket = 0; bucket <= nht->hash_mask; bucket++) { + n = rcu_dereference_bh(nht->hash_buckets[bucket]); while (n) { if (!net_eq(dev_net(n->dev), net)) @@ -2219,8 +2341,8 @@ static struct neighbour *neigh_get_first(struct seq_file *seq) break; if (n->nud_state & ~NUD_NOARP) break; - next: - n = n->next; +next: + n = rcu_dereference_bh(n->next); } if (n) @@ -2237,14 +2359,14 @@ static struct neighbour *neigh_get_next(struct seq_file *seq, { struct neigh_seq_state *state = seq->private; struct net *net = seq_file_net(seq); - struct neigh_table *tbl = state->tbl; + struct neigh_hash_table *nht = state->nht; if (state->neigh_sub_iter) { void *v = state->neigh_sub_iter(state, n, pos); if (v) return n; } - n = n->next; + n = rcu_dereference_bh(n->next); while (1) { while (n) { @@ -2261,17 +2383,17 @@ static struct neighbour *neigh_get_next(struct seq_file *seq, if (n->nud_state & ~NUD_NOARP) break; - next: - n = n->next; +next: + n = rcu_dereference_bh(n->next); } if (n) break; - if (++state->bucket > tbl->hash_mask) + if (++state->bucket > nht->hash_mask) break; - n = tbl->hash_buckets[state->bucket]; + n = rcu_dereference_bh(nht->hash_buckets[state->bucket]); } if (n && pos) @@ -2369,7 +2491,7 @@ static void *neigh_get_idx_any(struct seq_file *seq, loff_t *pos) } void *neigh_seq_start(struct seq_file *seq, loff_t *pos, struct neigh_table *tbl, unsigned int neigh_seq_flags) - __acquires(tbl->lock) + __acquires(rcu_bh) { struct neigh_seq_state *state = seq->private; @@ -2377,7 +2499,8 @@ void *neigh_seq_start(struct seq_file *seq, loff_t *pos, struct neigh_table *tbl state->bucket = 0; state->flags = (neigh_seq_flags & ~NEIGH_SEQ_IS_PNEIGH); - read_lock_bh(&tbl->lock); + rcu_read_lock_bh(); + state->nht = rcu_dereference_bh(tbl->nht); return *pos ? neigh_get_idx_any(seq, pos) : SEQ_START_TOKEN; } @@ -2411,12 +2534,9 @@ out: EXPORT_SYMBOL(neigh_seq_next); void neigh_seq_stop(struct seq_file *seq, void *v) - __releases(tbl->lock) + __releases(rcu_bh) { - struct neigh_seq_state *state = seq->private; - struct neigh_table *tbl = state->tbl; - - read_unlock_bh(&tbl->lock); + rcu_read_unlock_bh(); } EXPORT_SYMBOL(neigh_seq_stop); diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index af4dfbadf2a..b143173e3eb 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -515,7 +515,7 @@ static ssize_t rx_queue_attr_store(struct kobject *kobj, struct attribute *attr, return attribute->store(queue, attribute, buf, count); } -static struct sysfs_ops rx_queue_sysfs_ops = { +static const struct sysfs_ops rx_queue_sysfs_ops = { .show = rx_queue_attr_show, .store = rx_queue_attr_store, }; @@ -726,6 +726,7 @@ static struct kobj_type rx_queue_ktype = { static int rx_queue_add_kobject(struct net_device *net, int index) { struct netdev_rx_queue *queue = net->_rx + index; + struct netdev_rx_queue *first = queue->first; struct kobject *kobj = &queue->kobj; int error = 0; @@ -738,38 +739,43 @@ static int rx_queue_add_kobject(struct net_device *net, int index) } kobject_uevent(kobj, KOBJ_ADD); + atomic_inc(&first->count); return error; } -static int rx_queue_register_kobjects(struct net_device *net) +int +net_rx_queue_update_kobjects(struct net_device *net, int old_num, int new_num) { int i; int error = 0; - net->queues_kset = kset_create_and_add("queues", - NULL, &net->dev.kobj); - if (!net->queues_kset) - return -ENOMEM; - for (i = 0; i < net->num_rx_queues; i++) { + for (i = old_num; i < new_num; i++) { error = rx_queue_add_kobject(net, i); - if (error) + if (error) { + new_num = old_num; break; + } } - if (error) - while (--i >= 0) - kobject_put(&net->_rx[i].kobj); + while (--i >= new_num) + kobject_put(&net->_rx[i].kobj); return error; } -static void rx_queue_remove_kobjects(struct net_device *net) +static int rx_queue_register_kobjects(struct net_device *net) { - int i; + net->queues_kset = kset_create_and_add("queues", + NULL, &net->dev.kobj); + if (!net->queues_kset) + return -ENOMEM; + return net_rx_queue_update_kobjects(net, 0, net->real_num_rx_queues); +} - for (i = 0; i < net->num_rx_queues; i++) - kobject_put(&net->_rx[i].kobj); +static void rx_queue_remove_kobjects(struct net_device *net) +{ + net_rx_queue_update_kobjects(net, net->real_num_rx_queues, 0); kset_unregister(net->queues_kset); } #endif /* CONFIG_RPS */ @@ -789,12 +795,13 @@ static const void *net_netlink_ns(struct sock *sk) return sock_net(sk); } -static struct kobj_ns_type_operations net_ns_type_operations = { +struct kobj_ns_type_operations net_ns_type_operations = { .type = KOBJ_NS_TYPE_NET, .current_ns = net_current_ns, .netlink_ns = net_netlink_ns, .initial_ns = net_initial_ns, }; +EXPORT_SYMBOL_GPL(net_ns_type_operations); static void net_kobj_ns_exit(struct net *net) { diff --git a/net/core/net-sysfs.h b/net/core/net-sysfs.h index 805555e8b18..778e1571548 100644 --- a/net/core/net-sysfs.h +++ b/net/core/net-sysfs.h @@ -4,4 +4,8 @@ int netdev_kobject_init(void); int netdev_register_kobject(struct net_device *); void netdev_unregister_kobject(struct net_device *); +#ifdef CONFIG_RPS +int net_rx_queue_update_kobjects(struct net_device *, int old_num, int new_num); +#endif + #endif diff --git a/net/core/net-traces.c b/net/core/net-traces.c index afa6380ed88..7f1bb2aba03 100644 --- a/net/core/net-traces.c +++ b/net/core/net-traces.c @@ -26,6 +26,7 @@ #define CREATE_TRACE_POINTS #include <trace/events/skb.h> +#include <trace/events/net.h> #include <trace/events/napi.h> EXPORT_TRACEPOINT_SYMBOL_GPL(kfree_skb); diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 537e01afd81..4e98ffac3af 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -288,11 +288,11 @@ static int netpoll_owner_active(struct net_device *dev) return 0; } -void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) +void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb, + struct net_device *dev) { int status = NETDEV_TX_BUSY; unsigned long tries; - struct net_device *dev = np->dev; const struct net_device_ops *ops = dev->netdev_ops; /* It is up to the caller to keep npinfo alive. */ struct netpoll_info *npinfo = np->dev->npinfo; @@ -346,7 +346,7 @@ void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) schedule_delayed_work(&npinfo->tx_work,0); } } -EXPORT_SYMBOL(netpoll_send_skb); +EXPORT_SYMBOL(netpoll_send_skb_on_dev); void netpoll_send_udp(struct netpoll *np, const char *msg, int len) { diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 10a1ea72010..2c0df0f95b3 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -729,16 +729,14 @@ static int hex32_arg(const char __user *user_buffer, unsigned long maxlen, *num = 0; for (; i < maxlen; i++) { + int value; char c; *num <<= 4; if (get_user(c, &user_buffer[i])) return -EFAULT; - if ((c >= '0') && (c <= '9')) - *num |= c - '0'; - else if ((c >= 'a') && (c <= 'f')) - *num |= c - 'a' + 10; - else if ((c >= 'A') && (c <= 'F')) - *num |= c - 'A' + 10; + value = hex_to_bin(c); + if (value >= 0) + *num |= value; else break; } @@ -3907,8 +3905,6 @@ static void __exit pg_cleanup(void) { struct pktgen_thread *t; struct list_head *q, *n; - wait_queue_head_t queue; - init_waitqueue_head(&queue); /* Stop all interfaces & threads */ diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index f78d821bd93..8121268ddbd 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -299,14 +299,6 @@ static void __rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops) unregister_netdevice_many(&list_kill); } -void rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops) -{ - rtnl_lock(); - __rtnl_kill_links(net, ops); - rtnl_unlock(); -} -EXPORT_SYMBOL_GPL(rtnl_kill_links); - /** * __rtnl_link_unregister - Unregister rtnl_link_ops from rtnetlink. * @ops: struct rtnl_link_ops * to unregister @@ -612,36 +604,7 @@ static void copy_rtnl_link_stats(struct rtnl_link_stats *a, static void copy_rtnl_link_stats64(void *v, const struct rtnl_link_stats64 *b) { - struct rtnl_link_stats64 a; - - a.rx_packets = b->rx_packets; - a.tx_packets = b->tx_packets; - a.rx_bytes = b->rx_bytes; - a.tx_bytes = b->tx_bytes; - a.rx_errors = b->rx_errors; - a.tx_errors = b->tx_errors; - a.rx_dropped = b->rx_dropped; - a.tx_dropped = b->tx_dropped; - - a.multicast = b->multicast; - a.collisions = b->collisions; - - a.rx_length_errors = b->rx_length_errors; - a.rx_over_errors = b->rx_over_errors; - a.rx_crc_errors = b->rx_crc_errors; - a.rx_frame_errors = b->rx_frame_errors; - a.rx_fifo_errors = b->rx_fifo_errors; - a.rx_missed_errors = b->rx_missed_errors; - - a.tx_aborted_errors = b->tx_aborted_errors; - a.tx_carrier_errors = b->tx_carrier_errors; - a.tx_fifo_errors = b->tx_fifo_errors; - a.tx_heartbeat_errors = b->tx_heartbeat_errors; - a.tx_window_errors = b->tx_window_errors; - - a.rx_compressed = b->rx_compressed; - a.tx_compressed = b->tx_compressed; - memcpy(v, &a, sizeof(a)); + memcpy(v, b, sizeof(*b)); } /* All VF info */ diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 3a2513f0d0c..104f8444754 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -202,8 +202,6 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask, skb->data = data; skb_reset_tail_pointer(skb); skb->end = skb->tail + size; - kmemcheck_annotate_bitfield(skb, flags1); - kmemcheck_annotate_bitfield(skb, flags2); #ifdef NET_SKBUFF_DATA_USES_OFFSET skb->mac_header = ~0U; #endif @@ -249,10 +247,9 @@ EXPORT_SYMBOL(__alloc_skb); struct sk_buff *__netdev_alloc_skb(struct net_device *dev, unsigned int length, gfp_t gfp_mask) { - int node = dev->dev.parent ? dev_to_node(dev->dev.parent) : -1; struct sk_buff *skb; - skb = __alloc_skb(length + NET_SKB_PAD, gfp_mask, 0, node); + skb = __alloc_skb(length + NET_SKB_PAD, gfp_mask, 0, NUMA_NO_NODE); if (likely(skb)) { skb_reserve(skb, NET_SKB_PAD); skb->dev = dev; @@ -261,16 +258,6 @@ struct sk_buff *__netdev_alloc_skb(struct net_device *dev, } EXPORT_SYMBOL(__netdev_alloc_skb); -struct page *__netdev_alloc_page(struct net_device *dev, gfp_t gfp_mask) -{ - int node = dev->dev.parent ? dev_to_node(dev->dev.parent) : -1; - struct page *page; - - page = alloc_pages_node(node, gfp_mask, 0); - return page; -} -EXPORT_SYMBOL(__netdev_alloc_page); - void skb_add_rx_frag(struct sk_buff *skb, int i, struct page *page, int off, int size) { @@ -340,7 +327,7 @@ static void skb_release_data(struct sk_buff *skb) put_page(skb_shinfo(skb)->frags[i].page); } - if (skb_has_frags(skb)) + if (skb_has_frag_list(skb)) skb_drop_fraglist(skb); kfree(skb->head); @@ -466,6 +453,7 @@ void consume_skb(struct sk_buff *skb) smp_rmb(); else if (likely(!atomic_dec_and_test(&skb->users))) return; + trace_consume_skb(skb); __kfree_skb(skb); } EXPORT_SYMBOL(consume_skb); @@ -685,16 +673,10 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old) struct sk_buff *skb_copy(const struct sk_buff *skb, gfp_t gfp_mask) { - int headerlen = skb->data - skb->head; - /* - * Allocate the copy buffer - */ - struct sk_buff *n; -#ifdef NET_SKBUFF_DATA_USES_OFFSET - n = alloc_skb(skb->end + skb->data_len, gfp_mask); -#else - n = alloc_skb(skb->end - skb->head + skb->data_len, gfp_mask); -#endif + int headerlen = skb_headroom(skb); + unsigned int size = (skb_end_pointer(skb) - skb->head) + skb->data_len; + struct sk_buff *n = alloc_skb(size, gfp_mask); + if (!n) return NULL; @@ -726,20 +708,14 @@ EXPORT_SYMBOL(skb_copy); struct sk_buff *pskb_copy(struct sk_buff *skb, gfp_t gfp_mask) { - /* - * Allocate the copy buffer - */ - struct sk_buff *n; -#ifdef NET_SKBUFF_DATA_USES_OFFSET - n = alloc_skb(skb->end, gfp_mask); -#else - n = alloc_skb(skb->end - skb->head, gfp_mask); -#endif + unsigned int size = skb_end_pointer(skb) - skb->head; + struct sk_buff *n = alloc_skb(size, gfp_mask); + if (!n) goto out; /* Set the data pointer */ - skb_reserve(n, skb->data - skb->head); + skb_reserve(n, skb_headroom(skb)); /* Set the tail pointer and length */ skb_put(n, skb_headlen(skb)); /* Copy the bytes */ @@ -759,7 +735,7 @@ struct sk_buff *pskb_copy(struct sk_buff *skb, gfp_t gfp_mask) skb_shinfo(n)->nr_frags = i; } - if (skb_has_frags(skb)) { + if (skb_has_frag_list(skb)) { skb_shinfo(n)->frag_list = skb_shinfo(skb)->frag_list; skb_clone_fraglist(n); } @@ -791,12 +767,9 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, { int i; u8 *data; -#ifdef NET_SKBUFF_DATA_USES_OFFSET - int size = nhead + skb->end + ntail; -#else - int size = nhead + (skb->end - skb->head) + ntail; -#endif + int size = nhead + (skb_end_pointer(skb) - skb->head) + ntail; long off; + bool fastpath; BUG_ON(nhead < 0); @@ -810,23 +783,36 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, goto nodata; /* Copy only real data... and, alas, header. This should be - * optimized for the cases when header is void. */ -#ifdef NET_SKBUFF_DATA_USES_OFFSET - memcpy(data + nhead, skb->head, skb->tail); -#else - memcpy(data + nhead, skb->head, skb->tail - skb->head); -#endif - memcpy(data + size, skb_end_pointer(skb), + * optimized for the cases when header is void. + */ + memcpy(data + nhead, skb->head, skb_tail_pointer(skb) - skb->head); + + memcpy((struct skb_shared_info *)(data + size), + skb_shinfo(skb), offsetof(struct skb_shared_info, frags[skb_shinfo(skb)->nr_frags])); - for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) - get_page(skb_shinfo(skb)->frags[i].page); + /* Check if we can avoid taking references on fragments if we own + * the last reference on skb->head. (see skb_release_data()) + */ + if (!skb->cloned) + fastpath = true; + else { + int delta = skb->nohdr ? (1 << SKB_DATAREF_SHIFT) + 1 : 1; - if (skb_has_frags(skb)) - skb_clone_fraglist(skb); + fastpath = atomic_read(&skb_shinfo(skb)->dataref) == delta; + } - skb_release_data(skb); + if (fastpath) { + kfree(skb->head); + } else { + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) + get_page(skb_shinfo(skb)->frags[i].page); + + if (skb_has_frag_list(skb)) + skb_clone_fraglist(skb); + skb_release_data(skb); + } off = (data + nhead) - skb->head; skb->head = data; @@ -1099,7 +1085,7 @@ drop_pages: for (; i < nfrags; i++) put_page(skb_shinfo(skb)->frags[i].page); - if (skb_has_frags(skb)) + if (skb_has_frag_list(skb)) skb_drop_fraglist(skb); goto done; } @@ -1194,7 +1180,7 @@ unsigned char *__pskb_pull_tail(struct sk_buff *skb, int delta) /* Optimization: no fragments, no reasons to preestimate * size of pulled pages. Superb. */ - if (!skb_has_frags(skb)) + if (!skb_has_frag_list(skb)) goto pull_pages; /* Estimate size of pulled pages. */ @@ -2323,7 +2309,7 @@ next_skb: st->frag_data = NULL; } - if (st->root_skb == st->cur_skb && skb_has_frags(st->root_skb)) { + if (st->root_skb == st->cur_skb && skb_has_frag_list(st->root_skb)) { st->cur_skb = skb_shinfo(st->root_skb)->frag_list; st->frag_idx = 0; goto next_skb; @@ -2573,6 +2559,10 @@ struct sk_buff *skb_segment(struct sk_buff *skb, int features) __copy_skb_header(nskb, skb); nskb->mac_len = skb->mac_len; + /* nskb and skb might have different headroom */ + if (nskb->ip_summed == CHECKSUM_PARTIAL) + nskb->csum_start += skb_headroom(nskb) - headroom; + skb_reset_mac_header(nskb); skb_set_network_header(nskb, skb->mac_len); nskb->transport_header = (nskb->network_header + @@ -2703,7 +2693,7 @@ int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb) return -E2BIG; headroom = skb_headroom(p); - nskb = netdev_alloc_skb(p->dev, headroom + skb_gro_offset(p)); + nskb = alloc_skb(headroom + skb_gro_offset(p), GFP_ATOMIC); if (unlikely(!nskb)) return -ENOMEM; @@ -2889,7 +2879,7 @@ int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer) return -ENOMEM; /* Easy case. Most of packets will go this way. */ - if (!skb_has_frags(skb)) { + if (!skb_has_frag_list(skb)) { /* A little of trouble, not enough of space for trailer. * This should not happen, when stack is tuned to generate * good frames. OK, on miss we reallocate and reserve even more @@ -2924,7 +2914,7 @@ int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer) if (skb1->next == NULL && tailbits) { if (skb_shinfo(skb1)->nr_frags || - skb_has_frags(skb1) || + skb_has_frag_list(skb1) || skb_tailroom(skb1) < tailbits) ntail = tailbits + 128; } @@ -2933,7 +2923,7 @@ int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer) skb_cloned(skb1) || ntail || skb_shinfo(skb1)->nr_frags || - skb_has_frags(skb1)) { + skb_has_frag_list(skb1)) { struct sk_buff *skb2; /* Fuck, we are miserable poor guys... */ @@ -3016,7 +3006,7 @@ void skb_tstamp_tx(struct sk_buff *orig_skb, } else { /* * no hardware time stamps available, - * so keep the skb_shared_tx and only + * so keep the shared tx_flags and only * store software time stamp */ skb->tstamp = ktime_get_real(); diff --git a/net/core/sock.c b/net/core/sock.c index b05b9b6ddb8..11db43632df 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1078,8 +1078,11 @@ static void sk_prot_free(struct proto *prot, struct sock *sk) #ifdef CONFIG_CGROUPS void sock_update_classid(struct sock *sk) { - u32 classid = task_cls_classid(current); + u32 classid; + rcu_read_lock(); /* doing current task, which cannot vanish. */ + classid = task_cls_classid(current); + rcu_read_unlock(); if (classid && classid != sk->sk_classid) sk->sk_classid = classid; } @@ -1351,9 +1354,9 @@ int sock_i_uid(struct sock *sk) { int uid; - read_lock(&sk->sk_callback_lock); + read_lock_bh(&sk->sk_callback_lock); uid = sk->sk_socket ? SOCK_INODE(sk->sk_socket)->i_uid : 0; - read_unlock(&sk->sk_callback_lock); + read_unlock_bh(&sk->sk_callback_lock); return uid; } EXPORT_SYMBOL(sock_i_uid); @@ -1362,9 +1365,9 @@ unsigned long sock_i_ino(struct sock *sk) { unsigned long ino; - read_lock(&sk->sk_callback_lock); + read_lock_bh(&sk->sk_callback_lock); ino = sk->sk_socket ? SOCK_INODE(sk->sk_socket)->i_ino : 0; - read_unlock(&sk->sk_callback_lock); + read_unlock_bh(&sk->sk_callback_lock); return ino; } EXPORT_SYMBOL(sock_i_ino); @@ -1557,6 +1560,8 @@ struct sk_buff *sock_alloc_send_skb(struct sock *sk, unsigned long size, EXPORT_SYMBOL(sock_alloc_send_skb); static void __lock_sock(struct sock *sk) + __releases(&sk->sk_lock.slock) + __acquires(&sk->sk_lock.slock) { DEFINE_WAIT(wait); @@ -1573,6 +1578,8 @@ static void __lock_sock(struct sock *sk) } static void __release_sock(struct sock *sk) + __releases(&sk->sk_lock.slock) + __acquires(&sk->sk_lock.slock) { struct sk_buff *skb = sk->sk_backlog.head; diff --git a/net/core/stream.c b/net/core/stream.c index d959e0f4152..f5df85dcd20 100644 --- a/net/core/stream.c +++ b/net/core/stream.c @@ -141,10 +141,10 @@ int sk_stream_wait_memory(struct sock *sk, long *timeo_p) set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); sk->sk_write_pending++; - sk_wait_event(sk, ¤t_timeo, !sk->sk_err && - !(sk->sk_shutdown & SEND_SHUTDOWN) && - sk_stream_memory_free(sk) && - vm_wait); + sk_wait_event(sk, ¤t_timeo, sk->sk_err || + (sk->sk_shutdown & SEND_SHUTDOWN) || + (sk_stream_memory_free(sk) && + !vm_wait)); sk->sk_write_pending--; if (vm_wait) { diff --git a/net/core/utils.c b/net/core/utils.c index f4185447053..5fea0ab2190 100644 --- a/net/core/utils.c +++ b/net/core/utils.c @@ -75,7 +75,7 @@ __be32 in_aton(const char *str) str++; } } - return(htonl(l)); + return htonl(l); } EXPORT_SYMBOL(in_aton); @@ -92,18 +92,19 @@ EXPORT_SYMBOL(in_aton); static inline int xdigit2bin(char c, int delim) { + int val; + if (c == delim || c == '\0') return IN6PTON_DELIM; if (c == ':') return IN6PTON_COLON_MASK; if (c == '.') return IN6PTON_DOT; - if (c >= '0' && c <= '9') - return (IN6PTON_XDIGIT | IN6PTON_DIGIT| (c - '0')); - if (c >= 'a' && c <= 'f') - return (IN6PTON_XDIGIT | (c - 'a' + 10)); - if (c >= 'A' && c <= 'F') - return (IN6PTON_XDIGIT | (c - 'A' + 10)); + + val = hex_to_bin(c); + if (val >= 0) + return val | IN6PTON_XDIGIT | (val < 10 ? IN6PTON_DIGIT : 0); + if (delim == -1) return IN6PTON_DELIM; return IN6PTON_UNKNOWN; diff --git a/net/dccp/ccid.h b/net/dccp/ccid.h index 6df6f8ac963..117fb093dca 100644 --- a/net/dccp/ccid.h +++ b/net/dccp/ccid.h @@ -62,22 +62,18 @@ struct ccid_operations { void (*ccid_hc_tx_exit)(struct sock *sk); void (*ccid_hc_rx_packet_recv)(struct sock *sk, struct sk_buff *skb); - int (*ccid_hc_rx_parse_options)(struct sock *sk, - unsigned char option, - unsigned char len, u16 idx, - unsigned char* value); + int (*ccid_hc_rx_parse_options)(struct sock *sk, u8 pkt, + u8 opt, u8 *val, u8 len); int (*ccid_hc_rx_insert_options)(struct sock *sk, struct sk_buff *skb); void (*ccid_hc_tx_packet_recv)(struct sock *sk, struct sk_buff *skb); - int (*ccid_hc_tx_parse_options)(struct sock *sk, - unsigned char option, - unsigned char len, u16 idx, - unsigned char* value); + int (*ccid_hc_tx_parse_options)(struct sock *sk, u8 pkt, + u8 opt, u8 *val, u8 len); int (*ccid_hc_tx_send_packet)(struct sock *sk, struct sk_buff *skb); void (*ccid_hc_tx_packet_sent)(struct sock *sk, - int more, unsigned int len); + unsigned int len); void (*ccid_hc_rx_get_info)(struct sock *sk, struct tcp_info *info); void (*ccid_hc_tx_get_info)(struct sock *sk, @@ -148,10 +144,10 @@ static inline int ccid_hc_tx_send_packet(struct ccid *ccid, struct sock *sk, } static inline void ccid_hc_tx_packet_sent(struct ccid *ccid, struct sock *sk, - int more, unsigned int len) + unsigned int len) { if (ccid->ccid_ops->ccid_hc_tx_packet_sent != NULL) - ccid->ccid_ops->ccid_hc_tx_packet_sent(sk, more, len); + ccid->ccid_ops->ccid_hc_tx_packet_sent(sk, len); } static inline void ccid_hc_rx_packet_recv(struct ccid *ccid, struct sock *sk, @@ -168,27 +164,31 @@ static inline void ccid_hc_tx_packet_recv(struct ccid *ccid, struct sock *sk, ccid->ccid_ops->ccid_hc_tx_packet_recv(sk, skb); } +/** + * ccid_hc_tx_parse_options - Parse CCID-specific options sent by the receiver + * @pkt: type of packet that @opt appears on (RFC 4340, 5.1) + * @opt: the CCID-specific option type (RFC 4340, 5.8 and 10.3) + * @val: value of @opt + * @len: length of @val in bytes + */ static inline int ccid_hc_tx_parse_options(struct ccid *ccid, struct sock *sk, - unsigned char option, - unsigned char len, u16 idx, - unsigned char* value) + u8 pkt, u8 opt, u8 *val, u8 len) { - int rc = 0; - if (ccid->ccid_ops->ccid_hc_tx_parse_options != NULL) - rc = ccid->ccid_ops->ccid_hc_tx_parse_options(sk, option, len, idx, - value); - return rc; + if (ccid->ccid_ops->ccid_hc_tx_parse_options == NULL) + return 0; + return ccid->ccid_ops->ccid_hc_tx_parse_options(sk, pkt, opt, val, len); } +/** + * ccid_hc_rx_parse_options - Parse CCID-specific options sent by the sender + * Arguments are analogous to ccid_hc_tx_parse_options() + */ static inline int ccid_hc_rx_parse_options(struct ccid *ccid, struct sock *sk, - unsigned char option, - unsigned char len, u16 idx, - unsigned char* value) + u8 pkt, u8 opt, u8 *val, u8 len) { - int rc = 0; - if (ccid->ccid_ops->ccid_hc_rx_parse_options != NULL) - rc = ccid->ccid_ops->ccid_hc_rx_parse_options(sk, option, len, idx, value); - return rc; + if (ccid->ccid_ops->ccid_hc_rx_parse_options == NULL) + return 0; + return ccid->ccid_ops->ccid_hc_rx_parse_options(sk, pkt, opt, val, len); } static inline int ccid_hc_rx_insert_options(struct ccid *ccid, struct sock *sk, diff --git a/net/dccp/ccids/Kconfig b/net/dccp/ccids/Kconfig index 8408398cd44..0581143cb80 100644 --- a/net/dccp/ccids/Kconfig +++ b/net/dccp/ccids/Kconfig @@ -47,37 +47,6 @@ config IP_DCCP_CCID3_DEBUG If in doubt, say N. -config IP_DCCP_CCID3_RTO - int "Use higher bound for nofeedback timer" - default 100 - depends on IP_DCCP_CCID3 && EXPERIMENTAL - ---help--- - Use higher lower bound for nofeedback timer expiration. - - The TFRC nofeedback timer normally expires after the maximum of 4 - RTTs and twice the current send interval (RFC 3448, 4.3). On LANs - with a small RTT this can mean a high processing load and reduced - performance, since then the nofeedback timer is triggered very - frequently. - - This option enables to set a higher lower bound for the nofeedback - value. Values in units of milliseconds can be set here. - - A value of 0 disables this feature by enforcing the value specified - in RFC 3448. The following values have been suggested as bounds for - experimental use: - * 16-20ms to match the typical multimedia inter-frame interval - * 100ms as a reasonable compromise [default] - * 1000ms corresponds to the lower TCP RTO bound (RFC 2988, 2.4) - - The default of 100ms is a compromise between a large value for - efficient DCCP implementations, and a small value to avoid disrupting - the network in times of congestion. - - The purpose of the nofeedback timer is to slow DCCP down when there - is serious network congestion: experimenting with larger values should - therefore not be performed on WANs. - config IP_DCCP_TFRC_LIB def_bool y if IP_DCCP_CCID3 diff --git a/net/dccp/ccids/ccid2.c b/net/dccp/ccids/ccid2.c index 9b3ae9922be..d850e291f87 100644 --- a/net/dccp/ccids/ccid2.c +++ b/net/dccp/ccids/ccid2.c @@ -25,59 +25,14 @@ */ #include <linux/slab.h> #include "../feat.h" -#include "../ccid.h" -#include "../dccp.h" #include "ccid2.h" #ifdef CONFIG_IP_DCCP_CCID2_DEBUG static int ccid2_debug; #define ccid2_pr_debug(format, a...) DCCP_PR_DEBUG(ccid2_debug, format, ##a) - -static void ccid2_hc_tx_check_sanity(const struct ccid2_hc_tx_sock *hc) -{ - int len = 0; - int pipe = 0; - struct ccid2_seq *seqp = hc->tx_seqh; - - /* there is data in the chain */ - if (seqp != hc->tx_seqt) { - seqp = seqp->ccid2s_prev; - len++; - if (!seqp->ccid2s_acked) - pipe++; - - while (seqp != hc->tx_seqt) { - struct ccid2_seq *prev = seqp->ccid2s_prev; - - len++; - if (!prev->ccid2s_acked) - pipe++; - - /* packets are sent sequentially */ - BUG_ON(dccp_delta_seqno(seqp->ccid2s_seq, - prev->ccid2s_seq ) >= 0); - BUG_ON(time_before(seqp->ccid2s_sent, - prev->ccid2s_sent)); - - seqp = prev; - } - } - - BUG_ON(pipe != hc->tx_pipe); - ccid2_pr_debug("len of chain=%d\n", len); - - do { - seqp = seqp->ccid2s_prev; - len++; - } while (seqp != hc->tx_seqh); - - ccid2_pr_debug("total len=%d\n", len); - BUG_ON(len != hc->tx_seqbufc * CCID2_SEQBUF_LEN); -} #else #define ccid2_pr_debug(format, a...) -#define ccid2_hc_tx_check_sanity(hc) #endif static int ccid2_hc_tx_alloc_seq(struct ccid2_hc_tx_sock *hc) @@ -156,19 +111,10 @@ static void ccid2_change_l_ack_ratio(struct sock *sk, u32 val) dp->dccps_l_ack_ratio = val; } -static void ccid2_change_srtt(struct ccid2_hc_tx_sock *hc, long val) -{ - ccid2_pr_debug("change SRTT to %ld\n", val); - hc->tx_srtt = val; -} - -static void ccid2_start_rto_timer(struct sock *sk); - static void ccid2_hc_tx_rto_expire(unsigned long data) { struct sock *sk = (struct sock *)data; struct ccid2_hc_tx_sock *hc = ccid2_hc_tx_sk(sk); - long s; bh_lock_sock(sk); if (sock_owned_by_user(sk)) { @@ -178,23 +124,19 @@ static void ccid2_hc_tx_rto_expire(unsigned long data) ccid2_pr_debug("RTO_EXPIRE\n"); - ccid2_hc_tx_check_sanity(hc); - /* back-off timer */ hc->tx_rto <<= 1; + if (hc->tx_rto > DCCP_RTO_MAX) + hc->tx_rto = DCCP_RTO_MAX; - s = hc->tx_rto / HZ; - if (s > 60) - hc->tx_rto = 60 * HZ; - - ccid2_start_rto_timer(sk); + sk_reset_timer(sk, &hc->tx_rtotimer, jiffies + hc->tx_rto); /* adjust pipe, cwnd etc */ hc->tx_ssthresh = hc->tx_cwnd / 2; if (hc->tx_ssthresh < 2) hc->tx_ssthresh = 2; - hc->tx_cwnd = 1; - hc->tx_pipe = 0; + hc->tx_cwnd = 1; + hc->tx_pipe = 0; /* clear state about stuff we sent */ hc->tx_seqt = hc->tx_seqh; @@ -204,23 +146,12 @@ static void ccid2_hc_tx_rto_expire(unsigned long data) hc->tx_rpseq = 0; hc->tx_rpdupack = -1; ccid2_change_l_ack_ratio(sk, 1); - ccid2_hc_tx_check_sanity(hc); out: bh_unlock_sock(sk); sock_put(sk); } -static void ccid2_start_rto_timer(struct sock *sk) -{ - struct ccid2_hc_tx_sock *hc = ccid2_hc_tx_sk(sk); - - ccid2_pr_debug("setting RTO timeout=%ld\n", hc->tx_rto); - - BUG_ON(timer_pending(&hc->tx_rtotimer)); - sk_reset_timer(sk, &hc->tx_rtotimer, jiffies + hc->tx_rto); -} - -static void ccid2_hc_tx_packet_sent(struct sock *sk, int more, unsigned int len) +static void ccid2_hc_tx_packet_sent(struct sock *sk, unsigned int len) { struct dccp_sock *dp = dccp_sk(sk); struct ccid2_hc_tx_sock *hc = ccid2_hc_tx_sk(sk); @@ -230,7 +161,7 @@ static void ccid2_hc_tx_packet_sent(struct sock *sk, int more, unsigned int len) hc->tx_seqh->ccid2s_seq = dp->dccps_gss; hc->tx_seqh->ccid2s_acked = 0; - hc->tx_seqh->ccid2s_sent = jiffies; + hc->tx_seqh->ccid2s_sent = ccid2_time_stamp; next = hc->tx_seqh->ccid2s_next; /* check if we need to alloc more space */ @@ -296,23 +227,20 @@ static void ccid2_hc_tx_packet_sent(struct sock *sk, int more, unsigned int len) } #endif - /* setup RTO timer */ - if (!timer_pending(&hc->tx_rtotimer)) - ccid2_start_rto_timer(sk); + sk_reset_timer(sk, &hc->tx_rtotimer, jiffies + hc->tx_rto); #ifdef CONFIG_IP_DCCP_CCID2_DEBUG do { struct ccid2_seq *seqp = hc->tx_seqt; while (seqp != hc->tx_seqh) { - ccid2_pr_debug("out seq=%llu acked=%d time=%lu\n", + ccid2_pr_debug("out seq=%llu acked=%d time=%u\n", (unsigned long long)seqp->ccid2s_seq, seqp->ccid2s_acked, seqp->ccid2s_sent); seqp = seqp->ccid2s_next; } } while (0); ccid2_pr_debug("=========\n"); - ccid2_hc_tx_check_sanity(hc); #endif } @@ -378,17 +306,87 @@ out_invalid_option: return -1; } -static void ccid2_hc_tx_kill_rto_timer(struct sock *sk) +/** + * ccid2_rtt_estimator - Sample RTT and compute RTO using RFC2988 algorithm + * This code is almost identical with TCP's tcp_rtt_estimator(), since + * - it has a higher sampling frequency (recommended by RFC 1323), + * - the RTO does not collapse into RTT due to RTTVAR going towards zero, + * - it is simple (cf. more complex proposals such as Eifel timer or research + * which suggests that the gain should be set according to window size), + * - in tests it was found to work well with CCID2 [gerrit]. + */ +static void ccid2_rtt_estimator(struct sock *sk, const long mrtt) { struct ccid2_hc_tx_sock *hc = ccid2_hc_tx_sk(sk); + long m = mrtt ? : 1; - sk_stop_timer(sk, &hc->tx_rtotimer); - ccid2_pr_debug("deleted RTO timer\n"); + if (hc->tx_srtt == 0) { + /* First measurement m */ + hc->tx_srtt = m << 3; + hc->tx_mdev = m << 1; + + hc->tx_mdev_max = max(hc->tx_mdev, tcp_rto_min(sk)); + hc->tx_rttvar = hc->tx_mdev_max; + + hc->tx_rtt_seq = dccp_sk(sk)->dccps_gss; + } else { + /* Update scaled SRTT as SRTT += 1/8 * (m - SRTT) */ + m -= (hc->tx_srtt >> 3); + hc->tx_srtt += m; + + /* Similarly, update scaled mdev with regard to |m| */ + if (m < 0) { + m = -m; + m -= (hc->tx_mdev >> 2); + /* + * This neutralises RTO increase when RTT < SRTT - mdev + * (see P. Sarolahti, A. Kuznetsov,"Congestion Control + * in Linux TCP", USENIX 2002, pp. 49-62). + */ + if (m > 0) + m >>= 3; + } else { + m -= (hc->tx_mdev >> 2); + } + hc->tx_mdev += m; + + if (hc->tx_mdev > hc->tx_mdev_max) { + hc->tx_mdev_max = hc->tx_mdev; + if (hc->tx_mdev_max > hc->tx_rttvar) + hc->tx_rttvar = hc->tx_mdev_max; + } + + /* + * Decay RTTVAR at most once per flight, exploiting that + * 1) pipe <= cwnd <= Sequence_Window = W (RFC 4340, 7.5.2) + * 2) AWL = GSS-W+1 <= GAR <= GSS (RFC 4340, 7.5.1) + * GAR is a useful bound for FlightSize = pipe. + * AWL is probably too low here, as it over-estimates pipe. + */ + if (after48(dccp_sk(sk)->dccps_gar, hc->tx_rtt_seq)) { + if (hc->tx_mdev_max < hc->tx_rttvar) + hc->tx_rttvar -= (hc->tx_rttvar - + hc->tx_mdev_max) >> 2; + hc->tx_rtt_seq = dccp_sk(sk)->dccps_gss; + hc->tx_mdev_max = tcp_rto_min(sk); + } + } + + /* + * Set RTO from SRTT and RTTVAR + * As in TCP, 4 * RTTVAR >= TCP_RTO_MIN, giving a minimum RTO of 200 ms. + * This agrees with RFC 4341, 5: + * "Because DCCP does not retransmit data, DCCP does not require + * TCP's recommended minimum timeout of one second". + */ + hc->tx_rto = (hc->tx_srtt >> 3) + hc->tx_rttvar; + + if (hc->tx_rto > DCCP_RTO_MAX) + hc->tx_rto = DCCP_RTO_MAX; } -static inline void ccid2_new_ack(struct sock *sk, - struct ccid2_seq *seqp, - unsigned int *maxincr) +static void ccid2_new_ack(struct sock *sk, struct ccid2_seq *seqp, + unsigned int *maxincr) { struct ccid2_hc_tx_sock *hc = ccid2_hc_tx_sk(sk); @@ -402,93 +400,27 @@ static inline void ccid2_new_ack(struct sock *sk, hc->tx_cwnd += 1; hc->tx_packets_acked = 0; } - - /* update RTO */ - if (hc->tx_srtt == -1 || - time_after(jiffies, hc->tx_lastrtt + hc->tx_srtt)) { - unsigned long r = (long)jiffies - (long)seqp->ccid2s_sent; - int s; - - /* first measurement */ - if (hc->tx_srtt == -1) { - ccid2_pr_debug("R: %lu Time=%lu seq=%llu\n", - r, jiffies, - (unsigned long long)seqp->ccid2s_seq); - ccid2_change_srtt(hc, r); - hc->tx_rttvar = r >> 1; - } else { - /* RTTVAR */ - long tmp = hc->tx_srtt - r; - long srtt; - - if (tmp < 0) - tmp *= -1; - - tmp >>= 2; - hc->tx_rttvar *= 3; - hc->tx_rttvar >>= 2; - hc->tx_rttvar += tmp; - - /* SRTT */ - srtt = hc->tx_srtt; - srtt *= 7; - srtt >>= 3; - tmp = r >> 3; - srtt += tmp; - ccid2_change_srtt(hc, srtt); - } - s = hc->tx_rttvar << 2; - /* clock granularity is 1 when based on jiffies */ - if (!s) - s = 1; - hc->tx_rto = hc->tx_srtt + s; - - /* must be at least a second */ - s = hc->tx_rto / HZ; - /* DCCP doesn't require this [but I like it cuz my code sux] */ -#if 1 - if (s < 1) - hc->tx_rto = HZ; -#endif - /* max 60 seconds */ - if (s > 60) - hc->tx_rto = HZ * 60; - - hc->tx_lastrtt = jiffies; - - ccid2_pr_debug("srtt: %ld rttvar: %ld rto: %ld (HZ=%d) R=%lu\n", - hc->tx_srtt, hc->tx_rttvar, - hc->tx_rto, HZ, r); - } - - /* we got a new ack, so re-start RTO timer */ - ccid2_hc_tx_kill_rto_timer(sk); - ccid2_start_rto_timer(sk); -} - -static void ccid2_hc_tx_dec_pipe(struct sock *sk) -{ - struct ccid2_hc_tx_sock *hc = ccid2_hc_tx_sk(sk); - - if (hc->tx_pipe == 0) - DCCP_BUG("pipe == 0"); - else - hc->tx_pipe--; - - if (hc->tx_pipe == 0) - ccid2_hc_tx_kill_rto_timer(sk); + /* + * FIXME: RTT is sampled several times per acknowledgment (for each + * entry in the Ack Vector), instead of once per Ack (as in TCP SACK). + * This causes the RTT to be over-estimated, since the older entries + * in the Ack Vector have earlier sending times. + * The cleanest solution is to not use the ccid2s_sent field at all + * and instead use DCCP timestamps: requires changes in other places. + */ + ccid2_rtt_estimator(sk, ccid2_time_stamp - seqp->ccid2s_sent); } static void ccid2_congestion_event(struct sock *sk, struct ccid2_seq *seqp) { struct ccid2_hc_tx_sock *hc = ccid2_hc_tx_sk(sk); - if (time_before(seqp->ccid2s_sent, hc->tx_last_cong)) { + if ((s32)(seqp->ccid2s_sent - hc->tx_last_cong) < 0) { ccid2_pr_debug("Multiple losses in an RTT---treating as one\n"); return; } - hc->tx_last_cong = jiffies; + hc->tx_last_cong = ccid2_time_stamp; hc->tx_cwnd = hc->tx_cwnd / 2 ? : 1U; hc->tx_ssthresh = max(hc->tx_cwnd, 2U); @@ -510,7 +442,6 @@ static void ccid2_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) int done = 0; unsigned int maxincr = 0; - ccid2_hc_tx_check_sanity(hc); /* check reverse path congestion */ seqno = DCCP_SKB_CB(skb)->dccpd_seq; @@ -620,7 +551,7 @@ static void ccid2_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) seqp->ccid2s_acked = 1; ccid2_pr_debug("Got ack for %llu\n", (unsigned long long)seqp->ccid2s_seq); - ccid2_hc_tx_dec_pipe(sk); + hc->tx_pipe--; } if (seqp == hc->tx_seqt) { done = 1; @@ -677,7 +608,7 @@ static void ccid2_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) * one ack vector. */ ccid2_congestion_event(sk, seqp); - ccid2_hc_tx_dec_pipe(sk); + hc->tx_pipe--; } if (seqp == hc->tx_seqt) break; @@ -695,7 +626,11 @@ static void ccid2_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) hc->tx_seqt = hc->tx_seqt->ccid2s_next; } - ccid2_hc_tx_check_sanity(hc); + /* restart RTO timer if not all outstanding data has been acked */ + if (hc->tx_pipe == 0) + sk_stop_timer(sk, &hc->tx_rtotimer); + else + sk_reset_timer(sk, &hc->tx_rtotimer, jiffies + hc->tx_rto); } static int ccid2_hc_tx_init(struct ccid *ccid, struct sock *sk) @@ -707,12 +642,8 @@ static int ccid2_hc_tx_init(struct ccid *ccid, struct sock *sk) /* RFC 4341, 5: initialise ssthresh to arbitrarily high (max) value */ hc->tx_ssthresh = ~0U; - /* - * RFC 4341, 5: "The cwnd parameter is initialized to at most four - * packets for new connections, following the rules from [RFC3390]". - * We need to convert the bytes of RFC3390 into the packets of RFC 4341. - */ - hc->tx_cwnd = clamp(4380U / dp->dccps_mss_cache, 2U, 4U); + /* Use larger initial windows (RFC 4341, section 5). */ + hc->tx_cwnd = rfc3390_bytes_to_packets(dp->dccps_mss_cache); /* Make sure that Ack Ratio is enabled and within bounds. */ max_ratio = DIV_ROUND_UP(hc->tx_cwnd, 2); @@ -723,15 +654,11 @@ static int ccid2_hc_tx_init(struct ccid *ccid, struct sock *sk) if (ccid2_hc_tx_alloc_seq(hc)) return -ENOMEM; - hc->tx_rto = 3 * HZ; - ccid2_change_srtt(hc, -1); - hc->tx_rttvar = -1; + hc->tx_rto = DCCP_TIMEOUT_INIT; hc->tx_rpdupack = -1; - hc->tx_last_cong = jiffies; + hc->tx_last_cong = ccid2_time_stamp; setup_timer(&hc->tx_rtotimer, ccid2_hc_tx_rto_expire, (unsigned long)sk); - - ccid2_hc_tx_check_sanity(hc); return 0; } @@ -740,7 +667,7 @@ static void ccid2_hc_tx_exit(struct sock *sk) struct ccid2_hc_tx_sock *hc = ccid2_hc_tx_sk(sk); int i; - ccid2_hc_tx_kill_rto_timer(sk); + sk_stop_timer(sk, &hc->tx_rtotimer); for (i = 0; i < hc->tx_seqbufc; i++) kfree(hc->tx_seqbuf[i]); diff --git a/net/dccp/ccids/ccid2.h b/net/dccp/ccids/ccid2.h index 1ec6a30103b..9731c2dc148 100644 --- a/net/dccp/ccids/ccid2.h +++ b/net/dccp/ccids/ccid2.h @@ -18,18 +18,23 @@ #ifndef _DCCP_CCID2_H_ #define _DCCP_CCID2_H_ -#include <linux/dccp.h> #include <linux/timer.h> #include <linux/types.h> #include "../ccid.h" +#include "../dccp.h" + +/* + * CCID-2 timestamping faces the same issues as TCP timestamping. + * Hence we reuse/share as much of the code as possible. + */ +#define ccid2_time_stamp tcp_time_stamp + /* NUMDUPACK parameter from RFC 4341, p. 6 */ #define NUMDUPACK 3 -struct sock; - struct ccid2_seq { u64 ccid2s_seq; - unsigned long ccid2s_sent; + u32 ccid2s_sent; int ccid2s_acked; struct ccid2_seq *ccid2s_prev; struct ccid2_seq *ccid2s_next; @@ -42,7 +47,12 @@ struct ccid2_seq { * struct ccid2_hc_tx_sock - CCID2 TX half connection * @tx_{cwnd,ssthresh,pipe}: as per RFC 4341, section 5 * @tx_packets_acked: Ack counter for deriving cwnd growth (RFC 3465) - * @tx_lastrtt: time RTT was last measured + * @tx_srtt: smoothed RTT estimate, scaled by 2^3 + * @tx_mdev: smoothed RTT variation, scaled by 2^2 + * @tx_mdev_max: maximum of @mdev during one flight + * @tx_rttvar: moving average/maximum of @mdev_max + * @tx_rto: RTO value deriving from SRTT and RTTVAR (RFC 2988) + * @tx_rtt_seq: to decay RTTVAR at most once per flight * @tx_rpseq: last consecutive seqno * @tx_rpdupack: dupacks since rpseq */ @@ -55,14 +65,19 @@ struct ccid2_hc_tx_sock { int tx_seqbufc; struct ccid2_seq *tx_seqh; struct ccid2_seq *tx_seqt; - long tx_rto; - long tx_srtt; - long tx_rttvar; - unsigned long tx_lastrtt; + + /* RTT measurement: variables/principles are the same as in TCP */ + u32 tx_srtt, + tx_mdev, + tx_mdev_max, + tx_rttvar, + tx_rto; + u64 tx_rtt_seq:48; struct timer_list tx_rtotimer; + u64 tx_rpseq; int tx_rpdupack; - unsigned long tx_last_cong; + u32 tx_last_cong; u64 tx_high_ack; }; diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 95f75298649..3060a60ed5a 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -54,7 +54,6 @@ static const char *ccid3_tx_state_name(enum ccid3_hc_tx_states state) [TFRC_SSTATE_NO_SENT] = "NO_SENT", [TFRC_SSTATE_NO_FBACK] = "NO_FBACK", [TFRC_SSTATE_FBACK] = "FBACK", - [TFRC_SSTATE_TERM] = "TERM", }; return ccid3_state_names[state]; @@ -91,19 +90,16 @@ static inline u64 rfc3390_initial_rate(struct sock *sk) return scaled_div(w_init << 6, hc->tx_rtt); } -/* - * Recalculate t_ipi and delta (should be called whenever X changes) +/** + * ccid3_update_send_interval - Calculate new t_ipi = s / X_inst + * This respects the granularity of X_inst (64 * bytes/second). */ static void ccid3_update_send_interval(struct ccid3_hc_tx_sock *hc) { - /* Calculate new t_ipi = s / X_inst (X_inst is in 64 * bytes/second) */ hc->tx_t_ipi = scaled_div32(((u64)hc->tx_s) << 6, hc->tx_x); - /* Calculate new delta by delta = min(t_ipi / 2, t_gran / 2) */ - hc->tx_delta = min_t(u32, hc->tx_t_ipi / 2, TFRC_OPSYS_HALF_TIME_GRAN); - - ccid3_pr_debug("t_ipi=%u, delta=%u, s=%u, X=%u\n", hc->tx_t_ipi, - hc->tx_delta, hc->tx_s, (unsigned)(hc->tx_x >> 6)); + ccid3_pr_debug("t_ipi=%u, s=%u, X=%u\n", hc->tx_t_ipi, + hc->tx_s, (unsigned)(hc->tx_x >> 6)); } static u32 ccid3_hc_tx_idle_rtt(struct ccid3_hc_tx_sock *hc, ktime_t now) @@ -211,16 +207,19 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) ccid3_pr_debug("%s(%p, state=%s) - entry\n", dccp_role(sk), sk, ccid3_tx_state_name(hc->tx_state)); + /* Ignore and do not restart after leaving the established state */ + if ((1 << sk->sk_state) & ~(DCCPF_OPEN | DCCPF_PARTOPEN)) + goto out; + + /* Reset feedback state to "no feedback received" */ if (hc->tx_state == TFRC_SSTATE_FBACK) ccid3_hc_tx_set_state(sk, TFRC_SSTATE_NO_FBACK); - else if (hc->tx_state != TFRC_SSTATE_NO_FBACK) - goto out; /* * Determine new allowed sending rate X as per draft rfc3448bis-00, 4.4 + * RTO is 0 if and only if no feedback has been received yet. */ - if (hc->tx_t_rto == 0 || /* no feedback received yet */ - hc->tx_p == 0) { + if (hc->tx_t_rto == 0 || hc->tx_p == 0) { /* halve send rate directly */ hc->tx_x = max(hc->tx_x / 2, @@ -256,7 +255,7 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) * Set new timeout for the nofeedback timer. * See comments in packet_recv() regarding the value of t_RTO. */ - if (unlikely(hc->tx_t_rto == 0)) /* no feedback yet */ + if (unlikely(hc->tx_t_rto == 0)) /* no feedback received yet */ t_nfb = TFRC_INITIAL_TIMEOUT; else t_nfb = max(hc->tx_t_rto, 2 * hc->tx_t_ipi); @@ -290,8 +289,7 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) if (unlikely(skb->len == 0)) return -EBADMSG; - switch (hc->tx_state) { - case TFRC_SSTATE_NO_SENT: + if (hc->tx_state == TFRC_SSTATE_NO_SENT) { sk_reset_timer(sk, &hc->tx_no_feedback_timer, (jiffies + usecs_to_jiffies(TFRC_INITIAL_TIMEOUT))); hc->tx_last_win_count = 0; @@ -326,27 +324,22 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) ccid3_update_send_interval(hc); ccid3_hc_tx_set_state(sk, TFRC_SSTATE_NO_FBACK); - break; - case TFRC_SSTATE_NO_FBACK: - case TFRC_SSTATE_FBACK: + + } else { delay = ktime_us_delta(hc->tx_t_nom, now); ccid3_pr_debug("delay=%ld\n", (long)delay); /* - * Scheduling of packet transmissions [RFC 3448, 4.6] + * Scheduling of packet transmissions (RFC 5348, 8.3) * * if (t_now > t_nom - delta) * // send the packet now * else * // send the packet in (t_nom - t_now) milliseconds. */ - if (delay - (s64)hc->tx_delta >= 1000) - return (u32)delay / 1000L; + if (delay >= TFRC_T_DELTA) + return (u32)delay / USEC_PER_MSEC; ccid3_hc_tx_update_win_count(hc, now); - break; - case TFRC_SSTATE_TERM: - DCCP_BUG("%s(%p) - Illegal state TERM", dccp_role(sk), sk); - return -EINVAL; } /* prepare to send now (add options etc.) */ @@ -358,8 +351,7 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) return 0; } -static void ccid3_hc_tx_packet_sent(struct sock *sk, int more, - unsigned int len) +static void ccid3_hc_tx_packet_sent(struct sock *sk, unsigned int len) { struct ccid3_hc_tx_sock *hc = ccid3_hc_tx_sk(sk); @@ -372,48 +364,34 @@ static void ccid3_hc_tx_packet_sent(struct sock *sk, int more, static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) { struct ccid3_hc_tx_sock *hc = ccid3_hc_tx_sk(sk); - struct ccid3_options_received *opt_recv; + struct tfrc_tx_hist_entry *acked; ktime_t now; unsigned long t_nfb; - u32 pinv, r_sample; + u32 r_sample; /* we are only interested in ACKs */ if (!(DCCP_SKB_CB(skb)->dccpd_type == DCCP_PKT_ACK || DCCP_SKB_CB(skb)->dccpd_type == DCCP_PKT_DATAACK)) return; - /* ... and only in the established state */ - if (hc->tx_state != TFRC_SSTATE_FBACK && - hc->tx_state != TFRC_SSTATE_NO_FBACK) - return; - - opt_recv = &hc->tx_options_received; - now = ktime_get_real(); - - /* Estimate RTT from history if ACK number is valid */ - r_sample = tfrc_tx_hist_rtt(hc->tx_hist, - DCCP_SKB_CB(skb)->dccpd_ack_seq, now); - if (r_sample == 0) { - DCCP_WARN("%s(%p): %s with bogus ACK-%llu\n", dccp_role(sk), sk, - dccp_packet_name(DCCP_SKB_CB(skb)->dccpd_type), - (unsigned long long)DCCP_SKB_CB(skb)->dccpd_ack_seq); - return; - } - - /* Update receive rate in units of 64 * bytes/second */ - hc->tx_x_recv = opt_recv->ccid3or_receive_rate; - hc->tx_x_recv <<= 6; - - /* Update loss event rate (which is scaled by 1e6) */ - pinv = opt_recv->ccid3or_loss_event_rate; - if (pinv == ~0U || pinv == 0) /* see RFC 4342, 8.5 */ - hc->tx_p = 0; - else /* can not exceed 100% */ - hc->tx_p = scaled_div(1, pinv); /* - * Validate new RTT sample and update moving average + * Locate the acknowledged packet in the TX history. + * + * Returning "entry not found" here can for instance happen when + * - the host has not sent out anything (e.g. a passive server), + * - the Ack is outdated (packet with higher Ack number was received), + * - it is a bogus Ack (for a packet not sent on this connection). */ - r_sample = dccp_sample_rtt(sk, r_sample); + acked = tfrc_tx_hist_find_entry(hc->tx_hist, dccp_hdr_ack_seq(skb)); + if (acked == NULL) + return; + /* For the sake of RTT sampling, ignore/remove all older entries */ + tfrc_tx_hist_purge(&acked->next); + + /* Update the moving average for the RTT estimate (RFC 3448, 4.3) */ + now = ktime_get_real(); + r_sample = dccp_sample_rtt(sk, ktime_us_delta(now, acked->stamp)); hc->tx_rtt = tfrc_ewma(hc->tx_rtt, r_sample, 9); + /* * Update allowed sending rate X as per draft rfc3448bis-00, 4.2/3 */ @@ -461,13 +439,12 @@ done_computing_x: sk->sk_write_space(sk); /* - * Update timeout interval for the nofeedback timer. - * We use a configuration option to increase the lower bound. - * This can help avoid triggering the nofeedback timer too - * often ('spinning') on LANs with small RTTs. + * Update timeout interval for the nofeedback timer. In order to control + * rate halving on networks with very low RTTs (<= 1 ms), use per-route + * tunable RTAX_RTO_MIN value as the lower bound. */ - hc->tx_t_rto = max_t(u32, 4 * hc->tx_rtt, (CONFIG_IP_DCCP_CCID3_RTO * - (USEC_PER_SEC / 1000))); + hc->tx_t_rto = max_t(u32, 4 * hc->tx_rtt, + USEC_PER_SEC/HZ * tcp_rto_min(sk)); /* * Schedule no feedback timer to expire in * max(t_RTO, 2 * s/X) = max(t_RTO, 2 * t_ipi) @@ -482,66 +459,41 @@ done_computing_x: jiffies + usecs_to_jiffies(t_nfb)); } -static int ccid3_hc_tx_parse_options(struct sock *sk, unsigned char option, - unsigned char len, u16 idx, - unsigned char *value) +static int ccid3_hc_tx_parse_options(struct sock *sk, u8 packet_type, + u8 option, u8 *optval, u8 optlen) { - int rc = 0; - const struct dccp_sock *dp = dccp_sk(sk); struct ccid3_hc_tx_sock *hc = ccid3_hc_tx_sk(sk); - struct ccid3_options_received *opt_recv; __be32 opt_val; - opt_recv = &hc->tx_options_received; - - if (opt_recv->ccid3or_seqno != dp->dccps_gsr) { - opt_recv->ccid3or_seqno = dp->dccps_gsr; - opt_recv->ccid3or_loss_event_rate = ~0; - opt_recv->ccid3or_loss_intervals_idx = 0; - opt_recv->ccid3or_loss_intervals_len = 0; - opt_recv->ccid3or_receive_rate = 0; - } - switch (option) { + case TFRC_OPT_RECEIVE_RATE: case TFRC_OPT_LOSS_EVENT_RATE: - if (unlikely(len != 4)) { - DCCP_WARN("%s(%p), invalid len %d " - "for TFRC_OPT_LOSS_EVENT_RATE\n", - dccp_role(sk), sk, len); - rc = -EINVAL; - } else { - opt_val = get_unaligned((__be32 *)value); - opt_recv->ccid3or_loss_event_rate = ntohl(opt_val); - ccid3_pr_debug("%s(%p), LOSS_EVENT_RATE=%u\n", - dccp_role(sk), sk, - opt_recv->ccid3or_loss_event_rate); + /* Must be ignored on Data packets, cf. RFC 4342 8.3 and 8.5 */ + if (packet_type == DCCP_PKT_DATA) + break; + if (unlikely(optlen != 4)) { + DCCP_WARN("%s(%p), invalid len %d for %u\n", + dccp_role(sk), sk, optlen, option); + return -EINVAL; } - break; - case TFRC_OPT_LOSS_INTERVALS: - opt_recv->ccid3or_loss_intervals_idx = idx; - opt_recv->ccid3or_loss_intervals_len = len; - ccid3_pr_debug("%s(%p), LOSS_INTERVALS=(%u, %u)\n", - dccp_role(sk), sk, - opt_recv->ccid3or_loss_intervals_idx, - opt_recv->ccid3or_loss_intervals_len); - break; - case TFRC_OPT_RECEIVE_RATE: - if (unlikely(len != 4)) { - DCCP_WARN("%s(%p), invalid len %d " - "for TFRC_OPT_RECEIVE_RATE\n", - dccp_role(sk), sk, len); - rc = -EINVAL; - } else { - opt_val = get_unaligned((__be32 *)value); - opt_recv->ccid3or_receive_rate = ntohl(opt_val); + opt_val = ntohl(get_unaligned((__be32 *)optval)); + + if (option == TFRC_OPT_RECEIVE_RATE) { + /* Receive Rate is kept in units of 64 bytes/second */ + hc->tx_x_recv = opt_val; + hc->tx_x_recv <<= 6; + ccid3_pr_debug("%s(%p), RECEIVE_RATE=%u\n", - dccp_role(sk), sk, - opt_recv->ccid3or_receive_rate); + dccp_role(sk), sk, opt_val); + } else { + /* Update the fixpoint Loss Event Rate fraction */ + hc->tx_p = tfrc_invert_loss_event_rate(opt_val); + + ccid3_pr_debug("%s(%p), LOSS_EVENT_RATE=%u\n", + dccp_role(sk), sk, opt_val); } - break; } - - return rc; + return 0; } static int ccid3_hc_tx_init(struct ccid *ccid, struct sock *sk) @@ -559,42 +511,36 @@ static void ccid3_hc_tx_exit(struct sock *sk) { struct ccid3_hc_tx_sock *hc = ccid3_hc_tx_sk(sk); - ccid3_hc_tx_set_state(sk, TFRC_SSTATE_TERM); sk_stop_timer(sk, &hc->tx_no_feedback_timer); - tfrc_tx_hist_purge(&hc->tx_hist); } static void ccid3_hc_tx_get_info(struct sock *sk, struct tcp_info *info) { - struct ccid3_hc_tx_sock *hc; - - /* Listen socks doesn't have a private CCID block */ - if (sk->sk_state == DCCP_LISTEN) - return; - - hc = ccid3_hc_tx_sk(sk); - info->tcpi_rto = hc->tx_t_rto; - info->tcpi_rtt = hc->tx_rtt; + info->tcpi_rto = ccid3_hc_tx_sk(sk)->tx_t_rto; + info->tcpi_rtt = ccid3_hc_tx_sk(sk)->tx_rtt; } static int ccid3_hc_tx_getsockopt(struct sock *sk, const int optname, int len, u32 __user *optval, int __user *optlen) { - const struct ccid3_hc_tx_sock *hc; + const struct ccid3_hc_tx_sock *hc = ccid3_hc_tx_sk(sk); + struct tfrc_tx_info tfrc; const void *val; - /* Listen socks doesn't have a private CCID block */ - if (sk->sk_state == DCCP_LISTEN) - return -EINVAL; - - hc = ccid3_hc_tx_sk(sk); switch (optname) { case DCCP_SOCKOPT_CCID_TX_INFO: - if (len < sizeof(hc->tx_tfrc)) + if (len < sizeof(tfrc)) return -EINVAL; - len = sizeof(hc->tx_tfrc); - val = &hc->tx_tfrc; + tfrc.tfrctx_x = hc->tx_x; + tfrc.tfrctx_x_recv = hc->tx_x_recv; + tfrc.tfrctx_x_calc = hc->tx_x_calc; + tfrc.tfrctx_rtt = hc->tx_rtt; + tfrc.tfrctx_p = hc->tx_p; + tfrc.tfrctx_rto = hc->tx_t_rto; + tfrc.tfrctx_ipi = hc->tx_t_ipi; + len = sizeof(tfrc); + val = &tfrc; break; default: return -ENOPROTOOPT; @@ -624,7 +570,6 @@ static const char *ccid3_rx_state_name(enum ccid3_hc_rx_states state) static const char *const ccid3_rx_state_names[] = { [TFRC_RSTATE_NO_DATA] = "NO_DATA", [TFRC_RSTATE_DATA] = "DATA", - [TFRC_RSTATE_TERM] = "TERM", }; return ccid3_rx_state_names[state]; @@ -650,14 +595,9 @@ static void ccid3_hc_rx_send_feedback(struct sock *sk, { struct ccid3_hc_rx_sock *hc = ccid3_hc_rx_sk(sk); struct dccp_sock *dp = dccp_sk(sk); - ktime_t now; + ktime_t now = ktime_get_real(); s64 delta = 0; - if (unlikely(hc->rx_state == TFRC_RSTATE_TERM)) - return; - - now = ktime_get_real(); - switch (fbtype) { case CCID3_FBACK_INITIAL: hc->rx_x_recv = 0; @@ -701,14 +641,12 @@ static void ccid3_hc_rx_send_feedback(struct sock *sk, static int ccid3_hc_rx_insert_options(struct sock *sk, struct sk_buff *skb) { - const struct ccid3_hc_rx_sock *hc; + const struct ccid3_hc_rx_sock *hc = ccid3_hc_rx_sk(sk); __be32 x_recv, pinv; if (!(sk->sk_state == DCCP_OPEN || sk->sk_state == DCCP_PARTOPEN)) return 0; - hc = ccid3_hc_rx_sk(sk); - if (dccp_packet_without_ack(skb)) return 0; @@ -749,10 +687,11 @@ static u32 ccid3_first_li(struct sock *sk) x_recv = scaled_div32(hc->rx_bytes_recv, delta); if (x_recv == 0) { /* would also trigger divide-by-zero */ DCCP_WARN("X_recv==0\n"); - if ((x_recv = hc->rx_x_recv) == 0) { + if (hc->rx_x_recv == 0) { DCCP_BUG("stored value of X_recv is zero"); return ~0U; } + x_recv = hc->rx_x_recv; } fval = scaled_div(hc->rx_s, hc->rx_rtt); @@ -862,46 +801,31 @@ static void ccid3_hc_rx_exit(struct sock *sk) { struct ccid3_hc_rx_sock *hc = ccid3_hc_rx_sk(sk); - ccid3_hc_rx_set_state(sk, TFRC_RSTATE_TERM); - tfrc_rx_hist_purge(&hc->rx_hist); tfrc_lh_cleanup(&hc->rx_li_hist); } static void ccid3_hc_rx_get_info(struct sock *sk, struct tcp_info *info) { - const struct ccid3_hc_rx_sock *hc; - - /* Listen socks doesn't have a private CCID block */ - if (sk->sk_state == DCCP_LISTEN) - return; - - hc = ccid3_hc_rx_sk(sk); - info->tcpi_ca_state = hc->rx_state; + info->tcpi_ca_state = ccid3_hc_rx_sk(sk)->rx_state; info->tcpi_options |= TCPI_OPT_TIMESTAMPS; - info->tcpi_rcv_rtt = hc->rx_rtt; + info->tcpi_rcv_rtt = ccid3_hc_rx_sk(sk)->rx_rtt; } static int ccid3_hc_rx_getsockopt(struct sock *sk, const int optname, int len, u32 __user *optval, int __user *optlen) { - const struct ccid3_hc_rx_sock *hc; + const struct ccid3_hc_rx_sock *hc = ccid3_hc_rx_sk(sk); struct tfrc_rx_info rx_info; const void *val; - /* Listen socks doesn't have a private CCID block */ - if (sk->sk_state == DCCP_LISTEN) - return -EINVAL; - - hc = ccid3_hc_rx_sk(sk); switch (optname) { case DCCP_SOCKOPT_CCID_RX_INFO: if (len < sizeof(rx_info)) return -EINVAL; rx_info.tfrcrx_x_recv = hc->rx_x_recv; rx_info.tfrcrx_rtt = hc->rx_rtt; - rx_info.tfrcrx_p = hc->rx_pinv == 0 ? ~0U : - scaled_div(1, hc->rx_pinv); + rx_info.tfrcrx_p = tfrc_invert_loss_event_rate(hc->rx_pinv); len = sizeof(rx_info); val = &rx_info; break; diff --git a/net/dccp/ccids/ccid3.h b/net/dccp/ccids/ccid3.h index 03263577665..1a9933c2967 100644 --- a/net/dccp/ccids/ccid3.h +++ b/net/dccp/ccids/ccid3.h @@ -42,35 +42,36 @@ #include "lib/tfrc.h" #include "../ccid.h" -/* Two seconds as per RFC 3448 4.2 */ +/* Two seconds as per RFC 5348, 4.2 */ #define TFRC_INITIAL_TIMEOUT (2 * USEC_PER_SEC) -/* In usecs - half the scheduling granularity as per RFC3448 4.6 */ -#define TFRC_OPSYS_HALF_TIME_GRAN (USEC_PER_SEC / (2 * HZ)) - /* Parameter t_mbi from [RFC 3448, 4.3]: backoff interval in seconds */ #define TFRC_T_MBI 64 +/* + * The t_delta parameter (RFC 5348, 8.3): delays of less than %USEC_PER_MSEC are + * rounded down to 0, since sk_reset_timer() here uses millisecond granularity. + * Hence we can use a constant t_delta = %USEC_PER_MSEC when HZ >= 500. A coarse + * resolution of HZ < 500 means that the error is below one timer tick (t_gran) + * when using the constant t_delta = t_gran / 2 = %USEC_PER_SEC / (2 * HZ). + */ +#if (HZ >= 500) +# define TFRC_T_DELTA USEC_PER_MSEC +#else +# define TFRC_T_DELTA (USEC_PER_SEC / (2 * HZ)) +#endif + enum ccid3_options { TFRC_OPT_LOSS_EVENT_RATE = 192, TFRC_OPT_LOSS_INTERVALS = 193, TFRC_OPT_RECEIVE_RATE = 194, }; -struct ccid3_options_received { - u64 ccid3or_seqno:48, - ccid3or_loss_intervals_idx:16; - u16 ccid3or_loss_intervals_len; - u32 ccid3or_loss_event_rate; - u32 ccid3or_receive_rate; -}; - /* TFRC sender states */ enum ccid3_hc_tx_states { TFRC_SSTATE_NO_SENT = 1, TFRC_SSTATE_NO_FBACK, TFRC_SSTATE_FBACK, - TFRC_SSTATE_TERM, }; /** @@ -90,19 +91,16 @@ enum ccid3_hc_tx_states { * @tx_no_feedback_timer: Handle to no feedback timer * @tx_t_ld: Time last doubled during slow start * @tx_t_nom: Nominal send time of next packet - * @tx_delta: Send timer delta (RFC 3448, 4.6) in usecs * @tx_hist: Packet history - * @tx_options_received: Parsed set of retrieved options */ struct ccid3_hc_tx_sock { - struct tfrc_tx_info tx_tfrc; -#define tx_x tx_tfrc.tfrctx_x -#define tx_x_recv tx_tfrc.tfrctx_x_recv -#define tx_x_calc tx_tfrc.tfrctx_x_calc -#define tx_rtt tx_tfrc.tfrctx_rtt -#define tx_p tx_tfrc.tfrctx_p -#define tx_t_rto tx_tfrc.tfrctx_rto -#define tx_t_ipi tx_tfrc.tfrctx_ipi + u64 tx_x; + u64 tx_x_recv; + u32 tx_x_calc; + u32 tx_rtt; + u32 tx_p; + u32 tx_t_rto; + u32 tx_t_ipi; u16 tx_s; enum ccid3_hc_tx_states tx_state:8; u8 tx_last_win_count; @@ -110,9 +108,7 @@ struct ccid3_hc_tx_sock { struct timer_list tx_no_feedback_timer; ktime_t tx_t_ld; ktime_t tx_t_nom; - u32 tx_delta; struct tfrc_tx_hist_entry *tx_hist; - struct ccid3_options_received tx_options_received; }; static inline struct ccid3_hc_tx_sock *ccid3_hc_tx_sk(const struct sock *sk) @@ -126,21 +122,16 @@ static inline struct ccid3_hc_tx_sock *ccid3_hc_tx_sk(const struct sock *sk) enum ccid3_hc_rx_states { TFRC_RSTATE_NO_DATA = 1, TFRC_RSTATE_DATA, - TFRC_RSTATE_TERM = 127, }; /** * struct ccid3_hc_rx_sock - CCID3 receiver half-connection socket - * @rx_x_recv: Receiver estimate of send rate (RFC 3448 4.3) - * @rx_rtt: Receiver estimate of rtt (non-standard) - * @rx_p: Current loss event rate (RFC 3448 5.4) * @rx_last_counter: Tracks window counter (RFC 4342, 8.1) * @rx_state: Receiver state, one of %ccid3_hc_rx_states * @rx_bytes_recv: Total sum of DCCP payload bytes * @rx_x_recv: Receiver estimate of send rate (RFC 3448, sec. 4.3) * @rx_rtt: Receiver estimate of RTT * @rx_tstamp_last_feedback: Time at which last feedback was sent - * @rx_tstamp_last_ack: Time at which last feedback was sent * @rx_hist: Packet history (loss detection + RTT sampling) * @rx_li_hist: Loss Interval database * @rx_s: Received packet size in bytes diff --git a/net/dccp/ccids/lib/loss_interval.c b/net/dccp/ccids/lib/loss_interval.c index 8fc3cbf7907..497723c4d4b 100644 --- a/net/dccp/ccids/lib/loss_interval.c +++ b/net/dccp/ccids/lib/loss_interval.c @@ -116,7 +116,7 @@ u8 tfrc_lh_update_i_mean(struct tfrc_loss_hist *lh, struct sk_buff *skb) cur->li_length = len; tfrc_lh_calc_i_mean(lh); - return (lh->i_mean < old_i_mean); + return lh->i_mean < old_i_mean; } /* Determine if `new_loss' does begin a new loss interval [RFC 4342, 10.2] */ diff --git a/net/dccp/ccids/lib/packet_history.c b/net/dccp/ccids/lib/packet_history.c index 3a4f414e94a..de8fe294bf0 100644 --- a/net/dccp/ccids/lib/packet_history.c +++ b/net/dccp/ccids/lib/packet_history.c @@ -38,18 +38,6 @@ #include "packet_history.h" #include "../../dccp.h" -/** - * tfrc_tx_hist_entry - Simple singly-linked TX history list - * @next: next oldest entry (LIFO order) - * @seqno: sequence number of this entry - * @stamp: send time of packet with sequence number @seqno - */ -struct tfrc_tx_hist_entry { - struct tfrc_tx_hist_entry *next; - u64 seqno; - ktime_t stamp; -}; - /* * Transmitter History Routines */ @@ -71,15 +59,6 @@ void tfrc_tx_packet_history_exit(void) } } -static struct tfrc_tx_hist_entry * - tfrc_tx_hist_find_entry(struct tfrc_tx_hist_entry *head, u64 seqno) -{ - while (head != NULL && head->seqno != seqno) - head = head->next; - - return head; -} - int tfrc_tx_hist_add(struct tfrc_tx_hist_entry **headp, u64 seqno) { struct tfrc_tx_hist_entry *entry = kmem_cache_alloc(tfrc_tx_hist_slab, gfp_any()); @@ -107,24 +86,6 @@ void tfrc_tx_hist_purge(struct tfrc_tx_hist_entry **headp) *headp = NULL; } -u32 tfrc_tx_hist_rtt(struct tfrc_tx_hist_entry *head, const u64 seqno, - const ktime_t now) -{ - u32 rtt = 0; - struct tfrc_tx_hist_entry *packet = tfrc_tx_hist_find_entry(head, seqno); - - if (packet != NULL) { - rtt = ktime_us_delta(now, packet->stamp); - /* - * Garbage-collect older (irrelevant) entries: - */ - tfrc_tx_hist_purge(&packet->next); - } - - return rtt; -} - - /* * Receiver History Routines */ diff --git a/net/dccp/ccids/lib/packet_history.h b/net/dccp/ccids/lib/packet_history.h index 7df6c529999..7ee4a9d9d33 100644 --- a/net/dccp/ccids/lib/packet_history.h +++ b/net/dccp/ccids/lib/packet_history.h @@ -40,12 +40,28 @@ #include <linux/slab.h> #include "tfrc.h" -struct tfrc_tx_hist_entry; +/** + * tfrc_tx_hist_entry - Simple singly-linked TX history list + * @next: next oldest entry (LIFO order) + * @seqno: sequence number of this entry + * @stamp: send time of packet with sequence number @seqno + */ +struct tfrc_tx_hist_entry { + struct tfrc_tx_hist_entry *next; + u64 seqno; + ktime_t stamp; +}; + +static inline struct tfrc_tx_hist_entry * + tfrc_tx_hist_find_entry(struct tfrc_tx_hist_entry *head, u64 seqno) +{ + while (head != NULL && head->seqno != seqno) + head = head->next; + return head; +} extern int tfrc_tx_hist_add(struct tfrc_tx_hist_entry **headp, u64 seqno); extern void tfrc_tx_hist_purge(struct tfrc_tx_hist_entry **headp); -extern u32 tfrc_tx_hist_rtt(struct tfrc_tx_hist_entry *head, - const u64 seqno, const ktime_t now); /* Subtraction a-b modulo-16, respects circular wrap-around */ #define SUB16(a, b) (((a) + 16 - (b)) & 0xF) diff --git a/net/dccp/ccids/lib/tfrc.h b/net/dccp/ccids/lib/tfrc.h index 01bb48e96c2..f8ee3f54977 100644 --- a/net/dccp/ccids/lib/tfrc.h +++ b/net/dccp/ccids/lib/tfrc.h @@ -57,6 +57,7 @@ static inline u32 tfrc_ewma(const u32 avg, const u32 newval, const u8 weight) extern u32 tfrc_calc_x(u16 s, u32 R, u32 p); extern u32 tfrc_calc_x_reverse_lookup(u32 fvalue); +extern u32 tfrc_invert_loss_event_rate(u32 loss_event_rate); extern int tfrc_tx_packet_history_init(void); extern void tfrc_tx_packet_history_exit(void); diff --git a/net/dccp/ccids/lib/tfrc_equation.c b/net/dccp/ccids/lib/tfrc_equation.c index 22ca1cf0eb5..a052a4377e2 100644 --- a/net/dccp/ccids/lib/tfrc_equation.c +++ b/net/dccp/ccids/lib/tfrc_equation.c @@ -687,3 +687,17 @@ u32 tfrc_calc_x_reverse_lookup(u32 fvalue) index = tfrc_binsearch(fvalue, 0); return (index + 1) * 1000000 / TFRC_CALC_X_ARRSIZE; } + +/** + * tfrc_invert_loss_event_rate - Compute p so that 10^6 corresponds to 100% + * When @loss_event_rate is large, there is a chance that p is truncated to 0. + * To avoid re-entering slow-start in that case, we set p = TFRC_SMALLEST_P > 0. + */ +u32 tfrc_invert_loss_event_rate(u32 loss_event_rate) +{ + if (loss_event_rate == UINT_MAX) /* see RFC 4342, 8.5 */ + return 0; + if (unlikely(loss_event_rate == 0)) /* map 1/0 into 100% */ + return 1000000; + return max_t(u32, scaled_div(1, loss_event_rate), TFRC_SMALLEST_P); +} diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h index 3ccef1b70fe..3eb264b6082 100644 --- a/net/dccp/dccp.h +++ b/net/dccp/dccp.h @@ -153,18 +153,27 @@ static inline u64 max48(const u64 seq1, const u64 seq2) } /** - * dccp_loss_free - Evaluates condition for data loss from RFC 4340, 7.7.1 - * @s1: start sequence number - * @s2: end sequence number + * dccp_loss_count - Approximate the number of lost data packets in a burst loss + * @s1: last known sequence number before the loss ('hole') + * @s2: first sequence number seen after the 'hole' * @ndp: NDP count on packet with sequence number @s2 - * Returns true if the sequence range s1...s2 has no data loss. */ -static inline bool dccp_loss_free(const u64 s1, const u64 s2, const u64 ndp) +static inline u64 dccp_loss_count(const u64 s1, const u64 s2, const u64 ndp) { s64 delta = dccp_delta_seqno(s1, s2); WARN_ON(delta < 0); - return (u64)delta <= ndp + 1; + delta -= ndp + 1; + + return delta > 0 ? delta : 0; +} + +/** + * dccp_loss_free - Evaluate condition for data loss from RFC 4340, 7.7.1 + */ +static inline bool dccp_loss_free(const u64 s1, const u64 s2, const u64 ndp) +{ + return dccp_loss_count(s1, s2, ndp) == 0; } enum { @@ -246,7 +255,6 @@ static inline void dccp_clear_xmit_timers(struct sock *sk) extern unsigned int dccp_sync_mss(struct sock *sk, u32 pmtu); extern const char *dccp_packet_name(const int type); -extern const char *dccp_state_name(const int state); extern void dccp_set_state(struct sock *sk, const int state); extern void dccp_done(struct sock *sk); @@ -415,6 +423,23 @@ static inline void dccp_update_gsr(struct sock *sk, u64 seq) dp->dccps_gsr = seq; /* Sequence validity window depends on remote Sequence Window (7.5.1) */ dp->dccps_swl = SUB48(ADD48(dp->dccps_gsr, 1), dp->dccps_r_seq_win / 4); + /* + * Adjust SWL so that it is not below ISR. In contrast to RFC 4340, + * 7.5.1 we perform this check beyond the initial handshake: W/W' are + * always > 32, so for the first W/W' packets in the lifetime of a + * connection we always have to adjust SWL. + * A second reason why we are doing this is that the window depends on + * the feature-remote value of Sequence Window: nothing stops the peer + * from updating this value while we are busy adjusting SWL for the + * first W packets (we would have to count from scratch again then). + * Therefore it is safer to always make sure that the Sequence Window + * is not artificially extended by a peer who grows SWL downwards by + * continually updating the feature-remote Sequence-Window. + * If sequence numbers wrap it is bad luck. But that will take a while + * (48 bit), and this measure prevents Sequence-number attacks. + */ + if (before48(dp->dccps_swl, dp->dccps_isr)) + dp->dccps_swl = dp->dccps_isr; dp->dccps_swh = ADD48(dp->dccps_gsr, (3 * dp->dccps_r_seq_win) / 4); } @@ -425,14 +450,16 @@ static inline void dccp_update_gss(struct sock *sk, u64 seq) dp->dccps_gss = seq; /* Ack validity window depends on local Sequence Window value (7.5.1) */ dp->dccps_awl = SUB48(ADD48(dp->dccps_gss, 1), dp->dccps_l_seq_win); + /* Adjust AWL so that it is not below ISS - see comment above for SWL */ + if (before48(dp->dccps_awl, dp->dccps_iss)) + dp->dccps_awl = dp->dccps_iss; dp->dccps_awh = dp->dccps_gss; } static inline int dccp_ack_pending(const struct sock *sk) { const struct dccp_sock *dp = dccp_sk(sk); - return dp->dccps_timestamp_echo != 0 || - (dp->dccps_hc_rx_ackvec != NULL && + return (dp->dccps_hc_rx_ackvec != NULL && dccp_ackvec_pending(dp->dccps_hc_rx_ackvec)) || inet_csk_ack_scheduled(sk); } @@ -449,7 +476,6 @@ extern int dccp_insert_options_rsk(struct dccp_request_sock*, struct sk_buff*); extern int dccp_insert_option_elapsed_time(struct sk_buff *skb, u32 elapsed); extern u32 dccp_timestamp(void); extern void dccp_timestamping_init(void); -extern int dccp_insert_option_timestamp(struct sk_buff *skb); extern int dccp_insert_option(struct sk_buff *skb, unsigned char option, const void *value, unsigned char len); diff --git a/net/dccp/feat.c b/net/dccp/feat.c index df7dd26cf07..568def95272 100644 --- a/net/dccp/feat.c +++ b/net/dccp/feat.c @@ -730,16 +730,6 @@ int dccp_feat_register_sp(struct sock *sk, u8 feat, u8 is_local, 0, list, len); } -/* Analogous to dccp_feat_register_sp(), but for non-negotiable values */ -int dccp_feat_register_nn(struct sock *sk, u8 feat, u64 val) -{ - /* any changes must be registered before establishing the connection */ - if (sk->sk_state != DCCP_CLOSED) - return -EISCONN; - if (dccp_feat_type(feat) != FEAT_NN) - return -EINVAL; - return __feat_register_nn(&dccp_sk(sk)->dccps_featneg, feat, 0, val); -} /* * Tracking features whose value depend on the choice of CCID diff --git a/net/dccp/feat.h b/net/dccp/feat.h index f96721619de..e56a4e5e634 100644 --- a/net/dccp/feat.h +++ b/net/dccp/feat.h @@ -111,7 +111,6 @@ extern int dccp_feat_init(struct sock *sk); extern void dccp_feat_initialise_sysctls(void); extern int dccp_feat_register_sp(struct sock *sk, u8 feat, u8 is_local, u8 const *list, u8 len); -extern int dccp_feat_register_nn(struct sock *sk, u8 feat, u64 val); extern int dccp_feat_parse_options(struct sock *, struct dccp_request_sock *, u8 mand, u8 opt, u8 feat, u8 *val, u8 len); extern int dccp_feat_clone_list(struct list_head const *, struct list_head *); diff --git a/net/dccp/input.c b/net/dccp/input.c index 10c957a88f4..265985370fa 100644 --- a/net/dccp/input.c +++ b/net/dccp/input.c @@ -259,7 +259,7 @@ static int dccp_check_seqno(struct sock *sk, struct sk_buff *skb) sysctl_dccp_sync_ratelimit))) return 0; - DCCP_WARN("DCCP: Step 6 failed for %s packet, " + DCCP_WARN("Step 6 failed for %s packet, " "(LSWL(%llu) <= P.seqno(%llu) <= S.SWH(%llu)) and " "(P.ackno %s or LAWL(%llu) <= P.ackno(%llu) <= S.AWH(%llu), " "sending SYNC...\n", dccp_packet_name(dh->dccph_type), @@ -441,20 +441,14 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk, kfree_skb(sk->sk_send_head); sk->sk_send_head = NULL; - dp->dccps_isr = DCCP_SKB_CB(skb)->dccpd_seq; - dccp_update_gsr(sk, dp->dccps_isr); /* - * SWL and AWL are initially adjusted so that they are not less than - * the initial Sequence Numbers received and sent, respectively: - * SWL := max(GSR + 1 - floor(W/4), ISR), - * AWL := max(GSS - W' + 1, ISS). - * These adjustments MUST be applied only at the beginning of the - * connection. - * - * AWL was adjusted in dccp_v4_connect -acme + * Set ISR, GSR from packet. ISS was set in dccp_v{4,6}_connect + * and GSS in dccp_transmit_skb(). Setting AWL/AWH and SWL/SWH + * is done as part of activating the feature values below, since + * these settings depend on the local/remote Sequence Window + * features, which were undefined or not confirmed until now. */ - dccp_set_seqno(&dp->dccps_swl, - max48(dp->dccps_swl, dp->dccps_isr)); + dp->dccps_gsr = dp->dccps_isr = DCCP_SKB_CB(skb)->dccpd_seq; dccp_sync_mss(sk, icsk->icsk_pmtu_cookie); diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index d4a166f0f39..3f69ea11482 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -392,7 +392,7 @@ struct sock *dccp_v4_request_recv_sock(struct sock *sk, struct sk_buff *skb, newsk = dccp_create_openreq_child(sk, req, skb); if (newsk == NULL) - goto exit; + goto exit_nonewsk; sk_setup_caps(newsk, dst); @@ -409,16 +409,20 @@ struct sock *dccp_v4_request_recv_sock(struct sock *sk, struct sk_buff *skb, dccp_sync_mss(newsk, dst_mtu(dst)); + if (__inet_inherit_port(sk, newsk) < 0) { + sock_put(newsk); + goto exit; + } __inet_hash_nolisten(newsk, NULL); - __inet_inherit_port(sk, newsk); return newsk; exit_overflow: NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENOVERFLOWS); +exit_nonewsk: + dst_release(dst); exit: NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS); - dst_release(dst); return NULL; } diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 6e3f32575df..dca711df9b6 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -564,7 +564,7 @@ static struct sock *dccp_v6_request_recv_sock(struct sock *sk, newsk = dccp_create_openreq_child(sk, req, skb); if (newsk == NULL) - goto out; + goto out_nonewsk; /* * No need to charge this sock to the relevant IPv6 refcnt debug socks @@ -632,18 +632,22 @@ static struct sock *dccp_v6_request_recv_sock(struct sock *sk, newinet->inet_daddr = newinet->inet_saddr = LOOPBACK4_IPV6; newinet->inet_rcv_saddr = LOOPBACK4_IPV6; + if (__inet_inherit_port(sk, newsk) < 0) { + sock_put(newsk); + goto out; + } __inet6_hash(newsk, NULL); - __inet_inherit_port(sk, newsk); return newsk; out_overflow: NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENOVERFLOWS); +out_nonewsk: + dst_release(dst); out: NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS); if (opt != NULL && opt != np->opt) sock_kfree_s(sk, opt, opt->tot_len); - dst_release(dst); return NULL; } diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 128b089d3ae..d7041a0963a 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -121,30 +121,18 @@ struct sock *dccp_create_openreq_child(struct sock *sk, * * Choose S.ISS (initial seqno) or set from Init Cookies * Initialize S.GAR := S.ISS - * Set S.ISR, S.GSR, S.SWL, S.SWH from packet or Init Cookies - */ - newdp->dccps_gar = newdp->dccps_iss = dreq->dreq_iss; - dccp_update_gss(newsk, dreq->dreq_iss); - - newdp->dccps_isr = dreq->dreq_isr; - dccp_update_gsr(newsk, dreq->dreq_isr); - - /* - * SWL and AWL are initially adjusted so that they are not less than - * the initial Sequence Numbers received and sent, respectively: - * SWL := max(GSR + 1 - floor(W/4), ISR), - * AWL := max(GSS - W' + 1, ISS). - * These adjustments MUST be applied only at the beginning of the - * connection. + * Set S.ISR, S.GSR from packet (or Init Cookies) + * + * Setting AWL/AWH and SWL/SWH happens as part of the feature + * activation below, as these windows all depend on the local + * and remote Sequence Window feature values (7.5.2). */ - dccp_set_seqno(&newdp->dccps_swl, - max48(newdp->dccps_swl, newdp->dccps_isr)); - dccp_set_seqno(&newdp->dccps_awl, - max48(newdp->dccps_awl, newdp->dccps_iss)); + newdp->dccps_gss = newdp->dccps_iss = dreq->dreq_iss; + newdp->dccps_gar = newdp->dccps_iss; + newdp->dccps_gsr = newdp->dccps_isr = dreq->dreq_isr; /* - * Activate features after initialising the sequence numbers, - * since CCID initialisation may depend on GSS, ISR, ISS etc. + * Activate features: initialise CCIDs, sequence windows etc. */ if (dccp_feat_activate_values(newsk, &dreq->dreq_featneg)) { /* It is still raw copy of parent, so invalidate diff --git a/net/dccp/options.c b/net/dccp/options.c index bfda087bd90..cd306181300 100644 --- a/net/dccp/options.c +++ b/net/dccp/options.c @@ -96,18 +96,11 @@ int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq, } /* - * CCID-Specific Options (from RFC 4340, sec. 10.3): - * - * Option numbers 128 through 191 are for options sent from the - * HC-Sender to the HC-Receiver; option numbers 192 through 255 - * are for options sent from the HC-Receiver to the HC-Sender. - * * CCID-specific options are ignored during connection setup, as * negotiation may still be in progress (see RFC 4340, 10.3). * The same applies to Ack Vectors, as these depend on the CCID. - * */ - if (dreq != NULL && (opt >= 128 || + if (dreq != NULL && (opt >= DCCPO_MIN_RX_CCID_SPECIFIC || opt == DCCPO_ACK_VECTOR_0 || opt == DCCPO_ACK_VECTOR_1)) goto ignore_option; @@ -170,6 +163,8 @@ int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq, dccp_role(sk), ntohl(opt_val), (unsigned long long) DCCP_SKB_CB(skb)->dccpd_ack_seq); + /* schedule an Ack in case this sender is quiescent */ + inet_csk_schedule_ack(sk); break; case DCCPO_TIMESTAMP_ECHO: if (len != 4 && len != 6 && len != 8) @@ -226,23 +221,15 @@ int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq, dccp_pr_debug("%s rx opt: ELAPSED_TIME=%d\n", dccp_role(sk), elapsed_time); break; - case 128 ... 191: { - const u16 idx = value - options; - + case DCCPO_MIN_RX_CCID_SPECIFIC ... DCCPO_MAX_RX_CCID_SPECIFIC: if (ccid_hc_rx_parse_options(dp->dccps_hc_rx_ccid, sk, - opt, len, idx, - value) != 0) + pkt_type, opt, value, len)) goto out_invalid_option; - } break; - case 192 ... 255: { - const u16 idx = value - options; - + case DCCPO_MIN_TX_CCID_SPECIFIC ... DCCPO_MAX_TX_CCID_SPECIFIC: if (ccid_hc_tx_parse_options(dp->dccps_hc_tx_ccid, sk, - opt, len, idx, - value) != 0) + pkt_type, opt, value, len)) goto out_invalid_option; - } break; default: DCCP_CRIT("DCCP(%p): option %d(len=%d) not " @@ -384,7 +371,7 @@ int dccp_insert_option_elapsed_time(struct sk_buff *skb, u32 elapsed_time) EXPORT_SYMBOL_GPL(dccp_insert_option_elapsed_time); -int dccp_insert_option_timestamp(struct sk_buff *skb) +static int dccp_insert_option_timestamp(struct sk_buff *skb) { __be32 now = htonl(dccp_timestamp()); /* yes this will overflow but that is the point as we want a @@ -393,8 +380,6 @@ int dccp_insert_option_timestamp(struct sk_buff *skb) return dccp_insert_option(skb, DCCPO_TIMESTAMP, &now, sizeof(now)); } -EXPORT_SYMBOL_GPL(dccp_insert_option_timestamp); - static int dccp_insert_option_timestamp_echo(struct dccp_sock *dp, struct dccp_request_sock *dreq, struct sk_buff *skb) diff --git a/net/dccp/output.c b/net/dccp/output.c index aadbdb58758..a988fe9ffcb 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -304,7 +304,7 @@ void dccp_write_xmit(struct sock *sk, int block) dcb->dccpd_type = DCCP_PKT_DATA; err = dccp_transmit_skb(sk, skb); - ccid_hc_tx_packet_sent(dp->dccps_hc_tx_ccid, sk, 0, len); + ccid_hc_tx_packet_sent(dp->dccps_hc_tx_ccid, sk, len); if (err) DCCP_BUG("err=%d after ccid_hc_tx_packet_sent", err); @@ -474,8 +474,9 @@ int dccp_send_reset(struct sock *sk, enum dccp_reset_codes code) /* * Do all connect socket setups that can be done AF independent. */ -static inline void dccp_connect_init(struct sock *sk) +int dccp_connect(struct sock *sk) { + struct sk_buff *skb; struct dccp_sock *dp = dccp_sk(sk); struct dst_entry *dst = __sk_dst_get(sk); struct inet_connection_sock *icsk = inet_csk(sk); @@ -485,22 +486,12 @@ static inline void dccp_connect_init(struct sock *sk) dccp_sync_mss(sk, dst_mtu(dst)); - /* Initialise GAR as per 8.5; AWL/AWH are set in dccp_transmit_skb() */ - dp->dccps_gar = dp->dccps_iss; - - icsk->icsk_retransmits = 0; -} - -int dccp_connect(struct sock *sk) -{ - struct sk_buff *skb; - struct inet_connection_sock *icsk = inet_csk(sk); - /* do not connect if feature negotiation setup fails */ if (dccp_feat_finalise_settings(dccp_sk(sk))) return -EPROTO; - dccp_connect_init(sk); + /* Initialise GAR as per 8.5; AWL/AWH are set in dccp_transmit_skb() */ + dp->dccps_gar = dp->dccps_iss; skb = alloc_skb(sk->sk_prot->max_header, sk->sk_allocation); if (unlikely(skb == NULL)) @@ -516,6 +507,7 @@ int dccp_connect(struct sock *sk) DCCP_INC_STATS(DCCP_MIB_ACTIVEOPENS); /* Timer for repeating the REQUEST until an answer. */ + icsk->icsk_retransmits = 0; inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, icsk->icsk_rto, DCCP_RTO_MAX); return 0; diff --git a/net/dccp/probe.c b/net/dccp/probe.c index 078e48d442f..33d0e6297c2 100644 --- a/net/dccp/probe.c +++ b/net/dccp/probe.c @@ -149,6 +149,7 @@ static const struct file_operations dccpprobe_fops = { .owner = THIS_MODULE, .open = dccpprobe_open, .read = dccpprobe_read, + .llseek = noop_llseek, }; static __init int dccpprobe_init(void) diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 096250d1323..7e5fc04eb6d 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -50,6 +50,30 @@ EXPORT_SYMBOL_GPL(dccp_hashinfo); /* the maximum queue length for tx in packets. 0 is no limit */ int sysctl_dccp_tx_qlen __read_mostly = 5; +#ifdef CONFIG_IP_DCCP_DEBUG +static const char *dccp_state_name(const int state) +{ + static const char *const dccp_state_names[] = { + [DCCP_OPEN] = "OPEN", + [DCCP_REQUESTING] = "REQUESTING", + [DCCP_PARTOPEN] = "PARTOPEN", + [DCCP_LISTEN] = "LISTEN", + [DCCP_RESPOND] = "RESPOND", + [DCCP_CLOSING] = "CLOSING", + [DCCP_ACTIVE_CLOSEREQ] = "CLOSEREQ", + [DCCP_PASSIVE_CLOSE] = "PASSIVE_CLOSE", + [DCCP_PASSIVE_CLOSEREQ] = "PASSIVE_CLOSEREQ", + [DCCP_TIME_WAIT] = "TIME_WAIT", + [DCCP_CLOSED] = "CLOSED", + }; + + if (state >= DCCP_MAX_STATES) + return "INVALID STATE!"; + else + return dccp_state_names[state]; +} +#endif + void dccp_set_state(struct sock *sk, const int state) { const int oldstate = sk->sk_state; @@ -146,30 +170,6 @@ const char *dccp_packet_name(const int type) EXPORT_SYMBOL_GPL(dccp_packet_name); -const char *dccp_state_name(const int state) -{ - static const char *const dccp_state_names[] = { - [DCCP_OPEN] = "OPEN", - [DCCP_REQUESTING] = "REQUESTING", - [DCCP_PARTOPEN] = "PARTOPEN", - [DCCP_LISTEN] = "LISTEN", - [DCCP_RESPOND] = "RESPOND", - [DCCP_CLOSING] = "CLOSING", - [DCCP_ACTIVE_CLOSEREQ] = "CLOSEREQ", - [DCCP_PASSIVE_CLOSE] = "PASSIVE_CLOSE", - [DCCP_PASSIVE_CLOSEREQ] = "PASSIVE_CLOSEREQ", - [DCCP_TIME_WAIT] = "TIME_WAIT", - [DCCP_CLOSED] = "CLOSED", - }; - - if (state >= DCCP_MAX_STATES) - return "INVALID STATE!"; - else - return dccp_state_names[state]; -} - -EXPORT_SYMBOL_GPL(dccp_state_name); - int dccp_init_sock(struct sock *sk, const __u8 ctl_sock_initialized) { struct dccp_sock *dp = dccp_sk(sk); @@ -944,7 +944,7 @@ void dccp_close(struct sock *sk, long timeout) if (data_was_unread) { /* Unread data was tossed, send an appropriate Reset Code */ - DCCP_WARN("DCCP: ABORT -- %u bytes unread\n", data_was_unread); + DCCP_WARN("ABORT with %u bytes unread\n", data_was_unread); dccp_send_reset(sk, DCCP_RESET_CODE_ABORTED); dccp_set_state(sk, DCCP_CLOSED); } else if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) { diff --git a/net/decnet/dn_neigh.c b/net/decnet/dn_neigh.c index 0363bb95cc7..a085dbcf5c7 100644 --- a/net/decnet/dn_neigh.c +++ b/net/decnet/dn_neigh.c @@ -48,7 +48,6 @@ #include <net/dn_neigh.h> #include <net/dn_route.h> -static u32 dn_neigh_hash(const void *pkey, const struct net_device *dev); static int dn_neigh_construct(struct neighbour *); static void dn_long_error_report(struct neighbour *, struct sk_buff *); static void dn_short_error_report(struct neighbour *, struct sk_buff *); @@ -93,6 +92,13 @@ static const struct neigh_ops dn_phase3_ops = { .queue_xmit = dev_queue_xmit }; +static u32 dn_neigh_hash(const void *pkey, + const struct net_device *dev, + __u32 hash_rnd) +{ + return jhash_2words(*(__u16 *)pkey, 0, hash_rnd); +} + struct neigh_table dn_neigh_table = { .family = PF_DECnet, .entry_size = sizeof(struct dn_neigh), @@ -122,11 +128,6 @@ struct neigh_table dn_neigh_table = { .gc_thresh3 = 1024, }; -static u32 dn_neigh_hash(const void *pkey, const struct net_device *dev) -{ - return jhash_2words(*(__u16 *)pkey, 0, dn_neigh_table.hash_rnd); -} - static int dn_neigh_construct(struct neighbour *neigh) { struct net_device *dev = neigh->dev; diff --git a/net/decnet/dn_nsp_out.c b/net/decnet/dn_nsp_out.c index baeb1eaf011..2ef115277be 100644 --- a/net/decnet/dn_nsp_out.c +++ b/net/decnet/dn_nsp_out.c @@ -693,22 +693,22 @@ void dn_nsp_send_conninit(struct sock *sk, unsigned char msgflg) aux = scp->accessdata.acc_userl; *skb_put(skb, 1) = aux; if (aux > 0) - memcpy(skb_put(skb, aux), scp->accessdata.acc_user, aux); + memcpy(skb_put(skb, aux), scp->accessdata.acc_user, aux); aux = scp->accessdata.acc_passl; *skb_put(skb, 1) = aux; if (aux > 0) - memcpy(skb_put(skb, aux), scp->accessdata.acc_pass, aux); + memcpy(skb_put(skb, aux), scp->accessdata.acc_pass, aux); aux = scp->accessdata.acc_accl; *skb_put(skb, 1) = aux; if (aux > 0) - memcpy(skb_put(skb, aux), scp->accessdata.acc_acc, aux); + memcpy(skb_put(skb, aux), scp->accessdata.acc_acc, aux); aux = (__u8)le16_to_cpu(scp->conndata_out.opt_optl); *skb_put(skb, 1) = aux; if (aux > 0) - memcpy(skb_put(skb,aux), scp->conndata_out.opt_data, aux); + memcpy(skb_put(skb, aux), scp->conndata_out.opt_data, aux); scp->persist = dn_nsp_persist(sk); scp->persist_fxn = dn_nsp_retrans_conninit; diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 6585ea6d118..df0f3e54ff8 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -132,7 +132,6 @@ static struct dst_ops dn_dst_ops = { .negative_advice = dn_dst_negative_advice, .link_failure = dn_dst_link_failure, .update_pmtu = dn_dst_update_pmtu, - .entries = ATOMIC_INIT(0), }; static __inline__ unsigned dn_hash(__le16 src, __le16 dst) @@ -1758,6 +1757,7 @@ void __init dn_route_init(void) dn_dst_ops.kmem_cachep = kmem_cache_create("dn_dst_cache", sizeof(struct dn_route), 0, SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL); + dst_entries_init(&dn_dst_ops); setup_timer(&dn_route_timer, dn_dst_check_expire, 0); dn_route_timer.expires = jiffies + decnet_dst_gc_interval * HZ; add_timer(&dn_route_timer); @@ -1816,5 +1816,6 @@ void __exit dn_route_cleanup(void) dn_run_flush(0); proc_net_remove(&init_net, "decnet_cache"); + dst_entries_destroy(&dn_dst_ops); } diff --git a/net/dns_resolver/dns_key.c b/net/dns_resolver/dns_key.c index 400a04d5c9a..739435a6af3 100644 --- a/net/dns_resolver/dns_key.c +++ b/net/dns_resolver/dns_key.c @@ -29,6 +29,7 @@ #include <linux/kernel.h> #include <linux/keyctl.h> #include <linux/err.h> +#include <linux/seq_file.h> #include <keys/dns_resolver-type.h> #include <keys/user-type.h> #include "internal.h" @@ -43,6 +44,8 @@ MODULE_PARM_DESC(debug, "DNS Resolver debugging mask"); const struct cred *dns_resolver_cache; +#define DNS_ERRORNO_OPTION "dnserror" + /* * Instantiate a user defined key for dns_resolver. * @@ -59,9 +62,10 @@ static int dns_resolver_instantiate(struct key *key, const void *_data, size_t datalen) { struct user_key_payload *upayload; + unsigned long derrno; int ret; size_t result_len = 0; - const char *data = _data, *opt; + const char *data = _data, *end, *opt; kenter("%%%d,%s,'%s',%zu", key->serial, key->description, data, datalen); @@ -71,13 +75,77 @@ dns_resolver_instantiate(struct key *key, const void *_data, size_t datalen) datalen--; /* deal with any options embedded in the data */ + end = data + datalen; opt = memchr(data, '#', datalen); if (!opt) { - kdebug("no options currently supported"); - return -EINVAL; + /* no options: the entire data is the result */ + kdebug("no options"); + result_len = datalen; + } else { + const char *next_opt; + + result_len = opt - data; + opt++; + kdebug("options: '%s'", opt); + do { + const char *eq; + int opt_len, opt_nlen, opt_vlen, tmp; + + next_opt = memchr(opt, '#', end - opt) ?: end; + opt_len = next_opt - opt; + if (!opt_len) { + printk(KERN_WARNING + "Empty option to dns_resolver key %d\n", + key->serial); + return -EINVAL; + } + + eq = memchr(opt, '=', opt_len) ?: end; + opt_nlen = eq - opt; + eq++; + opt_vlen = next_opt - eq; /* will be -1 if no value */ + + tmp = opt_vlen >= 0 ? opt_vlen : 0; + kdebug("option '%*.*s' val '%*.*s'", + opt_nlen, opt_nlen, opt, tmp, tmp, eq); + + /* see if it's an error number representing a DNS error + * that's to be recorded as the result in this key */ + if (opt_nlen == sizeof(DNS_ERRORNO_OPTION) - 1 && + memcmp(opt, DNS_ERRORNO_OPTION, opt_nlen) == 0) { + kdebug("dns error number option"); + if (opt_vlen <= 0) + goto bad_option_value; + + ret = strict_strtoul(eq, 10, &derrno); + if (ret < 0) + goto bad_option_value; + + if (derrno < 1 || derrno > 511) + goto bad_option_value; + + kdebug("dns error no. = %lu", derrno); + key->type_data.x[0] = -derrno; + continue; + } + + bad_option_value: + printk(KERN_WARNING + "Option '%*.*s' to dns_resolver key %d:" + " bad/missing value\n", + opt_nlen, opt_nlen, opt, key->serial); + return -EINVAL; + } while (opt = next_opt + 1, opt < end); + } + + /* don't cache the result if we're caching an error saying there's no + * result */ + if (key->type_data.x[0]) { + kleave(" = 0 [h_error %ld]", key->type_data.x[0]); + return 0; } - result_len = datalen; + kdebug("store result"); ret = key_payload_reserve(key, result_len); if (ret < 0) return -EINVAL; @@ -135,13 +203,27 @@ no_match: return ret; } +/* + * Describe a DNS key + */ +static void dns_resolver_describe(const struct key *key, struct seq_file *m) +{ + int err = key->type_data.x[0]; + + seq_puts(m, key->description); + if (err) + seq_printf(m, ": %d", err); + else + seq_printf(m, ": %u", key->datalen); +} + struct key_type key_type_dns_resolver = { .name = "dns_resolver", .instantiate = dns_resolver_instantiate, .match = dns_resolver_match, .revoke = user_revoke, .destroy = user_destroy, - .describe = user_describe, + .describe = dns_resolver_describe, .read = user_read, }; diff --git a/net/dns_resolver/dns_query.c b/net/dns_resolver/dns_query.c index 03d5255f5cf..c32be292c7e 100644 --- a/net/dns_resolver/dns_query.c +++ b/net/dns_resolver/dns_query.c @@ -136,6 +136,11 @@ int dns_query(const char *type, const char *name, size_t namelen, if (ret < 0) goto put; + /* If the DNS server gave an error, return that to the caller */ + ret = rkey->type_data.x[0]; + if (ret) + goto put; + upayload = rcu_dereference_protected(rkey->payload.data, lockdep_is_held(&rkey->sem)); len = upayload->datalen; diff --git a/net/dsa/Kconfig b/net/dsa/Kconfig index 11201784d29..87bb5f4de0e 100644 --- a/net/dsa/Kconfig +++ b/net/dsa/Kconfig @@ -1,7 +1,7 @@ menuconfig NET_DSA bool "Distributed Switch Architecture support" default n - depends on EXPERIMENTAL && NET_ETHERNET && !S390 + depends on EXPERIMENTAL && NETDEVICES && !S390 select PHYLIB ---help--- This allows you to use hardware switch chips that use diff --git a/net/econet/af_econet.c b/net/econet/af_econet.c index dc54bd0d083..f8c1ae4b41f 100644 --- a/net/econet/af_econet.c +++ b/net/econet/af_econet.c @@ -392,7 +392,7 @@ static int econet_sendmsg(struct kiocb *iocb, struct socket *sock, dev_queue_xmit(skb); dev_put(dev); mutex_unlock(&econet_mutex); - return(len); + return len; out_free: kfree_skb(skb); @@ -637,7 +637,7 @@ static int econet_create(struct net *net, struct socket *sock, int protocol, eo->num = protocol; econet_insert_socket(&econet_sklist, sk); - return(0); + return 0; out: return err; } @@ -1009,7 +1009,6 @@ static int __init aun_udp_initialise(void) struct sockaddr_in sin; skb_queue_head_init(&aun_queue); - spin_lock_init(&aun_queue_lock); setup_timer(&ab_cleanup_timer, ab_cleanup, 0); ab_cleanup_timer.expires = jiffies + (HZ*2); add_timer(&ab_cleanup_timer); @@ -1167,7 +1166,6 @@ static int __init econet_proto_init(void) goto out; sock_register(&econet_family_ops); #ifdef CONFIG_ECONET_AUNUDP - spin_lock_init(&aun_queue_lock); aun_udp_initialise(); #endif #ifdef CONFIG_ECONET_NATIVE diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c index 215c83986a9..f00ef2f1d81 100644 --- a/net/ethernet/eth.c +++ b/net/ethernet/eth.c @@ -367,7 +367,7 @@ struct net_device *alloc_etherdev_mq(int sizeof_priv, unsigned int queue_count) EXPORT_SYMBOL(alloc_etherdev_mq); static size_t _format_mac_addr(char *buf, int buflen, - const unsigned char *addr, int len) + const unsigned char *addr, int len) { int i; char *cp = buf; @@ -376,7 +376,7 @@ static size_t _format_mac_addr(char *buf, int buflen, cp += scnprintf(cp, buflen - (cp - buf), "%02x", addr[i]); if (i == len - 1) break; - cp += strlcpy(cp, ":", buflen - (cp - buf)); + cp += scnprintf(cp, buflen - (cp - buf), ":"); } return cp - buf; } @@ -386,7 +386,7 @@ ssize_t sysfs_format_mac(char *buf, const unsigned char *addr, int len) size_t l; l = _format_mac_addr(buf, PAGE_SIZE, addr, len); - l += strlcpy(buf + l, "\n", PAGE_SIZE - l); - return ((ssize_t) l); + l += scnprintf(buf + l, PAGE_SIZE - l, "\n"); + return (ssize_t)l; } EXPORT_SYMBOL(sysfs_format_mac); diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index 1cc7ef270d5..9e95d7fb6d5 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -46,7 +46,7 @@ config IP_ADVANCED_ROUTER rp_filter on use: echo 1 > /proc/sys/net/ipv4/conf/<device>/rp_filter - and + or echo 1 > /proc/sys/net/ipv4/conf/all/rp_filter Note that some distributions enable it in startup scripts. @@ -215,8 +215,15 @@ config NET_IPIP be inserted in and removed from the running kernel whenever you want). Most people won't need this and can say N. +config NET_IPGRE_DEMUX + tristate "IP: GRE demultiplexer" + help + This is helper module to demultiplex GRE packets on GRE version field criteria. + Required by ip_gre and pptp modules. + config NET_IPGRE tristate "IP: GRE tunnels over IP" + depends on (IPV6 || IPV6=n) && NET_IPGRE_DEMUX help Tunneling means encapsulating data of one protocol type within another protocol and sending it over a channel that understands the @@ -412,7 +419,7 @@ config INET_XFRM_MODE_BEET If unsure, say Y. config INET_LRO - bool "Large Receive Offload (ipv4/tcp)" + tristate "Large Receive Offload (ipv4/tcp)" default y ---help--- Support for Large Receive Offload (ipv4/tcp). diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index 80ff87ce43a..4978d22f9a7 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -20,6 +20,7 @@ obj-$(CONFIG_PROC_FS) += proc.o obj-$(CONFIG_IP_MULTIPLE_TABLES) += fib_rules.o obj-$(CONFIG_IP_MROUTE) += ipmr.o obj-$(CONFIG_NET_IPIP) += ipip.o +obj-$(CONFIG_NET_IPGRE_DEMUX) += gre.o obj-$(CONFIG_NET_IPGRE) += ip_gre.o obj-$(CONFIG_SYN_COOKIES) += syncookies.o obj-$(CONFIG_INET_AH) += ah4.o diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 6a1100c25a9..f581f77d109 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -227,18 +227,16 @@ EXPORT_SYMBOL(inet_ehash_secret); /* * inet_ehash_secret must be set exactly once - * Instead of using a dedicated spinlock, we (ab)use inetsw_lock */ void build_ehash_secret(void) { u32 rnd; + do { get_random_bytes(&rnd, sizeof(rnd)); } while (rnd == 0); - spin_lock_bh(&inetsw_lock); - if (!inet_ehash_secret) - inet_ehash_secret = rnd; - spin_unlock_bh(&inetsw_lock); + + cmpxchg(&inet_ehash_secret, 0, rnd); } EXPORT_SYMBOL(build_ehash_secret); diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 96c1955b3e2..d8e540c5b07 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -55,7 +55,7 @@ * Stuart Cheshire : Metricom and grat arp fixes * *** FOR 2.1 clean this up *** * Lawrence V. Stefani: (08/12/96) Added FDDI support. - * Alan Cox : Took the AP1000 nasty FDDI hack and + * Alan Cox : Took the AP1000 nasty FDDI hack and * folded into the mainstream FDDI code. * Ack spit, Linus how did you allow that * one in... @@ -120,14 +120,14 @@ EXPORT_SYMBOL(clip_tbl_hook); #endif #include <asm/system.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> #include <linux/netfilter_arp.h> /* * Interface to generic neighbour cache. */ -static u32 arp_hash(const void *pkey, const struct net_device *dev); +static u32 arp_hash(const void *pkey, const struct net_device *dev, __u32 rnd); static int arp_constructor(struct neighbour *neigh); static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb); static void arp_error_report(struct neighbour *neigh, struct sk_buff *skb); @@ -161,7 +161,7 @@ static const struct neigh_ops arp_direct_ops = { .queue_xmit = dev_queue_xmit, }; -const struct neigh_ops arp_broken_ops = { +static const struct neigh_ops arp_broken_ops = { .family = AF_INET, .solicit = arp_solicit, .error_report = arp_error_report, @@ -170,35 +170,34 @@ const struct neigh_ops arp_broken_ops = { .hh_output = dev_queue_xmit, .queue_xmit = dev_queue_xmit, }; -EXPORT_SYMBOL(arp_broken_ops); struct neigh_table arp_tbl = { - .family = AF_INET, - .entry_size = sizeof(struct neighbour) + 4, - .key_len = 4, - .hash = arp_hash, - .constructor = arp_constructor, - .proxy_redo = parp_redo, - .id = "arp_cache", - .parms = { - .tbl = &arp_tbl, - .base_reachable_time = 30 * HZ, - .retrans_time = 1 * HZ, - .gc_staletime = 60 * HZ, - .reachable_time = 30 * HZ, - .delay_probe_time = 5 * HZ, - .queue_len = 3, - .ucast_probes = 3, - .mcast_probes = 3, - .anycast_delay = 1 * HZ, - .proxy_delay = (8 * HZ) / 10, - .proxy_qlen = 64, - .locktime = 1 * HZ, + .family = AF_INET, + .entry_size = sizeof(struct neighbour) + 4, + .key_len = 4, + .hash = arp_hash, + .constructor = arp_constructor, + .proxy_redo = parp_redo, + .id = "arp_cache", + .parms = { + .tbl = &arp_tbl, + .base_reachable_time = 30 * HZ, + .retrans_time = 1 * HZ, + .gc_staletime = 60 * HZ, + .reachable_time = 30 * HZ, + .delay_probe_time = 5 * HZ, + .queue_len = 3, + .ucast_probes = 3, + .mcast_probes = 3, + .anycast_delay = 1 * HZ, + .proxy_delay = (8 * HZ) / 10, + .proxy_qlen = 64, + .locktime = 1 * HZ, }, - .gc_interval = 30 * HZ, - .gc_thresh1 = 128, - .gc_thresh2 = 512, - .gc_thresh3 = 1024, + .gc_interval = 30 * HZ, + .gc_thresh1 = 128, + .gc_thresh2 = 512, + .gc_thresh3 = 1024, }; EXPORT_SYMBOL(arp_tbl); @@ -226,14 +225,16 @@ int arp_mc_map(__be32 addr, u8 *haddr, struct net_device *dev, int dir) } -static u32 arp_hash(const void *pkey, const struct net_device *dev) +static u32 arp_hash(const void *pkey, + const struct net_device *dev, + __u32 hash_rnd) { - return jhash_2words(*(u32 *)pkey, dev->ifindex, arp_tbl.hash_rnd); + return jhash_2words(*(u32 *)pkey, dev->ifindex, hash_rnd); } static int arp_constructor(struct neighbour *neigh) { - __be32 addr = *(__be32*)neigh->primary_key; + __be32 addr = *(__be32 *)neigh->primary_key; struct net_device *dev = neigh->dev; struct in_device *in_dev; struct neigh_parms *parms; @@ -296,16 +297,19 @@ static int arp_constructor(struct neighbour *neigh) neigh->ops = &arp_broken_ops; neigh->output = neigh->ops->output; return 0; +#else + break; #endif - ;} + } #endif if (neigh->type == RTN_MULTICAST) { neigh->nud_state = NUD_NOARP; arp_mc_map(addr, neigh->ha, dev, 1); - } else if (dev->flags&(IFF_NOARP|IFF_LOOPBACK)) { + } else if (dev->flags & (IFF_NOARP | IFF_LOOPBACK)) { neigh->nud_state = NUD_NOARP; memcpy(neigh->ha, dev->dev_addr, dev->addr_len); - } else if (neigh->type == RTN_BROADCAST || dev->flags&IFF_POINTOPOINT) { + } else if (neigh->type == RTN_BROADCAST || + (dev->flags & IFF_POINTOPOINT)) { neigh->nud_state = NUD_NOARP; memcpy(neigh->ha, dev->broadcast, dev->addr_len); } @@ -315,7 +319,7 @@ static int arp_constructor(struct neighbour *neigh) else neigh->ops = &arp_generic_ops; - if (neigh->nud_state&NUD_VALID) + if (neigh->nud_state & NUD_VALID) neigh->output = neigh->ops->connected_output; else neigh->output = neigh->ops->output; @@ -334,7 +338,7 @@ static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb) __be32 saddr = 0; u8 *dst_ha = NULL; struct net_device *dev = neigh->dev; - __be32 target = *(__be32*)neigh->primary_key; + __be32 target = *(__be32 *)neigh->primary_key; int probes = atomic_read(&neigh->probes); struct in_device *in_dev; @@ -347,7 +351,8 @@ static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb) switch (IN_DEV_ARP_ANNOUNCE(in_dev)) { default: case 0: /* By default announce any local IP */ - if (skb && inet_addr_type(dev_net(dev), ip_hdr(skb)->saddr) == RTN_LOCAL) + if (skb && inet_addr_type(dev_net(dev), + ip_hdr(skb)->saddr) == RTN_LOCAL) saddr = ip_hdr(skb)->saddr; break; case 1: /* Restrict announcements of saddr in same subnet */ @@ -369,16 +374,21 @@ static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb) if (!saddr) saddr = inet_select_addr(dev, target, RT_SCOPE_LINK); - if ((probes -= neigh->parms->ucast_probes) < 0) { - if (!(neigh->nud_state&NUD_VALID)) - printk(KERN_DEBUG "trying to ucast probe in NUD_INVALID\n"); + probes -= neigh->parms->ucast_probes; + if (probes < 0) { + if (!(neigh->nud_state & NUD_VALID)) + printk(KERN_DEBUG + "trying to ucast probe in NUD_INVALID\n"); dst_ha = neigh->ha; read_lock_bh(&neigh->lock); - } else if ((probes -= neigh->parms->app_probes) < 0) { + } else { + probes -= neigh->parms->app_probes; + if (probes < 0) { #ifdef CONFIG_ARPD - neigh_app_ns(neigh); + neigh_app_ns(neigh); #endif - return; + return; + } } arp_send(ARPOP_REQUEST, ETH_P_ARP, target, dev, saddr, @@ -451,7 +461,8 @@ static int arp_filter(__be32 sip, __be32 tip, struct net_device *dev) * is allowed to use this function, it is scheduled to be removed. --ANK */ -static int arp_set_predefined(int addr_hint, unsigned char * haddr, __be32 paddr, struct net_device * dev) +static int arp_set_predefined(int addr_hint, unsigned char *haddr, + __be32 paddr, struct net_device *dev) { switch (addr_hint) { case RTN_LOCAL: @@ -483,17 +494,16 @@ int arp_find(unsigned char *haddr, struct sk_buff *skb) paddr = skb_rtable(skb)->rt_gateway; - if (arp_set_predefined(inet_addr_type(dev_net(dev), paddr), haddr, paddr, dev)) + if (arp_set_predefined(inet_addr_type(dev_net(dev), paddr), haddr, + paddr, dev)) return 0; n = __neigh_lookup(&arp_tbl, &paddr, dev, 1); if (n) { n->used = jiffies; - if (n->nud_state&NUD_VALID || neigh_event_send(n, skb) == 0) { - read_lock_bh(&n->lock); - memcpy(haddr, n->ha, dev->addr_len); - read_unlock_bh(&n->lock); + if (n->nud_state & NUD_VALID || neigh_event_send(n, skb) == 0) { + neigh_ha_snapshot(haddr, n, dev); neigh_release(n); return 0; } @@ -515,13 +525,14 @@ int arp_bind_neighbour(struct dst_entry *dst) return -EINVAL; if (n == NULL) { __be32 nexthop = ((struct rtable *)dst)->rt_gateway; - if (dev->flags&(IFF_LOOPBACK|IFF_POINTOPOINT)) + if (dev->flags & (IFF_LOOPBACK | IFF_POINTOPOINT)) nexthop = 0; n = __neigh_lookup_errno( #if defined(CONFIG_ATM_CLIP) || defined(CONFIG_ATM_CLIP_MODULE) - dev->type == ARPHRD_ATM ? clip_tbl_hook : + dev->type == ARPHRD_ATM ? + clip_tbl_hook : #endif - &arp_tbl, &nexthop, dev); + &arp_tbl, &nexthop, dev); if (IS_ERR(n)) return PTR_ERR(n); dst->neighbour = n; @@ -543,8 +554,8 @@ static inline int arp_fwd_proxy(struct in_device *in_dev, if (!IN_DEV_PROXY_ARP(in_dev)) return 0; - - if ((imi = IN_DEV_MEDIUM_ID(in_dev)) == 0) + imi = IN_DEV_MEDIUM_ID(in_dev); + if (imi == 0) return 1; if (imi == -1) return 0; @@ -555,7 +566,7 @@ static inline int arp_fwd_proxy(struct in_device *in_dev, if (out_dev) omi = IN_DEV_MEDIUM_ID(out_dev); - return (omi != imi && omi != -1); + return omi != imi && omi != -1; } /* @@ -685,7 +696,7 @@ struct sk_buff *arp_create(int type, int ptype, __be32 dest_ip, arp->ar_pln = 4; arp->ar_op = htons(type); - arp_ptr=(unsigned char *)(arp+1); + arp_ptr = (unsigned char *)(arp + 1); memcpy(arp_ptr, src_hw, dev->addr_len); arp_ptr += dev->addr_len; @@ -735,9 +746,8 @@ void arp_send(int type, int ptype, __be32 dest_ip, skb = arp_create(type, ptype, dest_ip, dev, src_ip, dest_hw, src_hw, target_hw); - if (skb == NULL) { + if (skb == NULL) return; - } arp_xmit(skb); } @@ -815,7 +825,7 @@ static int arp_process(struct sk_buff *skb) /* * Extract fields */ - arp_ptr= (unsigned char *)(arp+1); + arp_ptr = (unsigned char *)(arp + 1); sha = arp_ptr; arp_ptr += dev->addr_len; memcpy(&sip, arp_ptr, 4); @@ -869,16 +879,17 @@ static int arp_process(struct sk_buff *skb) addr_type = rt->rt_type; if (addr_type == RTN_LOCAL) { - int dont_send = 0; + int dont_send; - if (!dont_send) - dont_send |= arp_ignore(in_dev,sip,tip); + dont_send = arp_ignore(in_dev, sip, tip); if (!dont_send && IN_DEV_ARPFILTER(in_dev)) - dont_send |= arp_filter(sip,tip,dev); + dont_send |= arp_filter(sip, tip, dev); if (!dont_send) { n = neigh_event_ns(&arp_tbl, sha, &sip, dev); if (n) { - arp_send(ARPOP_REPLY,ETH_P_ARP,sip,dev,tip,sha,dev->dev_addr,sha); + arp_send(ARPOP_REPLY, ETH_P_ARP, sip, + dev, tip, sha, dev->dev_addr, + sha); neigh_release(n); } } @@ -887,8 +898,7 @@ static int arp_process(struct sk_buff *skb) if (addr_type == RTN_UNICAST && (arp_fwd_proxy(in_dev, dev, rt) || arp_fwd_pvlan(in_dev, dev, rt, sip, tip) || - pneigh_lookup(&arp_tbl, net, &tip, dev, 0))) - { + pneigh_lookup(&arp_tbl, net, &tip, dev, 0))) { n = neigh_event_ns(&arp_tbl, sha, &sip, dev); if (n) neigh_release(n); @@ -896,9 +906,12 @@ static int arp_process(struct sk_buff *skb) if (NEIGH_CB(skb)->flags & LOCALLY_ENQUEUED || skb->pkt_type == PACKET_HOST || in_dev->arp_parms->proxy_delay == 0) { - arp_send(ARPOP_REPLY,ETH_P_ARP,sip,dev,tip,sha,dev->dev_addr,sha); + arp_send(ARPOP_REPLY, ETH_P_ARP, sip, + dev, tip, sha, dev->dev_addr, + sha); } else { - pneigh_enqueue(&arp_tbl, in_dev->arp_parms, skb); + pneigh_enqueue(&arp_tbl, + in_dev->arp_parms, skb); return 0; } goto out; @@ -939,7 +952,8 @@ static int arp_process(struct sk_buff *skb) if (arp->ar_op != htons(ARPOP_REPLY) || skb->pkt_type != PACKET_HOST) state = NUD_STALE; - neigh_update(n, sha, state, override ? NEIGH_UPDATE_F_OVERRIDE : 0); + neigh_update(n, sha, state, + override ? NEIGH_UPDATE_F_OVERRIDE : 0); neigh_release(n); } @@ -975,7 +989,8 @@ static int arp_rcv(struct sk_buff *skb, struct net_device *dev, arp->ar_pln != 4) goto freeskb; - if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) + skb = skb_share_check(skb, GFP_ATOMIC); + if (skb == NULL) goto out_of_mem; memset(NEIGH_CB(skb), 0, sizeof(struct neighbour_cb)); @@ -1019,7 +1034,7 @@ static int arp_req_set_public(struct net *net, struct arpreq *r, return -EINVAL; if (!dev && (r->arp_flags & ATF_COM)) { dev = dev_getbyhwaddr(net, r->arp_ha.sa_family, - r->arp_ha.sa_data); + r->arp_ha.sa_data); if (!dev) return -ENODEV; } @@ -1033,7 +1048,7 @@ static int arp_req_set_public(struct net *net, struct arpreq *r, } static int arp_req_set(struct net *net, struct arpreq *r, - struct net_device * dev) + struct net_device *dev) { __be32 ip; struct neighbour *neigh; @@ -1046,10 +1061,11 @@ static int arp_req_set(struct net *net, struct arpreq *r, if (r->arp_flags & ATF_PERM) r->arp_flags |= ATF_COM; if (dev == NULL) { - struct flowi fl = { .nl_u = { .ip4_u = { .daddr = ip, - .tos = RTO_ONLINK } } }; - struct rtable * rt; - if ((err = ip_route_output_key(net, &rt, &fl)) != 0) + struct flowi fl = { .nl_u.ip4_u = { .daddr = ip, + .tos = RTO_ONLINK } }; + struct rtable *rt; + err = ip_route_output_key(net, &rt, &fl); + if (err != 0) return err; dev = rt->dst.dev; ip_rt_put(rt); @@ -1083,9 +1099,9 @@ static int arp_req_set(struct net *net, struct arpreq *r, unsigned state = NUD_STALE; if (r->arp_flags & ATF_PERM) state = NUD_PERMANENT; - err = neigh_update(neigh, (r->arp_flags&ATF_COM) ? + err = neigh_update(neigh, (r->arp_flags & ATF_COM) ? r->arp_ha.sa_data : NULL, state, - NEIGH_UPDATE_F_OVERRIDE| + NEIGH_UPDATE_F_OVERRIDE | NEIGH_UPDATE_F_ADMIN); neigh_release(neigh); } @@ -1094,12 +1110,12 @@ static int arp_req_set(struct net *net, struct arpreq *r, static unsigned arp_state_to_flags(struct neighbour *neigh) { - unsigned flags = 0; if (neigh->nud_state&NUD_PERMANENT) - flags = ATF_PERM|ATF_COM; + return ATF_PERM | ATF_COM; else if (neigh->nud_state&NUD_VALID) - flags = ATF_COM; - return flags; + return ATF_COM; + else + return 0; } /* @@ -1142,7 +1158,7 @@ static int arp_req_delete_public(struct net *net, struct arpreq *r, } static int arp_req_delete(struct net *net, struct arpreq *r, - struct net_device * dev) + struct net_device *dev) { int err; __be32 ip; @@ -1153,10 +1169,11 @@ static int arp_req_delete(struct net *net, struct arpreq *r, ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr; if (dev == NULL) { - struct flowi fl = { .nl_u = { .ip4_u = { .daddr = ip, - .tos = RTO_ONLINK } } }; - struct rtable * rt; - if ((err = ip_route_output_key(net, &rt, &fl)) != 0) + struct flowi fl = { .nl_u.ip4_u = { .daddr = ip, + .tos = RTO_ONLINK } }; + struct rtable *rt; + err = ip_route_output_key(net, &rt, &fl); + if (err != 0) return err; dev = rt->dst.dev; ip_rt_put(rt); @@ -1166,7 +1183,7 @@ static int arp_req_delete(struct net *net, struct arpreq *r, err = -ENXIO; neigh = neigh_lookup(&arp_tbl, &ip, dev); if (neigh) { - if (neigh->nud_state&~NUD_NOARP) + if (neigh->nud_state & ~NUD_NOARP) err = neigh_update(neigh, NULL, NUD_FAILED, NEIGH_UPDATE_F_OVERRIDE| NEIGH_UPDATE_F_ADMIN); @@ -1186,24 +1203,24 @@ int arp_ioctl(struct net *net, unsigned int cmd, void __user *arg) struct net_device *dev = NULL; switch (cmd) { - case SIOCDARP: - case SIOCSARP: - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - case SIOCGARP: - err = copy_from_user(&r, arg, sizeof(struct arpreq)); - if (err) - return -EFAULT; - break; - default: - return -EINVAL; + case SIOCDARP: + case SIOCSARP: + if (!capable(CAP_NET_ADMIN)) + return -EPERM; + case SIOCGARP: + err = copy_from_user(&r, arg, sizeof(struct arpreq)); + if (err) + return -EFAULT; + break; + default: + return -EINVAL; } if (r.arp_pa.sa_family != AF_INET) return -EPFNOSUPPORT; if (!(r.arp_flags & ATF_PUBL) && - (r.arp_flags & (ATF_NETMASK|ATF_DONTPUB))) + (r.arp_flags & (ATF_NETMASK | ATF_DONTPUB))) return -EINVAL; if (!(r.arp_flags & ATF_NETMASK)) ((struct sockaddr_in *)&r.arp_netmask)->sin_addr.s_addr = @@ -1211,7 +1228,8 @@ int arp_ioctl(struct net *net, unsigned int cmd, void __user *arg) rtnl_lock(); if (r.arp_dev[0]) { err = -ENODEV; - if ((dev = __dev_get_by_name(net, r.arp_dev)) == NULL) + dev = __dev_get_by_name(net, r.arp_dev); + if (dev == NULL) goto out; /* Mmmm... It is wrong... ARPHRD_NETROM==0 */ @@ -1243,7 +1261,8 @@ out: return err; } -static int arp_netdev_event(struct notifier_block *this, unsigned long event, void *ptr) +static int arp_netdev_event(struct notifier_block *this, unsigned long event, + void *ptr) { struct net_device *dev = ptr; @@ -1311,12 +1330,13 @@ static char *ax2asc2(ax25_address *a, char *buf) for (n = 0, s = buf; n < 6; n++) { c = (a->ax25_call[n] >> 1) & 0x7F; - if (c != ' ') *s++ = c; + if (c != ' ') + *s++ = c; } *s++ = '-'; - - if ((n = ((a->ax25_call[6] >> 1) & 0x0F)) > 9) { + n = (a->ax25_call[6] >> 1) & 0x0F; + if (n > 9) { *s++ = '1'; n -= 10; } @@ -1325,10 +1345,9 @@ static char *ax2asc2(ax25_address *a, char *buf) *s++ = '\0'; if (*buf == '\0' || *buf == '-') - return "*"; + return "*"; return buf; - } #endif /* CONFIG_AX25 */ @@ -1408,10 +1427,10 @@ static void *arp_seq_start(struct seq_file *seq, loff_t *pos) /* ------------------------------------------------------------------------ */ static const struct seq_operations arp_seq_ops = { - .start = arp_seq_start, - .next = neigh_seq_next, - .stop = neigh_seq_stop, - .show = arp_seq_show, + .start = arp_seq_start, + .next = neigh_seq_next, + .stop = neigh_seq_stop, + .show = arp_seq_show, }; static int arp_seq_open(struct inode *inode, struct file *file) diff --git a/net/ipv4/datagram.c b/net/ipv4/datagram.c index f0550941df7..174be6caa5c 100644 --- a/net/ipv4/datagram.c +++ b/net/ipv4/datagram.c @@ -62,14 +62,17 @@ int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) } if (!inet->inet_saddr) inet->inet_saddr = rt->rt_src; /* Update source address */ - if (!inet->inet_rcv_saddr) + if (!inet->inet_rcv_saddr) { inet->inet_rcv_saddr = rt->rt_src; + if (sk->sk_prot->rehash) + sk->sk_prot->rehash(sk); + } inet->inet_daddr = rt->rt_dst; inet->inet_dport = usin->sin_port; sk->sk_state = TCP_ESTABLISHED; inet->inet_id = jiffies; sk_dst_set(sk, &rt->dst); - return(0); + return 0; } EXPORT_SYMBOL(ip4_datagram_connect); diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index da14c49284f..dc94b0316b7 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -209,7 +209,7 @@ static void inetdev_destroy(struct in_device *in_dev) inet_free_ifa(ifa); } - dev->ip_ptr = NULL; + rcu_assign_pointer(dev->ip_ptr, NULL); devinet_sysctl_unregister(in_dev); neigh_parms_release(&arp_tbl, in_dev->arp_parms); @@ -403,6 +403,9 @@ static int inet_set_ifa(struct net_device *dev, struct in_ifaddr *ifa) return inet_insert_ifa(ifa); } +/* Caller must hold RCU or RTNL : + * We dont take a reference on found in_device + */ struct in_device *inetdev_by_index(struct net *net, int ifindex) { struct net_device *dev; @@ -411,7 +414,7 @@ struct in_device *inetdev_by_index(struct net *net, int ifindex) rcu_read_lock(); dev = dev_get_by_index_rcu(net, ifindex); if (dev) - in_dev = in_dev_get(dev); + in_dev = rcu_dereference_rtnl(dev->ip_ptr); rcu_read_unlock(); return in_dev; } @@ -453,8 +456,6 @@ static int inet_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg goto errout; } - __in_dev_put(in_dev); - for (ifap = &in_dev->ifa_list; (ifa = *ifap) != NULL; ifap = &ifa->ifa_next) { if (tb[IFA_LOCAL] && @@ -1059,7 +1060,7 @@ static int inetdev_event(struct notifier_block *this, unsigned long event, switch (event) { case NETDEV_REGISTER: printk(KERN_DEBUG "inetdev_event: bug\n"); - dev->ip_ptr = NULL; + rcu_assign_pointer(dev->ip_ptr, NULL); break; case NETDEV_UP: if (!inetdev_valid_mtu(dev->mtu)) diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index a4396891835..36e27c2107d 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -147,35 +147,43 @@ static void fib_flush(struct net *net) rt_cache_flush(net, -1); } -/* - * Find the first device with a given source address. +/** + * __ip_dev_find - find the first device with a given source address. + * @net: the net namespace + * @addr: the source address + * @devref: if true, take a reference on the found device + * + * If a caller uses devref=false, it should be protected by RCU, or RTNL */ - -struct net_device * ip_dev_find(struct net *net, __be32 addr) +struct net_device *__ip_dev_find(struct net *net, __be32 addr, bool devref) { - struct flowi fl = { .nl_u = { .ip4_u = { .daddr = addr } } }; - struct fib_result res; + struct flowi fl = { + .nl_u = { + .ip4_u = { + .daddr = addr + } + }, + .flags = FLOWI_FLAG_MATCH_ANY_IIF + }; + struct fib_result res = { 0 }; struct net_device *dev = NULL; - struct fib_table *local_table; - -#ifdef CONFIG_IP_MULTIPLE_TABLES - res.r = NULL; -#endif - local_table = fib_get_table(net, RT_TABLE_LOCAL); - if (!local_table || fib_table_lookup(local_table, &fl, &res)) + rcu_read_lock(); + if (fib_lookup(net, &fl, &res)) { + rcu_read_unlock(); return NULL; + } if (res.type != RTN_LOCAL) goto out; dev = FIB_RES_DEV(res); - if (dev) + if (dev && devref) dev_hold(dev); out: - fib_res_put(&res); + rcu_read_unlock(); return dev; } -EXPORT_SYMBOL(ip_dev_find); +EXPORT_SYMBOL(__ip_dev_find); /* * Find address type as if only "dev" was present in the system. If @@ -202,11 +210,12 @@ static inline unsigned __inet_dev_addr_type(struct net *net, local_table = fib_get_table(net, RT_TABLE_LOCAL); if (local_table) { ret = RTN_UNICAST; - if (!fib_table_lookup(local_table, &fl, &res)) { + rcu_read_lock(); + if (!fib_table_lookup(local_table, &fl, &res, FIB_LOOKUP_NOREF)) { if (!dev || dev == res.fi->fib_dev) ret = res.type; - fib_res_put(&res); } + rcu_read_unlock(); } return ret; } @@ -220,37 +229,41 @@ EXPORT_SYMBOL(inet_addr_type); unsigned int inet_dev_addr_type(struct net *net, const struct net_device *dev, __be32 addr) { - return __inet_dev_addr_type(net, dev, addr); + return __inet_dev_addr_type(net, dev, addr); } EXPORT_SYMBOL(inet_dev_addr_type); /* Given (packet source, input interface) and optional (dst, oif, tos): - - (main) check, that source is valid i.e. not broadcast or our local - address. - - figure out what "logical" interface this packet arrived - and calculate "specific destination" address. - - check, that packet arrived from expected physical interface. + * - (main) check, that source is valid i.e. not broadcast or our local + * address. + * - figure out what "logical" interface this packet arrived + * and calculate "specific destination" address. + * - check, that packet arrived from expected physical interface. + * called with rcu_read_lock() */ - int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif, struct net_device *dev, __be32 *spec_dst, u32 *itag, u32 mark) { struct in_device *in_dev; - struct flowi fl = { .nl_u = { .ip4_u = - { .daddr = src, - .saddr = dst, - .tos = tos } }, - .mark = mark, - .iif = oif }; - + struct flowi fl = { + .nl_u = { + .ip4_u = { + .daddr = src, + .saddr = dst, + .tos = tos + } + }, + .mark = mark, + .iif = oif + }; struct fib_result res; int no_addr, rpf, accept_local; + bool dev_match; int ret; struct net *net; no_addr = rpf = accept_local = 0; - rcu_read_lock(); in_dev = __in_dev_get_rcu(dev); if (in_dev) { no_addr = in_dev->ifa_list == NULL; @@ -259,7 +272,6 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif, if (mark && !IN_DEV_SRC_VMARK(in_dev)) fl.mark = 0; } - rcu_read_unlock(); if (in_dev == NULL) goto e_inval; @@ -269,21 +281,29 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif, goto last_resort; if (res.type != RTN_UNICAST) { if (res.type != RTN_LOCAL || !accept_local) - goto e_inval_res; + goto e_inval; } *spec_dst = FIB_RES_PREFSRC(res); fib_combine_itag(itag, &res); + dev_match = false; + #ifdef CONFIG_IP_ROUTE_MULTIPATH - if (FIB_RES_DEV(res) == dev || res.fi->fib_nhs > 1) + for (ret = 0; ret < res.fi->fib_nhs; ret++) { + struct fib_nh *nh = &res.fi->fib_nh[ret]; + + if (nh->nh_dev == dev) { + dev_match = true; + break; + } + } #else if (FIB_RES_DEV(res) == dev) + dev_match = true; #endif - { + if (dev_match) { ret = FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST; - fib_res_put(&res); return ret; } - fib_res_put(&res); if (no_addr) goto last_resort; if (rpf == 1) @@ -296,7 +316,6 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif, *spec_dst = FIB_RES_PREFSRC(res); ret = FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST; } - fib_res_put(&res); } return ret; @@ -307,8 +326,6 @@ last_resort: *itag = 0; return 0; -e_inval_res: - fib_res_put(&res); e_inval: return -EINVAL; e_rpf: @@ -461,9 +478,9 @@ static int rtentry_to_fib_config(struct net *net, int cmd, struct rtentry *rt, } /* - * Handle IP routing ioctl calls. These are used to manipulate the routing tables + * Handle IP routing ioctl calls. + * These are used to manipulate the routing tables */ - int ip_rt_ioctl(struct net *net, unsigned int cmd, void __user *arg) { struct fib_config cfg; @@ -507,7 +524,7 @@ int ip_rt_ioctl(struct net *net, unsigned int cmd, void __user *arg) return -EINVAL; } -const struct nla_policy rtm_ipv4_policy[RTA_MAX+1] = { +const struct nla_policy rtm_ipv4_policy[RTA_MAX + 1] = { [RTA_DST] = { .type = NLA_U32 }, [RTA_SRC] = { .type = NLA_U32 }, [RTA_IIF] = { .type = NLA_U32 }, @@ -521,7 +538,7 @@ const struct nla_policy rtm_ipv4_policy[RTA_MAX+1] = { }; static int rtm_to_fib_config(struct net *net, struct sk_buff *skb, - struct nlmsghdr *nlh, struct fib_config *cfg) + struct nlmsghdr *nlh, struct fib_config *cfg) { struct nlattr *attr; int err, remaining; @@ -676,12 +693,11 @@ out: } /* Prepare and feed intra-kernel routing request. - Really, it should be netlink message, but :-( netlink - can be not configured, so that we feed it directly - to fib engine. It is legal, because all events occur - only when netlink is already locked. + * Really, it should be netlink message, but :-( netlink + * can be not configured, so that we feed it directly + * to fib engine. It is legal, because all events occur + * only when netlink is already locked. */ - static void fib_magic(int cmd, int type, __be32 dst, int dst_len, struct in_ifaddr *ifa) { struct net *net = dev_net(ifa->ifa_dev->dev); @@ -727,9 +743,9 @@ void fib_add_ifaddr(struct in_ifaddr *ifa) struct in_ifaddr *prim = ifa; __be32 mask = ifa->ifa_mask; __be32 addr = ifa->ifa_local; - __be32 prefix = ifa->ifa_address&mask; + __be32 prefix = ifa->ifa_address & mask; - if (ifa->ifa_flags&IFA_F_SECONDARY) { + if (ifa->ifa_flags & IFA_F_SECONDARY) { prim = inet_ifa_byprefix(in_dev, prefix, mask); if (prim == NULL) { printk(KERN_WARNING "fib_add_ifaddr: bug: prim == NULL\n"); @@ -739,22 +755,24 @@ void fib_add_ifaddr(struct in_ifaddr *ifa) fib_magic(RTM_NEWROUTE, RTN_LOCAL, addr, 32, prim); - if (!(dev->flags&IFF_UP)) + if (!(dev->flags & IFF_UP)) return; /* Add broadcast address, if it is explicitly assigned. */ if (ifa->ifa_broadcast && ifa->ifa_broadcast != htonl(0xFFFFFFFF)) fib_magic(RTM_NEWROUTE, RTN_BROADCAST, ifa->ifa_broadcast, 32, prim); - if (!ipv4_is_zeronet(prefix) && !(ifa->ifa_flags&IFA_F_SECONDARY) && + if (!ipv4_is_zeronet(prefix) && !(ifa->ifa_flags & IFA_F_SECONDARY) && (prefix != addr || ifa->ifa_prefixlen < 32)) { - fib_magic(RTM_NEWROUTE, dev->flags&IFF_LOOPBACK ? RTN_LOCAL : - RTN_UNICAST, prefix, ifa->ifa_prefixlen, prim); + fib_magic(RTM_NEWROUTE, + dev->flags & IFF_LOOPBACK ? RTN_LOCAL : RTN_UNICAST, + prefix, ifa->ifa_prefixlen, prim); /* Add network specific broadcasts, when it takes a sense */ if (ifa->ifa_prefixlen < 31) { fib_magic(RTM_NEWROUTE, RTN_BROADCAST, prefix, 32, prim); - fib_magic(RTM_NEWROUTE, RTN_BROADCAST, prefix|~mask, 32, prim); + fib_magic(RTM_NEWROUTE, RTN_BROADCAST, prefix | ~mask, + 32, prim); } } } @@ -765,17 +783,18 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa) struct net_device *dev = in_dev->dev; struct in_ifaddr *ifa1; struct in_ifaddr *prim = ifa; - __be32 brd = ifa->ifa_address|~ifa->ifa_mask; - __be32 any = ifa->ifa_address&ifa->ifa_mask; + __be32 brd = ifa->ifa_address | ~ifa->ifa_mask; + __be32 any = ifa->ifa_address & ifa->ifa_mask; #define LOCAL_OK 1 #define BRD_OK 2 #define BRD0_OK 4 #define BRD1_OK 8 unsigned ok = 0; - if (!(ifa->ifa_flags&IFA_F_SECONDARY)) - fib_magic(RTM_DELROUTE, dev->flags&IFF_LOOPBACK ? RTN_LOCAL : - RTN_UNICAST, any, ifa->ifa_prefixlen, prim); + if (!(ifa->ifa_flags & IFA_F_SECONDARY)) + fib_magic(RTM_DELROUTE, + dev->flags & IFF_LOOPBACK ? RTN_LOCAL : RTN_UNICAST, + any, ifa->ifa_prefixlen, prim); else { prim = inet_ifa_byprefix(in_dev, any, ifa->ifa_mask); if (prim == NULL) { @@ -785,9 +804,9 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa) } /* Deletion is more complicated than add. - We should take care of not to delete too much :-) - - Scan address list to be sure that addresses are really gone. + * We should take care of not to delete too much :-) + * + * Scan address list to be sure that addresses are really gone. */ for (ifa1 = in_dev->ifa_list; ifa1; ifa1 = ifa1->ifa_next) { @@ -801,23 +820,23 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa) ok |= BRD0_OK; } - if (!(ok&BRD_OK)) + if (!(ok & BRD_OK)) fib_magic(RTM_DELROUTE, RTN_BROADCAST, ifa->ifa_broadcast, 32, prim); - if (!(ok&BRD1_OK)) + if (!(ok & BRD1_OK)) fib_magic(RTM_DELROUTE, RTN_BROADCAST, brd, 32, prim); - if (!(ok&BRD0_OK)) + if (!(ok & BRD0_OK)) fib_magic(RTM_DELROUTE, RTN_BROADCAST, any, 32, prim); - if (!(ok&LOCAL_OK)) { + if (!(ok & LOCAL_OK)) { fib_magic(RTM_DELROUTE, RTN_LOCAL, ifa->ifa_local, 32, prim); /* Check, that this local address finally disappeared. */ if (inet_addr_type(dev_net(dev), ifa->ifa_local) != RTN_LOCAL) { /* And the last, but not the least thing. - We must flush stray FIB entries. - - First of all, we scan fib_info list searching - for stray nexthop entries, then ignite fib_flush. - */ + * We must flush stray FIB entries. + * + * First of all, we scan fib_info list searching + * for stray nexthop entries, then ignite fib_flush. + */ if (fib_sync_down_addr(dev_net(dev), ifa->ifa_local)) fib_flush(dev_net(dev)); } @@ -828,14 +847,20 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa) #undef BRD1_OK } -static void nl_fib_lookup(struct fib_result_nl *frn, struct fib_table *tb ) +static void nl_fib_lookup(struct fib_result_nl *frn, struct fib_table *tb) { struct fib_result res; - struct flowi fl = { .mark = frn->fl_mark, - .nl_u = { .ip4_u = { .daddr = frn->fl_addr, - .tos = frn->fl_tos, - .scope = frn->fl_scope } } }; + struct flowi fl = { + .mark = frn->fl_mark, + .nl_u = { + .ip4_u = { + .daddr = frn->fl_addr, + .tos = frn->fl_tos, + .scope = frn->fl_scope + } + } + }; #ifdef CONFIG_IP_MULTIPLE_TABLES res.r = NULL; @@ -846,15 +871,16 @@ static void nl_fib_lookup(struct fib_result_nl *frn, struct fib_table *tb ) local_bh_disable(); frn->tb_id = tb->tb_id; - frn->err = fib_table_lookup(tb, &fl, &res); + rcu_read_lock(); + frn->err = fib_table_lookup(tb, &fl, &res, FIB_LOOKUP_NOREF); if (!frn->err) { frn->prefixlen = res.prefixlen; frn->nh_sel = res.nh_sel; frn->type = res.type; frn->scope = res.scope; - fib_res_put(&res); } + rcu_read_unlock(); local_bh_enable(); } } @@ -883,8 +909,8 @@ static void nl_fib_input(struct sk_buff *skb) nl_fib_lookup(frn, tb); - pid = NETLINK_CB(skb).pid; /* pid of sending process */ - NETLINK_CB(skb).pid = 0; /* from kernel */ + pid = NETLINK_CB(skb).pid; /* pid of sending process */ + NETLINK_CB(skb).pid = 0; /* from kernel */ NETLINK_CB(skb).dst_group = 0; /* unicast */ netlink_unicast(net->ipv4.fibnl, skb, pid, MSG_DONTWAIT); } @@ -931,7 +957,7 @@ static int fib_inetaddr_event(struct notifier_block *this, unsigned long event, fib_del_ifaddr(ifa); if (ifa->ifa_dev->ifa_list == NULL) { /* Last address was deleted from this interface. - Disable IP. + * Disable IP. */ fib_disable_ip(dev, 1, 0); } else { @@ -990,16 +1016,15 @@ static struct notifier_block fib_netdev_notifier = { static int __net_init ip_fib_net_init(struct net *net) { int err; - unsigned int i; + size_t size = sizeof(struct hlist_head) * FIB_TABLE_HASHSZ; - net->ipv4.fib_table_hash = kzalloc( - sizeof(struct hlist_head)*FIB_TABLE_HASHSZ, GFP_KERNEL); + /* Avoid false sharing : Use at least a full cache line */ + size = max_t(size_t, size, L1_CACHE_BYTES); + + net->ipv4.fib_table_hash = kzalloc(size, GFP_KERNEL); if (net->ipv4.fib_table_hash == NULL) return -ENOMEM; - for (i = 0; i < FIB_TABLE_HASHSZ; i++) - INIT_HLIST_HEAD(&net->ipv4.fib_table_hash[i]); - err = fib4_rules_init(net); if (err < 0) goto fail; diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c index 4ed7e0dea1b..43e1c594ce8 100644 --- a/net/ipv4/fib_hash.c +++ b/net/ipv4/fib_hash.c @@ -54,36 +54,37 @@ struct fib_node { struct fib_alias fn_embedded_alias; }; -struct fn_zone { - struct fn_zone *fz_next; /* Next not empty zone */ - struct hlist_head *fz_hash; /* Hash table pointer */ - int fz_nent; /* Number of entries */ +#define EMBEDDED_HASH_SIZE (L1_CACHE_BYTES / sizeof(struct hlist_head)) - int fz_divisor; /* Hash divisor */ +struct fn_zone { + struct fn_zone __rcu *fz_next; /* Next not empty zone */ + struct hlist_head __rcu *fz_hash; /* Hash table pointer */ + seqlock_t fz_lock; u32 fz_hashmask; /* (fz_divisor - 1) */ -#define FZ_HASHMASK(fz) ((fz)->fz_hashmask) - int fz_order; /* Zone order */ - __be32 fz_mask; + u8 fz_order; /* Zone order (0..32) */ + u8 fz_revorder; /* 32 - fz_order */ + __be32 fz_mask; /* inet_make_mask(order) */ #define FZ_MASK(fz) ((fz)->fz_mask) -}; -/* NOTE. On fast computers evaluation of fz_hashmask and fz_mask - * can be cheaper than memory lookup, so that FZ_* macros are used. - */ + struct hlist_head fz_embedded_hash[EMBEDDED_HASH_SIZE]; + + int fz_nent; /* Number of entries */ + int fz_divisor; /* Hash size (mask+1) */ +}; struct fn_hash { - struct fn_zone *fn_zones[33]; - struct fn_zone *fn_zone_list; + struct fn_zone *fn_zones[33]; + struct fn_zone __rcu *fn_zone_list; }; static inline u32 fn_hash(__be32 key, struct fn_zone *fz) { - u32 h = ntohl(key)>>(32 - fz->fz_order); + u32 h = ntohl(key) >> fz->fz_revorder; h ^= (h>>20); h ^= (h>>10); h ^= (h>>5); - h &= FZ_HASHMASK(fz); + h &= fz->fz_hashmask; return h; } @@ -92,7 +93,6 @@ static inline __be32 fz_key(__be32 dst, struct fn_zone *fz) return dst & FZ_MASK(fz); } -static DEFINE_RWLOCK(fib_hash_lock); static unsigned int fib_hash_genid; #define FZ_MAX_DIVISOR ((PAGE_SIZE<<MAX_ORDER) / sizeof(struct hlist_head)) @@ -101,12 +101,11 @@ static struct hlist_head *fz_hash_alloc(int divisor) { unsigned long size = divisor * sizeof(struct hlist_head); - if (size <= PAGE_SIZE) { + if (size <= PAGE_SIZE) return kzalloc(size, GFP_KERNEL); - } else { - return (struct hlist_head *) - __get_free_pages(GFP_KERNEL | __GFP_ZERO, get_order(size)); - } + + return (struct hlist_head *) + __get_free_pages(GFP_KERNEL | __GFP_ZERO, get_order(size)); } /* The fib hash lock must be held when this is called. */ @@ -121,12 +120,12 @@ static inline void fn_rebuild_zone(struct fn_zone *fz, struct fib_node *f; hlist_for_each_entry_safe(f, node, n, &old_ht[i], fn_hash) { - struct hlist_head *new_head; + struct hlist_head __rcu *new_head; - hlist_del(&f->fn_hash); + hlist_del_rcu(&f->fn_hash); new_head = &fz->fz_hash[fn_hash(f->fn_key, fz)]; - hlist_add_head(&f->fn_hash, new_head); + hlist_add_head_rcu(&f->fn_hash, new_head); } } } @@ -147,14 +146,14 @@ static void fn_rehash_zone(struct fn_zone *fz) int old_divisor, new_divisor; u32 new_hashmask; - old_divisor = fz->fz_divisor; + new_divisor = old_divisor = fz->fz_divisor; switch (old_divisor) { - case 16: - new_divisor = 256; + case EMBEDDED_HASH_SIZE: + new_divisor *= EMBEDDED_HASH_SIZE; break; - case 256: - new_divisor = 1024; + case EMBEDDED_HASH_SIZE*EMBEDDED_HASH_SIZE: + new_divisor *= (EMBEDDED_HASH_SIZE/2); break; default: if ((old_divisor << 1) > FZ_MAX_DIVISOR) { @@ -175,31 +174,55 @@ static void fn_rehash_zone(struct fn_zone *fz) ht = fz_hash_alloc(new_divisor); if (ht) { - write_lock_bh(&fib_hash_lock); + struct fn_zone nfz; + + memcpy(&nfz, fz, sizeof(nfz)); + + write_seqlock_bh(&fz->fz_lock); old_ht = fz->fz_hash; - fz->fz_hash = ht; + nfz.fz_hash = ht; + nfz.fz_hashmask = new_hashmask; + nfz.fz_divisor = new_divisor; + fn_rebuild_zone(&nfz, old_ht, old_divisor); + fib_hash_genid++; + rcu_assign_pointer(fz->fz_hash, ht); fz->fz_hashmask = new_hashmask; fz->fz_divisor = new_divisor; - fn_rebuild_zone(fz, old_ht, old_divisor); - fib_hash_genid++; - write_unlock_bh(&fib_hash_lock); + write_sequnlock_bh(&fz->fz_lock); - fz_hash_free(old_ht, old_divisor); + if (old_ht != fz->fz_embedded_hash) { + synchronize_rcu(); + fz_hash_free(old_ht, old_divisor); + } } } -static inline void fn_free_node(struct fib_node * f) +static void fn_free_node_rcu(struct rcu_head *head) { + struct fib_node *f = container_of(head, struct fib_node, fn_embedded_alias.rcu); + kmem_cache_free(fn_hash_kmem, f); } +static inline void fn_free_node(struct fib_node *f) +{ + call_rcu(&f->fn_embedded_alias.rcu, fn_free_node_rcu); +} + +static void fn_free_alias_rcu(struct rcu_head *head) +{ + struct fib_alias *fa = container_of(head, struct fib_alias, rcu); + + kmem_cache_free(fn_alias_kmem, fa); +} + static inline void fn_free_alias(struct fib_alias *fa, struct fib_node *f) { fib_release_info(fa->fa_info); if (fa == &f->fn_embedded_alias) fa->fa_info = NULL; else - kmem_cache_free(fn_alias_kmem, fa); + call_rcu(&fa->rcu, fn_free_alias_rcu); } static struct fn_zone * @@ -210,68 +233,71 @@ fn_new_zone(struct fn_hash *table, int z) if (!fz) return NULL; - if (z) { - fz->fz_divisor = 16; - } else { - fz->fz_divisor = 1; - } - fz->fz_hashmask = (fz->fz_divisor - 1); - fz->fz_hash = fz_hash_alloc(fz->fz_divisor); - if (!fz->fz_hash) { - kfree(fz); - return NULL; - } + seqlock_init(&fz->fz_lock); + fz->fz_divisor = z ? EMBEDDED_HASH_SIZE : 1; + fz->fz_hashmask = fz->fz_divisor - 1; + fz->fz_hash = fz->fz_embedded_hash; fz->fz_order = z; + fz->fz_revorder = 32 - z; fz->fz_mask = inet_make_mask(z); /* Find the first not empty zone with more specific mask */ - for (i=z+1; i<=32; i++) + for (i = z + 1; i <= 32; i++) if (table->fn_zones[i]) break; - write_lock_bh(&fib_hash_lock); - if (i>32) { + if (i > 32) { /* No more specific masks, we are the first. */ - fz->fz_next = table->fn_zone_list; - table->fn_zone_list = fz; + rcu_assign_pointer(fz->fz_next, + rtnl_dereference(table->fn_zone_list)); + rcu_assign_pointer(table->fn_zone_list, fz); } else { - fz->fz_next = table->fn_zones[i]->fz_next; - table->fn_zones[i]->fz_next = fz; + rcu_assign_pointer(fz->fz_next, + rtnl_dereference(table->fn_zones[i]->fz_next)); + rcu_assign_pointer(table->fn_zones[i]->fz_next, fz); } table->fn_zones[z] = fz; fib_hash_genid++; - write_unlock_bh(&fib_hash_lock); return fz; } int fib_table_lookup(struct fib_table *tb, - const struct flowi *flp, struct fib_result *res) + const struct flowi *flp, struct fib_result *res, + int fib_flags) { int err; struct fn_zone *fz; struct fn_hash *t = (struct fn_hash *)tb->tb_data; - read_lock(&fib_hash_lock); - for (fz = t->fn_zone_list; fz; fz = fz->fz_next) { - struct hlist_head *head; + rcu_read_lock(); + for (fz = rcu_dereference(t->fn_zone_list); + fz != NULL; + fz = rcu_dereference(fz->fz_next)) { + struct hlist_head __rcu *head; struct hlist_node *node; struct fib_node *f; - __be32 k = fz_key(flp->fl4_dst, fz); + __be32 k; + unsigned int seq; - head = &fz->fz_hash[fn_hash(k, fz)]; - hlist_for_each_entry(f, node, head, fn_hash) { - if (f->fn_key != k) - continue; + do { + seq = read_seqbegin(&fz->fz_lock); + k = fz_key(flp->fl4_dst, fz); + + head = &fz->fz_hash[fn_hash(k, fz)]; + hlist_for_each_entry_rcu(f, node, head, fn_hash) { + if (f->fn_key != k) + continue; - err = fib_semantic_match(&f->fn_alias, + err = fib_semantic_match(&f->fn_alias, flp, res, - fz->fz_order); - if (err <= 0) - goto out; - } + fz->fz_order, fib_flags); + if (err <= 0) + goto out; + } + } while (read_seqretry(&fz->fz_lock, seq)); } err = 1; out: - read_unlock(&fib_hash_lock); + rcu_read_unlock(); return err; } @@ -293,11 +319,11 @@ void fib_table_select_default(struct fib_table *tb, last_resort = NULL; order = -1; - read_lock(&fib_hash_lock); - hlist_for_each_entry(f, node, &fz->fz_hash[0], fn_hash) { + rcu_read_lock(); + hlist_for_each_entry_rcu(f, node, &fz->fz_hash[0], fn_hash) { struct fib_alias *fa; - list_for_each_entry(fa, &f->fn_alias, fa_list) { + list_for_each_entry_rcu(fa, &f->fn_alias, fa_list) { struct fib_info *next_fi = fa->fa_info; if (fa->fa_scope != res->scope || @@ -309,7 +335,8 @@ void fib_table_select_default(struct fib_table *tb, if (!next_fi->fib_nh[0].nh_gw || next_fi->fib_nh[0].nh_scope != RT_SCOPE_LINK) continue; - fa->fa_state |= FA_S_ACCESSED; + + fib_alias_accessed(fa); if (fi == NULL) { if (next_fi != res->fi) @@ -341,7 +368,7 @@ void fib_table_select_default(struct fib_table *tb, fib_result_assign(res, last_resort); tb->tb_default = last_idx; out: - read_unlock(&fib_hash_lock); + rcu_read_unlock(); } /* Insert node F to FZ. */ @@ -349,7 +376,7 @@ static inline void fib_insert_node(struct fn_zone *fz, struct fib_node *f) { struct hlist_head *head = &fz->fz_hash[fn_hash(f->fn_key, fz)]; - hlist_add_head(&f->fn_hash, head); + hlist_add_head_rcu(&f->fn_hash, head); } /* Return the node in FZ matching KEY. */ @@ -359,7 +386,7 @@ static struct fib_node *fib_find_node(struct fn_zone *fz, __be32 key) struct hlist_node *node; struct fib_node *f; - hlist_for_each_entry(f, node, head, fn_hash) { + hlist_for_each_entry_rcu(f, node, head, fn_hash) { if (f->fn_key == key) return f; } @@ -367,6 +394,17 @@ static struct fib_node *fib_find_node(struct fn_zone *fz, __be32 key) return NULL; } + +static struct fib_alias *fib_fast_alloc(struct fib_node *f) +{ + struct fib_alias *fa = &f->fn_embedded_alias; + + if (fa->fa_info != NULL) + fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL); + return fa; +} + +/* Caller must hold RTNL. */ int fib_table_insert(struct fib_table *tb, struct fib_config *cfg) { struct fn_hash *table = (struct fn_hash *) tb->tb_data; @@ -451,7 +489,6 @@ int fib_table_insert(struct fib_table *tb, struct fib_config *cfg) } if (cfg->fc_nlflags & NLM_F_REPLACE) { - struct fib_info *fi_drop; u8 state; fa = fa_first; @@ -460,21 +497,25 @@ int fib_table_insert(struct fib_table *tb, struct fib_config *cfg) err = 0; goto out; } - write_lock_bh(&fib_hash_lock); - fi_drop = fa->fa_info; - fa->fa_info = fi; - fa->fa_type = cfg->fc_type; - fa->fa_scope = cfg->fc_scope; + err = -ENOBUFS; + new_fa = fib_fast_alloc(f); + if (new_fa == NULL) + goto out; + + new_fa->fa_tos = fa->fa_tos; + new_fa->fa_info = fi; + new_fa->fa_type = cfg->fc_type; + new_fa->fa_scope = cfg->fc_scope; state = fa->fa_state; - fa->fa_state &= ~FA_S_ACCESSED; + new_fa->fa_state = state & ~FA_S_ACCESSED; fib_hash_genid++; - write_unlock_bh(&fib_hash_lock); + list_replace_rcu(&fa->fa_list, &new_fa->fa_list); - fib_release_info(fi_drop); + fn_free_alias(fa, f); if (state & FA_S_ACCESSED) rt_cache_flush(cfg->fc_nlinfo.nl_net, -1); - rtmsg_fib(RTM_NEWROUTE, key, fa, cfg->fc_dst_len, tb->tb_id, - &cfg->fc_nlinfo, NLM_F_REPLACE); + rtmsg_fib(RTM_NEWROUTE, key, new_fa, cfg->fc_dst_len, + tb->tb_id, &cfg->fc_nlinfo, NLM_F_REPLACE); return 0; } @@ -506,12 +547,10 @@ int fib_table_insert(struct fib_table *tb, struct fib_config *cfg) f = new_f; } - new_fa = &f->fn_embedded_alias; - if (new_fa->fa_info != NULL) { - new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL); - if (new_fa == NULL) - goto out; - } + new_fa = fib_fast_alloc(f); + if (new_fa == NULL) + goto out; + new_fa->fa_info = fi; new_fa->fa_tos = tos; new_fa->fa_type = cfg->fc_type; @@ -522,13 +561,11 @@ int fib_table_insert(struct fib_table *tb, struct fib_config *cfg) * Insert new entry to the list. */ - write_lock_bh(&fib_hash_lock); if (new_f) fib_insert_node(fz, new_f); - list_add_tail(&new_fa->fa_list, + list_add_tail_rcu(&new_fa->fa_list, (fa ? &fa->fa_list : &f->fn_alias)); fib_hash_genid++; - write_unlock_bh(&fib_hash_lock); if (new_f) fz->fz_nent++; @@ -603,14 +640,12 @@ int fib_table_delete(struct fib_table *tb, struct fib_config *cfg) tb->tb_id, &cfg->fc_nlinfo, 0); kill_fn = 0; - write_lock_bh(&fib_hash_lock); - list_del(&fa->fa_list); + list_del_rcu(&fa->fa_list); if (list_empty(&f->fn_alias)) { - hlist_del(&f->fn_hash); + hlist_del_rcu(&f->fn_hash); kill_fn = 1; } fib_hash_genid++; - write_unlock_bh(&fib_hash_lock); if (fa->fa_state & FA_S_ACCESSED) rt_cache_flush(cfg->fc_nlinfo.nl_net, -1); @@ -641,14 +676,12 @@ static int fn_flush_list(struct fn_zone *fz, int idx) struct fib_info *fi = fa->fa_info; if (fi && (fi->fib_flags&RTNH_F_DEAD)) { - write_lock_bh(&fib_hash_lock); - list_del(&fa->fa_list); + list_del_rcu(&fa->fa_list); if (list_empty(&f->fn_alias)) { - hlist_del(&f->fn_hash); + hlist_del_rcu(&f->fn_hash); kill_f = 1; } fib_hash_genid++; - write_unlock_bh(&fib_hash_lock); fn_free_alias(fa, f); found++; @@ -662,13 +695,16 @@ static int fn_flush_list(struct fn_zone *fz, int idx) return found; } +/* caller must hold RTNL. */ int fib_table_flush(struct fib_table *tb) { struct fn_hash *table = (struct fn_hash *) tb->tb_data; struct fn_zone *fz; int found = 0; - for (fz = table->fn_zone_list; fz; fz = fz->fz_next) { + for (fz = rtnl_dereference(table->fn_zone_list); + fz != NULL; + fz = rtnl_dereference(fz->fz_next)) { int i; for (i = fz->fz_divisor - 1; i >= 0; i--) @@ -690,10 +726,10 @@ fn_hash_dump_bucket(struct sk_buff *skb, struct netlink_callback *cb, s_i = cb->args[4]; i = 0; - hlist_for_each_entry(f, node, head, fn_hash) { + hlist_for_each_entry_rcu(f, node, head, fn_hash) { struct fib_alias *fa; - list_for_each_entry(fa, &f->fn_alias, fa_list) { + list_for_each_entry_rcu(fa, &f->fn_alias, fa_list) { if (i < s_i) goto next; @@ -711,7 +747,7 @@ fn_hash_dump_bucket(struct sk_buff *skb, struct netlink_callback *cb, cb->args[4] = i; return -1; } - next: +next: i++; } } @@ -746,23 +782,26 @@ fn_hash_dump_zone(struct sk_buff *skb, struct netlink_callback *cb, int fib_table_dump(struct fib_table *tb, struct sk_buff *skb, struct netlink_callback *cb) { - int m, s_m; + int m = 0, s_m; struct fn_zone *fz; struct fn_hash *table = (struct fn_hash *)tb->tb_data; s_m = cb->args[2]; - read_lock(&fib_hash_lock); - for (fz = table->fn_zone_list, m=0; fz; fz = fz->fz_next, m++) { - if (m < s_m) continue; + rcu_read_lock(); + for (fz = rcu_dereference(table->fn_zone_list); + fz != NULL; + fz = rcu_dereference(fz->fz_next), m++) { + if (m < s_m) + continue; if (fn_hash_dump_zone(skb, cb, tb, fz) < 0) { cb->args[2] = m; - read_unlock(&fib_hash_lock); + rcu_read_unlock(); return -1; } memset(&cb->args[3], 0, sizeof(cb->args) - 3*sizeof(cb->args[0])); } - read_unlock(&fib_hash_lock); + rcu_read_unlock(); cb->args[2] = m; return skb->len; } @@ -825,8 +864,9 @@ static struct fib_alias *fib_get_first(struct seq_file *seq) iter->genid = fib_hash_genid; iter->valid = 1; - for (iter->zone = table->fn_zone_list; iter->zone; - iter->zone = iter->zone->fz_next) { + for (iter->zone = rcu_dereference(table->fn_zone_list); + iter->zone != NULL; + iter->zone = rcu_dereference(iter->zone->fz_next)) { int maxslot; if (!iter->zone->fz_nent) @@ -911,7 +951,7 @@ static struct fib_alias *fib_get_next(struct seq_file *seq) } } - iter->zone = iter->zone->fz_next; + iter->zone = rcu_dereference(iter->zone->fz_next); if (!iter->zone) goto out; @@ -950,11 +990,11 @@ static struct fib_alias *fib_get_idx(struct seq_file *seq, loff_t pos) } static void *fib_seq_start(struct seq_file *seq, loff_t *pos) - __acquires(fib_hash_lock) + __acquires(RCU) { void *v = NULL; - read_lock(&fib_hash_lock); + rcu_read_lock(); if (fib_get_table(seq_file_net(seq), RT_TABLE_MAIN)) v = *pos ? fib_get_idx(seq, *pos - 1) : SEQ_START_TOKEN; return v; @@ -967,15 +1007,16 @@ static void *fib_seq_next(struct seq_file *seq, void *v, loff_t *pos) } static void fib_seq_stop(struct seq_file *seq, void *v) - __releases(fib_hash_lock) + __releases(RCU) { - read_unlock(&fib_hash_lock); + rcu_read_unlock(); } static unsigned fib_flag_trans(int type, __be32 mask, struct fib_info *fi) { static const unsigned type2flags[RTN_MAX + 1] = { - [7] = RTF_REJECT, [8] = RTF_REJECT, + [7] = RTF_REJECT, + [8] = RTF_REJECT, }; unsigned flags = type2flags[type]; diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h index 637b133973b..a29edf2219c 100644 --- a/net/ipv4/fib_lookup.h +++ b/net/ipv4/fib_lookup.h @@ -12,17 +12,22 @@ struct fib_alias { u8 fa_type; u8 fa_scope; u8 fa_state; -#ifdef CONFIG_IP_FIB_TRIE struct rcu_head rcu; -#endif }; #define FA_S_ACCESSED 0x01 +/* Dont write on fa_state unless needed, to keep it shared on all cpus */ +static inline void fib_alias_accessed(struct fib_alias *fa) +{ + if (!(fa->fa_state & FA_S_ACCESSED)) + fa->fa_state |= FA_S_ACCESSED; +} + /* Exported by fib_semantics.c */ extern int fib_semantic_match(struct list_head *head, const struct flowi *flp, - struct fib_result *res, int prefixlen); + struct fib_result *res, int prefixlen, int fib_flags); extern void fib_release_info(struct fib_info *); extern struct fib_info *fib_create_info(struct fib_config *cfg); extern int fib_nh_match(struct fib_config *cfg, struct fib_info *fi); diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index 76daeb5ff56..7981a24f5c7 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -6,7 +6,7 @@ * IPv4 Forwarding Information Base: policy rules. * * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> - * Thomas Graf <tgraf@suug.ch> + * Thomas Graf <tgraf@suug.ch> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -14,7 +14,7 @@ * 2 of the License, or (at your option) any later version. * * Fixes: - * Rani Assaf : local_rule cannot be deleted + * Rani Assaf : local_rule cannot be deleted * Marc Boucher : routing by fwmark */ @@ -32,8 +32,7 @@ #include <net/ip_fib.h> #include <net/fib_rules.h> -struct fib4_rule -{ +struct fib4_rule { struct fib_rule common; u8 dst_len; u8 src_len; @@ -58,6 +57,7 @@ int fib_lookup(struct net *net, struct flowi *flp, struct fib_result *res) { struct fib_lookup_arg arg = { .result = res, + .flags = FIB_LOOKUP_NOREF, }; int err; @@ -91,10 +91,11 @@ static int fib4_rule_action(struct fib_rule *rule, struct flowi *flp, goto errout; } - if ((tbl = fib_get_table(rule->fr_net, rule->table)) == NULL) + tbl = fib_get_table(rule->fr_net, rule->table); + if (!tbl) goto errout; - err = fib_table_lookup(tbl, flp, (struct fib_result *) arg->result); + err = fib_table_lookup(tbl, flp, (struct fib_result *) arg->result, arg->flags); if (err > 0) err = -EAGAIN; errout: diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 20f09c5b31e..3e0da3ef611 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -60,21 +60,30 @@ static struct hlist_head fib_info_devhash[DEVINDEX_HASHSIZE]; static DEFINE_SPINLOCK(fib_multipath_lock); -#define for_nexthops(fi) { int nhsel; const struct fib_nh * nh; \ -for (nhsel=0, nh = (fi)->fib_nh; nhsel < (fi)->fib_nhs; nh++, nhsel++) - -#define change_nexthops(fi) { int nhsel; struct fib_nh *nexthop_nh; \ -for (nhsel=0, nexthop_nh = (struct fib_nh *)((fi)->fib_nh); nhsel < (fi)->fib_nhs; nexthop_nh++, nhsel++) +#define for_nexthops(fi) { \ + int nhsel; const struct fib_nh *nh; \ + for (nhsel = 0, nh = (fi)->fib_nh; \ + nhsel < (fi)->fib_nhs; \ + nh++, nhsel++) + +#define change_nexthops(fi) { \ + int nhsel; struct fib_nh *nexthop_nh; \ + for (nhsel = 0, nexthop_nh = (struct fib_nh *)((fi)->fib_nh); \ + nhsel < (fi)->fib_nhs; \ + nexthop_nh++, nhsel++) #else /* CONFIG_IP_ROUTE_MULTIPATH */ /* Hope, that gcc will optimize it to get rid of dummy loop */ -#define for_nexthops(fi) { int nhsel = 0; const struct fib_nh * nh = (fi)->fib_nh; \ -for (nhsel=0; nhsel < 1; nhsel++) +#define for_nexthops(fi) { \ + int nhsel; const struct fib_nh *nh = (fi)->fib_nh; \ + for (nhsel = 0; nhsel < 1; nhsel++) -#define change_nexthops(fi) { int nhsel = 0; struct fib_nh *nexthop_nh = (struct fib_nh *)((fi)->fib_nh); \ -for (nhsel=0; nhsel < 1; nhsel++) +#define change_nexthops(fi) { \ + int nhsel; \ + struct fib_nh *nexthop_nh = (struct fib_nh *)((fi)->fib_nh); \ + for (nhsel = 0; nhsel < 1; nhsel++) #endif /* CONFIG_IP_ROUTE_MULTIPATH */ @@ -86,63 +95,70 @@ static const struct int error; u8 scope; } fib_props[RTN_MAX + 1] = { - { + [RTN_UNSPEC] = { .error = 0, .scope = RT_SCOPE_NOWHERE, - }, /* RTN_UNSPEC */ - { + }, + [RTN_UNICAST] = { .error = 0, .scope = RT_SCOPE_UNIVERSE, - }, /* RTN_UNICAST */ - { + }, + [RTN_LOCAL] = { .error = 0, .scope = RT_SCOPE_HOST, - }, /* RTN_LOCAL */ - { + }, + [RTN_BROADCAST] = { .error = 0, .scope = RT_SCOPE_LINK, - }, /* RTN_BROADCAST */ - { + }, + [RTN_ANYCAST] = { .error = 0, .scope = RT_SCOPE_LINK, - }, /* RTN_ANYCAST */ - { + }, + [RTN_MULTICAST] = { .error = 0, .scope = RT_SCOPE_UNIVERSE, - }, /* RTN_MULTICAST */ - { + }, + [RTN_BLACKHOLE] = { .error = -EINVAL, .scope = RT_SCOPE_UNIVERSE, - }, /* RTN_BLACKHOLE */ - { + }, + [RTN_UNREACHABLE] = { .error = -EHOSTUNREACH, .scope = RT_SCOPE_UNIVERSE, - }, /* RTN_UNREACHABLE */ - { + }, + [RTN_PROHIBIT] = { .error = -EACCES, .scope = RT_SCOPE_UNIVERSE, - }, /* RTN_PROHIBIT */ - { + }, + [RTN_THROW] = { .error = -EAGAIN, .scope = RT_SCOPE_UNIVERSE, - }, /* RTN_THROW */ - { + }, + [RTN_NAT] = { .error = -EINVAL, .scope = RT_SCOPE_NOWHERE, - }, /* RTN_NAT */ - { + }, + [RTN_XRESOLVE] = { .error = -EINVAL, .scope = RT_SCOPE_NOWHERE, - }, /* RTN_XRESOLVE */ + }, }; /* Release a nexthop info record */ +static void free_fib_info_rcu(struct rcu_head *head) +{ + struct fib_info *fi = container_of(head, struct fib_info, rcu); + + kfree(fi); +} + void free_fib_info(struct fib_info *fi) { if (fi->fib_dead == 0) { - printk(KERN_WARNING "Freeing alive fib_info %p\n", fi); + pr_warning("Freeing alive fib_info %p\n", fi); return; } change_nexthops(fi) { @@ -152,7 +168,7 @@ void free_fib_info(struct fib_info *fi) } endfor_nexthops(fi); fib_info_cnt--; release_net(fi->fib_net); - kfree(fi); + call_rcu(&fi->rcu, free_fib_info_rcu); } void fib_release_info(struct fib_info *fi) @@ -173,7 +189,7 @@ void fib_release_info(struct fib_info *fi) spin_unlock_bh(&fib_info_lock); } -static __inline__ int nh_comp(const struct fib_info *fi, const struct fib_info *ofi) +static inline int nh_comp(const struct fib_info *fi, const struct fib_info *ofi) { const struct fib_nh *onh = ofi->fib_nh; @@ -187,7 +203,7 @@ static __inline__ int nh_comp(const struct fib_info *fi, const struct fib_info * #ifdef CONFIG_NET_CLS_ROUTE nh->nh_tclassid != onh->nh_tclassid || #endif - ((nh->nh_flags^onh->nh_flags)&~RTNH_F_DEAD)) + ((nh->nh_flags ^ onh->nh_flags) & ~RTNH_F_DEAD)) return -1; onh++; } endfor_nexthops(fi); @@ -238,7 +254,7 @@ static struct fib_info *fib_find_info(const struct fib_info *nfi) nfi->fib_priority == fi->fib_priority && memcmp(nfi->fib_metrics, fi->fib_metrics, sizeof(fi->fib_metrics)) == 0 && - ((nfi->fib_flags^fi->fib_flags)&~RTNH_F_DEAD) == 0 && + ((nfi->fib_flags ^ fi->fib_flags) & ~RTNH_F_DEAD) == 0 && (nfi->fib_nhs == 0 || nh_comp(fi, nfi) == 0)) return fi; } @@ -247,9 +263,8 @@ static struct fib_info *fib_find_info(const struct fib_info *nfi) } /* Check, that the gateway is already configured. - Used only by redirect accept routine. + * Used only by redirect accept routine. */ - int ip_fib_check_default(__be32 gw, struct net_device *dev) { struct hlist_head *head; @@ -264,7 +279,7 @@ int ip_fib_check_default(__be32 gw, struct net_device *dev) hlist_for_each_entry(nh, node, head, nh_hash) { if (nh->nh_dev == dev && nh->nh_gw == gw && - !(nh->nh_flags&RTNH_F_DEAD)) { + !(nh->nh_flags & RTNH_F_DEAD)) { spin_unlock(&fib_info_lock); return 0; } @@ -362,10 +377,10 @@ int fib_detect_death(struct fib_info *fi, int order, } if (state == NUD_REACHABLE) return 0; - if ((state&NUD_VALID) && order != dflt) + if ((state & NUD_VALID) && order != dflt) return 0; - if ((state&NUD_VALID) || - (*last_idx<0 && order > dflt)) { + if ((state & NUD_VALID) || + (*last_idx < 0 && order > dflt)) { *last_resort = fi; *last_idx = order; } @@ -476,75 +491,76 @@ int fib_nh_match(struct fib_config *cfg, struct fib_info *fi) /* - Picture - ------- - - Semantics of nexthop is very messy by historical reasons. - We have to take into account, that: - a) gateway can be actually local interface address, - so that gatewayed route is direct. - b) gateway must be on-link address, possibly - described not by an ifaddr, but also by a direct route. - c) If both gateway and interface are specified, they should not - contradict. - d) If we use tunnel routes, gateway could be not on-link. - - Attempt to reconcile all of these (alas, self-contradictory) conditions - results in pretty ugly and hairy code with obscure logic. - - I chose to generalized it instead, so that the size - of code does not increase practically, but it becomes - much more general. - Every prefix is assigned a "scope" value: "host" is local address, - "link" is direct route, - [ ... "site" ... "interior" ... ] - and "universe" is true gateway route with global meaning. - - Every prefix refers to a set of "nexthop"s (gw, oif), - where gw must have narrower scope. This recursion stops - when gw has LOCAL scope or if "nexthop" is declared ONLINK, - which means that gw is forced to be on link. - - Code is still hairy, but now it is apparently logically - consistent and very flexible. F.e. as by-product it allows - to co-exists in peace independent exterior and interior - routing processes. - - Normally it looks as following. - - {universe prefix} -> (gw, oif) [scope link] - | - |-> {link prefix} -> (gw, oif) [scope local] - | - |-> {local prefix} (terminal node) + * Picture + * ------- + * + * Semantics of nexthop is very messy by historical reasons. + * We have to take into account, that: + * a) gateway can be actually local interface address, + * so that gatewayed route is direct. + * b) gateway must be on-link address, possibly + * described not by an ifaddr, but also by a direct route. + * c) If both gateway and interface are specified, they should not + * contradict. + * d) If we use tunnel routes, gateway could be not on-link. + * + * Attempt to reconcile all of these (alas, self-contradictory) conditions + * results in pretty ugly and hairy code with obscure logic. + * + * I chose to generalized it instead, so that the size + * of code does not increase practically, but it becomes + * much more general. + * Every prefix is assigned a "scope" value: "host" is local address, + * "link" is direct route, + * [ ... "site" ... "interior" ... ] + * and "universe" is true gateway route with global meaning. + * + * Every prefix refers to a set of "nexthop"s (gw, oif), + * where gw must have narrower scope. This recursion stops + * when gw has LOCAL scope or if "nexthop" is declared ONLINK, + * which means that gw is forced to be on link. + * + * Code is still hairy, but now it is apparently logically + * consistent and very flexible. F.e. as by-product it allows + * to co-exists in peace independent exterior and interior + * routing processes. + * + * Normally it looks as following. + * + * {universe prefix} -> (gw, oif) [scope link] + * | + * |-> {link prefix} -> (gw, oif) [scope local] + * | + * |-> {local prefix} (terminal node) */ - static int fib_check_nh(struct fib_config *cfg, struct fib_info *fi, struct fib_nh *nh) { int err; struct net *net; + struct net_device *dev; net = cfg->fc_nlinfo.nl_net; if (nh->nh_gw) { struct fib_result res; - if (nh->nh_flags&RTNH_F_ONLINK) { - struct net_device *dev; + if (nh->nh_flags & RTNH_F_ONLINK) { if (cfg->fc_scope >= RT_SCOPE_LINK) return -EINVAL; if (inet_addr_type(net, nh->nh_gw) != RTN_UNICAST) return -EINVAL; - if ((dev = __dev_get_by_index(net, nh->nh_oif)) == NULL) + dev = __dev_get_by_index(net, nh->nh_oif); + if (!dev) return -ENODEV; - if (!(dev->flags&IFF_UP)) + if (!(dev->flags & IFF_UP)) return -ENETDOWN; nh->nh_dev = dev; dev_hold(dev); nh->nh_scope = RT_SCOPE_LINK; return 0; } + rcu_read_lock(); { struct flowi fl = { .nl_u = { @@ -559,50 +575,53 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_info *fi, /* It is not necessary, but requires a bit of thinking */ if (fl.fl4_scope < RT_SCOPE_LINK) fl.fl4_scope = RT_SCOPE_LINK; - if ((err = fib_lookup(net, &fl, &res)) != 0) + err = fib_lookup(net, &fl, &res); + if (err) { + rcu_read_unlock(); return err; + } } err = -EINVAL; if (res.type != RTN_UNICAST && res.type != RTN_LOCAL) goto out; nh->nh_scope = res.scope; nh->nh_oif = FIB_RES_OIF(res); - if ((nh->nh_dev = FIB_RES_DEV(res)) == NULL) + nh->nh_dev = dev = FIB_RES_DEV(res); + if (!dev) goto out; - dev_hold(nh->nh_dev); - err = -ENETDOWN; - if (!(nh->nh_dev->flags & IFF_UP)) - goto out; - err = 0; -out: - fib_res_put(&res); - return err; + dev_hold(dev); + err = (dev->flags & IFF_UP) ? 0 : -ENETDOWN; } else { struct in_device *in_dev; - if (nh->nh_flags&(RTNH_F_PERVASIVE|RTNH_F_ONLINK)) + if (nh->nh_flags & (RTNH_F_PERVASIVE | RTNH_F_ONLINK)) return -EINVAL; + rcu_read_lock(); + err = -ENODEV; in_dev = inetdev_by_index(net, nh->nh_oif); if (in_dev == NULL) - return -ENODEV; - if (!(in_dev->dev->flags&IFF_UP)) { - in_dev_put(in_dev); - return -ENETDOWN; - } + goto out; + err = -ENETDOWN; + if (!(in_dev->dev->flags & IFF_UP)) + goto out; nh->nh_dev = in_dev->dev; dev_hold(nh->nh_dev); nh->nh_scope = RT_SCOPE_HOST; - in_dev_put(in_dev); + err = 0; } - return 0; +out: + rcu_read_unlock(); + return err; } static inline unsigned int fib_laddr_hashfn(__be32 val) { unsigned int mask = (fib_hash_size - 1); - return ((__force u32)val ^ ((__force u32)val >> 7) ^ ((__force u32)val >> 14)) & mask; + return ((__force u32)val ^ + ((__force u32)val >> 7) ^ + ((__force u32)val >> 14)) & mask; } static struct hlist_head *fib_hash_alloc(int bytes) @@ -611,7 +630,8 @@ static struct hlist_head *fib_hash_alloc(int bytes) return kzalloc(bytes, GFP_KERNEL); else return (struct hlist_head *) - __get_free_pages(GFP_KERNEL | __GFP_ZERO, get_order(bytes)); + __get_free_pages(GFP_KERNEL | __GFP_ZERO, + get_order(bytes)); } static void fib_hash_free(struct hlist_head *hash, int bytes) @@ -806,7 +826,8 @@ struct fib_info *fib_create_info(struct fib_config *cfg) goto failure; } else { change_nexthops(fi) { - if ((err = fib_check_nh(cfg, fi, nexthop_nh)) != 0) + err = fib_check_nh(cfg, fi, nexthop_nh); + if (err != 0) goto failure; } endfor_nexthops(fi) } @@ -819,7 +840,8 @@ struct fib_info *fib_create_info(struct fib_config *cfg) } link_it: - if ((ofi = fib_find_info(fi)) != NULL) { + ofi = fib_find_info(fi); + if (ofi) { fi->fib_dead = 1; free_fib_info(fi); ofi->fib_treeref++; @@ -864,7 +886,7 @@ failure: /* Note! fib_semantic_match intentionally uses RCU list functions. */ int fib_semantic_match(struct list_head *head, const struct flowi *flp, - struct fib_result *res, int prefixlen) + struct fib_result *res, int prefixlen, int fib_flags) { struct fib_alias *fa; int nh_sel = 0; @@ -879,7 +901,7 @@ int fib_semantic_match(struct list_head *head, const struct flowi *flp, if (fa->fa_scope < flp->fl4_scope) continue; - fa->fa_state |= FA_S_ACCESSED; + fib_alias_accessed(fa); err = fib_props[fa->fa_type].error; if (err == 0) { @@ -895,7 +917,7 @@ int fib_semantic_match(struct list_head *head, const struct flowi *flp, case RTN_ANYCAST: case RTN_MULTICAST: for_nexthops(fi) { - if (nh->nh_flags&RTNH_F_DEAD) + if (nh->nh_flags & RTNH_F_DEAD) continue; if (!flp->oif || flp->oif == nh->nh_oif) break; @@ -906,16 +928,15 @@ int fib_semantic_match(struct list_head *head, const struct flowi *flp, goto out_fill_res; } #else - if (nhsel < 1) { + if (nhsel < 1) goto out_fill_res; - } #endif endfor_nexthops(fi); continue; default: - printk(KERN_WARNING "fib_semantic_match bad type %#x\n", - fa->fa_type); + pr_warning("fib_semantic_match bad type %#x\n", + fa->fa_type); return -EINVAL; } } @@ -929,7 +950,8 @@ out_fill_res: res->type = fa->fa_type; res->scope = fa->fa_scope; res->fi = fa->fa_info; - atomic_inc(&res->fi->fib_clntref); + if (!(fib_flags & FIB_LOOKUP_NOREF)) + atomic_inc(&res->fi->fib_clntref); return 0; } @@ -1028,10 +1050,10 @@ nla_put_failure: } /* - Update FIB if: - - local address disappeared -> we must delete all the entries - referring to it. - - device went down -> we must shutdown all nexthops going via it. + * Update FIB if: + * - local address disappeared -> we must delete all the entries + * referring to it. + * - device went down -> we must shutdown all nexthops going via it. */ int fib_sync_down_addr(struct net *net, __be32 local) { @@ -1078,7 +1100,7 @@ int fib_sync_down_dev(struct net_device *dev, int force) prev_fi = fi; dead = 0; change_nexthops(fi) { - if (nexthop_nh->nh_flags&RTNH_F_DEAD) + if (nexthop_nh->nh_flags & RTNH_F_DEAD) dead++; else if (nexthop_nh->nh_dev == dev && nexthop_nh->nh_scope != scope) { @@ -1110,10 +1132,9 @@ int fib_sync_down_dev(struct net_device *dev, int force) #ifdef CONFIG_IP_ROUTE_MULTIPATH /* - Dead device goes up. We wake up dead nexthops. - It takes sense only on multipath routes. + * Dead device goes up. We wake up dead nexthops. + * It takes sense only on multipath routes. */ - int fib_sync_up(struct net_device *dev) { struct fib_info *prev_fi; @@ -1123,7 +1144,7 @@ int fib_sync_up(struct net_device *dev) struct fib_nh *nh; int ret; - if (!(dev->flags&IFF_UP)) + if (!(dev->flags & IFF_UP)) return 0; prev_fi = NULL; @@ -1142,12 +1163,12 @@ int fib_sync_up(struct net_device *dev) prev_fi = fi; alive = 0; change_nexthops(fi) { - if (!(nexthop_nh->nh_flags&RTNH_F_DEAD)) { + if (!(nexthop_nh->nh_flags & RTNH_F_DEAD)) { alive++; continue; } if (nexthop_nh->nh_dev == NULL || - !(nexthop_nh->nh_dev->flags&IFF_UP)) + !(nexthop_nh->nh_dev->flags & IFF_UP)) continue; if (nexthop_nh->nh_dev != dev || !__in_dev_get_rtnl(dev)) @@ -1169,10 +1190,9 @@ int fib_sync_up(struct net_device *dev) } /* - The algorithm is suboptimal, but it provides really - fair weighted route distribution. + * The algorithm is suboptimal, but it provides really + * fair weighted route distribution. */ - void fib_select_multipath(const struct flowi *flp, struct fib_result *res) { struct fib_info *fi = res->fi; @@ -1182,7 +1202,7 @@ void fib_select_multipath(const struct flowi *flp, struct fib_result *res) if (fi->fib_power <= 0) { int power = 0; change_nexthops(fi) { - if (!(nexthop_nh->nh_flags&RTNH_F_DEAD)) { + if (!(nexthop_nh->nh_flags & RTNH_F_DEAD)) { power += nexthop_nh->nh_weight; nexthop_nh->nh_power = nexthop_nh->nh_weight; } @@ -1198,15 +1218,16 @@ void fib_select_multipath(const struct flowi *flp, struct fib_result *res) /* w should be random number [0..fi->fib_power-1], - it is pretty bad approximation. + * it is pretty bad approximation. */ w = jiffies % fi->fib_power; change_nexthops(fi) { - if (!(nexthop_nh->nh_flags&RTNH_F_DEAD) && + if (!(nexthop_nh->nh_flags & RTNH_F_DEAD) && nexthop_nh->nh_power) { - if ((w -= nexthop_nh->nh_power) <= 0) { + w -= nexthop_nh->nh_power; + if (w <= 0) { nexthop_nh->nh_power--; fi->fib_power--; res->nh_sel = nhsel; diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 2230ae3bf20..b1445089510 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -186,7 +186,7 @@ static inline struct tnode *node_parent_rcu(struct node *node) { struct tnode *ret = node_parent(node); - return rcu_dereference(ret); + return rcu_dereference_rtnl(ret); } /* Same as rcu_assign_pointer @@ -209,9 +209,7 @@ static inline struct node *tnode_get_child_rcu(struct tnode *tn, unsigned int i) { struct node *ret = tnode_get_child(tn, i); - return rcu_dereference_check(ret, - rcu_read_lock_held() || - lockdep_rtnl_is_held()); + return rcu_dereference_rtnl(ret); } static inline int tnode_child_length(const struct tnode *tn) @@ -457,8 +455,8 @@ static struct tnode *tnode_new(t_key key, int pos, int bits) tn->empty_children = 1<<bits; } - pr_debug("AT %p s=%u %lu\n", tn, (unsigned int) sizeof(struct tnode), - (unsigned long) (sizeof(struct node) << bits)); + pr_debug("AT %p s=%zu %zu\n", tn, sizeof(struct tnode), + sizeof(struct node) << bits); return tn; } @@ -607,11 +605,10 @@ static struct node *resize(struct trie *t, struct tnode *tn) /* Keep root node larger */ - if (!node_parent((struct node*) tn)) { + if (!node_parent((struct node *)tn)) { inflate_threshold_use = inflate_threshold_root; halve_threshold_use = halve_threshold_root; - } - else { + } else { inflate_threshold_use = inflate_threshold; halve_threshold_use = halve_threshold; } @@ -637,7 +634,7 @@ static struct node *resize(struct trie *t, struct tnode *tn) check_tnode(tn); /* Return if at least one inflate is run */ - if( max_work != MAX_WORK) + if (max_work != MAX_WORK) return (struct node *) tn; /* @@ -964,9 +961,7 @@ fib_find_node(struct trie *t, u32 key) struct node *n; pos = 0; - n = rcu_dereference_check(t->trie, - rcu_read_lock_held() || - lockdep_rtnl_is_held()); + n = rcu_dereference_rtnl(t->trie); while (n != NULL && NODE_TYPE(n) == T_TNODE) { tn = (struct tnode *) n; @@ -1347,7 +1342,7 @@ err: /* should be called with rcu_read_lock */ static int check_leaf(struct trie *t, struct leaf *l, t_key key, const struct flowi *flp, - struct fib_result *res) + struct fib_result *res, int fib_flags) { struct leaf_info *li; struct hlist_head *hhead = &l->list; @@ -1361,7 +1356,7 @@ static int check_leaf(struct trie *t, struct leaf *l, if (l->key != (key & ntohl(mask))) continue; - err = fib_semantic_match(&li->falh, flp, res, plen); + err = fib_semantic_match(&li->falh, flp, res, plen, fib_flags); #ifdef CONFIG_IP_FIB_TRIE_STATS if (err <= 0) @@ -1377,7 +1372,7 @@ static int check_leaf(struct trie *t, struct leaf *l, } int fib_table_lookup(struct fib_table *tb, const struct flowi *flp, - struct fib_result *res) + struct fib_result *res, int fib_flags) { struct trie *t = (struct trie *) tb->tb_data; int ret; @@ -1389,8 +1384,7 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi *flp, t_key cindex = 0; int current_prefix_length = KEYLENGTH; struct tnode *cn; - t_key node_prefix, key_prefix, pref_mismatch; - int mp; + t_key pref_mismatch; rcu_read_lock(); @@ -1404,7 +1398,7 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi *flp, /* Just a leaf? */ if (IS_LEAF(n)) { - ret = check_leaf(t, (struct leaf *)n, key, flp, res); + ret = check_leaf(t, (struct leaf *)n, key, flp, res, fib_flags); goto found; } @@ -1429,7 +1423,7 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi *flp, } if (IS_LEAF(n)) { - ret = check_leaf(t, (struct leaf *)n, key, flp, res); + ret = check_leaf(t, (struct leaf *)n, key, flp, res, fib_flags); if (ret > 0) goto backtrace; goto found; @@ -1505,10 +1499,7 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi *flp, * matching prefix. */ - node_prefix = mask_pfx(cn->key, cn->pos); - key_prefix = mask_pfx(key, cn->pos); - pref_mismatch = key_prefix^node_prefix; - mp = 0; + pref_mismatch = mask_pfx(cn->key ^ key, cn->pos); /* * In short: If skipped bits in this node do not match @@ -1516,13 +1507,9 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi *flp, * state.directly. */ if (pref_mismatch) { - while (!(pref_mismatch & (1<<(KEYLENGTH-1)))) { - mp++; - pref_mismatch = pref_mismatch << 1; - } - key_prefix = tkey_extract_bits(cn->key, mp, cn->pos-mp); + int mp = KEYLENGTH - fls(pref_mismatch); - if (key_prefix != 0) + if (tkey_extract_bits(cn->key, mp, cn->pos - mp) != 0) goto backtrace; if (current_prefix_length >= cn->pos) @@ -1746,14 +1733,14 @@ static struct leaf *leaf_walk_rcu(struct tnode *p, struct node *c) /* Node empty, walk back up to parent */ c = (struct node *) p; - } while ( (p = node_parent_rcu(c)) != NULL); + } while ((p = node_parent_rcu(c)) != NULL); return NULL; /* Root of trie */ } static struct leaf *trie_firstleaf(struct trie *t) { - struct tnode *n = (struct tnode *) rcu_dereference(t->trie); + struct tnode *n = (struct tnode *)rcu_dereference_rtnl(t->trie); if (!n) return NULL; @@ -1851,7 +1838,8 @@ void fib_table_select_default(struct fib_table *tb, if (!next_fi->fib_nh[0].nh_gw || next_fi->fib_nh[0].nh_scope != RT_SCOPE_LINK) continue; - fa->fa_state |= FA_S_ACCESSED; + + fib_alias_accessed(fa); if (fi == NULL) { if (next_fi != res->fi) @@ -2039,14 +2027,14 @@ struct fib_trie_iter { struct seq_net_private p; struct fib_table *tb; struct tnode *tnode; - unsigned index; - unsigned depth; + unsigned int index; + unsigned int depth; }; static struct node *fib_trie_get_next(struct fib_trie_iter *iter) { struct tnode *tn = iter->tnode; - unsigned cindex = iter->index; + unsigned int cindex = iter->index; struct tnode *p; /* A single entry routing table */ @@ -2155,7 +2143,7 @@ static void trie_collect_stats(struct trie *t, struct trie_stat *s) */ static void trie_show_stats(struct seq_file *seq, struct trie_stat *stat) { - unsigned i, max, pointers, bytes, avdepth; + unsigned int i, max, pointers, bytes, avdepth; if (stat->leaves) avdepth = stat->totdepth*100 / stat->leaves; @@ -2352,7 +2340,8 @@ static void fib_trie_seq_stop(struct seq_file *seq, void *v) static void seq_indent(struct seq_file *seq, int n) { - while (n-- > 0) seq_puts(seq, " "); + while (n-- > 0) + seq_puts(seq, " "); } static inline const char *rtn_scope(char *buf, size_t len, enum rt_scope_t s) @@ -2384,7 +2373,7 @@ static const char *const rtn_type_names[__RTN_MAX] = { [RTN_XRESOLVE] = "XRESOLVE", }; -static inline const char *rtn_type(char *buf, size_t len, unsigned t) +static inline const char *rtn_type(char *buf, size_t len, unsigned int t) { if (t < __RTN_MAX && rtn_type_names[t]) return rtn_type_names[t]; @@ -2540,13 +2529,12 @@ static void fib_route_seq_stop(struct seq_file *seq, void *v) rcu_read_unlock(); } -static unsigned fib_flag_trans(int type, __be32 mask, const struct fib_info *fi) +static unsigned int fib_flag_trans(int type, __be32 mask, const struct fib_info *fi) { - static unsigned type2flags[RTN_MAX + 1] = { - [7] = RTF_REJECT, [8] = RTF_REJECT, - }; - unsigned flags = type2flags[type]; + unsigned int flags = 0; + if (type == RTN_UNREACHABLE || type == RTN_PROHIBIT) + flags = RTF_REJECT; if (fi && fi->fib_nh->nh_gw) flags |= RTF_GATEWAY; if (mask == htonl(0xFFFFFFFF)) @@ -2558,7 +2546,7 @@ static unsigned fib_flag_trans(int type, __be32 mask, const struct fib_info *fi) /* * This outputs /proc/net/route. * The format of the file is not supposed to be changed - * and needs to be same as fib_hash output to avoid breaking + * and needs to be same as fib_hash output to avoid breaking * legacy utilities */ static int fib_route_seq_show(struct seq_file *seq, void *v) @@ -2583,7 +2571,7 @@ static int fib_route_seq_show(struct seq_file *seq, void *v) list_for_each_entry_rcu(fa, &li->falh, fa_list) { const struct fib_info *fi = fa->fa_info; - unsigned flags = fib_flag_trans(fa->fa_type, mask, fi); + unsigned int flags = fib_flag_trans(fa->fa_type, mask, fi); int len; if (fa->fa_type == RTN_BROADCAST diff --git a/net/ipv4/gre.c b/net/ipv4/gre.c new file mode 100644 index 00000000000..caea6885fdb --- /dev/null +++ b/net/ipv4/gre.c @@ -0,0 +1,151 @@ +/* + * GRE over IPv4 demultiplexer driver + * + * Authors: Dmitry Kozlov (xeb@mail.ru) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + */ + +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/kmod.h> +#include <linux/skbuff.h> +#include <linux/in.h> +#include <linux/netdevice.h> +#include <linux/version.h> +#include <linux/spinlock.h> +#include <net/protocol.h> +#include <net/gre.h> + + +static const struct gre_protocol *gre_proto[GREPROTO_MAX] __read_mostly; +static DEFINE_SPINLOCK(gre_proto_lock); + +int gre_add_protocol(const struct gre_protocol *proto, u8 version) +{ + if (version >= GREPROTO_MAX) + goto err_out; + + spin_lock(&gre_proto_lock); + if (gre_proto[version]) + goto err_out_unlock; + + rcu_assign_pointer(gre_proto[version], proto); + spin_unlock(&gre_proto_lock); + return 0; + +err_out_unlock: + spin_unlock(&gre_proto_lock); +err_out: + return -1; +} +EXPORT_SYMBOL_GPL(gre_add_protocol); + +int gre_del_protocol(const struct gre_protocol *proto, u8 version) +{ + if (version >= GREPROTO_MAX) + goto err_out; + + spin_lock(&gre_proto_lock); + if (gre_proto[version] != proto) + goto err_out_unlock; + rcu_assign_pointer(gre_proto[version], NULL); + spin_unlock(&gre_proto_lock); + synchronize_rcu(); + return 0; + +err_out_unlock: + spin_unlock(&gre_proto_lock); +err_out: + return -1; +} +EXPORT_SYMBOL_GPL(gre_del_protocol); + +static int gre_rcv(struct sk_buff *skb) +{ + const struct gre_protocol *proto; + u8 ver; + int ret; + + if (!pskb_may_pull(skb, 12)) + goto drop; + + ver = skb->data[1]&0x7f; + if (ver >= GREPROTO_MAX) + goto drop; + + rcu_read_lock(); + proto = rcu_dereference(gre_proto[ver]); + if (!proto || !proto->handler) + goto drop_unlock; + ret = proto->handler(skb); + rcu_read_unlock(); + return ret; + +drop_unlock: + rcu_read_unlock(); +drop: + kfree_skb(skb); + return NET_RX_DROP; +} + +static void gre_err(struct sk_buff *skb, u32 info) +{ + const struct gre_protocol *proto; + u8 ver; + + if (!pskb_may_pull(skb, 12)) + goto drop; + + ver = skb->data[1]&0x7f; + if (ver >= GREPROTO_MAX) + goto drop; + + rcu_read_lock(); + proto = rcu_dereference(gre_proto[ver]); + if (!proto || !proto->err_handler) + goto drop_unlock; + proto->err_handler(skb, info); + rcu_read_unlock(); + return; + +drop_unlock: + rcu_read_unlock(); +drop: + kfree_skb(skb); +} + +static const struct net_protocol net_gre_protocol = { + .handler = gre_rcv, + .err_handler = gre_err, + .netns_ok = 1, +}; + +static int __init gre_init(void) +{ + pr_info("GRE over IPv4 demultiplexor driver"); + + if (inet_add_protocol(&net_gre_protocol, IPPROTO_GRE) < 0) { + pr_err("gre: can't add protocol\n"); + return -EAGAIN; + } + + return 0; +} + +static void __exit gre_exit(void) +{ + inet_del_protocol(&net_gre_protocol, IPPROTO_GRE); +} + +module_init(gre_init); +module_exit(gre_exit); + +MODULE_DESCRIPTION("GRE over IPv4 demultiplexer driver"); +MODULE_AUTHOR("D. Kozlov (xeb@mail.ru)"); +MODULE_LICENSE("GPL"); + diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index a0d847c7cba..96bc7f9475a 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -379,7 +379,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) inet->tos = ip_hdr(skb)->tos; daddr = ipc.addr = rt->rt_src; ipc.opt = NULL; - ipc.shtx.flags = 0; + ipc.tx_flags = 0; if (icmp_param->replyopts.optlen) { ipc.opt = &icmp_param->replyopts; if (ipc.opt->srr) @@ -538,7 +538,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) inet_sk(sk)->tos = tos; ipc.addr = iph->saddr; ipc.opt = &icmp_param.replyopts; - ipc.shtx.flags = 0; + ipc.tx_flags = 0; { struct flowi fl = { diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index a1ad0e7180d..c8877c6c721 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -856,6 +856,18 @@ static void igmp_heard_query(struct in_device *in_dev, struct sk_buff *skb, igmpv3_clear_delrec(in_dev); } else if (len < 12) { return; /* ignore bogus packet; freed by caller */ + } else if (IGMP_V1_SEEN(in_dev)) { + /* This is a v3 query with v1 queriers present */ + max_delay = IGMP_Query_Response_Interval; + group = 0; + } else if (IGMP_V2_SEEN(in_dev)) { + /* this is a v3 query with v2 queriers present; + * Interpretation of the max_delay code is problematic here. + * A real v2 host would use ih_code directly, while v3 has a + * different encoding. We use the v3 encoding as more likely + * to be intended in a v3 query. + */ + max_delay = IGMPV3_MRC(ih3->code)*(HZ/IGMP_TIMER_SCALE); } else { /* v3 */ if (!pskb_may_pull(skb, sizeof(struct igmpv3_query))) return; @@ -1257,14 +1269,14 @@ void ip_mc_rejoin_group(struct ip_mc_list *im) if (im->multiaddr == IGMP_ALL_HOSTS) return; - if (IGMP_V1_SEEN(in_dev) || IGMP_V2_SEEN(in_dev)) { - igmp_mod_timer(im, IGMP_Initial_Report_Delay); - return; - } - /* else, v3 */ - im->crcount = in_dev->mr_qrv ? in_dev->mr_qrv : - IGMP_Unsolicited_Report_Count; - igmp_ifc_event(in_dev); + /* a failover is happening and switches + * must be notified immediately */ + if (IGMP_V1_SEEN(in_dev)) + igmp_send_report(in_dev, im, IGMP_HOST_MEMBERSHIP_REPORT); + else if (IGMP_V2_SEEN(in_dev)) + igmp_send_report(in_dev, im, IGMPV2_HOST_MEMBERSHIP_REPORT); + else + igmp_send_report(in_dev, im, IGMPV3_HOST_MEMBERSHIP_REPORT); #endif } EXPORT_SYMBOL(ip_mc_rejoin_group); @@ -1406,6 +1418,7 @@ void ip_mc_destroy_dev(struct in_device *in_dev) write_unlock_bh(&in_dev->mc_list_lock); } +/* RTNL is locked */ static struct in_device *ip_mc_find_dev(struct net *net, struct ip_mreqn *imr) { struct flowi fl = { .nl_u = { .ip4_u = @@ -1416,15 +1429,12 @@ static struct in_device *ip_mc_find_dev(struct net *net, struct ip_mreqn *imr) if (imr->imr_ifindex) { idev = inetdev_by_index(net, imr->imr_ifindex); - if (idev) - __in_dev_put(idev); return idev; } if (imr->imr_address.s_addr) { - dev = ip_dev_find(net, imr->imr_address.s_addr); + dev = __ip_dev_find(net, imr->imr_address.s_addr, false); if (!dev) return NULL; - dev_put(dev); } if (!dev && !ip_route_output_key(net, &rt, &fl)) { diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index e5fa2ddce32..ba804266584 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -425,7 +425,7 @@ static int inet_diag_bc_run(const void *bc, int len, bc += op->no; } } - return (len == 0); + return len == 0; } static int valid_cc(const void *bc, int len, int cc) diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index fb7ad5a21ff..1b344f30b46 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -101,19 +101,43 @@ void inet_put_port(struct sock *sk) } EXPORT_SYMBOL(inet_put_port); -void __inet_inherit_port(struct sock *sk, struct sock *child) +int __inet_inherit_port(struct sock *sk, struct sock *child) { struct inet_hashinfo *table = sk->sk_prot->h.hashinfo; - const int bhash = inet_bhashfn(sock_net(sk), inet_sk(child)->inet_num, + unsigned short port = inet_sk(child)->inet_num; + const int bhash = inet_bhashfn(sock_net(sk), port, table->bhash_size); struct inet_bind_hashbucket *head = &table->bhash[bhash]; struct inet_bind_bucket *tb; spin_lock(&head->lock); tb = inet_csk(sk)->icsk_bind_hash; + if (tb->port != port) { + /* NOTE: using tproxy and redirecting skbs to a proxy + * on a different listener port breaks the assumption + * that the listener socket's icsk_bind_hash is the same + * as that of the child socket. We have to look up or + * create a new bind bucket for the child here. */ + struct hlist_node *node; + inet_bind_bucket_for_each(tb, node, &head->chain) { + if (net_eq(ib_net(tb), sock_net(sk)) && + tb->port == port) + break; + } + if (!node) { + tb = inet_bind_bucket_create(table->bind_bucket_cachep, + sock_net(sk), head, port); + if (!tb) { + spin_unlock(&head->lock); + return -ENOMEM; + } + } + } sk_add_bind_node(child, &tb->owners); inet_csk(child)->icsk_bind_hash = tb; spin_unlock(&head->lock); + + return 0; } EXPORT_SYMBOL_GPL(__inet_inherit_port); diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index b7c41654dde..168440834ad 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -116,11 +116,11 @@ static int ip4_frag_match(struct inet_frag_queue *q, void *a) struct ip4_create_arg *arg = a; qp = container_of(q, struct ipq, q); - return (qp->id == arg->iph->id && + return qp->id == arg->iph->id && qp->saddr == arg->iph->saddr && qp->daddr == arg->iph->daddr && qp->protocol == arg->iph->protocol && - qp->user == arg->user); + qp->user == arg->user; } /* Memory Tracking Functions. */ @@ -542,7 +542,7 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *prev, /* If the first fragment is fragmented itself, we split * it to two chunks: the first with data and paged part * and the second, holding only fragments. */ - if (skb_has_frags(head)) { + if (skb_has_frag_list(head)) { struct sk_buff *clone; int i, plen = 0; diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 945b20a5ad5..d0ffcbe369b 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -44,8 +44,9 @@ #include <net/net_namespace.h> #include <net/netns/generic.h> #include <net/rtnetlink.h> +#include <net/gre.h> -#ifdef CONFIG_IPV6 +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) #include <net/ipv6.h> #include <net/ip6_fib.h> #include <net/ip6_route.h> @@ -63,13 +64,13 @@ We cannot track such dead loops during route installation, it is infeasible task. The most general solutions would be to keep skb->encapsulation counter (sort of local ttl), - and silently drop packet when it expires. It is the best + and silently drop packet when it expires. It is a good solution, but it supposes maintaing new variable in ALL skb, even if no tunneling is used. - Current solution: HARD_TX_LOCK lock breaks dead loops. - - + Current solution: xmit_recursion breaks dead loops. This is a percpu + counter, since when we enter the first ndo_xmit(), cpu migration is + forbidden. We force an exit if this counter reaches RECURSION_LIMIT 2. Networking dead loops would not kill routers, but would really kill network. IP hop limit plays role of "t->recursion" in this case, @@ -128,7 +129,7 @@ static int ipgre_tunnel_bind_dev(struct net_device *dev); static int ipgre_net_id __read_mostly; struct ipgre_net { - struct ip_tunnel *tunnels[4][HASH_SIZE]; + struct ip_tunnel __rcu *tunnels[4][HASH_SIZE]; struct net_device *fb_tunnel_dev; }; @@ -158,13 +159,40 @@ struct ipgre_net { #define tunnels_l tunnels[1] #define tunnels_wc tunnels[0] /* - * Locking : hash tables are protected by RCU and a spinlock + * Locking : hash tables are protected by RCU and RTNL */ -static DEFINE_SPINLOCK(ipgre_lock); #define for_each_ip_tunnel_rcu(start) \ for (t = rcu_dereference(start); t; t = rcu_dereference(t->next)) +/* often modified stats are per cpu, other are shared (netdev->stats) */ +struct pcpu_tstats { + unsigned long rx_packets; + unsigned long rx_bytes; + unsigned long tx_packets; + unsigned long tx_bytes; +}; + +static struct net_device_stats *ipgre_get_stats(struct net_device *dev) +{ + struct pcpu_tstats sum = { 0 }; + int i; + + for_each_possible_cpu(i) { + const struct pcpu_tstats *tstats = per_cpu_ptr(dev->tstats, i); + + sum.rx_packets += tstats->rx_packets; + sum.rx_bytes += tstats->rx_bytes; + sum.tx_packets += tstats->tx_packets; + sum.tx_bytes += tstats->tx_bytes; + } + dev->stats.rx_packets = sum.rx_packets; + dev->stats.rx_bytes = sum.rx_bytes; + dev->stats.tx_packets = sum.tx_packets; + dev->stats.tx_bytes = sum.tx_bytes; + return &dev->stats; +} + /* Given src, dst and key, find appropriate for input tunnel. */ static struct ip_tunnel * ipgre_tunnel_lookup(struct net_device *dev, @@ -173,8 +201,8 @@ static struct ip_tunnel * ipgre_tunnel_lookup(struct net_device *dev, { struct net *net = dev_net(dev); int link = dev->ifindex; - unsigned h0 = HASH(remote); - unsigned h1 = HASH(key); + unsigned int h0 = HASH(remote); + unsigned int h1 = HASH(key); struct ip_tunnel *t, *cand = NULL; struct ipgre_net *ign = net_generic(net, ipgre_net_id); int dev_type = (gre_proto == htons(ETH_P_TEB)) ? @@ -289,13 +317,13 @@ static struct ip_tunnel * ipgre_tunnel_lookup(struct net_device *dev, return NULL; } -static struct ip_tunnel **__ipgre_bucket(struct ipgre_net *ign, +static struct ip_tunnel __rcu **__ipgre_bucket(struct ipgre_net *ign, struct ip_tunnel_parm *parms) { __be32 remote = parms->iph.daddr; __be32 local = parms->iph.saddr; __be32 key = parms->i_key; - unsigned h = HASH(key); + unsigned int h = HASH(key); int prio = 0; if (local) @@ -308,7 +336,7 @@ static struct ip_tunnel **__ipgre_bucket(struct ipgre_net *ign, return &ign->tunnels[prio][h]; } -static inline struct ip_tunnel **ipgre_bucket(struct ipgre_net *ign, +static inline struct ip_tunnel __rcu **ipgre_bucket(struct ipgre_net *ign, struct ip_tunnel *t) { return __ipgre_bucket(ign, &t->parms); @@ -316,23 +344,22 @@ static inline struct ip_tunnel **ipgre_bucket(struct ipgre_net *ign, static void ipgre_tunnel_link(struct ipgre_net *ign, struct ip_tunnel *t) { - struct ip_tunnel **tp = ipgre_bucket(ign, t); + struct ip_tunnel __rcu **tp = ipgre_bucket(ign, t); - spin_lock_bh(&ipgre_lock); - t->next = *tp; + rcu_assign_pointer(t->next, rtnl_dereference(*tp)); rcu_assign_pointer(*tp, t); - spin_unlock_bh(&ipgre_lock); } static void ipgre_tunnel_unlink(struct ipgre_net *ign, struct ip_tunnel *t) { - struct ip_tunnel **tp; - - for (tp = ipgre_bucket(ign, t); *tp; tp = &(*tp)->next) { - if (t == *tp) { - spin_lock_bh(&ipgre_lock); - *tp = t->next; - spin_unlock_bh(&ipgre_lock); + struct ip_tunnel __rcu **tp; + struct ip_tunnel *iter; + + for (tp = ipgre_bucket(ign, t); + (iter = rtnl_dereference(*tp)) != NULL; + tp = &iter->next) { + if (t == iter) { + rcu_assign_pointer(*tp, t->next); break; } } @@ -346,10 +373,13 @@ static struct ip_tunnel *ipgre_tunnel_find(struct net *net, __be32 local = parms->iph.saddr; __be32 key = parms->i_key; int link = parms->link; - struct ip_tunnel *t, **tp; + struct ip_tunnel *t; + struct ip_tunnel __rcu **tp; struct ipgre_net *ign = net_generic(net, ipgre_net_id); - for (tp = __ipgre_bucket(ign, parms); (t = *tp) != NULL; tp = &t->next) + for (tp = __ipgre_bucket(ign, parms); + (t = rtnl_dereference(*tp)) != NULL; + tp = &t->next) if (local == t->parms.iph.saddr && remote == t->parms.iph.daddr && key == t->parms.i_key && @@ -360,7 +390,7 @@ static struct ip_tunnel *ipgre_tunnel_find(struct net *net, return t; } -static struct ip_tunnel * ipgre_tunnel_locate(struct net *net, +static struct ip_tunnel *ipgre_tunnel_locate(struct net *net, struct ip_tunnel_parm *parms, int create) { struct ip_tunnel *t, *nt; @@ -582,7 +612,7 @@ static int ipgre_rcv(struct sk_buff *skb) if ((tunnel = ipgre_tunnel_lookup(skb->dev, iph->saddr, iph->daddr, key, gre_proto))) { - struct net_device_stats *stats = &tunnel->dev->stats; + struct pcpu_tstats *tstats; secpath_reset(skb); @@ -606,22 +636,22 @@ static int ipgre_rcv(struct sk_buff *skb) /* Looped back packet, drop it! */ if (skb_rtable(skb)->fl.iif == 0) goto drop; - stats->multicast++; + tunnel->dev->stats.multicast++; skb->pkt_type = PACKET_BROADCAST; } #endif if (((flags&GRE_CSUM) && csum) || (!(flags&GRE_CSUM) && tunnel->parms.i_flags&GRE_CSUM)) { - stats->rx_crc_errors++; - stats->rx_errors++; + tunnel->dev->stats.rx_crc_errors++; + tunnel->dev->stats.rx_errors++; goto drop; } if (tunnel->parms.i_flags&GRE_SEQ) { if (!(flags&GRE_SEQ) || (tunnel->i_seqno && (s32)(seqno - tunnel->i_seqno) < 0)) { - stats->rx_fifo_errors++; - stats->rx_errors++; + tunnel->dev->stats.rx_fifo_errors++; + tunnel->dev->stats.rx_errors++; goto drop; } tunnel->i_seqno = seqno + 1; @@ -630,8 +660,8 @@ static int ipgre_rcv(struct sk_buff *skb) /* Warning: All skb pointers will be invalidated! */ if (tunnel->dev->type == ARPHRD_ETHER) { if (!pskb_may_pull(skb, ETH_HLEN)) { - stats->rx_length_errors++; - stats->rx_errors++; + tunnel->dev->stats.rx_length_errors++; + tunnel->dev->stats.rx_errors++; goto drop; } @@ -640,14 +670,19 @@ static int ipgre_rcv(struct sk_buff *skb) skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN); } - skb_tunnel_rx(skb, tunnel->dev); + tstats = this_cpu_ptr(tunnel->dev->tstats); + tstats->rx_packets++; + tstats->rx_bytes += skb->len; + + __skb_tunnel_rx(skb, tunnel->dev); skb_reset_network_header(skb); ipgre_ecn_decapsulate(iph, skb); netif_rx(skb); + rcu_read_unlock(); - return(0); + return 0; } icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); @@ -655,20 +690,19 @@ drop: rcu_read_unlock(); drop_nolock: kfree_skb(skb); - return(0); + return 0; } static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); - struct net_device_stats *stats = &dev->stats; - struct netdev_queue *txq = netdev_get_tx_queue(dev, 0); + struct pcpu_tstats *tstats; struct iphdr *old_iph = ip_hdr(skb); struct iphdr *tiph; u8 tos; __be16 df; struct rtable *rt; /* Route to the other host */ - struct net_device *tdev; /* Device to other host */ + struct net_device *tdev; /* Device to other host */ struct iphdr *iph; /* Our new IP header */ unsigned int max_headroom; /* The extra header space needed */ int gre_hlen; @@ -690,7 +724,7 @@ static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev /* NBMA tunnel */ if (skb_dst(skb) == NULL) { - stats->tx_fifo_errors++; + dev->stats.tx_fifo_errors++; goto tx_error; } @@ -699,7 +733,7 @@ static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev if ((dst = rt->rt_gateway) == 0) goto tx_error_icmp; } -#ifdef CONFIG_IPV6 +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) else if (skb->protocol == htons(ETH_P_IPV6)) { struct in6_addr *addr6; int addr_type; @@ -736,14 +770,20 @@ static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev } { - struct flowi fl = { .oif = tunnel->parms.link, - .nl_u = { .ip4_u = - { .daddr = dst, - .saddr = tiph->saddr, - .tos = RT_TOS(tos) } }, - .proto = IPPROTO_GRE }; + struct flowi fl = { + .oif = tunnel->parms.link, + .nl_u = { + .ip4_u = { + .daddr = dst, + .saddr = tiph->saddr, + .tos = RT_TOS(tos) + } + }, + .proto = IPPROTO_GRE + } +; if (ip_route_output_key(dev_net(dev), &rt, &fl)) { - stats->tx_carrier_errors++; + dev->stats.tx_carrier_errors++; goto tx_error; } } @@ -751,7 +791,7 @@ static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev if (tdev == dev) { ip_rt_put(rt); - stats->collisions++; + dev->stats.collisions++; goto tx_error; } @@ -774,7 +814,7 @@ static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev goto tx_error; } } -#ifdef CONFIG_IPV6 +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) else if (skb->protocol == htons(ETH_P_IPV6)) { struct rt6_info *rt6 = (struct rt6_info *)skb_dst(skb); @@ -814,7 +854,7 @@ static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev dev->needed_headroom = max_headroom; if (!new_skb) { ip_rt_put(rt); - txq->tx_dropped++; + dev->stats.tx_dropped++; dev_kfree_skb(skb); return NETDEV_TX_OK; } @@ -850,7 +890,7 @@ static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev if ((iph->ttl = tiph->ttl) == 0) { if (skb->protocol == htons(ETH_P_IP)) iph->ttl = old_iph->ttl; -#ifdef CONFIG_IPV6 +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) else if (skb->protocol == htons(ETH_P_IPV6)) iph->ttl = ((struct ipv6hdr *)old_iph)->hop_limit; #endif @@ -881,15 +921,15 @@ static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev } nf_reset(skb); - - IPTUNNEL_XMIT(); + tstats = this_cpu_ptr(dev->tstats); + __IPTUNNEL_XMIT(tstats, &dev->stats); return NETDEV_TX_OK; tx_error_icmp: dst_link_failure(skb); tx_error: - stats->tx_errors++; + dev->stats.tx_errors++; dev_kfree_skb(skb); return NETDEV_TX_OK; } @@ -909,13 +949,19 @@ static int ipgre_tunnel_bind_dev(struct net_device *dev) /* Guess output device to choose reasonable mtu and needed_headroom */ if (iph->daddr) { - struct flowi fl = { .oif = tunnel->parms.link, - .nl_u = { .ip4_u = - { .daddr = iph->daddr, - .saddr = iph->saddr, - .tos = RT_TOS(iph->tos) } }, - .proto = IPPROTO_GRE }; + struct flowi fl = { + .oif = tunnel->parms.link, + .nl_u = { + .ip4_u = { + .daddr = iph->daddr, + .saddr = iph->saddr, + .tos = RT_TOS(iph->tos) + } + }, + .proto = IPPROTO_GRE + }; struct rtable *rt; + if (!ip_route_output_key(dev_net(dev), &rt, &fl)) { tdev = rt->dst.dev; ip_rt_put(rt); @@ -1012,7 +1058,7 @@ ipgre_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd) break; } } else { - unsigned nflags = 0; + unsigned int nflags = 0; t = netdev_priv(dev); @@ -1125,7 +1171,7 @@ static int ipgre_tunnel_change_mtu(struct net_device *dev, int new_mtu) static int ipgre_header(struct sk_buff *skb, struct net_device *dev, unsigned short type, - const void *daddr, const void *saddr, unsigned len) + const void *daddr, const void *saddr, unsigned int len) { struct ip_tunnel *t = netdev_priv(dev); struct iphdr *iph = (struct iphdr *)skb_push(skb, t->hlen); @@ -1167,13 +1213,19 @@ static int ipgre_open(struct net_device *dev) struct ip_tunnel *t = netdev_priv(dev); if (ipv4_is_multicast(t->parms.iph.daddr)) { - struct flowi fl = { .oif = t->parms.link, - .nl_u = { .ip4_u = - { .daddr = t->parms.iph.daddr, - .saddr = t->parms.iph.saddr, - .tos = RT_TOS(t->parms.iph.tos) } }, - .proto = IPPROTO_GRE }; + struct flowi fl = { + .oif = t->parms.link, + .nl_u = { + .ip4_u = { + .daddr = t->parms.iph.daddr, + .saddr = t->parms.iph.saddr, + .tos = RT_TOS(t->parms.iph.tos) + } + }, + .proto = IPPROTO_GRE + }; struct rtable *rt; + if (ip_route_output_key(dev_net(dev), &rt, &fl)) return -EADDRNOTAVAIL; dev = rt->dst.dev; @@ -1193,10 +1245,8 @@ static int ipgre_close(struct net_device *dev) if (ipv4_is_multicast(t->parms.iph.daddr) && t->mlink) { struct in_device *in_dev; in_dev = inetdev_by_index(dev_net(dev), t->mlink); - if (in_dev) { + if (in_dev) ip_mc_dec_group(in_dev, t->parms.iph.daddr); - in_dev_put(in_dev); - } } return 0; } @@ -1213,12 +1263,19 @@ static const struct net_device_ops ipgre_netdev_ops = { .ndo_start_xmit = ipgre_tunnel_xmit, .ndo_do_ioctl = ipgre_tunnel_ioctl, .ndo_change_mtu = ipgre_tunnel_change_mtu, + .ndo_get_stats = ipgre_get_stats, }; +static void ipgre_dev_free(struct net_device *dev) +{ + free_percpu(dev->tstats); + free_netdev(dev); +} + static void ipgre_tunnel_setup(struct net_device *dev) { dev->netdev_ops = &ipgre_netdev_ops; - dev->destructor = free_netdev; + dev->destructor = ipgre_dev_free; dev->type = ARPHRD_IPGRE; dev->needed_headroom = LL_MAX_HEADER + sizeof(struct iphdr) + 4; @@ -1256,6 +1313,10 @@ static int ipgre_tunnel_init(struct net_device *dev) } else dev->header_ops = &ipgre_header_ops; + dev->tstats = alloc_percpu(struct pcpu_tstats); + if (!dev->tstats) + return -ENOMEM; + return 0; } @@ -1274,14 +1335,13 @@ static void ipgre_fb_tunnel_init(struct net_device *dev) tunnel->hlen = sizeof(struct iphdr) + 4; dev_hold(dev); - ign->tunnels_wc[0] = tunnel; + rcu_assign_pointer(ign->tunnels_wc[0], tunnel); } -static const struct net_protocol ipgre_protocol = { - .handler = ipgre_rcv, - .err_handler = ipgre_err, - .netns_ok = 1, +static const struct gre_protocol ipgre_protocol = { + .handler = ipgre_rcv, + .err_handler = ipgre_err, }; static void ipgre_destroy_tunnels(struct ipgre_net *ign, struct list_head *head) @@ -1291,11 +1351,13 @@ static void ipgre_destroy_tunnels(struct ipgre_net *ign, struct list_head *head) for (prio = 0; prio < 4; prio++) { int h; for (h = 0; h < HASH_SIZE; h++) { - struct ip_tunnel *t = ign->tunnels[prio][h]; + struct ip_tunnel *t; + + t = rtnl_dereference(ign->tunnels[prio][h]); while (t != NULL) { unregister_netdevice_queue(t->dev, head); - t = t->next; + t = rtnl_dereference(t->next); } } } @@ -1441,6 +1503,10 @@ static int ipgre_tap_init(struct net_device *dev) ipgre_tunnel_bind_dev(dev); + dev->tstats = alloc_percpu(struct pcpu_tstats); + if (!dev->tstats) + return -ENOMEM; + return 0; } @@ -1451,6 +1517,7 @@ static const struct net_device_ops ipgre_tap_netdev_ops = { .ndo_set_mac_address = eth_mac_addr, .ndo_validate_addr = eth_validate_addr, .ndo_change_mtu = ipgre_tunnel_change_mtu, + .ndo_get_stats = ipgre_get_stats, }; static void ipgre_tap_setup(struct net_device *dev) @@ -1459,7 +1526,7 @@ static void ipgre_tap_setup(struct net_device *dev) ether_setup(dev); dev->netdev_ops = &ipgre_tap_netdev_ops; - dev->destructor = free_netdev; + dev->destructor = ipgre_dev_free; dev->iflink = 0; dev->features |= NETIF_F_NETNS_LOCAL; @@ -1487,6 +1554,10 @@ static int ipgre_newlink(struct net *src_net, struct net_device *dev, struct nla if (!tb[IFLA_MTU]) dev->mtu = mtu; + /* Can use a lockless transmit, unless we generate output sequences */ + if (!(nt->parms.o_flags & GRE_SEQ)) + dev->features |= NETIF_F_LLTX; + err = register_netdevice(dev); if (err) goto out; @@ -1522,7 +1593,7 @@ static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[], t = nt; if (dev->type != ARPHRD_ETHER) { - unsigned nflags = 0; + unsigned int nflags = 0; if (ipv4_is_multicast(p.iph.daddr)) nflags = IFF_BROADCAST; @@ -1663,7 +1734,7 @@ static int __init ipgre_init(void) if (err < 0) return err; - err = inet_add_protocol(&ipgre_protocol, IPPROTO_GRE); + err = gre_add_protocol(&ipgre_protocol, GREPROTO_CISCO); if (err < 0) { printk(KERN_INFO "ipgre init: can't add protocol\n"); goto add_proto_failed; @@ -1683,7 +1754,7 @@ out: tap_ops_failed: rtnl_link_unregister(&ipgre_link_ops); rtnl_link_failed: - inet_del_protocol(&ipgre_protocol, IPPROTO_GRE); + gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO); add_proto_failed: unregister_pernet_device(&ipgre_net_ops); goto out; @@ -1693,7 +1764,7 @@ static void __exit ipgre_fini(void) { rtnl_link_unregister(&ipgre_tap_ops); rtnl_link_unregister(&ipgre_link_ops); - if (inet_del_protocol(&ipgre_protocol, IPPROTO_GRE) < 0) + if (gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO) < 0) printk(KERN_INFO "ipgre close: can't remove protocol\n"); unregister_pernet_device(&ipgre_net_ops); } diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index ba9836c488e..1906fa35860 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -466,7 +466,7 @@ error: } return -EINVAL; } - +EXPORT_SYMBOL(ip_options_compile); /* * Undo all the changes done by ip_options_compile(). @@ -646,3 +646,4 @@ int ip_options_rcv_srr(struct sk_buff *skb) } return 0; } +EXPORT_SYMBOL(ip_options_rcv_srr); diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 04b69896df5..439d2a34ee4 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -487,10 +487,9 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) * LATER: this step can be merged to real generation of fragments, * we can switch to copy when see the first bad fragment. */ - if (skb_has_frags(skb)) { - struct sk_buff *frag; + if (skb_has_frag_list(skb)) { + struct sk_buff *frag, *frag2; int first_len = skb_pagelen(skb); - int truesizes = 0; if (first_len - hlen > mtu || ((first_len - hlen) & 7) || @@ -503,18 +502,18 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) if (frag->len > mtu || ((frag->len & 7) && frag->next) || skb_headroom(frag) < hlen) - goto slow_path; + goto slow_path_clean; /* Partially cloned skb? */ if (skb_shared(frag)) - goto slow_path; + goto slow_path_clean; BUG_ON(frag->sk); if (skb->sk) { frag->sk = skb->sk; frag->destructor = sock_wfree; } - truesizes += frag->truesize; + skb->truesize -= frag->truesize; } /* Everything is OK. Generate! */ @@ -524,7 +523,6 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) frag = skb_shinfo(skb)->frag_list; skb_frag_list_init(skb); skb->data_len = first_len - skb_headlen(skb); - skb->truesize -= truesizes; skb->len = first_len; iph->tot_len = htons(first_len); iph->frag_off = htons(IP_MF); @@ -576,6 +574,15 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) } IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS); return err; + +slow_path_clean: + skb_walk_frags(skb, frag2) { + if (frag2 == frag) + break; + frag2->sk = NULL; + frag2->destructor = NULL; + skb->truesize += frag2->truesize; + } } slow_path: @@ -837,10 +844,9 @@ int ip_append_data(struct sock *sk, inet->cork.length = 0; sk->sk_sndmsg_page = NULL; sk->sk_sndmsg_off = 0; - if ((exthdrlen = rt->dst.header_len) != 0) { - length += exthdrlen; - transhdrlen += exthdrlen; - } + exthdrlen = rt->dst.header_len; + length += exthdrlen; + transhdrlen += exthdrlen; } else { rt = (struct rtable *)inet->cork.dst; if (inet->cork.flags & IPCORK_OPT) @@ -927,16 +933,19 @@ alloc_new_skb: !(rt->dst.dev->features&NETIF_F_SG)) alloclen = mtu; else - alloclen = datalen + fragheaderlen; + alloclen = fraglen; /* The last fragment gets additional space at tail. * Note, with MSG_MORE we overallocate on fragments, * because we have no idea what fragment will be * the last. */ - if (datalen == length + fraggap) + if (datalen == length + fraggap) { alloclen += rt->dst.trailer_len; - + /* make sure mtu is not reached */ + if (datalen > mtu - fragheaderlen - rt->dst.trailer_len) + datalen -= ALIGN(rt->dst.trailer_len, 8); + } if (transhdrlen) { skb = sock_alloc_send_skb(sk, alloclen + hh_len + 15, @@ -953,7 +962,7 @@ alloc_new_skb: else /* only the initial fragment is time stamped */ - ipc->shtx.flags = 0; + ipc->tx_flags = 0; } if (skb == NULL) goto error; @@ -964,7 +973,7 @@ alloc_new_skb: skb->ip_summed = csummode; skb->csum = 0; skb_reserve(skb, hh_len); - *skb_tx(skb) = ipc->shtx; + skb_shinfo(skb)->tx_flags = ipc->tx_flags; /* * Find where to start putting bytes. @@ -1384,7 +1393,7 @@ void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *ar daddr = ipc.addr = rt->rt_src; ipc.opt = NULL; - ipc.shtx.flags = 0; + ipc.tx_flags = 0; if (replyopts.opt.optlen) { ipc.opt = &replyopts.opt; diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index 6c40a8c46e7..64b70ad162e 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -1129,6 +1129,9 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, case IP_HDRINCL: val = inet->hdrincl; break; + case IP_NODEFRAG: + val = inet->nodefrag; + break; case IP_MTU_DISCOVER: val = inet->pmtudisc; break; diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index ec036731a70..e9b816e6cd7 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -122,31 +122,59 @@ static int ipip_net_id __read_mostly; struct ipip_net { - struct ip_tunnel *tunnels_r_l[HASH_SIZE]; - struct ip_tunnel *tunnels_r[HASH_SIZE]; - struct ip_tunnel *tunnels_l[HASH_SIZE]; - struct ip_tunnel *tunnels_wc[1]; - struct ip_tunnel **tunnels[4]; + struct ip_tunnel __rcu *tunnels_r_l[HASH_SIZE]; + struct ip_tunnel __rcu *tunnels_r[HASH_SIZE]; + struct ip_tunnel __rcu *tunnels_l[HASH_SIZE]; + struct ip_tunnel __rcu *tunnels_wc[1]; + struct ip_tunnel __rcu **tunnels[4]; struct net_device *fb_tunnel_dev; }; -static void ipip_tunnel_init(struct net_device *dev); +static int ipip_tunnel_init(struct net_device *dev); static void ipip_tunnel_setup(struct net_device *dev); +static void ipip_dev_free(struct net_device *dev); /* - * Locking : hash tables are protected by RCU and a spinlock + * Locking : hash tables are protected by RCU and RTNL */ -static DEFINE_SPINLOCK(ipip_lock); #define for_each_ip_tunnel_rcu(start) \ for (t = rcu_dereference(start); t; t = rcu_dereference(t->next)) +/* often modified stats are per cpu, other are shared (netdev->stats) */ +struct pcpu_tstats { + unsigned long rx_packets; + unsigned long rx_bytes; + unsigned long tx_packets; + unsigned long tx_bytes; +}; + +static struct net_device_stats *ipip_get_stats(struct net_device *dev) +{ + struct pcpu_tstats sum = { 0 }; + int i; + + for_each_possible_cpu(i) { + const struct pcpu_tstats *tstats = per_cpu_ptr(dev->tstats, i); + + sum.rx_packets += tstats->rx_packets; + sum.rx_bytes += tstats->rx_bytes; + sum.tx_packets += tstats->tx_packets; + sum.tx_bytes += tstats->tx_bytes; + } + dev->stats.rx_packets = sum.rx_packets; + dev->stats.rx_bytes = sum.rx_bytes; + dev->stats.tx_packets = sum.tx_packets; + dev->stats.tx_bytes = sum.tx_bytes; + return &dev->stats; +} + static struct ip_tunnel * ipip_tunnel_lookup(struct net *net, __be32 remote, __be32 local) { - unsigned h0 = HASH(remote); - unsigned h1 = HASH(local); + unsigned int h0 = HASH(remote); + unsigned int h1 = HASH(local); struct ip_tunnel *t; struct ipip_net *ipn = net_generic(net, ipip_net_id); @@ -169,12 +197,12 @@ static struct ip_tunnel * ipip_tunnel_lookup(struct net *net, return NULL; } -static struct ip_tunnel **__ipip_bucket(struct ipip_net *ipn, +static struct ip_tunnel __rcu **__ipip_bucket(struct ipip_net *ipn, struct ip_tunnel_parm *parms) { __be32 remote = parms->iph.daddr; __be32 local = parms->iph.saddr; - unsigned h = 0; + unsigned int h = 0; int prio = 0; if (remote) { @@ -188,7 +216,7 @@ static struct ip_tunnel **__ipip_bucket(struct ipip_net *ipn, return &ipn->tunnels[prio][h]; } -static inline struct ip_tunnel **ipip_bucket(struct ipip_net *ipn, +static inline struct ip_tunnel __rcu **ipip_bucket(struct ipip_net *ipn, struct ip_tunnel *t) { return __ipip_bucket(ipn, &t->parms); @@ -196,13 +224,14 @@ static inline struct ip_tunnel **ipip_bucket(struct ipip_net *ipn, static void ipip_tunnel_unlink(struct ipip_net *ipn, struct ip_tunnel *t) { - struct ip_tunnel **tp; - - for (tp = ipip_bucket(ipn, t); *tp; tp = &(*tp)->next) { - if (t == *tp) { - spin_lock_bh(&ipip_lock); - *tp = t->next; - spin_unlock_bh(&ipip_lock); + struct ip_tunnel __rcu **tp; + struct ip_tunnel *iter; + + for (tp = ipip_bucket(ipn, t); + (iter = rtnl_dereference(*tp)) != NULL; + tp = &iter->next) { + if (t == iter) { + rcu_assign_pointer(*tp, t->next); break; } } @@ -210,12 +239,10 @@ static void ipip_tunnel_unlink(struct ipip_net *ipn, struct ip_tunnel *t) static void ipip_tunnel_link(struct ipip_net *ipn, struct ip_tunnel *t) { - struct ip_tunnel **tp = ipip_bucket(ipn, t); + struct ip_tunnel __rcu **tp = ipip_bucket(ipn, t); - spin_lock_bh(&ipip_lock); - t->next = *tp; + rcu_assign_pointer(t->next, rtnl_dereference(*tp)); rcu_assign_pointer(*tp, t); - spin_unlock_bh(&ipip_lock); } static struct ip_tunnel * ipip_tunnel_locate(struct net *net, @@ -223,12 +250,15 @@ static struct ip_tunnel * ipip_tunnel_locate(struct net *net, { __be32 remote = parms->iph.daddr; __be32 local = parms->iph.saddr; - struct ip_tunnel *t, **tp, *nt; + struct ip_tunnel *t, *nt; + struct ip_tunnel __rcu **tp; struct net_device *dev; char name[IFNAMSIZ]; struct ipip_net *ipn = net_generic(net, ipip_net_id); - for (tp = __ipip_bucket(ipn, parms); (t = *tp) != NULL; tp = &t->next) { + for (tp = __ipip_bucket(ipn, parms); + (t = rtnl_dereference(*tp)) != NULL; + tp = &t->next) { if (local == t->parms.iph.saddr && remote == t->parms.iph.daddr) return t; } @@ -238,7 +268,7 @@ static struct ip_tunnel * ipip_tunnel_locate(struct net *net, if (parms->name[0]) strlcpy(name, parms->name, IFNAMSIZ); else - sprintf(name, "tunl%%d"); + strcpy(name, "tunl%d"); dev = alloc_netdev(sizeof(*t), name, ipip_tunnel_setup); if (dev == NULL) @@ -254,7 +284,8 @@ static struct ip_tunnel * ipip_tunnel_locate(struct net *net, nt = netdev_priv(dev); nt->parms = *parms; - ipip_tunnel_init(dev); + if (ipip_tunnel_init(dev) < 0) + goto failed_free; if (register_netdevice(dev) < 0) goto failed_free; @@ -264,20 +295,19 @@ static struct ip_tunnel * ipip_tunnel_locate(struct net *net, return nt; failed_free: - free_netdev(dev); + ipip_dev_free(dev); return NULL; } +/* called with RTNL */ static void ipip_tunnel_uninit(struct net_device *dev) { struct net *net = dev_net(dev); struct ipip_net *ipn = net_generic(net, ipip_net_id); - if (dev == ipn->fb_tunnel_dev) { - spin_lock_bh(&ipip_lock); - ipn->tunnels_wc[0] = NULL; - spin_unlock_bh(&ipip_lock); - } else + if (dev == ipn->fb_tunnel_dev) + rcu_assign_pointer(ipn->tunnels_wc[0], NULL); + else ipip_tunnel_unlink(ipn, netdev_priv(dev)); dev_put(dev); } @@ -359,8 +389,10 @@ static int ipip_rcv(struct sk_buff *skb) const struct iphdr *iph = ip_hdr(skb); rcu_read_lock(); - if ((tunnel = ipip_tunnel_lookup(dev_net(skb->dev), - iph->saddr, iph->daddr)) != NULL) { + tunnel = ipip_tunnel_lookup(dev_net(skb->dev), iph->saddr, iph->daddr); + if (tunnel != NULL) { + struct pcpu_tstats *tstats; + if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) { rcu_read_unlock(); kfree_skb(skb); @@ -374,10 +406,16 @@ static int ipip_rcv(struct sk_buff *skb) skb->protocol = htons(ETH_P_IP); skb->pkt_type = PACKET_HOST; - skb_tunnel_rx(skb, tunnel->dev); + tstats = this_cpu_ptr(tunnel->dev->tstats); + tstats->rx_packets++; + tstats->rx_bytes += skb->len; + + __skb_tunnel_rx(skb, tunnel->dev); ipip_ecn_decapsulate(iph, skb); + netif_rx(skb); + rcu_read_unlock(); return 0; } @@ -394,13 +432,12 @@ static int ipip_rcv(struct sk_buff *skb) static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); - struct net_device_stats *stats = &dev->stats; - struct netdev_queue *txq = netdev_get_tx_queue(dev, 0); + struct pcpu_tstats *tstats; struct iphdr *tiph = &tunnel->parms.iph; u8 tos = tunnel->parms.iph.tos; __be16 df = tiph->frag_off; struct rtable *rt; /* Route to the other host */ - struct net_device *tdev; /* Device to other host */ + struct net_device *tdev; /* Device to other host */ struct iphdr *old_iph = ip_hdr(skb); struct iphdr *iph; /* Our new IP header */ unsigned int max_headroom; /* The extra header space needed */ @@ -410,13 +447,13 @@ static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if (skb->protocol != htons(ETH_P_IP)) goto tx_error; - if (tos&1) + if (tos & 1) tos = old_iph->tos; if (!dst) { /* NBMA tunnel */ if ((rt = skb_rtable(skb)) == NULL) { - stats->tx_fifo_errors++; + dev->stats.tx_fifo_errors++; goto tx_error; } if ((dst = rt->rt_gateway) == 0) @@ -424,14 +461,20 @@ static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) } { - struct flowi fl = { .oif = tunnel->parms.link, - .nl_u = { .ip4_u = - { .daddr = dst, - .saddr = tiph->saddr, - .tos = RT_TOS(tos) } }, - .proto = IPPROTO_IPIP }; + struct flowi fl = { + .oif = tunnel->parms.link, + .nl_u = { + .ip4_u = { + .daddr = dst, + .saddr = tiph->saddr, + .tos = RT_TOS(tos) + } + }, + .proto = IPPROTO_IPIP + }; + if (ip_route_output_key(dev_net(dev), &rt, &fl)) { - stats->tx_carrier_errors++; + dev->stats.tx_carrier_errors++; goto tx_error_icmp; } } @@ -439,7 +482,7 @@ static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if (tdev == dev) { ip_rt_put(rt); - stats->collisions++; + dev->stats.collisions++; goto tx_error; } @@ -449,7 +492,7 @@ static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) mtu = dst_mtu(&rt->dst) - sizeof(struct iphdr); if (mtu < 68) { - stats->collisions++; + dev->stats.collisions++; ip_rt_put(rt); goto tx_error; } @@ -485,7 +528,7 @@ static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) struct sk_buff *new_skb = skb_realloc_headroom(skb, max_headroom); if (!new_skb) { ip_rt_put(rt); - txq->tx_dropped++; + dev->stats.tx_dropped++; dev_kfree_skb(skb); return NETDEV_TX_OK; } @@ -522,14 +565,14 @@ static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) iph->ttl = old_iph->ttl; nf_reset(skb); - - IPTUNNEL_XMIT(); + tstats = this_cpu_ptr(dev->tstats); + __IPTUNNEL_XMIT(tstats, &dev->stats); return NETDEV_TX_OK; tx_error_icmp: dst_link_failure(skb); tx_error: - stats->tx_errors++; + dev->stats.tx_errors++; dev_kfree_skb(skb); return NETDEV_TX_OK; } @@ -544,13 +587,19 @@ static void ipip_tunnel_bind_dev(struct net_device *dev) iph = &tunnel->parms.iph; if (iph->daddr) { - struct flowi fl = { .oif = tunnel->parms.link, - .nl_u = { .ip4_u = - { .daddr = iph->daddr, - .saddr = iph->saddr, - .tos = RT_TOS(iph->tos) } }, - .proto = IPPROTO_IPIP }; + struct flowi fl = { + .oif = tunnel->parms.link, + .nl_u = { + .ip4_u = { + .daddr = iph->daddr, + .saddr = iph->saddr, + .tos = RT_TOS(iph->tos) + } + }, + .proto = IPPROTO_IPIP + }; struct rtable *rt; + if (!ip_route_output_key(dev_net(dev), &rt, &fl)) { tdev = rt->dst.dev; ip_rt_put(rt); @@ -696,13 +745,19 @@ static const struct net_device_ops ipip_netdev_ops = { .ndo_start_xmit = ipip_tunnel_xmit, .ndo_do_ioctl = ipip_tunnel_ioctl, .ndo_change_mtu = ipip_tunnel_change_mtu, - + .ndo_get_stats = ipip_get_stats, }; +static void ipip_dev_free(struct net_device *dev) +{ + free_percpu(dev->tstats); + free_netdev(dev); +} + static void ipip_tunnel_setup(struct net_device *dev) { dev->netdev_ops = &ipip_netdev_ops; - dev->destructor = free_netdev; + dev->destructor = ipip_dev_free; dev->type = ARPHRD_TUNNEL; dev->hard_header_len = LL_MAX_HEADER + sizeof(struct iphdr); @@ -711,10 +766,11 @@ static void ipip_tunnel_setup(struct net_device *dev) dev->iflink = 0; dev->addr_len = 4; dev->features |= NETIF_F_NETNS_LOCAL; + dev->features |= NETIF_F_LLTX; dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; } -static void ipip_tunnel_init(struct net_device *dev) +static int ipip_tunnel_init(struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); @@ -725,9 +781,15 @@ static void ipip_tunnel_init(struct net_device *dev) memcpy(dev->broadcast, &tunnel->parms.iph.daddr, 4); ipip_tunnel_bind_dev(dev); + + dev->tstats = alloc_percpu(struct pcpu_tstats); + if (!dev->tstats) + return -ENOMEM; + + return 0; } -static void __net_init ipip_fb_tunnel_init(struct net_device *dev) +static int __net_init ipip_fb_tunnel_init(struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); struct iphdr *iph = &tunnel->parms.iph; @@ -740,11 +802,16 @@ static void __net_init ipip_fb_tunnel_init(struct net_device *dev) iph->protocol = IPPROTO_IPIP; iph->ihl = 5; + dev->tstats = alloc_percpu(struct pcpu_tstats); + if (!dev->tstats) + return -ENOMEM; + dev_hold(dev); - ipn->tunnels_wc[0] = tunnel; + rcu_assign_pointer(ipn->tunnels_wc[0], tunnel); + return 0; } -static struct xfrm_tunnel ipip_handler = { +static struct xfrm_tunnel ipip_handler __read_mostly = { .handler = ipip_rcv, .err_handler = ipip_err, .priority = 1, @@ -760,11 +827,12 @@ static void ipip_destroy_tunnels(struct ipip_net *ipn, struct list_head *head) for (prio = 1; prio < 4; prio++) { int h; for (h = 0; h < HASH_SIZE; h++) { - struct ip_tunnel *t = ipn->tunnels[prio][h]; + struct ip_tunnel *t; + t = rtnl_dereference(ipn->tunnels[prio][h]); while (t != NULL) { unregister_netdevice_queue(t->dev, head); - t = t->next; + t = rtnl_dereference(t->next); } } } @@ -789,7 +857,9 @@ static int __net_init ipip_init_net(struct net *net) } dev_net_set(ipn->fb_tunnel_dev, net); - ipip_fb_tunnel_init(ipn->fb_tunnel_dev); + err = ipip_fb_tunnel_init(ipn->fb_tunnel_dev); + if (err) + goto err_reg_dev; if ((err = register_netdev(ipn->fb_tunnel_dev))) goto err_reg_dev; @@ -797,7 +867,7 @@ static int __net_init ipip_init_net(struct net *net) return 0; err_reg_dev: - free_netdev(ipn->fb_tunnel_dev); + ipip_dev_free(ipn->fb_tunnel_dev); err_alloc_dev: /* nothing */ return err; diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 179fcab866f..86dd5691af4 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -75,7 +75,7 @@ struct mr_table { struct net *net; #endif u32 id; - struct sock *mroute_sk; + struct sock __rcu *mroute_sk; struct timer_list ipmr_expire_timer; struct list_head mfc_unres_queue; struct list_head mfc_cache_array[MFC_LINES]; @@ -98,7 +98,7 @@ struct ipmr_result { }; /* Big lock, protecting vif table, mrt cache and mroute socket state. - Note that the changes are semaphored via rtnl_lock. + * Note that the changes are semaphored via rtnl_lock. */ static DEFINE_RWLOCK(mrt_lock); @@ -113,11 +113,11 @@ static DEFINE_RWLOCK(mrt_lock); static DEFINE_SPINLOCK(mfc_unres_lock); /* We return to original Alan's scheme. Hash table of resolved - entries is changed only in process context and protected - with weak lock mrt_lock. Queue of unresolved entries is protected - with strong spinlock mfc_unres_lock. - - In this case data path is free of exclusive locks at all. + * entries is changed only in process context and protected + * with weak lock mrt_lock. Queue of unresolved entries is protected + * with strong spinlock mfc_unres_lock. + * + * In this case data path is free of exclusive locks at all. */ static struct kmem_cache *mrt_cachep __read_mostly; @@ -396,9 +396,9 @@ struct net_device *ipmr_new_tunnel(struct net *net, struct vifctl *v) set_fs(KERNEL_DS); err = ops->ndo_do_ioctl(dev, &ifr, SIOCADDTUNNEL); set_fs(oldfs); - } else + } else { err = -EOPNOTSUPP; - + } dev = NULL; if (err == 0 && @@ -495,7 +495,8 @@ static struct net_device *ipmr_reg_vif(struct net *net, struct mr_table *mrt) dev->iflink = 0; rcu_read_lock(); - if ((in_dev = __in_dev_get_rcu(dev)) == NULL) { + in_dev = __in_dev_get_rcu(dev); + if (!in_dev) { rcu_read_unlock(); goto failure; } @@ -552,9 +553,10 @@ static int vif_delete(struct mr_table *mrt, int vifi, int notify, mrt->mroute_reg_vif_num = -1; #endif - if (vifi+1 == mrt->maxvif) { + if (vifi + 1 == mrt->maxvif) { int tmp; - for (tmp=vifi-1; tmp>=0; tmp--) { + + for (tmp = vifi - 1; tmp >= 0; tmp--) { if (VIF_EXISTS(mrt, tmp)) break; } @@ -565,25 +567,33 @@ static int vif_delete(struct mr_table *mrt, int vifi, int notify, dev_set_allmulti(dev, -1); - if ((in_dev = __in_dev_get_rtnl(dev)) != NULL) { + in_dev = __in_dev_get_rtnl(dev); + if (in_dev) { IPV4_DEVCONF(in_dev->cnf, MC_FORWARDING)--; ip_rt_multicast_event(in_dev); } - if (v->flags&(VIFF_TUNNEL|VIFF_REGISTER) && !notify) + if (v->flags & (VIFF_TUNNEL | VIFF_REGISTER) && !notify) unregister_netdevice_queue(dev, head); dev_put(dev); return 0; } -static inline void ipmr_cache_free(struct mfc_cache *c) +static void ipmr_cache_free_rcu(struct rcu_head *head) { + struct mfc_cache *c = container_of(head, struct mfc_cache, rcu); + kmem_cache_free(mrt_cachep, c); } +static inline void ipmr_cache_free(struct mfc_cache *c) +{ + call_rcu(&c->rcu, ipmr_cache_free_rcu); +} + /* Destroy an unresolved cache entry, killing queued skbs - and reporting error to netlink readers. + * and reporting error to netlink readers. */ static void ipmr_destroy_unres(struct mr_table *mrt, struct mfc_cache *c) @@ -605,8 +615,9 @@ static void ipmr_destroy_unres(struct mr_table *mrt, struct mfc_cache *c) memset(&e->msg, 0, sizeof(e->msg)); rtnl_unicast(skb, net, NETLINK_CB(skb).pid); - } else + } else { kfree_skb(skb); + } } ipmr_cache_free(c); @@ -724,13 +735,13 @@ static int vif_add(struct net *net, struct mr_table *mrt, case 0: if (vifc->vifc_flags == VIFF_USE_IFINDEX) { dev = dev_get_by_index(net, vifc->vifc_lcl_ifindex); - if (dev && dev->ip_ptr == NULL) { + if (dev && __in_dev_get_rtnl(dev) == NULL) { dev_put(dev); return -EADDRNOTAVAIL; } - } else + } else { dev = ip_dev_find(net, vifc->vifc_lcl_addr.s_addr); - + } if (!dev) return -EADDRNOTAVAIL; err = dev_set_allmulti(dev, 1); @@ -743,16 +754,16 @@ static int vif_add(struct net *net, struct mr_table *mrt, return -EINVAL; } - if ((in_dev = __in_dev_get_rtnl(dev)) == NULL) { + in_dev = __in_dev_get_rtnl(dev); + if (!in_dev) { dev_put(dev); return -EADDRNOTAVAIL; } IPV4_DEVCONF(in_dev->cnf, MC_FORWARDING)++; ip_rt_multicast_event(in_dev); - /* - * Fill in the VIF structures - */ + /* Fill in the VIF structures */ + v->rate_limit = vifc->vifc_rate_limit; v->local = vifc->vifc_lcl_addr.s_addr; v->remote = vifc->vifc_rmt_addr.s_addr; @@ -765,14 +776,14 @@ static int vif_add(struct net *net, struct mr_table *mrt, v->pkt_in = 0; v->pkt_out = 0; v->link = dev->ifindex; - if (v->flags&(VIFF_TUNNEL|VIFF_REGISTER)) + if (v->flags & (VIFF_TUNNEL | VIFF_REGISTER)) v->link = dev->iflink; /* And finish update writing critical data */ write_lock_bh(&mrt_lock); v->dev = dev; #ifdef CONFIG_IP_PIMSM - if (v->flags&VIFF_REGISTER) + if (v->flags & VIFF_REGISTER) mrt->mroute_reg_vif_num = vifi; #endif if (vifi+1 > mrt->maxvif) @@ -781,6 +792,7 @@ static int vif_add(struct net *net, struct mr_table *mrt, return 0; } +/* called with rcu_read_lock() */ static struct mfc_cache *ipmr_cache_find(struct mr_table *mrt, __be32 origin, __be32 mcastgrp) @@ -788,7 +800,7 @@ static struct mfc_cache *ipmr_cache_find(struct mr_table *mrt, int line = MFC_HASH(mcastgrp, origin); struct mfc_cache *c; - list_for_each_entry(c, &mrt->mfc_cache_array[line], list) { + list_for_each_entry_rcu(c, &mrt->mfc_cache_array[line], list) { if (c->mfc_origin == origin && c->mfc_mcastgrp == mcastgrp) return c; } @@ -801,19 +813,20 @@ static struct mfc_cache *ipmr_cache_find(struct mr_table *mrt, static struct mfc_cache *ipmr_cache_alloc(void) { struct mfc_cache *c = kmem_cache_zalloc(mrt_cachep, GFP_KERNEL); - if (c == NULL) - return NULL; - c->mfc_un.res.minvif = MAXVIFS; + + if (c) + c->mfc_un.res.minvif = MAXVIFS; return c; } static struct mfc_cache *ipmr_cache_alloc_unres(void) { struct mfc_cache *c = kmem_cache_zalloc(mrt_cachep, GFP_ATOMIC); - if (c == NULL) - return NULL; - skb_queue_head_init(&c->mfc_un.unres.unresolved); - c->mfc_un.unres.expires = jiffies + 10*HZ; + + if (c) { + skb_queue_head_init(&c->mfc_un.unres.unresolved); + c->mfc_un.unres.expires = jiffies + 10*HZ; + } return c; } @@ -827,17 +840,15 @@ static void ipmr_cache_resolve(struct net *net, struct mr_table *mrt, struct sk_buff *skb; struct nlmsgerr *e; - /* - * Play the pending entries through our router - */ + /* Play the pending entries through our router */ while ((skb = __skb_dequeue(&uc->mfc_un.unres.unresolved))) { if (ip_hdr(skb)->version == 0) { struct nlmsghdr *nlh = (struct nlmsghdr *)skb_pull(skb, sizeof(struct iphdr)); if (__ipmr_fill_mroute(mrt, skb, c, NLMSG_DATA(nlh)) > 0) { - nlh->nlmsg_len = (skb_tail_pointer(skb) - - (u8 *)nlh); + nlh->nlmsg_len = skb_tail_pointer(skb) - + (u8 *)nlh; } else { nlh->nlmsg_type = NLMSG_ERROR; nlh->nlmsg_len = NLMSG_LENGTH(sizeof(struct nlmsgerr)); @@ -848,8 +859,9 @@ static void ipmr_cache_resolve(struct net *net, struct mr_table *mrt, } rtnl_unicast(skb, net, NETLINK_CB(skb).pid); - } else + } else { ip_mr_forward(net, mrt, skb, c, 0); + } } } @@ -867,6 +879,7 @@ static int ipmr_cache_report(struct mr_table *mrt, const int ihl = ip_hdrlen(pkt); struct igmphdr *igmp; struct igmpmsg *msg; + struct sock *mroute_sk; int ret; #ifdef CONFIG_IP_PIMSM @@ -882,9 +895,9 @@ static int ipmr_cache_report(struct mr_table *mrt, #ifdef CONFIG_IP_PIMSM if (assert == IGMPMSG_WHOLEPKT) { /* Ugly, but we have no choice with this interface. - Duplicate old header, fix ihl, length etc. - And all this only to mangle msg->im_msgtype and - to set msg->im_mbz to "mbz" :-) + * Duplicate old header, fix ihl, length etc. + * And all this only to mangle msg->im_msgtype and + * to set msg->im_mbz to "mbz" :-) */ skb_push(skb, sizeof(struct iphdr)); skb_reset_network_header(skb); @@ -901,39 +914,38 @@ static int ipmr_cache_report(struct mr_table *mrt, #endif { - /* - * Copy the IP header - */ + /* Copy the IP header */ skb->network_header = skb->tail; skb_put(skb, ihl); skb_copy_to_linear_data(skb, pkt->data, ihl); - ip_hdr(skb)->protocol = 0; /* Flag to the kernel this is a route add */ + ip_hdr(skb)->protocol = 0; /* Flag to the kernel this is a route add */ msg = (struct igmpmsg *)skb_network_header(skb); msg->im_vif = vifi; skb_dst_set(skb, dst_clone(skb_dst(pkt))); - /* - * Add our header - */ + /* Add our header */ - igmp=(struct igmphdr *)skb_put(skb, sizeof(struct igmphdr)); + igmp = (struct igmphdr *)skb_put(skb, sizeof(struct igmphdr)); igmp->type = msg->im_msgtype = assert; - igmp->code = 0; - ip_hdr(skb)->tot_len = htons(skb->len); /* Fix the length */ + igmp->code = 0; + ip_hdr(skb)->tot_len = htons(skb->len); /* Fix the length */ skb->transport_header = skb->network_header; } - if (mrt->mroute_sk == NULL) { + rcu_read_lock(); + mroute_sk = rcu_dereference(mrt->mroute_sk); + if (mroute_sk == NULL) { + rcu_read_unlock(); kfree_skb(skb); return -EINVAL; } - /* - * Deliver to mrouted - */ - ret = sock_queue_rcv_skb(mrt->mroute_sk, skb); + /* Deliver to mrouted */ + + ret = sock_queue_rcv_skb(mroute_sk, skb); + rcu_read_unlock(); if (ret < 0) { if (net_ratelimit()) printk(KERN_WARNING "mroute: pending queue full, dropping entries.\n"); @@ -965,9 +977,7 @@ ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb) } if (!found) { - /* - * Create a new entry if allowable - */ + /* Create a new entry if allowable */ if (atomic_read(&mrt->cache_resolve_queue_len) >= 10 || (c = ipmr_cache_alloc_unres()) == NULL) { @@ -977,16 +987,14 @@ ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb) return -ENOBUFS; } - /* - * Fill in the new cache entry - */ + /* Fill in the new cache entry */ + c->mfc_parent = -1; c->mfc_origin = iph->saddr; c->mfc_mcastgrp = iph->daddr; - /* - * Reflect first query at mrouted. - */ + /* Reflect first query at mrouted. */ + err = ipmr_cache_report(mrt, skb, vifi, IGMPMSG_NOCACHE); if (err < 0) { /* If the report failed throw the cache entry @@ -1006,10 +1014,9 @@ ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb) mod_timer(&mrt->ipmr_expire_timer, c->mfc_un.unres.expires); } - /* - * See if we can append the packet - */ - if (c->mfc_un.unres.unresolved.qlen>3) { + /* See if we can append the packet */ + + if (c->mfc_un.unres.unresolved.qlen > 3) { kfree_skb(skb); err = -ENOBUFS; } else { @@ -1035,9 +1042,7 @@ static int ipmr_mfc_delete(struct mr_table *mrt, struct mfcctl *mfc) list_for_each_entry_safe(c, next, &mrt->mfc_cache_array[line], list) { if (c->mfc_origin == mfc->mfcc_origin.s_addr && c->mfc_mcastgrp == mfc->mfcc_mcastgrp.s_addr) { - write_lock_bh(&mrt_lock); - list_del(&c->list); - write_unlock_bh(&mrt_lock); + list_del_rcu(&c->list); ipmr_cache_free(c); return 0; @@ -1090,9 +1095,7 @@ static int ipmr_mfc_add(struct net *net, struct mr_table *mrt, if (!mrtsock) c->mfc_flags |= MFC_STATIC; - write_lock_bh(&mrt_lock); - list_add(&c->list, &mrt->mfc_cache_array[line]); - write_unlock_bh(&mrt_lock); + list_add_rcu(&c->list, &mrt->mfc_cache_array[line]); /* * Check to see if we resolved a queued list. If so we @@ -1130,26 +1133,21 @@ static void mroute_clean_tables(struct mr_table *mrt) LIST_HEAD(list); struct mfc_cache *c, *next; - /* - * Shut down all active vif entries - */ + /* Shut down all active vif entries */ + for (i = 0; i < mrt->maxvif; i++) { - if (!(mrt->vif_table[i].flags&VIFF_STATIC)) + if (!(mrt->vif_table[i].flags & VIFF_STATIC)) vif_delete(mrt, i, 0, &list); } unregister_netdevice_many(&list); - /* - * Wipe the cache - */ + /* Wipe the cache */ + for (i = 0; i < MFC_LINES; i++) { list_for_each_entry_safe(c, next, &mrt->mfc_cache_array[i], list) { - if (c->mfc_flags&MFC_STATIC) + if (c->mfc_flags & MFC_STATIC) continue; - write_lock_bh(&mrt_lock); - list_del(&c->list); - write_unlock_bh(&mrt_lock); - + list_del_rcu(&c->list); ipmr_cache_free(c); } } @@ -1164,6 +1162,9 @@ static void mroute_clean_tables(struct mr_table *mrt) } } +/* called from ip_ra_control(), before an RCU grace period, + * we dont need to call synchronize_rcu() here + */ static void mrtsock_destruct(struct sock *sk) { struct net *net = sock_net(sk); @@ -1171,13 +1172,9 @@ static void mrtsock_destruct(struct sock *sk) rtnl_lock(); ipmr_for_each_table(mrt, net) { - if (sk == mrt->mroute_sk) { + if (sk == rtnl_dereference(mrt->mroute_sk)) { IPV4_DEVCONF_ALL(net, MC_FORWARDING)--; - - write_lock_bh(&mrt_lock); - mrt->mroute_sk = NULL; - write_unlock_bh(&mrt_lock); - + rcu_assign_pointer(mrt->mroute_sk, NULL); mroute_clean_tables(mrt); } } @@ -1204,7 +1201,8 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsi return -ENOENT; if (optname != MRT_INIT) { - if (sk != mrt->mroute_sk && !capable(CAP_NET_ADMIN)) + if (sk != rcu_dereference_raw(mrt->mroute_sk) && + !capable(CAP_NET_ADMIN)) return -EACCES; } @@ -1217,23 +1215,20 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsi return -ENOPROTOOPT; rtnl_lock(); - if (mrt->mroute_sk) { + if (rtnl_dereference(mrt->mroute_sk)) { rtnl_unlock(); return -EADDRINUSE; } ret = ip_ra_control(sk, 1, mrtsock_destruct); if (ret == 0) { - write_lock_bh(&mrt_lock); - mrt->mroute_sk = sk; - write_unlock_bh(&mrt_lock); - + rcu_assign_pointer(mrt->mroute_sk, sk); IPV4_DEVCONF_ALL(net, MC_FORWARDING)++; } rtnl_unlock(); return ret; case MRT_DONE: - if (sk != mrt->mroute_sk) + if (sk != rcu_dereference_raw(mrt->mroute_sk)) return -EACCES; return ip_ra_control(sk, 0, NULL); case MRT_ADD_VIF: @@ -1246,7 +1241,8 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsi return -ENFILE; rtnl_lock(); if (optname == MRT_ADD_VIF) { - ret = vif_add(net, mrt, &vif, sk == mrt->mroute_sk); + ret = vif_add(net, mrt, &vif, + sk == rtnl_dereference(mrt->mroute_sk)); } else { ret = vif_delete(mrt, vif.vifc_vifi, 0, NULL); } @@ -1267,7 +1263,8 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsi if (optname == MRT_DEL_MFC) ret = ipmr_mfc_delete(mrt, &mfc); else - ret = ipmr_mfc_add(net, mrt, &mfc, sk == mrt->mroute_sk); + ret = ipmr_mfc_add(net, mrt, &mfc, + sk == rtnl_dereference(mrt->mroute_sk)); rtnl_unlock(); return ret; /* @@ -1276,7 +1273,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsi case MRT_ASSERT: { int v; - if (get_user(v,(int __user *)optval)) + if (get_user(v, (int __user *)optval)) return -EFAULT; mrt->mroute_do_assert = (v) ? 1 : 0; return 0; @@ -1286,7 +1283,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsi { int v; - if (get_user(v,(int __user *)optval)) + if (get_user(v, (int __user *)optval)) return -EFAULT; v = (v) ? 1 : 0; @@ -1309,14 +1306,16 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsi return -EINVAL; if (get_user(v, (u32 __user *)optval)) return -EFAULT; - if (sk == mrt->mroute_sk) - return -EBUSY; rtnl_lock(); ret = 0; - if (!ipmr_new_table(net, v)) - ret = -ENOMEM; - raw_sk(sk)->ipmr_table = v; + if (sk == rtnl_dereference(mrt->mroute_sk)) { + ret = -EBUSY; + } else { + if (!ipmr_new_table(net, v)) + ret = -ENOMEM; + raw_sk(sk)->ipmr_table = v; + } rtnl_unlock(); return ret; } @@ -1347,9 +1346,9 @@ int ip_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, int if (optname != MRT_VERSION && #ifdef CONFIG_IP_PIMSM - optname!=MRT_PIM && + optname != MRT_PIM && #endif - optname!=MRT_ASSERT) + optname != MRT_ASSERT) return -ENOPROTOOPT; if (get_user(olr, optlen)) @@ -1416,19 +1415,19 @@ int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg) if (copy_from_user(&sr, arg, sizeof(sr))) return -EFAULT; - read_lock(&mrt_lock); + rcu_read_lock(); c = ipmr_cache_find(mrt, sr.src.s_addr, sr.grp.s_addr); if (c) { sr.pktcnt = c->mfc_un.res.pkt; sr.bytecnt = c->mfc_un.res.bytes; sr.wrong_if = c->mfc_un.res.wrong_if; - read_unlock(&mrt_lock); + rcu_read_unlock(); if (copy_to_user(arg, &sr, sizeof(sr))) return -EFAULT; return 0; } - read_unlock(&mrt_lock); + rcu_read_unlock(); return -EADDRNOTAVAIL; default: return -ENOIOCTLCMD; @@ -1465,7 +1464,7 @@ static struct notifier_block ip_mr_notifier = { }; /* - * Encapsulate a packet by attaching a valid IPIP header to it. + * Encapsulate a packet by attaching a valid IPIP header to it. * This avoids tunnel drivers and other mess and gives us the speed so * important for multicast video. */ @@ -1480,7 +1479,7 @@ static void ip_encap(struct sk_buff *skb, __be32 saddr, __be32 daddr) skb_reset_network_header(skb); iph = ip_hdr(skb); - iph->version = 4; + iph->version = 4; iph->tos = old_iph->tos; iph->ttl = old_iph->ttl; iph->frag_off = 0; @@ -1498,7 +1497,7 @@ static void ip_encap(struct sk_buff *skb, __be32 saddr, __be32 daddr) static inline int ipmr_forward_finish(struct sk_buff *skb) { - struct ip_options * opt = &(IPCB(skb)->opt); + struct ip_options *opt = &(IPCB(skb)->opt); IP_INC_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_OUTFORWDATAGRAMS); @@ -1535,22 +1534,34 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, } #endif - if (vif->flags&VIFF_TUNNEL) { - struct flowi fl = { .oif = vif->link, - .nl_u = { .ip4_u = - { .daddr = vif->remote, - .saddr = vif->local, - .tos = RT_TOS(iph->tos) } }, - .proto = IPPROTO_IPIP }; + if (vif->flags & VIFF_TUNNEL) { + struct flowi fl = { + .oif = vif->link, + .nl_u = { + .ip4_u = { + .daddr = vif->remote, + .saddr = vif->local, + .tos = RT_TOS(iph->tos) + } + }, + .proto = IPPROTO_IPIP + }; + if (ip_route_output_key(net, &rt, &fl)) goto out_free; encap = sizeof(struct iphdr); } else { - struct flowi fl = { .oif = vif->link, - .nl_u = { .ip4_u = - { .daddr = iph->daddr, - .tos = RT_TOS(iph->tos) } }, - .proto = IPPROTO_IPIP }; + struct flowi fl = { + .oif = vif->link, + .nl_u = { + .ip4_u = { + .daddr = iph->daddr, + .tos = RT_TOS(iph->tos) + } + }, + .proto = IPPROTO_IPIP + }; + if (ip_route_output_key(net, &rt, &fl)) goto out_free; } @@ -1559,8 +1570,8 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, if (skb->len+encap > dst_mtu(&rt->dst) && (ntohs(iph->frag_off) & IP_DF)) { /* Do not fragment multicasts. Alas, IPv4 does not - allow to send ICMP, so that packets will disappear - to blackhole. + * allow to send ICMP, so that packets will disappear + * to blackhole. */ IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_FRAGFAILS); @@ -1583,7 +1594,8 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, ip_decrease_ttl(ip_hdr(skb)); /* FIXME: forward and output firewalls used to be called here. - * What do we do with netfilter? -- RR */ + * What do we do with netfilter? -- RR + */ if (vif->flags & VIFF_TUNNEL) { ip_encap(skb, vif->local, vif->remote); /* FIXME: extra output firewall step used to be here. --RR */ @@ -1644,15 +1656,15 @@ static int ip_mr_forward(struct net *net, struct mr_table *mrt, if (skb_rtable(skb)->fl.iif == 0) { /* It is our own packet, looped back. - Very complicated situation... - - The best workaround until routing daemons will be - fixed is not to redistribute packet, if it was - send through wrong interface. It means, that - multicast applications WILL NOT work for - (S,G), which have default multicast route pointing - to wrong oif. In any case, it is not a good - idea to use multicasting applications on router. + * Very complicated situation... + * + * The best workaround until routing daemons will be + * fixed is not to redistribute packet, if it was + * send through wrong interface. It means, that + * multicast applications WILL NOT work for + * (S,G), which have default multicast route pointing + * to wrong oif. In any case, it is not a good + * idea to use multicasting applications on router. */ goto dont_forward; } @@ -1662,9 +1674,9 @@ static int ip_mr_forward(struct net *net, struct mr_table *mrt, if (true_vifi >= 0 && mrt->mroute_do_assert && /* pimsm uses asserts, when switching from RPT to SPT, - so that we cannot check that packet arrived on an oif. - It is bad, but otherwise we would need to move pretty - large chunk of pimd to kernel. Ough... --ANK + * so that we cannot check that packet arrived on an oif. + * It is bad, but otherwise we would need to move pretty + * large chunk of pimd to kernel. Ough... --ANK */ (mrt->mroute_do_pim || cache->mfc_un.res.ttls[true_vifi] < 255) && @@ -1682,10 +1694,12 @@ static int ip_mr_forward(struct net *net, struct mr_table *mrt, /* * Forward the frame */ - for (ct = cache->mfc_un.res.maxvif-1; ct >= cache->mfc_un.res.minvif; ct--) { + for (ct = cache->mfc_un.res.maxvif - 1; + ct >= cache->mfc_un.res.minvif; ct--) { if (ip_hdr(skb)->ttl > cache->mfc_un.res.ttls[ct]) { if (psend != -1) { struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC); + if (skb2) ipmr_queue_xmit(net, mrt, skb2, cache, psend); @@ -1696,6 +1710,7 @@ static int ip_mr_forward(struct net *net, struct mr_table *mrt, if (psend != -1) { if (local) { struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC); + if (skb2) ipmr_queue_xmit(net, mrt, skb2, cache, psend); } else { @@ -1713,6 +1728,7 @@ dont_forward: /* * Multicast packets for forwarding arrive here + * Called with rcu_read_lock(); */ int ip_mr_input(struct sk_buff *skb) @@ -1724,9 +1740,9 @@ int ip_mr_input(struct sk_buff *skb) int err; /* Packet is looped back after forward, it should not be - forwarded second time, but still can be delivered locally. + * forwarded second time, but still can be delivered locally. */ - if (IPCB(skb)->flags&IPSKB_FORWARDED) + if (IPCB(skb)->flags & IPSKB_FORWARDED) goto dont_forward; err = ipmr_fib_lookup(net, &skb_rtable(skb)->fl, &mrt); @@ -1736,28 +1752,28 @@ int ip_mr_input(struct sk_buff *skb) } if (!local) { - if (IPCB(skb)->opt.router_alert) { - if (ip_call_ra_chain(skb)) - return 0; - } else if (ip_hdr(skb)->protocol == IPPROTO_IGMP){ - /* IGMPv1 (and broken IGMPv2 implementations sort of - Cisco IOS <= 11.2(8)) do not put router alert - option to IGMP packets destined to routable - groups. It is very bad, because it means - that we can forward NO IGMP messages. - */ - read_lock(&mrt_lock); - if (mrt->mroute_sk) { - nf_reset(skb); - raw_rcv(mrt->mroute_sk, skb); - read_unlock(&mrt_lock); - return 0; - } - read_unlock(&mrt_lock); + if (IPCB(skb)->opt.router_alert) { + if (ip_call_ra_chain(skb)) + return 0; + } else if (ip_hdr(skb)->protocol == IPPROTO_IGMP) { + /* IGMPv1 (and broken IGMPv2 implementations sort of + * Cisco IOS <= 11.2(8)) do not put router alert + * option to IGMP packets destined to routable + * groups. It is very bad, because it means + * that we can forward NO IGMP messages. + */ + struct sock *mroute_sk; + + mroute_sk = rcu_dereference(mrt->mroute_sk); + if (mroute_sk) { + nf_reset(skb); + raw_rcv(mroute_sk, skb); + return 0; + } } } - read_lock(&mrt_lock); + /* already under rcu_read_lock() */ cache = ipmr_cache_find(mrt, ip_hdr(skb)->saddr, ip_hdr(skb)->daddr); /* @@ -1769,13 +1785,12 @@ int ip_mr_input(struct sk_buff *skb) if (local) { struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC); ip_local_deliver(skb); - if (skb2 == NULL) { - read_unlock(&mrt_lock); + if (skb2 == NULL) return -ENOBUFS; - } skb = skb2; } + read_lock(&mrt_lock); vif = ipmr_find_vif(mrt, skb->dev); if (vif >= 0) { int err2 = ipmr_cache_unresolved(mrt, vif, skb); @@ -1788,8 +1803,8 @@ int ip_mr_input(struct sk_buff *skb) return -ENODEV; } + read_lock(&mrt_lock); ip_mr_forward(net, mrt, skb, cache, local); - read_unlock(&mrt_lock); if (local) @@ -1805,6 +1820,7 @@ dont_forward: } #ifdef CONFIG_IP_PIMSM +/* called with rcu_read_lock() */ static int __pim_rcv(struct mr_table *mrt, struct sk_buff *skb, unsigned int pimlen) { @@ -1813,10 +1829,10 @@ static int __pim_rcv(struct mr_table *mrt, struct sk_buff *skb, encap = (struct iphdr *)(skb_transport_header(skb) + pimlen); /* - Check that: - a. packet is really destinted to a multicast group - b. packet is not a NULL-REGISTER - c. packet is not truncated + * Check that: + * a. packet is really sent to a multicast group + * b. packet is not a NULL-REGISTER + * c. packet is not truncated */ if (!ipv4_is_multicast(encap->daddr) || encap->tot_len == 0 || @@ -1826,26 +1842,23 @@ static int __pim_rcv(struct mr_table *mrt, struct sk_buff *skb, read_lock(&mrt_lock); if (mrt->mroute_reg_vif_num >= 0) reg_dev = mrt->vif_table[mrt->mroute_reg_vif_num].dev; - if (reg_dev) - dev_hold(reg_dev); read_unlock(&mrt_lock); if (reg_dev == NULL) return 1; skb->mac_header = skb->network_header; - skb_pull(skb, (u8*)encap - skb->data); + skb_pull(skb, (u8 *)encap - skb->data); skb_reset_network_header(skb); skb->protocol = htons(ETH_P_IP); - skb->ip_summed = 0; + skb->ip_summed = CHECKSUM_NONE; skb->pkt_type = PACKET_HOST; skb_tunnel_rx(skb, reg_dev); netif_rx(skb); - dev_put(reg_dev); - return 0; + return NET_RX_SUCCESS; } #endif @@ -1854,7 +1867,7 @@ static int __pim_rcv(struct mr_table *mrt, struct sk_buff *skb, * Handle IGMP messages of PIMv1 */ -int pim_rcv_v1(struct sk_buff * skb) +int pim_rcv_v1(struct sk_buff *skb) { struct igmphdr *pim; struct net *net = dev_net(skb->dev); @@ -1881,7 +1894,7 @@ drop: #endif #ifdef CONFIG_IP_PIMSM_V2 -static int pim_rcv(struct sk_buff * skb) +static int pim_rcv(struct sk_buff *skb) { struct pimreghdr *pim; struct net *net = dev_net(skb->dev); @@ -1891,8 +1904,8 @@ static int pim_rcv(struct sk_buff * skb) goto drop; pim = (struct pimreghdr *)skb_transport_header(skb); - if (pim->type != ((PIM_VERSION<<4)|(PIM_REGISTER)) || - (pim->flags&PIM_NULL_REGISTER) || + if (pim->type != ((PIM_VERSION << 4) | (PIM_REGISTER)) || + (pim->flags & PIM_NULL_REGISTER) || (ip_compute_csum((void *)pim, sizeof(*pim)) != 0 && csum_fold(skb_checksum(skb, 0, skb->len, 0)))) goto drop; @@ -1958,28 +1971,33 @@ int ipmr_get_route(struct net *net, if (mrt == NULL) return -ENOENT; - read_lock(&mrt_lock); + rcu_read_lock(); cache = ipmr_cache_find(mrt, rt->rt_src, rt->rt_dst); if (cache == NULL) { struct sk_buff *skb2; struct iphdr *iph; struct net_device *dev; - int vif; + int vif = -1; if (nowait) { - read_unlock(&mrt_lock); + rcu_read_unlock(); return -EAGAIN; } dev = skb->dev; - if (dev == NULL || (vif = ipmr_find_vif(mrt, dev)) < 0) { + read_lock(&mrt_lock); + if (dev) + vif = ipmr_find_vif(mrt, dev); + if (vif < 0) { read_unlock(&mrt_lock); + rcu_read_unlock(); return -ENODEV; } skb2 = skb_clone(skb, GFP_ATOMIC); if (!skb2) { read_unlock(&mrt_lock); + rcu_read_unlock(); return -ENOMEM; } @@ -1992,13 +2010,16 @@ int ipmr_get_route(struct net *net, iph->version = 0; err = ipmr_cache_unresolved(mrt, vif, skb2); read_unlock(&mrt_lock); + rcu_read_unlock(); return err; } - if (!nowait && (rtm->rtm_flags&RTM_F_NOTIFY)) + read_lock(&mrt_lock); + if (!nowait && (rtm->rtm_flags & RTM_F_NOTIFY)) cache->mfc_flags |= MFC_NOTIFY; err = __ipmr_fill_mroute(mrt, skb, cache, rtm); read_unlock(&mrt_lock); + rcu_read_unlock(); return err; } @@ -2050,14 +2071,14 @@ static int ipmr_rtm_dumproute(struct sk_buff *skb, struct netlink_callback *cb) s_h = cb->args[1]; s_e = cb->args[2]; - read_lock(&mrt_lock); + rcu_read_lock(); ipmr_for_each_table(mrt, net) { if (t < s_t) goto next_table; if (t > s_t) s_h = 0; for (h = s_h; h < MFC_LINES; h++) { - list_for_each_entry(mfc, &mrt->mfc_cache_array[h], list) { + list_for_each_entry_rcu(mfc, &mrt->mfc_cache_array[h], list) { if (e < s_e) goto next_entry; if (ipmr_fill_mroute(mrt, skb, @@ -2075,7 +2096,7 @@ next_table: t++; } done: - read_unlock(&mrt_lock); + rcu_read_unlock(); cb->args[2] = e; cb->args[1] = h; @@ -2086,7 +2107,8 @@ done: #ifdef CONFIG_PROC_FS /* - * The /proc interfaces to multicast routing /proc/ip_mr_cache /proc/ip_mr_vif + * The /proc interfaces to multicast routing : + * /proc/net/ip_mr_cache & /proc/net/ip_mr_vif */ struct ipmr_vif_iter { struct seq_net_private p; @@ -2208,14 +2230,14 @@ static struct mfc_cache *ipmr_mfc_seq_idx(struct net *net, struct mr_table *mrt = it->mrt; struct mfc_cache *mfc; - read_lock(&mrt_lock); + rcu_read_lock(); for (it->ct = 0; it->ct < MFC_LINES; it->ct++) { it->cache = &mrt->mfc_cache_array[it->ct]; - list_for_each_entry(mfc, it->cache, list) + list_for_each_entry_rcu(mfc, it->cache, list) if (pos-- == 0) return mfc; } - read_unlock(&mrt_lock); + rcu_read_unlock(); spin_lock_bh(&mfc_unres_lock); it->cache = &mrt->mfc_unres_queue; @@ -2274,7 +2296,7 @@ static void *ipmr_mfc_seq_next(struct seq_file *seq, void *v, loff_t *pos) } /* exhausted cache_array, show unresolved */ - read_unlock(&mrt_lock); + rcu_read_unlock(); it->cache = &mrt->mfc_unres_queue; it->ct = 0; @@ -2282,7 +2304,7 @@ static void *ipmr_mfc_seq_next(struct seq_file *seq, void *v, loff_t *pos) if (!list_empty(it->cache)) return list_first_entry(it->cache, struct mfc_cache, list); - end_of_list: +end_of_list: spin_unlock_bh(&mfc_unres_lock); it->cache = NULL; @@ -2297,7 +2319,7 @@ static void ipmr_mfc_seq_stop(struct seq_file *seq, void *v) if (it->cache == &mrt->mfc_unres_queue) spin_unlock_bh(&mfc_unres_lock); else if (it->cache == &mrt->mfc_cache_array[it->ct]) - read_unlock(&mrt_lock); + rcu_read_unlock(); } static int ipmr_mfc_seq_show(struct seq_file *seq, void *v) @@ -2323,7 +2345,7 @@ static int ipmr_mfc_seq_show(struct seq_file *seq, void *v) mfc->mfc_un.res.bytes, mfc->mfc_un.res.wrong_if); for (n = mfc->mfc_un.res.minvif; - n < mfc->mfc_un.res.maxvif; n++ ) { + n < mfc->mfc_un.res.maxvif; n++) { if (VIF_EXISTS(mrt, n) && mfc->mfc_un.res.ttls[n] < 255) seq_printf(seq, @@ -2421,7 +2443,7 @@ int __init ip_mr_init(void) mrt_cachep = kmem_cache_create("ip_mrt_cache", sizeof(struct mfc_cache), - 0, SLAB_HWCACHE_ALIGN|SLAB_PANIC, + 0, SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL); if (!mrt_cachep) return -ENOMEM; diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index d048275a62c..babd1a2bae5 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -324,10 +324,10 @@ config IP_NF_TARGET_ECN config IP_NF_TARGET_TTL tristate '"TTL" target support' - depends on NETFILTER_ADVANCED + depends on NETFILTER_ADVANCED && IP_NF_MANGLE select NETFILTER_XT_TARGET_HL ---help--- - This is a backwards-compat option for the user's convenience + This is a backwards-compatible option for the user's convenience (e.g. when running oldconfig). It selects CONFIG_NETFILTER_XT_TARGET_HL. diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 6bccba31d13..3cad2591ace 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -72,7 +72,7 @@ static inline int arp_devaddr_compare(const struct arpt_devaddr_info *ap, for (i = 0; i < len; i++) ret |= (hdr_addr[i] ^ ap->addr[i]) & ap->mask[i]; - return (ret != 0); + return ret != 0; } /* @@ -228,7 +228,7 @@ arpt_error(struct sk_buff *skb, const struct xt_action_param *par) return NF_DROP; } -static inline const struct arpt_entry_target * +static inline const struct xt_entry_target * arpt_get_target_c(const struct arpt_entry *e) { return arpt_get_target((struct arpt_entry *)e); @@ -282,7 +282,7 @@ unsigned int arpt_do_table(struct sk_buff *skb, arp = arp_hdr(skb); do { - const struct arpt_entry_target *t; + const struct xt_entry_target *t; if (!arp_packet_match(arp, skb->dev, indev, outdev, &e->arp)) { e = arpt_next_entry(e); @@ -297,10 +297,10 @@ unsigned int arpt_do_table(struct sk_buff *skb, if (!t->u.kernel.target->target) { int v; - v = ((struct arpt_standard_target *)t)->verdict; + v = ((struct xt_standard_target *)t)->verdict; if (v < 0) { /* Pop from stack? */ - if (v != ARPT_RETURN) { + if (v != XT_RETURN) { verdict = (unsigned)(-v) - 1; break; } @@ -332,7 +332,7 @@ unsigned int arpt_do_table(struct sk_buff *skb, /* Target might have changed stuff. */ arp = arp_hdr(skb); - if (verdict == ARPT_CONTINUE) + if (verdict == XT_CONTINUE) e = arpt_next_entry(e); else /* Verdict */ @@ -377,7 +377,7 @@ static int mark_source_chains(const struct xt_table_info *newinfo, e->counters.pcnt = pos; for (;;) { - const struct arpt_standard_target *t + const struct xt_standard_target *t = (void *)arpt_get_target_c(e); int visited = e->comefrom & (1 << hook); @@ -392,13 +392,13 @@ static int mark_source_chains(const struct xt_table_info *newinfo, /* Unconditional return/END. */ if ((e->target_offset == sizeof(struct arpt_entry) && (strcmp(t->target.u.user.name, - ARPT_STANDARD_TARGET) == 0) && + XT_STANDARD_TARGET) == 0) && t->verdict < 0 && unconditional(&e->arp)) || visited) { unsigned int oldpos, size; if ((strcmp(t->target.u.user.name, - ARPT_STANDARD_TARGET) == 0) && + XT_STANDARD_TARGET) == 0) && t->verdict < -NF_MAX_VERDICT - 1) { duprintf("mark_source_chains: bad " "negative verdict (%i)\n", @@ -433,7 +433,7 @@ static int mark_source_chains(const struct xt_table_info *newinfo, int newpos = t->verdict; if (strcmp(t->target.u.user.name, - ARPT_STANDARD_TARGET) == 0 && + XT_STANDARD_TARGET) == 0 && newpos >= 0) { if (newpos > newinfo->size - sizeof(struct arpt_entry)) { @@ -464,14 +464,14 @@ static int mark_source_chains(const struct xt_table_info *newinfo, static inline int check_entry(const struct arpt_entry *e, const char *name) { - const struct arpt_entry_target *t; + const struct xt_entry_target *t; if (!arp_checkentry(&e->arp)) { duprintf("arp_tables: arp check failed %p %s.\n", e, name); return -EINVAL; } - if (e->target_offset + sizeof(struct arpt_entry_target) > e->next_offset) + if (e->target_offset + sizeof(struct xt_entry_target) > e->next_offset) return -EINVAL; t = arpt_get_target_c(e); @@ -483,7 +483,7 @@ static inline int check_entry(const struct arpt_entry *e, const char *name) static inline int check_target(struct arpt_entry *e, const char *name) { - struct arpt_entry_target *t = arpt_get_target(e); + struct xt_entry_target *t = arpt_get_target(e); int ret; struct xt_tgchk_param par = { .table = name, @@ -506,7 +506,7 @@ static inline int check_target(struct arpt_entry *e, const char *name) static inline int find_check_entry(struct arpt_entry *e, const char *name, unsigned int size) { - struct arpt_entry_target *t; + struct xt_entry_target *t; struct xt_target *target; int ret; @@ -536,7 +536,7 @@ out: static bool check_underflow(const struct arpt_entry *e) { - const struct arpt_entry_target *t; + const struct xt_entry_target *t; unsigned int verdict; if (!unconditional(&e->arp)) @@ -544,7 +544,7 @@ static bool check_underflow(const struct arpt_entry *e) t = arpt_get_target_c(e); if (strcmp(t->u.user.name, XT_STANDARD_TARGET) != 0) return false; - verdict = ((struct arpt_standard_target *)t)->verdict; + verdict = ((struct xt_standard_target *)t)->verdict; verdict = -verdict - 1; return verdict == NF_DROP || verdict == NF_ACCEPT; } @@ -566,7 +566,7 @@ static inline int check_entry_size_and_hooks(struct arpt_entry *e, } if (e->next_offset - < sizeof(struct arpt_entry) + sizeof(struct arpt_entry_target)) { + < sizeof(struct arpt_entry) + sizeof(struct xt_entry_target)) { duprintf("checking: element %p size %u\n", e, e->next_offset); return -EINVAL; @@ -598,7 +598,7 @@ static inline int check_entry_size_and_hooks(struct arpt_entry *e, static inline void cleanup_entry(struct arpt_entry *e) { struct xt_tgdtor_param par; - struct arpt_entry_target *t; + struct xt_entry_target *t; t = arpt_get_target(e); par.target = t->u.kernel.target; @@ -735,6 +735,7 @@ static void get_counters(const struct xt_table_info *t, if (cpu == curcpu) continue; i = 0; + local_bh_disable(); xt_info_wrlock(cpu); xt_entry_foreach(iter, t->entries[cpu], t->size) { ADD_COUNTER(counters[i], iter->counters.bcnt, @@ -742,6 +743,7 @@ static void get_counters(const struct xt_table_info *t, ++i; } xt_info_wrunlock(cpu); + local_bh_enable(); } put_cpu(); } @@ -792,7 +794,7 @@ static int copy_entries_to_user(unsigned int total_size, /* FIXME: use iterator macros --RR */ /* ... then go back and fix counters and names */ for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){ - const struct arpt_entry_target *t; + const struct xt_entry_target *t; e = (struct arpt_entry *)(loc_cpu_entry + off); if (copy_to_user(userptr + off @@ -805,7 +807,7 @@ static int copy_entries_to_user(unsigned int total_size, t = arpt_get_target_c(e); if (copy_to_user(userptr + off + e->target_offset - + offsetof(struct arpt_entry_target, + + offsetof(struct xt_entry_target, u.user.name), t->u.kernel.target->name, strlen(t->u.kernel.target->name)+1) != 0) { @@ -842,7 +844,7 @@ static int compat_calc_entry(const struct arpt_entry *e, const struct xt_table_info *info, const void *base, struct xt_table_info *newinfo) { - const struct arpt_entry_target *t; + const struct xt_entry_target *t; unsigned int entry_offset; int off, i, ret; @@ -893,7 +895,7 @@ static int compat_table_info(const struct xt_table_info *info, static int get_info(struct net *net, void __user *user, const int *len, int compat) { - char name[ARPT_TABLE_MAXNAMELEN]; + char name[XT_TABLE_MAXNAMELEN]; struct xt_table *t; int ret; @@ -906,7 +908,7 @@ static int get_info(struct net *net, void __user *user, if (copy_from_user(name, user, sizeof(name)) != 0) return -EFAULT; - name[ARPT_TABLE_MAXNAMELEN-1] = '\0'; + name[XT_TABLE_MAXNAMELEN-1] = '\0'; #ifdef CONFIG_COMPAT if (compat) xt_compat_lock(NFPROTO_ARP); @@ -1202,7 +1204,7 @@ static int do_add_counters(struct net *net, const void __user *user, #ifdef CONFIG_COMPAT static inline void compat_release_entry(struct compat_arpt_entry *e) { - struct arpt_entry_target *t; + struct xt_entry_target *t; t = compat_arpt_get_target(e); module_put(t->u.kernel.target->me); @@ -1218,7 +1220,7 @@ check_compat_entry_size_and_hooks(struct compat_arpt_entry *e, const unsigned int *underflows, const char *name) { - struct arpt_entry_target *t; + struct xt_entry_target *t; struct xt_target *target; unsigned int entry_offset; int ret, off, h; @@ -1286,7 +1288,7 @@ compat_copy_entry_from_user(struct compat_arpt_entry *e, void **dstptr, unsigned int *size, const char *name, struct xt_table_info *newinfo, unsigned char *base) { - struct arpt_entry_target *t; + struct xt_entry_target *t; struct xt_target *target; struct arpt_entry *de; unsigned int origsize; @@ -1418,6 +1420,9 @@ static int translate_compat_table(const char *name, if (ret != 0) break; ++i; + if (strcmp(arpt_get_target(iter1)->u.user.name, + XT_ERROR_TARGET) == 0) + ++newinfo->stacksize; } if (ret) { /* @@ -1469,7 +1474,7 @@ out_unlock: } struct compat_arpt_replace { - char name[ARPT_TABLE_MAXNAMELEN]; + char name[XT_TABLE_MAXNAMELEN]; u32 valid_hooks; u32 num_entries; u32 size; @@ -1562,7 +1567,7 @@ static int compat_copy_entry_to_user(struct arpt_entry *e, void __user **dstptr, struct xt_counters *counters, unsigned int i) { - struct arpt_entry_target *t; + struct xt_entry_target *t; struct compat_arpt_entry __user *ce; u_int16_t target_offset, next_offset; compat_uint_t origsize; @@ -1623,7 +1628,7 @@ static int compat_copy_entries_to_user(unsigned int total_size, } struct compat_arpt_get_entries { - char name[ARPT_TABLE_MAXNAMELEN]; + char name[XT_TABLE_MAXNAMELEN]; compat_uint_t size; struct compat_arpt_entry entrytable[0]; }; @@ -1823,7 +1828,7 @@ void arpt_unregister_table(struct xt_table *table) /* The built-in targets: standard (NULL) and error. */ static struct xt_target arpt_builtin_tg[] __read_mostly = { { - .name = ARPT_STANDARD_TARGET, + .name = XT_STANDARD_TARGET, .targetsize = sizeof(int), .family = NFPROTO_ARP, #ifdef CONFIG_COMPAT @@ -1833,9 +1838,9 @@ static struct xt_target arpt_builtin_tg[] __read_mostly = { #endif }, { - .name = ARPT_ERROR_TARGET, + .name = XT_ERROR_TARGET, .target = arpt_error, - .targetsize = ARPT_FUNCTION_MAXNAMELEN, + .targetsize = XT_FUNCTION_MAXNAMELEN, .family = NFPROTO_ARP, }, }; diff --git a/net/ipv4/netfilter/arpt_mangle.c b/net/ipv4/netfilter/arpt_mangle.c index e1be7dd1171..b8ddcc480ed 100644 --- a/net/ipv4/netfilter/arpt_mangle.c +++ b/net/ipv4/netfilter/arpt_mangle.c @@ -63,7 +63,7 @@ static int checkentry(const struct xt_tgchk_param *par) return false; if (mangle->target != NF_DROP && mangle->target != NF_ACCEPT && - mangle->target != ARPT_CONTINUE) + mangle->target != XT_CONTINUE) return false; return true; } diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index c439721b165..d31b007a6d8 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -186,7 +186,7 @@ static inline bool unconditional(const struct ipt_ip *ip) } /* for const-correctness */ -static inline const struct ipt_entry_target * +static inline const struct xt_entry_target * ipt_get_target_c(const struct ipt_entry *e) { return ipt_get_target((struct ipt_entry *)e); @@ -230,9 +230,9 @@ get_chainname_rulenum(const struct ipt_entry *s, const struct ipt_entry *e, const char *hookname, const char **chainname, const char **comment, unsigned int *rulenum) { - const struct ipt_standard_target *t = (void *)ipt_get_target_c(s); + const struct xt_standard_target *t = (void *)ipt_get_target_c(s); - if (strcmp(t->target.u.kernel.target->name, IPT_ERROR_TARGET) == 0) { + if (strcmp(t->target.u.kernel.target->name, XT_ERROR_TARGET) == 0) { /* Head of user chain: ERROR target with chainname */ *chainname = t->target.data; (*rulenum) = 0; @@ -241,7 +241,7 @@ get_chainname_rulenum(const struct ipt_entry *s, const struct ipt_entry *e, if (s->target_offset == sizeof(struct ipt_entry) && strcmp(t->target.u.kernel.target->name, - IPT_STANDARD_TARGET) == 0 && + XT_STANDARD_TARGET) == 0 && t->verdict < 0 && unconditional(&s->ip)) { /* Tail of chains: STANDARD target (return/policy) */ @@ -346,7 +346,7 @@ ipt_do_table(struct sk_buff *skb, get_entry(table_base, private->underflow[hook])); do { - const struct ipt_entry_target *t; + const struct xt_entry_target *t; const struct xt_entry_match *ematch; IP_NF_ASSERT(e); @@ -380,10 +380,10 @@ ipt_do_table(struct sk_buff *skb, if (!t->u.kernel.target->target) { int v; - v = ((struct ipt_standard_target *)t)->verdict; + v = ((struct xt_standard_target *)t)->verdict; if (v < 0) { /* Pop from stack? */ - if (v != IPT_RETURN) { + if (v != XT_RETURN) { verdict = (unsigned)(-v) - 1; break; } @@ -421,7 +421,7 @@ ipt_do_table(struct sk_buff *skb, verdict = t->u.kernel.target->target(skb, &acpar); /* Target might have changed stuff. */ ip = ip_hdr(skb); - if (verdict == IPT_CONTINUE) + if (verdict == XT_CONTINUE) e = ipt_next_entry(e); else /* Verdict */ @@ -461,7 +461,7 @@ mark_source_chains(const struct xt_table_info *newinfo, e->counters.pcnt = pos; for (;;) { - const struct ipt_standard_target *t + const struct xt_standard_target *t = (void *)ipt_get_target_c(e); int visited = e->comefrom & (1 << hook); @@ -475,13 +475,13 @@ mark_source_chains(const struct xt_table_info *newinfo, /* Unconditional return/END. */ if ((e->target_offset == sizeof(struct ipt_entry) && (strcmp(t->target.u.user.name, - IPT_STANDARD_TARGET) == 0) && + XT_STANDARD_TARGET) == 0) && t->verdict < 0 && unconditional(&e->ip)) || visited) { unsigned int oldpos, size; if ((strcmp(t->target.u.user.name, - IPT_STANDARD_TARGET) == 0) && + XT_STANDARD_TARGET) == 0) && t->verdict < -NF_MAX_VERDICT - 1) { duprintf("mark_source_chains: bad " "negative verdict (%i)\n", @@ -524,7 +524,7 @@ mark_source_chains(const struct xt_table_info *newinfo, int newpos = t->verdict; if (strcmp(t->target.u.user.name, - IPT_STANDARD_TARGET) == 0 && + XT_STANDARD_TARGET) == 0 && newpos >= 0) { if (newpos > newinfo->size - sizeof(struct ipt_entry)) { @@ -552,7 +552,7 @@ mark_source_chains(const struct xt_table_info *newinfo, return 1; } -static void cleanup_match(struct ipt_entry_match *m, struct net *net) +static void cleanup_match(struct xt_entry_match *m, struct net *net) { struct xt_mtdtor_param par; @@ -568,14 +568,14 @@ static void cleanup_match(struct ipt_entry_match *m, struct net *net) static int check_entry(const struct ipt_entry *e, const char *name) { - const struct ipt_entry_target *t; + const struct xt_entry_target *t; if (!ip_checkentry(&e->ip)) { duprintf("ip check failed %p %s.\n", e, par->match->name); return -EINVAL; } - if (e->target_offset + sizeof(struct ipt_entry_target) > + if (e->target_offset + sizeof(struct xt_entry_target) > e->next_offset) return -EINVAL; @@ -587,7 +587,7 @@ check_entry(const struct ipt_entry *e, const char *name) } static int -check_match(struct ipt_entry_match *m, struct xt_mtchk_param *par) +check_match(struct xt_entry_match *m, struct xt_mtchk_param *par) { const struct ipt_ip *ip = par->entryinfo; int ret; @@ -605,7 +605,7 @@ check_match(struct ipt_entry_match *m, struct xt_mtchk_param *par) } static int -find_check_match(struct ipt_entry_match *m, struct xt_mtchk_param *par) +find_check_match(struct xt_entry_match *m, struct xt_mtchk_param *par) { struct xt_match *match; int ret; @@ -630,7 +630,7 @@ err: static int check_target(struct ipt_entry *e, struct net *net, const char *name) { - struct ipt_entry_target *t = ipt_get_target(e); + struct xt_entry_target *t = ipt_get_target(e); struct xt_tgchk_param par = { .net = net, .table = name, @@ -656,7 +656,7 @@ static int find_check_entry(struct ipt_entry *e, struct net *net, const char *name, unsigned int size) { - struct ipt_entry_target *t; + struct xt_entry_target *t; struct xt_target *target; int ret; unsigned int j; @@ -707,7 +707,7 @@ find_check_entry(struct ipt_entry *e, struct net *net, const char *name, static bool check_underflow(const struct ipt_entry *e) { - const struct ipt_entry_target *t; + const struct xt_entry_target *t; unsigned int verdict; if (!unconditional(&e->ip)) @@ -715,7 +715,7 @@ static bool check_underflow(const struct ipt_entry *e) t = ipt_get_target_c(e); if (strcmp(t->u.user.name, XT_STANDARD_TARGET) != 0) return false; - verdict = ((struct ipt_standard_target *)t)->verdict; + verdict = ((struct xt_standard_target *)t)->verdict; verdict = -verdict - 1; return verdict == NF_DROP || verdict == NF_ACCEPT; } @@ -738,7 +738,7 @@ check_entry_size_and_hooks(struct ipt_entry *e, } if (e->next_offset - < sizeof(struct ipt_entry) + sizeof(struct ipt_entry_target)) { + < sizeof(struct ipt_entry) + sizeof(struct xt_entry_target)) { duprintf("checking: element %p size %u\n", e, e->next_offset); return -EINVAL; @@ -771,7 +771,7 @@ static void cleanup_entry(struct ipt_entry *e, struct net *net) { struct xt_tgdtor_param par; - struct ipt_entry_target *t; + struct xt_entry_target *t; struct xt_entry_match *ematch; /* Cleanup all matches */ @@ -909,6 +909,7 @@ get_counters(const struct xt_table_info *t, if (cpu == curcpu) continue; i = 0; + local_bh_disable(); xt_info_wrlock(cpu); xt_entry_foreach(iter, t->entries[cpu], t->size) { ADD_COUNTER(counters[i], iter->counters.bcnt, @@ -916,6 +917,7 @@ get_counters(const struct xt_table_info *t, ++i; /* macro does multi eval of i */ } xt_info_wrunlock(cpu); + local_bh_enable(); } put_cpu(); } @@ -970,8 +972,8 @@ copy_entries_to_user(unsigned int total_size, /* ... then go back and fix counters and names */ for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){ unsigned int i; - const struct ipt_entry_match *m; - const struct ipt_entry_target *t; + const struct xt_entry_match *m; + const struct xt_entry_target *t; e = (struct ipt_entry *)(loc_cpu_entry + off); if (copy_to_user(userptr + off @@ -988,7 +990,7 @@ copy_entries_to_user(unsigned int total_size, m = (void *)e + i; if (copy_to_user(userptr + off + i - + offsetof(struct ipt_entry_match, + + offsetof(struct xt_entry_match, u.user.name), m->u.kernel.match->name, strlen(m->u.kernel.match->name)+1) @@ -1000,7 +1002,7 @@ copy_entries_to_user(unsigned int total_size, t = ipt_get_target_c(e); if (copy_to_user(userptr + off + e->target_offset - + offsetof(struct ipt_entry_target, + + offsetof(struct xt_entry_target, u.user.name), t->u.kernel.target->name, strlen(t->u.kernel.target->name)+1) != 0) { @@ -1038,7 +1040,7 @@ static int compat_calc_entry(const struct ipt_entry *e, const void *base, struct xt_table_info *newinfo) { const struct xt_entry_match *ematch; - const struct ipt_entry_target *t; + const struct xt_entry_target *t; unsigned int entry_offset; int off, i, ret; @@ -1090,7 +1092,7 @@ static int compat_table_info(const struct xt_table_info *info, static int get_info(struct net *net, void __user *user, const int *len, int compat) { - char name[IPT_TABLE_MAXNAMELEN]; + char name[XT_TABLE_MAXNAMELEN]; struct xt_table *t; int ret; @@ -1103,7 +1105,7 @@ static int get_info(struct net *net, void __user *user, if (copy_from_user(name, user, sizeof(name)) != 0) return -EFAULT; - name[IPT_TABLE_MAXNAMELEN-1] = '\0'; + name[XT_TABLE_MAXNAMELEN-1] = '\0'; #ifdef CONFIG_COMPAT if (compat) xt_compat_lock(AF_INET); @@ -1398,14 +1400,14 @@ do_add_counters(struct net *net, const void __user *user, #ifdef CONFIG_COMPAT struct compat_ipt_replace { - char name[IPT_TABLE_MAXNAMELEN]; + char name[XT_TABLE_MAXNAMELEN]; u32 valid_hooks; u32 num_entries; u32 size; u32 hook_entry[NF_INET_NUMHOOKS]; u32 underflow[NF_INET_NUMHOOKS]; u32 num_counters; - compat_uptr_t counters; /* struct ipt_counters * */ + compat_uptr_t counters; /* struct xt_counters * */ struct compat_ipt_entry entries[0]; }; @@ -1414,7 +1416,7 @@ compat_copy_entry_to_user(struct ipt_entry *e, void __user **dstptr, unsigned int *size, struct xt_counters *counters, unsigned int i) { - struct ipt_entry_target *t; + struct xt_entry_target *t; struct compat_ipt_entry __user *ce; u_int16_t target_offset, next_offset; compat_uint_t origsize; @@ -1449,7 +1451,7 @@ compat_copy_entry_to_user(struct ipt_entry *e, void __user **dstptr, } static int -compat_find_calc_match(struct ipt_entry_match *m, +compat_find_calc_match(struct xt_entry_match *m, const char *name, const struct ipt_ip *ip, unsigned int hookmask, @@ -1471,7 +1473,7 @@ compat_find_calc_match(struct ipt_entry_match *m, static void compat_release_entry(struct compat_ipt_entry *e) { - struct ipt_entry_target *t; + struct xt_entry_target *t; struct xt_entry_match *ematch; /* Cleanup all matches */ @@ -1492,7 +1494,7 @@ check_compat_entry_size_and_hooks(struct compat_ipt_entry *e, const char *name) { struct xt_entry_match *ematch; - struct ipt_entry_target *t; + struct xt_entry_target *t; struct xt_target *target; unsigned int entry_offset; unsigned int j; @@ -1574,7 +1576,7 @@ compat_copy_entry_from_user(struct compat_ipt_entry *e, void **dstptr, unsigned int *size, const char *name, struct xt_table_info *newinfo, unsigned char *base) { - struct ipt_entry_target *t; + struct xt_entry_target *t; struct xt_target *target; struct ipt_entry *de; unsigned int origsize; @@ -1749,6 +1751,9 @@ translate_compat_table(struct net *net, if (ret != 0) break; ++i; + if (strcmp(ipt_get_target(iter1)->u.user.name, + XT_ERROR_TARGET) == 0) + ++newinfo->stacksize; } if (ret) { /* @@ -1879,7 +1884,7 @@ compat_do_ipt_set_ctl(struct sock *sk, int cmd, void __user *user, } struct compat_ipt_get_entries { - char name[IPT_TABLE_MAXNAMELEN]; + char name[XT_TABLE_MAXNAMELEN]; compat_uint_t size; struct compat_ipt_entry entrytable[0]; }; @@ -2034,7 +2039,7 @@ do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) case IPT_SO_GET_REVISION_MATCH: case IPT_SO_GET_REVISION_TARGET: { - struct ipt_get_revision rev; + struct xt_get_revision rev; int target; if (*len != sizeof(rev)) { @@ -2171,7 +2176,7 @@ static int icmp_checkentry(const struct xt_mtchk_param *par) static struct xt_target ipt_builtin_tg[] __read_mostly = { { - .name = IPT_STANDARD_TARGET, + .name = XT_STANDARD_TARGET, .targetsize = sizeof(int), .family = NFPROTO_IPV4, #ifdef CONFIG_COMPAT @@ -2181,9 +2186,9 @@ static struct xt_target ipt_builtin_tg[] __read_mostly = { #endif }, { - .name = IPT_ERROR_TARGET, + .name = XT_ERROR_TARGET, .target = ipt_error, - .targetsize = IPT_FUNCTION_MAXNAMELEN, + .targetsize = XT_FUNCTION_MAXNAMELEN, .family = NFPROTO_IPV4, }, }; diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index 3a43cf36db8..1e26a489765 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -29,6 +29,7 @@ #include <net/netfilter/nf_conntrack.h> #include <net/net_namespace.h> #include <net/checksum.h> +#include <net/ip.h> #define CLUSTERIP_VERSION "0.8" @@ -231,24 +232,22 @@ clusterip_hashfn(const struct sk_buff *skb, { const struct iphdr *iph = ip_hdr(skb); unsigned long hashval; - u_int16_t sport, dport; - const u_int16_t *ports; - - switch (iph->protocol) { - case IPPROTO_TCP: - case IPPROTO_UDP: - case IPPROTO_UDPLITE: - case IPPROTO_SCTP: - case IPPROTO_DCCP: - case IPPROTO_ICMP: - ports = (const void *)iph+iph->ihl*4; - sport = ports[0]; - dport = ports[1]; - break; - default: + u_int16_t sport = 0, dport = 0; + int poff; + + poff = proto_ports_offset(iph->protocol); + if (poff >= 0) { + const u_int16_t *ports; + u16 _ports[2]; + + ports = skb_header_pointer(skb, iph->ihl * 4 + poff, 4, _ports); + if (ports) { + sport = ports[0]; + dport = ports[1]; + } + } else { if (net_ratelimit()) pr_info("unknown protocol %u\n", iph->protocol); - sport = dport = 0; } switch (config->hash_mode) { diff --git a/net/ipv4/netfilter/ipt_LOG.c b/net/ipv4/netfilter/ipt_LOG.c index 915fc17d7ce..72ffc8fda2e 100644 --- a/net/ipv4/netfilter/ipt_LOG.c +++ b/net/ipv4/netfilter/ipt_LOG.c @@ -24,16 +24,15 @@ #include <linux/netfilter/x_tables.h> #include <linux/netfilter_ipv4/ipt_LOG.h> #include <net/netfilter/nf_log.h> +#include <net/netfilter/xt_log.h> MODULE_LICENSE("GPL"); MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>"); MODULE_DESCRIPTION("Xtables: IPv4 packet logging to syslog"); -/* Use lock to serialize, so printks don't overlap */ -static DEFINE_SPINLOCK(log_lock); - /* One level of recursion won't kill us */ -static void dump_packet(const struct nf_loginfo *info, +static void dump_packet(struct sbuff *m, + const struct nf_loginfo *info, const struct sk_buff *skb, unsigned int iphoff) { @@ -48,32 +47,32 @@ static void dump_packet(const struct nf_loginfo *info, ih = skb_header_pointer(skb, iphoff, sizeof(_iph), &_iph); if (ih == NULL) { - printk("TRUNCATED"); + sb_add(m, "TRUNCATED"); return; } /* Important fields: * TOS, len, DF/MF, fragment offset, TTL, src, dst, options. */ /* Max length: 40 "SRC=255.255.255.255 DST=255.255.255.255 " */ - printk("SRC=%pI4 DST=%pI4 ", + sb_add(m, "SRC=%pI4 DST=%pI4 ", &ih->saddr, &ih->daddr); /* Max length: 46 "LEN=65535 TOS=0xFF PREC=0xFF TTL=255 ID=65535 " */ - printk("LEN=%u TOS=0x%02X PREC=0x%02X TTL=%u ID=%u ", + sb_add(m, "LEN=%u TOS=0x%02X PREC=0x%02X TTL=%u ID=%u ", ntohs(ih->tot_len), ih->tos & IPTOS_TOS_MASK, ih->tos & IPTOS_PREC_MASK, ih->ttl, ntohs(ih->id)); /* Max length: 6 "CE DF MF " */ if (ntohs(ih->frag_off) & IP_CE) - printk("CE "); + sb_add(m, "CE "); if (ntohs(ih->frag_off) & IP_DF) - printk("DF "); + sb_add(m, "DF "); if (ntohs(ih->frag_off) & IP_MF) - printk("MF "); + sb_add(m, "MF "); /* Max length: 11 "FRAG:65535 " */ if (ntohs(ih->frag_off) & IP_OFFSET) - printk("FRAG:%u ", ntohs(ih->frag_off) & IP_OFFSET); + sb_add(m, "FRAG:%u ", ntohs(ih->frag_off) & IP_OFFSET); if ((logflags & IPT_LOG_IPOPT) && ih->ihl * 4 > sizeof(struct iphdr)) { @@ -85,15 +84,15 @@ static void dump_packet(const struct nf_loginfo *info, op = skb_header_pointer(skb, iphoff+sizeof(_iph), optsize, _opt); if (op == NULL) { - printk("TRUNCATED"); + sb_add(m, "TRUNCATED"); return; } /* Max length: 127 "OPT (" 15*4*2chars ") " */ - printk("OPT ("); + sb_add(m, "OPT ("); for (i = 0; i < optsize; i++) - printk("%02X", op[i]); - printk(") "); + sb_add(m, "%02X", op[i]); + sb_add(m, ") "); } switch (ih->protocol) { @@ -102,7 +101,7 @@ static void dump_packet(const struct nf_loginfo *info, const struct tcphdr *th; /* Max length: 10 "PROTO=TCP " */ - printk("PROTO=TCP "); + sb_add(m, "PROTO=TCP "); if (ntohs(ih->frag_off) & IP_OFFSET) break; @@ -111,41 +110,41 @@ static void dump_packet(const struct nf_loginfo *info, th = skb_header_pointer(skb, iphoff + ih->ihl * 4, sizeof(_tcph), &_tcph); if (th == NULL) { - printk("INCOMPLETE [%u bytes] ", + sb_add(m, "INCOMPLETE [%u bytes] ", skb->len - iphoff - ih->ihl*4); break; } /* Max length: 20 "SPT=65535 DPT=65535 " */ - printk("SPT=%u DPT=%u ", + sb_add(m, "SPT=%u DPT=%u ", ntohs(th->source), ntohs(th->dest)); /* Max length: 30 "SEQ=4294967295 ACK=4294967295 " */ if (logflags & IPT_LOG_TCPSEQ) - printk("SEQ=%u ACK=%u ", + sb_add(m, "SEQ=%u ACK=%u ", ntohl(th->seq), ntohl(th->ack_seq)); /* Max length: 13 "WINDOW=65535 " */ - printk("WINDOW=%u ", ntohs(th->window)); + sb_add(m, "WINDOW=%u ", ntohs(th->window)); /* Max length: 9 "RES=0x3F " */ - printk("RES=0x%02x ", (u8)(ntohl(tcp_flag_word(th) & TCP_RESERVED_BITS) >> 22)); + sb_add(m, "RES=0x%02x ", (u8)(ntohl(tcp_flag_word(th) & TCP_RESERVED_BITS) >> 22)); /* Max length: 32 "CWR ECE URG ACK PSH RST SYN FIN " */ if (th->cwr) - printk("CWR "); + sb_add(m, "CWR "); if (th->ece) - printk("ECE "); + sb_add(m, "ECE "); if (th->urg) - printk("URG "); + sb_add(m, "URG "); if (th->ack) - printk("ACK "); + sb_add(m, "ACK "); if (th->psh) - printk("PSH "); + sb_add(m, "PSH "); if (th->rst) - printk("RST "); + sb_add(m, "RST "); if (th->syn) - printk("SYN "); + sb_add(m, "SYN "); if (th->fin) - printk("FIN "); + sb_add(m, "FIN "); /* Max length: 11 "URGP=65535 " */ - printk("URGP=%u ", ntohs(th->urg_ptr)); + sb_add(m, "URGP=%u ", ntohs(th->urg_ptr)); if ((logflags & IPT_LOG_TCPOPT) && th->doff * 4 > sizeof(struct tcphdr)) { @@ -158,15 +157,15 @@ static void dump_packet(const struct nf_loginfo *info, iphoff+ih->ihl*4+sizeof(_tcph), optsize, _opt); if (op == NULL) { - printk("TRUNCATED"); + sb_add(m, "TRUNCATED"); return; } /* Max length: 127 "OPT (" 15*4*2chars ") " */ - printk("OPT ("); + sb_add(m, "OPT ("); for (i = 0; i < optsize; i++) - printk("%02X", op[i]); - printk(") "); + sb_add(m, "%02X", op[i]); + sb_add(m, ") "); } break; } @@ -177,9 +176,9 @@ static void dump_packet(const struct nf_loginfo *info, if (ih->protocol == IPPROTO_UDP) /* Max length: 10 "PROTO=UDP " */ - printk("PROTO=UDP " ); + sb_add(m, "PROTO=UDP " ); else /* Max length: 14 "PROTO=UDPLITE " */ - printk("PROTO=UDPLITE "); + sb_add(m, "PROTO=UDPLITE "); if (ntohs(ih->frag_off) & IP_OFFSET) break; @@ -188,13 +187,13 @@ static void dump_packet(const struct nf_loginfo *info, uh = skb_header_pointer(skb, iphoff+ih->ihl*4, sizeof(_udph), &_udph); if (uh == NULL) { - printk("INCOMPLETE [%u bytes] ", + sb_add(m, "INCOMPLETE [%u bytes] ", skb->len - iphoff - ih->ihl*4); break; } /* Max length: 20 "SPT=65535 DPT=65535 " */ - printk("SPT=%u DPT=%u LEN=%u ", + sb_add(m, "SPT=%u DPT=%u LEN=%u ", ntohs(uh->source), ntohs(uh->dest), ntohs(uh->len)); break; @@ -221,7 +220,7 @@ static void dump_packet(const struct nf_loginfo *info, [ICMP_ADDRESSREPLY] = 12 }; /* Max length: 11 "PROTO=ICMP " */ - printk("PROTO=ICMP "); + sb_add(m, "PROTO=ICMP "); if (ntohs(ih->frag_off) & IP_OFFSET) break; @@ -230,19 +229,19 @@ static void dump_packet(const struct nf_loginfo *info, ich = skb_header_pointer(skb, iphoff + ih->ihl * 4, sizeof(_icmph), &_icmph); if (ich == NULL) { - printk("INCOMPLETE [%u bytes] ", + sb_add(m, "INCOMPLETE [%u bytes] ", skb->len - iphoff - ih->ihl*4); break; } /* Max length: 18 "TYPE=255 CODE=255 " */ - printk("TYPE=%u CODE=%u ", ich->type, ich->code); + sb_add(m, "TYPE=%u CODE=%u ", ich->type, ich->code); /* Max length: 25 "INCOMPLETE [65535 bytes] " */ if (ich->type <= NR_ICMP_TYPES && required_len[ich->type] && skb->len-iphoff-ih->ihl*4 < required_len[ich->type]) { - printk("INCOMPLETE [%u bytes] ", + sb_add(m, "INCOMPLETE [%u bytes] ", skb->len - iphoff - ih->ihl*4); break; } @@ -251,35 +250,35 @@ static void dump_packet(const struct nf_loginfo *info, case ICMP_ECHOREPLY: case ICMP_ECHO: /* Max length: 19 "ID=65535 SEQ=65535 " */ - printk("ID=%u SEQ=%u ", + sb_add(m, "ID=%u SEQ=%u ", ntohs(ich->un.echo.id), ntohs(ich->un.echo.sequence)); break; case ICMP_PARAMETERPROB: /* Max length: 14 "PARAMETER=255 " */ - printk("PARAMETER=%u ", + sb_add(m, "PARAMETER=%u ", ntohl(ich->un.gateway) >> 24); break; case ICMP_REDIRECT: /* Max length: 24 "GATEWAY=255.255.255.255 " */ - printk("GATEWAY=%pI4 ", &ich->un.gateway); + sb_add(m, "GATEWAY=%pI4 ", &ich->un.gateway); /* Fall through */ case ICMP_DEST_UNREACH: case ICMP_SOURCE_QUENCH: case ICMP_TIME_EXCEEDED: /* Max length: 3+maxlen */ if (!iphoff) { /* Only recurse once. */ - printk("["); - dump_packet(info, skb, + sb_add(m, "["); + dump_packet(m, info, skb, iphoff + ih->ihl*4+sizeof(_icmph)); - printk("] "); + sb_add(m, "] "); } /* Max length: 10 "MTU=65535 " */ if (ich->type == ICMP_DEST_UNREACH && ich->code == ICMP_FRAG_NEEDED) - printk("MTU=%u ", ntohs(ich->un.frag.mtu)); + sb_add(m, "MTU=%u ", ntohs(ich->un.frag.mtu)); } break; } @@ -292,19 +291,19 @@ static void dump_packet(const struct nf_loginfo *info, break; /* Max length: 9 "PROTO=AH " */ - printk("PROTO=AH "); + sb_add(m, "PROTO=AH "); /* Max length: 25 "INCOMPLETE [65535 bytes] " */ ah = skb_header_pointer(skb, iphoff+ih->ihl*4, sizeof(_ahdr), &_ahdr); if (ah == NULL) { - printk("INCOMPLETE [%u bytes] ", + sb_add(m, "INCOMPLETE [%u bytes] ", skb->len - iphoff - ih->ihl*4); break; } /* Length: 15 "SPI=0xF1234567 " */ - printk("SPI=0x%x ", ntohl(ah->spi)); + sb_add(m, "SPI=0x%x ", ntohl(ah->spi)); break; } case IPPROTO_ESP: { @@ -312,7 +311,7 @@ static void dump_packet(const struct nf_loginfo *info, const struct ip_esp_hdr *eh; /* Max length: 10 "PROTO=ESP " */ - printk("PROTO=ESP "); + sb_add(m, "PROTO=ESP "); if (ntohs(ih->frag_off) & IP_OFFSET) break; @@ -321,25 +320,25 @@ static void dump_packet(const struct nf_loginfo *info, eh = skb_header_pointer(skb, iphoff+ih->ihl*4, sizeof(_esph), &_esph); if (eh == NULL) { - printk("INCOMPLETE [%u bytes] ", + sb_add(m, "INCOMPLETE [%u bytes] ", skb->len - iphoff - ih->ihl*4); break; } /* Length: 15 "SPI=0xF1234567 " */ - printk("SPI=0x%x ", ntohl(eh->spi)); + sb_add(m, "SPI=0x%x ", ntohl(eh->spi)); break; } /* Max length: 10 "PROTO 255 " */ default: - printk("PROTO=%u ", ih->protocol); + sb_add(m, "PROTO=%u ", ih->protocol); } /* Max length: 15 "UID=4294967295 " */ if ((logflags & IPT_LOG_UID) && !iphoff && skb->sk) { read_lock_bh(&skb->sk->sk_callback_lock); if (skb->sk->sk_socket && skb->sk->sk_socket->file) - printk("UID=%u GID=%u ", + sb_add(m, "UID=%u GID=%u ", skb->sk->sk_socket->file->f_cred->fsuid, skb->sk->sk_socket->file->f_cred->fsgid); read_unlock_bh(&skb->sk->sk_callback_lock); @@ -347,7 +346,7 @@ static void dump_packet(const struct nf_loginfo *info, /* Max length: 16 "MARK=0xFFFFFFFF " */ if (!iphoff && skb->mark) - printk("MARK=0x%x ", skb->mark); + sb_add(m, "MARK=0x%x ", skb->mark); /* Proto Max log string length */ /* IP: 40+46+6+11+127 = 230 */ @@ -364,7 +363,8 @@ static void dump_packet(const struct nf_loginfo *info, /* maxlen = 230+ 91 + 230 + 252 = 803 */ } -static void dump_mac_header(const struct nf_loginfo *info, +static void dump_mac_header(struct sbuff *m, + const struct nf_loginfo *info, const struct sk_buff *skb) { struct net_device *dev = skb->dev; @@ -378,7 +378,7 @@ static void dump_mac_header(const struct nf_loginfo *info, switch (dev->type) { case ARPHRD_ETHER: - printk("MACSRC=%pM MACDST=%pM MACPROTO=%04x ", + sb_add(m, "MACSRC=%pM MACDST=%pM MACPROTO=%04x ", eth_hdr(skb)->h_source, eth_hdr(skb)->h_dest, ntohs(eth_hdr(skb)->h_proto)); return; @@ -387,17 +387,17 @@ static void dump_mac_header(const struct nf_loginfo *info, } fallback: - printk("MAC="); + sb_add(m, "MAC="); if (dev->hard_header_len && skb->mac_header != skb->network_header) { const unsigned char *p = skb_mac_header(skb); unsigned int i; - printk("%02x", *p++); + sb_add(m, "%02x", *p++); for (i = 1; i < dev->hard_header_len; i++, p++) - printk(":%02x", *p); + sb_add(m, ":%02x", *p); } - printk(" "); + sb_add(m, " "); } static struct nf_loginfo default_loginfo = { @@ -419,11 +419,12 @@ ipt_log_packet(u_int8_t pf, const struct nf_loginfo *loginfo, const char *prefix) { + struct sbuff *m = sb_open(); + if (!loginfo) loginfo = &default_loginfo; - spin_lock_bh(&log_lock); - printk("<%d>%sIN=%s OUT=%s ", loginfo->u.log.level, + sb_add(m, "<%d>%sIN=%s OUT=%s ", loginfo->u.log.level, prefix, in ? in->name : "", out ? out->name : ""); @@ -434,20 +435,20 @@ ipt_log_packet(u_int8_t pf, physindev = skb->nf_bridge->physindev; if (physindev && in != physindev) - printk("PHYSIN=%s ", physindev->name); + sb_add(m, "PHYSIN=%s ", physindev->name); physoutdev = skb->nf_bridge->physoutdev; if (physoutdev && out != physoutdev) - printk("PHYSOUT=%s ", physoutdev->name); + sb_add(m, "PHYSOUT=%s ", physoutdev->name); } #endif /* MAC logging for input path only. */ if (in && !out) - dump_mac_header(loginfo, skb); + dump_mac_header(m, loginfo, skb); + + dump_packet(m, loginfo, skb, 0); - dump_packet(loginfo, skb, 0); - printk("\n"); - spin_unlock_bh(&log_lock); + sb_close(m); } static unsigned int diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c index b254dafaf42..43eec80c0e7 100644 --- a/net/ipv4/netfilter/ipt_REJECT.c +++ b/net/ipv4/netfilter/ipt_REJECT.c @@ -112,6 +112,7 @@ static void send_reset(struct sk_buff *oldskb, int hook) /* ip_route_me_harder expects skb->dst to be set */ skb_dst_set_noref(nskb, skb_dst(oldskb)); + nskb->protocol = htons(ETH_P_IP); if (ip_route_me_harder(nskb, addr_type)) goto free_nskb; diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c index 244f7cb08d6..37f8adb68c7 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c @@ -11,6 +11,7 @@ #include <linux/proc_fs.h> #include <linux/seq_file.h> #include <linux/percpu.h> +#include <linux/security.h> #include <net/net_namespace.h> #include <linux/netfilter.h> @@ -87,6 +88,29 @@ static void ct_seq_stop(struct seq_file *s, void *v) rcu_read_unlock(); } +#ifdef CONFIG_NF_CONNTRACK_SECMARK +static int ct_show_secctx(struct seq_file *s, const struct nf_conn *ct) +{ + int ret; + u32 len; + char *secctx; + + ret = security_secid_to_secctx(ct->secmark, &secctx, &len); + if (ret) + return ret; + + ret = seq_printf(s, "secctx=%s ", secctx); + + security_release_secctx(secctx, len); + return ret; +} +#else +static inline int ct_show_secctx(struct seq_file *s, const struct nf_conn *ct) +{ + return 0; +} +#endif + static int ct_seq_show(struct seq_file *s, void *v) { struct nf_conntrack_tuple_hash *hash = v; @@ -148,10 +172,8 @@ static int ct_seq_show(struct seq_file *s, void *v) goto release; #endif -#ifdef CONFIG_NF_CONNTRACK_SECMARK - if (seq_printf(s, "secmark=%u ", ct->secmark)) + if (ct_show_secctx(s, ct)) goto release; -#endif if (seq_printf(s, "use=%u\n", atomic_read(&ct->ct_general.use))) goto release; diff --git a/net/ipv4/netfilter/nf_defrag_ipv4.c b/net/ipv4/netfilter/nf_defrag_ipv4.c index eab8de32f20..f3a9b42b16c 100644 --- a/net/ipv4/netfilter/nf_defrag_ipv4.c +++ b/net/ipv4/netfilter/nf_defrag_ipv4.c @@ -66,9 +66,11 @@ static unsigned int ipv4_conntrack_defrag(unsigned int hooknum, const struct net_device *out, int (*okfn)(struct sk_buff *)) { + struct sock *sk = skb->sk; struct inet_sock *inet = inet_sk(skb->sk); - if (inet && inet->nodefrag) + if (sk && (sk->sk_family == PF_INET) && + inet->nodefrag) return NF_ACCEPT; #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) diff --git a/net/ipv4/netfilter/nf_nat_amanda.c b/net/ipv4/netfilter/nf_nat_amanda.c index c31b8766825..0f23b3f06df 100644 --- a/net/ipv4/netfilter/nf_nat_amanda.c +++ b/net/ipv4/netfilter/nf_nat_amanda.c @@ -44,9 +44,16 @@ static unsigned int help(struct sk_buff *skb, /* Try to get same port: if not, try to change it. */ for (port = ntohs(exp->saved_proto.tcp.port); port != 0; port++) { + int ret; + exp->tuple.dst.u.tcp.port = htons(port); - if (nf_ct_expect_related(exp) == 0) + ret = nf_ct_expect_related(exp); + if (ret == 0) + break; + else if (ret != -EBUSY) { + port = 0; break; + } } if (port == 0) diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c index 8c8632d9b93..295c97431e4 100644 --- a/net/ipv4/netfilter/nf_nat_core.c +++ b/net/ipv4/netfilter/nf_nat_core.c @@ -38,7 +38,7 @@ static DEFINE_SPINLOCK(nf_nat_lock); static struct nf_conntrack_l3proto *l3proto __read_mostly; #define MAX_IP_NAT_PROTO 256 -static const struct nf_nat_protocol *nf_nat_protos[MAX_IP_NAT_PROTO] +static const struct nf_nat_protocol __rcu *nf_nat_protos[MAX_IP_NAT_PROTO] __read_mostly; static inline const struct nf_nat_protocol * @@ -47,7 +47,7 @@ __nf_nat_proto_find(u_int8_t protonum) return rcu_dereference(nf_nat_protos[protonum]); } -const struct nf_nat_protocol * +static const struct nf_nat_protocol * nf_nat_proto_find_get(u_int8_t protonum) { const struct nf_nat_protocol *p; @@ -60,14 +60,12 @@ nf_nat_proto_find_get(u_int8_t protonum) return p; } -EXPORT_SYMBOL_GPL(nf_nat_proto_find_get); -void +static void nf_nat_proto_put(const struct nf_nat_protocol *p) { module_put(p->me); } -EXPORT_SYMBOL_GPL(nf_nat_proto_put); /* We keep an extra hash for each conntrack, for fast searching. */ static inline unsigned int @@ -262,11 +260,17 @@ get_unique_tuple(struct nf_conntrack_tuple *tuple, proto = __nf_nat_proto_find(orig_tuple->dst.protonum); /* Only bother mapping if it's not already in range and unique */ - if (!(range->flags & IP_NAT_RANGE_PROTO_RANDOM) && - (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED) || - proto->in_range(tuple, maniptype, &range->min, &range->max)) && - !nf_nat_used_tuple(tuple, ct)) - goto out; + if (!(range->flags & IP_NAT_RANGE_PROTO_RANDOM)) { + if (range->flags & IP_NAT_RANGE_PROTO_SPECIFIED) { + if (proto->in_range(tuple, maniptype, &range->min, + &range->max) && + (range->min.all == range->max.all || + !nf_nat_used_tuple(tuple, ct))) + goto out; + } else if (!nf_nat_used_tuple(tuple, ct)) { + goto out; + } + } /* Last change: get protocol to try to obtain unique tuple. */ proto->unique_tuple(tuple, range, maniptype, ct); @@ -458,6 +462,18 @@ int nf_nat_icmp_reply_translation(struct nf_conn *ct, return 0; } + if (manip == IP_NAT_MANIP_SRC) + statusbit = IPS_SRC_NAT; + else + statusbit = IPS_DST_NAT; + + /* Invert if this is reply dir. */ + if (dir == IP_CT_DIR_REPLY) + statusbit ^= IPS_NAT_MASK; + + if (!(ct->status & statusbit)) + return 1; + pr_debug("icmp_reply_translation: translating error %p manip %u " "dir %s\n", skb, manip, dir == IP_CT_DIR_ORIGINAL ? "ORIG" : "REPLY"); @@ -492,20 +508,9 @@ int nf_nat_icmp_reply_translation(struct nf_conn *ct, /* Change outer to look the reply to an incoming packet * (proto 0 means don't invert per-proto part). */ - if (manip == IP_NAT_MANIP_SRC) - statusbit = IPS_SRC_NAT; - else - statusbit = IPS_DST_NAT; - - /* Invert if this is reply dir. */ - if (dir == IP_CT_DIR_REPLY) - statusbit ^= IPS_NAT_MASK; - - if (ct->status & statusbit) { - nf_ct_invert_tuplepr(&target, &ct->tuplehash[!dir].tuple); - if (!manip_pkt(0, skb, 0, &target, manip)) - return 0; - } + nf_ct_invert_tuplepr(&target, &ct->tuplehash[!dir].tuple); + if (!manip_pkt(0, skb, 0, &target, manip)) + return 0; return 1; } diff --git a/net/ipv4/netfilter/nf_nat_ftp.c b/net/ipv4/netfilter/nf_nat_ftp.c index 86e0e84ff0a..dc73abb3fe2 100644 --- a/net/ipv4/netfilter/nf_nat_ftp.c +++ b/net/ipv4/netfilter/nf_nat_ftp.c @@ -79,9 +79,16 @@ static unsigned int nf_nat_ftp(struct sk_buff *skb, /* Try to get same port: if not, try to change it. */ for (port = ntohs(exp->saved_proto.tcp.port); port != 0; port++) { + int ret; + exp->tuple.dst.u.tcp.port = htons(port); - if (nf_ct_expect_related(exp) == 0) + ret = nf_ct_expect_related(exp); + if (ret == 0) + break; + else if (ret != -EBUSY) { + port = 0; break; + } } if (port == 0) diff --git a/net/ipv4/netfilter/nf_nat_h323.c b/net/ipv4/netfilter/nf_nat_h323.c index 5045196d853..790f3160e01 100644 --- a/net/ipv4/netfilter/nf_nat_h323.c +++ b/net/ipv4/netfilter/nf_nat_h323.c @@ -222,13 +222,24 @@ static int nat_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct, /* Try to get a pair of ports. */ for (nated_port = ntohs(rtp_exp->tuple.dst.u.udp.port); nated_port != 0; nated_port += 2) { + int ret; + rtp_exp->tuple.dst.u.udp.port = htons(nated_port); - if (nf_ct_expect_related(rtp_exp) == 0) { + ret = nf_ct_expect_related(rtp_exp); + if (ret == 0) { rtcp_exp->tuple.dst.u.udp.port = htons(nated_port + 1); - if (nf_ct_expect_related(rtcp_exp) == 0) + ret = nf_ct_expect_related(rtcp_exp); + if (ret == 0) + break; + else if (ret != -EBUSY) { + nf_ct_unexpect_related(rtp_exp); + nated_port = 0; break; - nf_ct_unexpect_related(rtp_exp); + } + } else if (ret != -EBUSY) { + nated_port = 0; + break; } } @@ -284,9 +295,16 @@ static int nat_t120(struct sk_buff *skb, struct nf_conn *ct, /* Try to get same port: if not, try to change it. */ for (; nated_port != 0; nated_port++) { + int ret; + exp->tuple.dst.u.tcp.port = htons(nated_port); - if (nf_ct_expect_related(exp) == 0) + ret = nf_ct_expect_related(exp); + if (ret == 0) + break; + else if (ret != -EBUSY) { + nated_port = 0; break; + } } if (nated_port == 0) { /* No port available */ @@ -334,9 +352,16 @@ static int nat_h245(struct sk_buff *skb, struct nf_conn *ct, /* Try to get same port: if not, try to change it. */ for (; nated_port != 0; nated_port++) { + int ret; + exp->tuple.dst.u.tcp.port = htons(nated_port); - if (nf_ct_expect_related(exp) == 0) + ret = nf_ct_expect_related(exp); + if (ret == 0) break; + else if (ret != -EBUSY) { + nated_port = 0; + break; + } } if (nated_port == 0) { /* No port available */ @@ -418,9 +443,16 @@ static int nat_q931(struct sk_buff *skb, struct nf_conn *ct, /* Try to get same port: if not, try to change it. */ for (; nated_port != 0; nated_port++) { + int ret; + exp->tuple.dst.u.tcp.port = htons(nated_port); - if (nf_ct_expect_related(exp) == 0) + ret = nf_ct_expect_related(exp); + if (ret == 0) + break; + else if (ret != -EBUSY) { + nated_port = 0; break; + } } if (nated_port == 0) { /* No port available */ @@ -500,9 +532,16 @@ static int nat_callforwarding(struct sk_buff *skb, struct nf_conn *ct, /* Try to get same port: if not, try to change it. */ for (nated_port = ntohs(port); nated_port != 0; nated_port++) { + int ret; + exp->tuple.dst.u.tcp.port = htons(nated_port); - if (nf_ct_expect_related(exp) == 0) + ret = nf_ct_expect_related(exp); + if (ret == 0) break; + else if (ret != -EBUSY) { + nated_port = 0; + break; + } } if (nated_port == 0) { /* No port available */ diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c index 4a0c6b548ee..31427fb57aa 100644 --- a/net/ipv4/netfilter/nf_nat_helper.c +++ b/net/ipv4/netfilter/nf_nat_helper.c @@ -153,6 +153,35 @@ void nf_nat_set_seq_adjust(struct nf_conn *ct, enum ip_conntrack_info ctinfo, } EXPORT_SYMBOL_GPL(nf_nat_set_seq_adjust); +static void nf_nat_csum(struct sk_buff *skb, struct iphdr *iph, void *data, + int datalen, __sum16 *check, int oldlen) +{ + struct rtable *rt = skb_rtable(skb); + + if (skb->ip_summed != CHECKSUM_PARTIAL) { + if (!(rt->rt_flags & RTCF_LOCAL) && + skb->dev->features & NETIF_F_V4_CSUM) { + skb->ip_summed = CHECKSUM_PARTIAL; + skb->csum_start = skb_headroom(skb) + + skb_network_offset(skb) + + iph->ihl * 4; + skb->csum_offset = (void *)check - data; + *check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, + datalen, iph->protocol, 0); + } else { + *check = 0; + *check = csum_tcpudp_magic(iph->saddr, iph->daddr, + datalen, iph->protocol, + csum_partial(data, datalen, + 0)); + if (iph->protocol == IPPROTO_UDP && !*check) + *check = CSUM_MANGLED_0; + } + } else + inet_proto_csum_replace2(check, skb, + htons(oldlen), htons(datalen), 1); +} + /* Generic function for mangling variable-length address changes inside * NATed TCP connections (like the PORT XXX,XXX,XXX,XXX,XXX,XXX * command in FTP). @@ -169,7 +198,6 @@ int __nf_nat_mangle_tcp_packet(struct sk_buff *skb, const char *rep_buffer, unsigned int rep_len, bool adjust) { - struct rtable *rt = skb_rtable(skb); struct iphdr *iph; struct tcphdr *tcph; int oldlen, datalen; @@ -192,26 +220,7 @@ int __nf_nat_mangle_tcp_packet(struct sk_buff *skb, match_offset, match_len, rep_buffer, rep_len); datalen = skb->len - iph->ihl*4; - if (skb->ip_summed != CHECKSUM_PARTIAL) { - if (!(rt->rt_flags & RTCF_LOCAL) && - skb->dev->features & NETIF_F_V4_CSUM) { - skb->ip_summed = CHECKSUM_PARTIAL; - skb->csum_start = skb_headroom(skb) + - skb_network_offset(skb) + - iph->ihl * 4; - skb->csum_offset = offsetof(struct tcphdr, check); - tcph->check = ~tcp_v4_check(datalen, - iph->saddr, iph->daddr, 0); - } else { - tcph->check = 0; - tcph->check = tcp_v4_check(datalen, - iph->saddr, iph->daddr, - csum_partial(tcph, - datalen, 0)); - } - } else - inet_proto_csum_replace2(&tcph->check, skb, - htons(oldlen), htons(datalen), 1); + nf_nat_csum(skb, iph, tcph, datalen, &tcph->check, oldlen); if (adjust && rep_len != match_len) nf_nat_set_seq_adjust(ct, ctinfo, tcph->seq, @@ -240,7 +249,6 @@ nf_nat_mangle_udp_packet(struct sk_buff *skb, const char *rep_buffer, unsigned int rep_len) { - struct rtable *rt = skb_rtable(skb); struct iphdr *iph; struct udphdr *udph; int datalen, oldlen; @@ -274,29 +282,7 @@ nf_nat_mangle_udp_packet(struct sk_buff *skb, if (!udph->check && skb->ip_summed != CHECKSUM_PARTIAL) return 1; - if (skb->ip_summed != CHECKSUM_PARTIAL) { - if (!(rt->rt_flags & RTCF_LOCAL) && - skb->dev->features & NETIF_F_V4_CSUM) { - skb->ip_summed = CHECKSUM_PARTIAL; - skb->csum_start = skb_headroom(skb) + - skb_network_offset(skb) + - iph->ihl * 4; - skb->csum_offset = offsetof(struct udphdr, check); - udph->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, - datalen, IPPROTO_UDP, - 0); - } else { - udph->check = 0; - udph->check = csum_tcpudp_magic(iph->saddr, iph->daddr, - datalen, IPPROTO_UDP, - csum_partial(udph, - datalen, 0)); - if (!udph->check) - udph->check = CSUM_MANGLED_0; - } - } else - inet_proto_csum_replace2(&udph->check, skb, - htons(oldlen), htons(datalen), 1); + nf_nat_csum(skb, iph, udph, datalen, &udph->check, oldlen); return 1; } diff --git a/net/ipv4/netfilter/nf_nat_irc.c b/net/ipv4/netfilter/nf_nat_irc.c index ea83a886b03..535e1a80235 100644 --- a/net/ipv4/netfilter/nf_nat_irc.c +++ b/net/ipv4/netfilter/nf_nat_irc.c @@ -45,9 +45,16 @@ static unsigned int help(struct sk_buff *skb, /* Try to get same port: if not, try to change it. */ for (port = ntohs(exp->saved_proto.tcp.port); port != 0; port++) { + int ret; + exp->tuple.dst.u.tcp.port = htons(port); - if (nf_ct_expect_related(exp) == 0) + ret = nf_ct_expect_related(exp); + if (ret == 0) + break; + else if (ret != -EBUSY) { + port = 0; break; + } } if (port == 0) diff --git a/net/ipv4/netfilter/nf_nat_rule.c b/net/ipv4/netfilter/nf_nat_rule.c index ebbd319f62f..21c30426480 100644 --- a/net/ipv4/netfilter/nf_nat_rule.c +++ b/net/ipv4/netfilter/nf_nat_rule.c @@ -106,16 +106,15 @@ alloc_null_binding(struct nf_conn *ct, unsigned int hooknum) { /* Force range to this IP; let proto decide mapping for per-proto parts (hence not IP_NAT_RANGE_PROTO_SPECIFIED). - Use reply in case it's already been mangled (eg local packet). */ - __be32 ip - = (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC - ? ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip - : ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip); - struct nf_nat_range range - = { IP_NAT_RANGE_MAP_IPS, ip, ip, { 0 }, { 0 } }; - - pr_debug("Allocating NULL binding for %p (%pI4)\n", ct, &ip); + struct nf_nat_range range; + + range.flags = 0; + pr_debug("Allocating NULL binding for %p (%pI4)\n", ct, + HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC ? + &ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip : + &ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip); + return nf_nat_setup_info(ct, &range, HOOK2MANIP(hooknum)); } diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index 11b538deaae..e40cf7816fd 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -307,9 +307,16 @@ static unsigned int ip_nat_sip_expect(struct sk_buff *skb, unsigned int dataoff, exp->expectfn = ip_nat_sip_expected; for (; port != 0; port++) { + int ret; + exp->tuple.dst.u.udp.port = htons(port); - if (nf_ct_expect_related(exp) == 0) + ret = nf_ct_expect_related(exp); + if (ret == 0) + break; + else if (ret != -EBUSY) { + port = 0; break; + } } if (port == 0) @@ -480,13 +487,25 @@ static unsigned int ip_nat_sdp_media(struct sk_buff *skb, unsigned int dataoff, /* Try to get same pair of ports: if not, try to change them. */ for (port = ntohs(rtp_exp->tuple.dst.u.udp.port); port != 0; port += 2) { + int ret; + rtp_exp->tuple.dst.u.udp.port = htons(port); - if (nf_ct_expect_related(rtp_exp) != 0) + ret = nf_ct_expect_related(rtp_exp); + if (ret == -EBUSY) continue; + else if (ret < 0) { + port = 0; + break; + } rtcp_exp->tuple.dst.u.udp.port = htons(port + 1); - if (nf_ct_expect_related(rtcp_exp) == 0) + ret = nf_ct_expect_related(rtcp_exp); + if (ret == 0) break; - nf_ct_unexpect_related(rtp_exp); + else if (ret != -EBUSY) { + nf_ct_unexpect_related(rtp_exp); + port = 0; + break; + } } if (port == 0) diff --git a/net/ipv4/netfilter/nf_nat_snmp_basic.c b/net/ipv4/netfilter/nf_nat_snmp_basic.c index 1679e2c0963..ee5f419d0a5 100644 --- a/net/ipv4/netfilter/nf_nat_snmp_basic.c +++ b/net/ipv4/netfilter/nf_nat_snmp_basic.c @@ -893,13 +893,15 @@ static void fast_csum(__sum16 *csum, unsigned char s[4]; if (offset & 1) { - s[0] = s[2] = 0; + s[0] = ~0; s[1] = ~*optr; + s[2] = 0; s[3] = *nptr; } else { - s[1] = s[3] = 0; s[0] = ~*optr; + s[1] = ~0; s[2] = *nptr; + s[3] = 0; } *csum = csum_fold(csum_partial(s, 4, ~csum_unfold(*csum))); diff --git a/net/ipv4/protocol.c b/net/ipv4/protocol.c index f2d29735140..65699c24411 100644 --- a/net/ipv4/protocol.c +++ b/net/ipv4/protocol.c @@ -28,8 +28,7 @@ #include <linux/spinlock.h> #include <net/protocol.h> -const struct net_protocol *inet_protos[MAX_INET_PROTOS] ____cacheline_aligned_in_smp; -static DEFINE_SPINLOCK(inet_proto_lock); +const struct net_protocol *inet_protos[MAX_INET_PROTOS] __read_mostly; /* * Add a protocol handler to the hash tables @@ -37,20 +36,9 @@ static DEFINE_SPINLOCK(inet_proto_lock); int inet_add_protocol(const struct net_protocol *prot, unsigned char protocol) { - int hash, ret; + int hash = protocol & (MAX_INET_PROTOS - 1); - hash = protocol & (MAX_INET_PROTOS - 1); - - spin_lock_bh(&inet_proto_lock); - if (inet_protos[hash]) { - ret = -1; - } else { - inet_protos[hash] = prot; - ret = 0; - } - spin_unlock_bh(&inet_proto_lock); - - return ret; + return !cmpxchg(&inet_protos[hash], NULL, prot) ? 0 : -1; } EXPORT_SYMBOL(inet_add_protocol); @@ -60,18 +48,9 @@ EXPORT_SYMBOL(inet_add_protocol); int inet_del_protocol(const struct net_protocol *prot, unsigned char protocol) { - int hash, ret; - - hash = protocol & (MAX_INET_PROTOS - 1); + int ret, hash = protocol & (MAX_INET_PROTOS - 1); - spin_lock_bh(&inet_proto_lock); - if (inet_protos[hash] == prot) { - inet_protos[hash] = NULL; - ret = 0; - } else { - ret = -1; - } - spin_unlock_bh(&inet_proto_lock); + ret = (cmpxchg(&inet_protos[hash], prot, NULL) == prot) ? 0 : -1; synchronize_net(); diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 009a7b2aa1e..1f85ef28989 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -505,7 +505,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, ipc.addr = inet->inet_saddr; ipc.opt = NULL; - ipc.shtx.flags = 0; + ipc.tx_flags = 0; ipc.oif = sk->sk_bound_dev_if; if (msg->msg_controllen) { diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 3f56b6e6c6a..d6cb2bfcd8e 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -159,7 +159,6 @@ static struct dst_ops ipv4_dst_ops = { .link_failure = ipv4_link_failure, .update_pmtu = ip_rt_update_pmtu, .local_out = __ip_local_out, - .entries = ATOMIC_INIT(0), }; #define ECN_OR_COST(class) TC_PRIO_##class @@ -466,7 +465,7 @@ static int rt_cpu_seq_show(struct seq_file *seq, void *v) seq_printf(seq,"%08x %08x %08x %08x %08x %08x %08x %08x " " %08x %08x %08x %08x %08x %08x %08x %08x %08x \n", - atomic_read(&ipv4_dst_ops.entries), + dst_entries_get_slow(&ipv4_dst_ops), st->in_hit, st->in_slow_tot, st->in_slow_mc, @@ -945,6 +944,7 @@ static int rt_garbage_collect(struct dst_ops *ops) struct rtable *rth, **rthp; unsigned long now = jiffies; int goal; + int entries = dst_entries_get_fast(&ipv4_dst_ops); /* * Garbage collection is pretty expensive, @@ -954,28 +954,28 @@ static int rt_garbage_collect(struct dst_ops *ops) RT_CACHE_STAT_INC(gc_total); if (now - last_gc < ip_rt_gc_min_interval && - atomic_read(&ipv4_dst_ops.entries) < ip_rt_max_size) { + entries < ip_rt_max_size) { RT_CACHE_STAT_INC(gc_ignored); goto out; } + entries = dst_entries_get_slow(&ipv4_dst_ops); /* Calculate number of entries, which we want to expire now. */ - goal = atomic_read(&ipv4_dst_ops.entries) - - (ip_rt_gc_elasticity << rt_hash_log); + goal = entries - (ip_rt_gc_elasticity << rt_hash_log); if (goal <= 0) { if (equilibrium < ipv4_dst_ops.gc_thresh) equilibrium = ipv4_dst_ops.gc_thresh; - goal = atomic_read(&ipv4_dst_ops.entries) - equilibrium; + goal = entries - equilibrium; if (goal > 0) { equilibrium += min_t(unsigned int, goal >> 1, rt_hash_mask + 1); - goal = atomic_read(&ipv4_dst_ops.entries) - equilibrium; + goal = entries - equilibrium; } } else { /* We are in dangerous area. Try to reduce cache really * aggressively. */ goal = max_t(unsigned int, goal >> 1, rt_hash_mask + 1); - equilibrium = atomic_read(&ipv4_dst_ops.entries) - goal; + equilibrium = entries - goal; } if (now - last_gc >= ip_rt_gc_min_interval) @@ -1032,14 +1032,16 @@ static int rt_garbage_collect(struct dst_ops *ops) expire >>= 1; #if RT_CACHE_DEBUG >= 2 printk(KERN_DEBUG "expire>> %u %d %d %d\n", expire, - atomic_read(&ipv4_dst_ops.entries), goal, i); + dst_entries_get_fast(&ipv4_dst_ops), goal, i); #endif - if (atomic_read(&ipv4_dst_ops.entries) < ip_rt_max_size) + if (dst_entries_get_fast(&ipv4_dst_ops) < ip_rt_max_size) goto out; } while (!in_softirq() && time_before_eq(jiffies, now)); - if (atomic_read(&ipv4_dst_ops.entries) < ip_rt_max_size) + if (dst_entries_get_fast(&ipv4_dst_ops) < ip_rt_max_size) + goto out; + if (dst_entries_get_slow(&ipv4_dst_ops) < ip_rt_max_size) goto out; if (net_ratelimit()) printk(KERN_WARNING "dst cache overflow\n"); @@ -1049,11 +1051,12 @@ static int rt_garbage_collect(struct dst_ops *ops) work_done: expire += ip_rt_gc_min_interval; if (expire > ip_rt_gc_timeout || - atomic_read(&ipv4_dst_ops.entries) < ipv4_dst_ops.gc_thresh) + dst_entries_get_fast(&ipv4_dst_ops) < ipv4_dst_ops.gc_thresh || + dst_entries_get_slow(&ipv4_dst_ops) < ipv4_dst_ops.gc_thresh) expire = ip_rt_gc_timeout; #if RT_CACHE_DEBUG >= 2 printk(KERN_DEBUG "expire++ %u %d %d %d\n", expire, - atomic_read(&ipv4_dst_ops.entries), goal, rover); + dst_entries_get_fast(&ipv4_dst_ops), goal, rover); #endif out: return 0; } @@ -1102,23 +1105,23 @@ restart: * Note that we do rt_free on this new route entry, so that * once its refcount hits zero, we are still able to reap it * (Thanks Alexey) - * Note also the rt_free uses call_rcu. We don't actually - * need rcu protection here, this is just our path to get - * on the route gc list. + * Note: To avoid expensive rcu stuff for this uncached dst, + * we set DST_NOCACHE so that dst_release() can free dst without + * waiting a grace period. */ + rt->dst.flags |= DST_NOCACHE; if (rt->rt_type == RTN_UNICAST || rt->fl.iif == 0) { int err = arp_bind_neighbour(&rt->dst); if (err) { if (net_ratelimit()) printk(KERN_WARNING "Neighbour table failure & not caching routes.\n"); - rt_drop(rt); + ip_rt_put(rt); return err; } } - rt_free(rt); goto skip_hashing; } @@ -1231,7 +1234,7 @@ restart: } if (net_ratelimit()) - printk(KERN_WARNING "Neighbour table overflow.\n"); + printk(KERN_WARNING "ipv4: Neighbour table overflow.\n"); rt_drop(rt); return -ENOBUFS; } @@ -1268,18 +1271,11 @@ skip_hashing: void rt_bind_peer(struct rtable *rt, int create) { - static DEFINE_SPINLOCK(rt_peer_lock); struct inet_peer *peer; peer = inet_getpeer(rt->rt_dst, create); - spin_lock_bh(&rt_peer_lock); - if (rt->peer == NULL) { - rt->peer = peer; - peer = NULL; - } - spin_unlock_bh(&rt_peer_lock); - if (peer) + if (peer && cmpxchg(&rt->peer, NULL, peer) != NULL) inet_putpeer(peer); } @@ -1779,12 +1775,15 @@ void ip_rt_get_source(u8 *addr, struct rtable *rt) if (rt->fl.iif == 0) src = rt->rt_src; - else if (fib_lookup(dev_net(rt->dst.dev), &rt->fl, &res) == 0) { - src = FIB_RES_PREFSRC(res); - fib_res_put(&res); - } else - src = inet_select_addr(rt->dst.dev, rt->rt_gateway, + else { + rcu_read_lock(); + if (fib_lookup(dev_net(rt->dst.dev), &rt->fl, &res) == 0) + src = FIB_RES_PREFSRC(res); + else + src = inet_select_addr(rt->dst.dev, rt->rt_gateway, RT_SCOPE_UNIVERSE); + rcu_read_unlock(); + } memcpy(addr, &src, 4); } @@ -2087,6 +2086,7 @@ static int ip_mkroute_input(struct sk_buff *skb, * Such approach solves two big problems: * 1. Not simplex devices are handled properly. * 2. IP spoofing attempts are filtered with 100% of guarantee. + * called with rcu_read_lock() */ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, @@ -2108,7 +2108,6 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, unsigned hash; __be32 spec_dst; int err = -EINVAL; - int free_res = 0; struct net * net = dev_net(dev); /* IP on this device is disabled. */ @@ -2124,7 +2123,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, ipv4_is_loopback(saddr)) goto martian_source; - if (daddr == htonl(0xFFFFFFFF) || (saddr == 0 && daddr == 0)) + if (ipv4_is_lbcast(daddr) || (saddr == 0 && daddr == 0)) goto brd_input; /* Accept zero addresses only to limited broadcast; @@ -2133,19 +2132,18 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, if (ipv4_is_zeronet(saddr)) goto martian_source; - if (ipv4_is_lbcast(daddr) || ipv4_is_zeronet(daddr) || - ipv4_is_loopback(daddr)) + if (ipv4_is_zeronet(daddr) || ipv4_is_loopback(daddr)) goto martian_destination; /* * Now we are ready to route packet. */ - if ((err = fib_lookup(net, &fl, &res)) != 0) { + err = fib_lookup(net, &fl, &res); + if (err != 0) { if (!IN_DEV_FORWARD(in_dev)) goto e_hostunreach; goto no_route; } - free_res = 1; RT_CACHE_STAT_INC(in_slow_tot); @@ -2154,8 +2152,8 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, if (res.type == RTN_LOCAL) { err = fib_validate_source(saddr, daddr, tos, - net->loopback_dev->ifindex, - dev, &spec_dst, &itag, skb->mark); + net->loopback_dev->ifindex, + dev, &spec_dst, &itag, skb->mark); if (err < 0) goto martian_source_keep_err; if (err) @@ -2170,9 +2168,6 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, goto martian_destination; err = ip_mkroute_input(skb, &res, &fl, in_dev, daddr, saddr, tos); -done: - if (free_res) - fib_res_put(&res); out: return err; brd_input: @@ -2232,7 +2227,7 @@ local_input: rth->rt_type = res.type; hash = rt_hash(daddr, saddr, fl.iif, rt_genid(net)); err = rt_intern_hash(hash, rth, NULL, skb, fl.iif); - goto done; + goto out; no_route: RT_CACHE_STAT_INC(in_no_route); @@ -2255,21 +2250,21 @@ martian_destination: e_hostunreach: err = -EHOSTUNREACH; - goto done; + goto out; e_inval: err = -EINVAL; - goto done; + goto out; e_nobufs: err = -ENOBUFS; - goto done; + goto out; martian_source: err = -EINVAL; martian_source_keep_err: ip_handle_martian_source(dev, in_dev, skb, daddr, saddr); - goto done; + goto out; } int ip_route_input_common(struct sk_buff *skb, __be32 daddr, __be32 saddr, @@ -2355,6 +2350,7 @@ skip_cache: } EXPORT_SYMBOL(ip_route_input_common); +/* called with rcu_read_lock() */ static int __mkroute_output(struct rtable **result, struct fib_result *res, const struct flowi *fl, @@ -2365,53 +2361,47 @@ static int __mkroute_output(struct rtable **result, struct rtable *rth; struct in_device *in_dev; u32 tos = RT_FL_TOS(oldflp); - int err = 0; - if (ipv4_is_loopback(fl->fl4_src) && !(dev_out->flags&IFF_LOOPBACK)) + if (ipv4_is_loopback(fl->fl4_src) && !(dev_out->flags & IFF_LOOPBACK)) return -EINVAL; - if (fl->fl4_dst == htonl(0xFFFFFFFF)) + if (ipv4_is_lbcast(fl->fl4_dst)) res->type = RTN_BROADCAST; else if (ipv4_is_multicast(fl->fl4_dst)) res->type = RTN_MULTICAST; - else if (ipv4_is_lbcast(fl->fl4_dst) || ipv4_is_zeronet(fl->fl4_dst)) + else if (ipv4_is_zeronet(fl->fl4_dst)) return -EINVAL; if (dev_out->flags & IFF_LOOPBACK) flags |= RTCF_LOCAL; - /* get work reference to inet device */ - in_dev = in_dev_get(dev_out); + in_dev = __in_dev_get_rcu(dev_out); if (!in_dev) return -EINVAL; if (res->type == RTN_BROADCAST) { flags |= RTCF_BROADCAST | RTCF_LOCAL; - if (res->fi) { - fib_info_put(res->fi); - res->fi = NULL; - } + res->fi = NULL; } else if (res->type == RTN_MULTICAST) { - flags |= RTCF_MULTICAST|RTCF_LOCAL; + flags |= RTCF_MULTICAST | RTCF_LOCAL; if (!ip_check_mc(in_dev, oldflp->fl4_dst, oldflp->fl4_src, oldflp->proto)) flags &= ~RTCF_LOCAL; /* If multicast route do not exist use - default one, but do not gateway in this case. - Yes, it is hack. + * default one, but do not gateway in this case. + * Yes, it is hack. */ - if (res->fi && res->prefixlen < 4) { - fib_info_put(res->fi); + if (res->fi && res->prefixlen < 4) res->fi = NULL; - } } rth = dst_alloc(&ipv4_dst_ops); - if (!rth) { - err = -ENOBUFS; - goto cleanup; - } + if (!rth) + return -ENOBUFS; + + in_dev_hold(in_dev); + rth->idev = in_dev; atomic_set(&rth->dst.__refcnt, 1); rth->dst.flags= DST_HOST; @@ -2432,7 +2422,6 @@ static int __mkroute_output(struct rtable **result, cache entry */ rth->dst.dev = dev_out; dev_hold(dev_out); - rth->idev = in_dev_get(dev_out); rth->rt_gateway = fl->fl4_dst; rth->rt_spec_dst= fl->fl4_src; @@ -2467,15 +2456,11 @@ static int __mkroute_output(struct rtable **result, rt_set_nexthop(rth, res, 0); rth->rt_flags = flags; - *result = rth; - cleanup: - /* release work reference to inet device */ - in_dev_put(in_dev); - - return err; + return 0; } +/* called with rcu_read_lock() */ static int ip_mkroute_output(struct rtable **rp, struct fib_result *res, const struct flowi *fl, @@ -2497,6 +2482,7 @@ static int ip_mkroute_output(struct rtable **rp, /* * Major route resolver routine. + * called with rcu_read_lock(); */ static int ip_route_output_slow(struct net *net, struct rtable **rp, @@ -2515,9 +2501,8 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp, .iif = net->loopback_dev->ifindex, .oif = oldflp->oif }; struct fib_result res; - unsigned flags = 0; + unsigned int flags = 0; struct net_device *dev_out = NULL; - int free_res = 0; int err; @@ -2543,9 +2528,9 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp, if (oldflp->oif == 0 && (ipv4_is_multicast(oldflp->fl4_dst) || - oldflp->fl4_dst == htonl(0xFFFFFFFF))) { + ipv4_is_lbcast(oldflp->fl4_dst))) { /* It is equivalent to inet_addr_type(saddr) == RTN_LOCAL */ - dev_out = ip_dev_find(net, oldflp->fl4_src); + dev_out = __ip_dev_find(net, oldflp->fl4_src, false); if (dev_out == NULL) goto out; @@ -2570,29 +2555,24 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp, if (!(oldflp->flags & FLOWI_FLAG_ANYSRC)) { /* It is equivalent to inet_addr_type(saddr) == RTN_LOCAL */ - dev_out = ip_dev_find(net, oldflp->fl4_src); - if (dev_out == NULL) + if (!__ip_dev_find(net, oldflp->fl4_src, false)) goto out; - dev_put(dev_out); - dev_out = NULL; } } if (oldflp->oif) { - dev_out = dev_get_by_index(net, oldflp->oif); + dev_out = dev_get_by_index_rcu(net, oldflp->oif); err = -ENODEV; if (dev_out == NULL) goto out; /* RACE: Check return value of inet_select_addr instead. */ - if (__in_dev_get_rtnl(dev_out) == NULL) { - dev_put(dev_out); + if (rcu_dereference(dev_out->ip_ptr) == NULL) goto out; /* Wrong error code */ - } if (ipv4_is_local_multicast(oldflp->fl4_dst) || - oldflp->fl4_dst == htonl(0xFFFFFFFF)) { + ipv4_is_lbcast(oldflp->fl4_dst)) { if (!fl.fl4_src) fl.fl4_src = inet_select_addr(dev_out, 0, RT_SCOPE_LINK); @@ -2612,10 +2592,7 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp, fl.fl4_dst = fl.fl4_src; if (!fl.fl4_dst) fl.fl4_dst = fl.fl4_src = htonl(INADDR_LOOPBACK); - if (dev_out) - dev_put(dev_out); dev_out = net->loopback_dev; - dev_hold(dev_out); fl.oif = net->loopback_dev->ifindex; res.type = RTN_LOCAL; flags |= RTCF_LOCAL; @@ -2649,23 +2626,15 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp, res.type = RTN_UNICAST; goto make_route; } - if (dev_out) - dev_put(dev_out); err = -ENETUNREACH; goto out; } - free_res = 1; if (res.type == RTN_LOCAL) { if (!fl.fl4_src) fl.fl4_src = fl.fl4_dst; - if (dev_out) - dev_put(dev_out); dev_out = net->loopback_dev; - dev_hold(dev_out); fl.oif = dev_out->ifindex; - if (res.fi) - fib_info_put(res.fi); res.fi = NULL; flags |= RTCF_LOCAL; goto make_route; @@ -2682,28 +2651,21 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp, if (!fl.fl4_src) fl.fl4_src = FIB_RES_PREFSRC(res); - if (dev_out) - dev_put(dev_out); dev_out = FIB_RES_DEV(res); - dev_hold(dev_out); fl.oif = dev_out->ifindex; make_route: err = ip_mkroute_output(rp, &res, &fl, oldflp, dev_out, flags); - - if (free_res) - fib_res_put(&res); - if (dev_out) - dev_put(dev_out); out: return err; } int __ip_route_output_key(struct net *net, struct rtable **rp, const struct flowi *flp) { - unsigned hash; + unsigned int hash; + int res; struct rtable *rth; if (!rt_caching(net)) @@ -2734,10 +2696,18 @@ int __ip_route_output_key(struct net *net, struct rtable **rp, rcu_read_unlock_bh(); slow_output: - return ip_route_output_slow(net, rp, flp); + rcu_read_lock(); + res = ip_route_output_slow(net, rp, flp); + rcu_read_unlock(); + return res; } EXPORT_SYMBOL_GPL(__ip_route_output_key); +static struct dst_entry *ipv4_blackhole_dst_check(struct dst_entry *dst, u32 cookie) +{ + return NULL; +} + static void ipv4_rt_blackhole_update_pmtu(struct dst_entry *dst, u32 mtu) { } @@ -2746,9 +2716,8 @@ static struct dst_ops ipv4_dst_blackhole_ops = { .family = AF_INET, .protocol = cpu_to_be16(ETH_P_IP), .destroy = ipv4_dst_destroy, - .check = ipv4_dst_check, + .check = ipv4_blackhole_dst_check, .update_pmtu = ipv4_rt_blackhole_update_pmtu, - .entries = ATOMIC_INIT(0), }; @@ -2793,7 +2762,7 @@ static int ipv4_dst_blackhole(struct net *net, struct rtable **rp, struct flowi dst_release(&(*rp)->dst); *rp = rt; - return (rt ? 0 : -ENOMEM); + return rt ? 0 : -ENOMEM; } int ip_route_output_flow(struct net *net, struct rtable **rp, struct flowi *flp, @@ -3318,6 +3287,12 @@ int __init ip_rt_init(void) ipv4_dst_blackhole_ops.kmem_cachep = ipv4_dst_ops.kmem_cachep; + if (dst_entries_init(&ipv4_dst_ops) < 0) + panic("IP: failed to allocate ipv4_dst_ops counter\n"); + + if (dst_entries_init(&ipv4_dst_blackhole_ops) < 0) + panic("IP: failed to allocate ipv4_dst_blackhole_ops counter\n"); + rt_hash_table = (struct rt_hash_bucket *) alloc_large_system_hash("IP route cache", sizeof(struct rt_hash_bucket), diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 176e11aaea7..1664a0590bb 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -386,8 +386,6 @@ unsigned int tcp_poll(struct file *file, struct socket *sock, poll_table *wait) */ mask = 0; - if (sk->sk_err) - mask = POLLERR; /* * POLLHUP is certainly not done right. But poll() doesn't @@ -451,11 +449,17 @@ unsigned int tcp_poll(struct file *file, struct socket *sock, poll_table *wait) if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk)) mask |= POLLOUT | POLLWRNORM; } - } + } else + mask |= POLLOUT | POLLWRNORM; if (tp->urg_data & TCP_URG_VALID) mask |= POLLPRI; } + /* This barrier is coupled with smp_wmb() in tcp_reset() */ + smp_rmb(); + if (sk->sk_err) + mask |= POLLERR; + return mask; } EXPORT_SYMBOL(tcp_poll); @@ -939,7 +943,7 @@ int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, sg = sk->sk_route_caps & NETIF_F_SG; while (--iovlen >= 0) { - int seglen = iov->iov_len; + size_t seglen = iov->iov_len; unsigned char __user *from = iov->iov_base; iov++; @@ -2011,11 +2015,8 @@ adjudge_to_death: } } if (sk->sk_state != TCP_CLOSE) { - int orphan_count = percpu_counter_read_positive( - sk->sk_prot->orphan_count); - sk_mem_reclaim(sk); - if (tcp_too_many_orphans(sk, orphan_count)) { + if (tcp_too_many_orphans(sk, 0)) { if (net_ratelimit()) printk(KERN_INFO "TCP: too many of orphaned " "sockets\n"); @@ -2391,7 +2392,12 @@ static int do_tcp_setsockopt(struct sock *sk, int level, err = tp->af_specific->md5_parse(sk, optval, optlen); break; #endif - + case TCP_USER_TIMEOUT: + /* Cap the max timeout in ms TCP will retry/retrans + * before giving up and aborting (ETIMEDOUT) a connection. + */ + icsk->icsk_user_timeout = msecs_to_jiffies(val); + break; default: err = -ENOPROTOOPT; break; @@ -2610,6 +2616,10 @@ static int do_tcp_getsockopt(struct sock *sk, int level, case TCP_THIN_DUPACK: val = tp->thin_dupack; break; + + case TCP_USER_TIMEOUT: + val = jiffies_to_msecs(icsk->icsk_user_timeout); + break; default: return -ENOPROTOOPT; } @@ -3212,7 +3222,7 @@ void __init tcp_init(void) { struct sk_buff *skb = NULL; unsigned long nr_pages, limit; - int order, i, max_share; + int i, max_share, cnt; unsigned long jiffy = jiffies; BUILD_BUG_ON(sizeof(struct tcp_skb_cb) > sizeof(skb->cb)); @@ -3261,22 +3271,12 @@ void __init tcp_init(void) INIT_HLIST_HEAD(&tcp_hashinfo.bhash[i].chain); } - /* Try to be a bit smarter and adjust defaults depending - * on available memory. - */ - for (order = 0; ((1 << order) << PAGE_SHIFT) < - (tcp_hashinfo.bhash_size * sizeof(struct inet_bind_hashbucket)); - order++) - ; - if (order >= 4) { - tcp_death_row.sysctl_max_tw_buckets = 180000; - sysctl_tcp_max_orphans = 4096 << (order - 4); - sysctl_max_syn_backlog = 1024; - } else if (order < 3) { - tcp_death_row.sysctl_max_tw_buckets >>= (3 - order); - sysctl_tcp_max_orphans >>= (3 - order); - sysctl_max_syn_backlog = 128; - } + + cnt = tcp_hashinfo.ehash_mask + 1; + + tcp_death_row.sysctl_max_tw_buckets = cnt / 2; + sysctl_tcp_max_orphans = cnt / 2; + sysctl_max_syn_backlog = max(128, cnt / 256); /* Set the pressure threshold to be a fraction of global memory that * is up to 1/2 at 256 MB, decreasing toward zero with the amount of diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index 0ec9bd0ae94..850c737e08e 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -196,10 +196,10 @@ void tcp_get_allowed_congestion_control(char *buf, size_t maxlen) int tcp_set_allowed_congestion_control(char *val) { struct tcp_congestion_ops *ca; - char *clone, *name; + char *saved_clone, *clone, *name; int ret = 0; - clone = kstrdup(val, GFP_USER); + saved_clone = clone = kstrdup(val, GFP_USER); if (!clone) return -ENOMEM; @@ -226,6 +226,7 @@ int tcp_set_allowed_congestion_control(char *val) } out: spin_unlock(&tcp_cong_list_lock); + kfree(saved_clone); return ret; } diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index bccce3424a6..3357f69e353 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -182,7 +182,7 @@ static void tcp_incr_quickack(struct sock *sk) icsk->icsk_ack.quick = min(quickacks, TCP_MAX_QUICKACKS); } -void tcp_enter_quickack_mode(struct sock *sk) +static void tcp_enter_quickack_mode(struct sock *sk) { struct inet_connection_sock *icsk = inet_csk(sk); tcp_incr_quickack(sk); @@ -805,25 +805,12 @@ void tcp_update_metrics(struct sock *sk) } } -/* Numbers are taken from RFC3390. - * - * John Heffner states: - * - * The RFC specifies a window of no more than 4380 bytes - * unless 2*MSS > 4380. Reading the pseudocode in the RFC - * is a bit misleading because they use a clamp at 4380 bytes - * rather than use a multiplier in the relevant range. - */ __u32 tcp_init_cwnd(struct tcp_sock *tp, struct dst_entry *dst) { __u32 cwnd = (dst ? dst_metric(dst, RTAX_INITCWND) : 0); - if (!cwnd) { - if (tp->mss_cache > 1460) - cwnd = 2; - else - cwnd = (tp->mss_cache > 1095) ? 3 : 4; - } + if (!cwnd) + cwnd = rfc3390_bytes_to_packets(tp->mss_cache); return min_t(__u32, cwnd, tp->snd_cwnd_clamp); } @@ -2314,7 +2301,7 @@ static inline int tcp_dupack_heuristics(struct tcp_sock *tp) static inline int tcp_skb_timedout(struct sock *sk, struct sk_buff *skb) { - return (tcp_time_stamp - TCP_SKB_CB(skb)->when > inet_csk(sk)->icsk_rto); + return tcp_time_stamp - TCP_SKB_CB(skb)->when > inet_csk(sk)->icsk_rto; } static inline int tcp_head_timedout(struct sock *sk) @@ -2508,7 +2495,7 @@ static void tcp_timeout_skbs(struct sock *sk) /* Mark head of queue up as lost. With RFC3517 SACK, the packets is * is against sacked "cnt", otherwise it's against facked "cnt" */ -static void tcp_mark_head_lost(struct sock *sk, int packets) +static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; @@ -2516,13 +2503,13 @@ static void tcp_mark_head_lost(struct sock *sk, int packets) int err; unsigned int mss; - if (packets == 0) - return; - WARN_ON(packets > tp->packets_out); if (tp->lost_skb_hint) { skb = tp->lost_skb_hint; cnt = tp->lost_cnt_hint; + /* Head already handled? */ + if (mark_head && skb != tcp_write_queue_head(sk)) + return; } else { skb = tcp_write_queue_head(sk); cnt = 0; @@ -2545,7 +2532,8 @@ static void tcp_mark_head_lost(struct sock *sk, int packets) cnt += tcp_skb_pcount(skb); if (cnt > packets) { - if (tcp_is_sack(tp) || (oldcnt >= packets)) + if ((tcp_is_sack(tp) && !tcp_is_fack(tp)) || + (oldcnt >= packets)) break; mss = skb_shinfo(skb)->gso_size; @@ -2556,6 +2544,9 @@ static void tcp_mark_head_lost(struct sock *sk, int packets) } tcp_skb_mark_lost(tp, skb); + + if (mark_head) + break; } tcp_verify_left_out(tp); } @@ -2567,17 +2558,18 @@ static void tcp_update_scoreboard(struct sock *sk, int fast_rexmit) struct tcp_sock *tp = tcp_sk(sk); if (tcp_is_reno(tp)) { - tcp_mark_head_lost(sk, 1); + tcp_mark_head_lost(sk, 1, 1); } else if (tcp_is_fack(tp)) { int lost = tp->fackets_out - tp->reordering; if (lost <= 0) lost = 1; - tcp_mark_head_lost(sk, lost); + tcp_mark_head_lost(sk, lost, 0); } else { int sacked_upto = tp->sacked_out - tp->reordering; - if (sacked_upto < fast_rexmit) - sacked_upto = fast_rexmit; - tcp_mark_head_lost(sk, sacked_upto); + if (sacked_upto >= 0) + tcp_mark_head_lost(sk, sacked_upto, 0); + else if (fast_rexmit) + tcp_mark_head_lost(sk, 1, 1); } tcp_timeout_skbs(sk); @@ -2886,7 +2878,7 @@ static void tcp_mtup_probe_success(struct sock *sk) icsk->icsk_mtup.probe_size; tp->snd_cwnd_cnt = 0; tp->snd_cwnd_stamp = tcp_time_stamp; - tp->rcv_ssthresh = tcp_current_ssthresh(sk); + tp->snd_ssthresh = tcp_current_ssthresh(sk); icsk->icsk_mtup.search_low = icsk->icsk_mtup.probe_size; icsk->icsk_mtup.probe_size = 0; @@ -2983,7 +2975,7 @@ static void tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag) before(tp->snd_una, tp->high_seq) && icsk->icsk_ca_state != TCP_CA_Open && tp->fackets_out > tp->reordering) { - tcp_mark_head_lost(sk, tp->fackets_out - tp->reordering); + tcp_mark_head_lost(sk, tp->fackets_out - tp->reordering, 0); NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPLOSS); } @@ -3411,8 +3403,8 @@ static void tcp_ack_probe(struct sock *sk) static inline int tcp_ack_is_dubious(const struct sock *sk, const int flag) { - return (!(flag & FLAG_NOT_DUP) || (flag & FLAG_CA_ALERT) || - inet_csk(sk)->icsk_ca_state != TCP_CA_Open); + return !(flag & FLAG_NOT_DUP) || (flag & FLAG_CA_ALERT) || + inet_csk(sk)->icsk_ca_state != TCP_CA_Open; } static inline int tcp_may_raise_cwnd(const struct sock *sk, const int flag) @@ -3429,9 +3421,9 @@ static inline int tcp_may_update_window(const struct tcp_sock *tp, const u32 ack, const u32 ack_seq, const u32 nwin) { - return (after(ack, tp->snd_una) || + return after(ack, tp->snd_una) || after(ack_seq, tp->snd_wl1) || - (ack_seq == tp->snd_wl1 && nwin > tp->snd_wnd)); + (ack_seq == tp->snd_wl1 && nwin > tp->snd_wnd); } /* Update our send window. @@ -4048,6 +4040,8 @@ static void tcp_reset(struct sock *sk) default: sk->sk_err = ECONNRESET; } + /* This barrier is coupled with smp_rmb() in tcp_poll() */ + smp_wmb(); if (!sock_flag(sk, SOCK_DEAD)) sk->sk_error_report(sk); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 020766292bb..8f8527d4168 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1422,7 +1422,7 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, newsk = tcp_create_openreq_child(sk, req, skb); if (!newsk) - goto exit; + goto exit_nonewsk; newsk->sk_gso_type = SKB_GSO_TCPV4; sk_setup_caps(newsk, dst); @@ -1469,16 +1469,20 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, } #endif + if (__inet_inherit_port(sk, newsk) < 0) { + sock_put(newsk); + goto exit; + } __inet_hash_nolisten(newsk, NULL); - __inet_inherit_port(sk, newsk); return newsk; exit_overflow: NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENOVERFLOWS); +exit_nonewsk: + dst_release(dst); exit: NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS); - dst_release(dst); return NULL; } EXPORT_SYMBOL(tcp_v4_syn_recv_sock); @@ -2571,7 +2575,6 @@ struct sk_buff **tcp4_gro_receive(struct sk_buff **head, struct sk_buff *skb) return tcp_gro_receive(head, skb); } -EXPORT_SYMBOL(tcp4_gro_receive); int tcp4_gro_complete(struct sk_buff *skb) { @@ -2584,7 +2587,6 @@ int tcp4_gro_complete(struct sk_buff *skb) return tcp_gro_complete(skb); } -EXPORT_SYMBOL(tcp4_gro_complete); struct proto tcp_prot = { .name = "TCP", diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index f25b56cb85c..43cf901d765 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -55,7 +55,7 @@ static __inline__ int tcp_in_window(u32 seq, u32 end_seq, u32 s_win, u32 e_win) return 1; if (after(end_seq, s_win) && before(seq, e_win)) return 1; - return (seq == e_win && seq == end_seq); + return seq == e_win && seq == end_seq; } /* diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index de3bd845858..05b1ecf3676 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -224,16 +224,10 @@ void tcp_select_initial_window(int __space, __u32 mss, } } - /* Set initial window to value enough for senders, - * following RFC2414. Senders, not following this RFC, - * will be satisfied with 2. - */ + /* Set initial window to value enough for senders, following RFC5681. */ if (mss > (1 << *rcv_wscale)) { - int init_cwnd = 4; - if (mss > 1460 * 3) - init_cwnd = 2; - else if (mss > 1460) - init_cwnd = 3; + int init_cwnd = rfc3390_bytes_to_packets(mss); + /* when initializing use the value from init_rcv_wnd * rather than the default from above */ @@ -1376,9 +1370,9 @@ static inline int tcp_nagle_check(const struct tcp_sock *tp, const struct sk_buff *skb, unsigned mss_now, int nonagle) { - return (skb->len < mss_now && + return skb->len < mss_now && ((nonagle & TCP_NAGLE_CORK) || - (!nonagle && tp->packets_out && tcp_minshall_check(tp)))); + (!nonagle && tp->packets_out && tcp_minshall_check(tp))); } /* Return non-zero if the Nagle test allows this packet to be @@ -1449,10 +1443,10 @@ int tcp_may_send_now(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb = tcp_send_head(sk); - return (skb && + return skb && tcp_snd_test(sk, skb, tcp_current_mss(sk), (tcp_skb_is_last(sk, skb) ? - tp->nonagle : TCP_NAGLE_PUSH))); + tp->nonagle : TCP_NAGLE_PUSH)); } /* Trim TSO SKB to LEN bytes, put the remaining data into a new packet @@ -2429,6 +2423,12 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst, __u8 rcv_wscale; /* Set this up on the first call only */ req->window_clamp = tp->window_clamp ? : dst_metric(dst, RTAX_WINDOW); + + /* limit the window selection if the user enforce a smaller rx buffer */ + if (sk->sk_userlocks & SOCK_RCVBUF_LOCK && + (req->window_clamp > tcp_full_space(sk) || req->window_clamp == 0)) + req->window_clamp = tcp_full_space(sk); + /* tcp_full_space because it is guaranteed to be the first packet */ tcp_select_initial_window(tcp_full_space(sk), mss - (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0), @@ -2555,6 +2555,11 @@ static void tcp_connect_init(struct sock *sk) tcp_initialize_rcv_mss(sk); + /* limit the window selection if the user enforce a smaller rx buffer */ + if (sk->sk_userlocks & SOCK_RCVBUF_LOCK && + (tp->window_clamp > tcp_full_space(sk) || tp->window_clamp == 0)) + tp->window_clamp = tcp_full_space(sk); + tcp_select_initial_window(tcp_full_space(sk), tp->advmss - (tp->rx_opt.ts_recent_stamp ? tp->tcp_header_len - sizeof(struct tcphdr) : 0), &tp->rcv_wnd, diff --git a/net/ipv4/tcp_probe.c b/net/ipv4/tcp_probe.c index f8efada580e..6211e211417 100644 --- a/net/ipv4/tcp_probe.c +++ b/net/ipv4/tcp_probe.c @@ -214,6 +214,7 @@ static const struct file_operations tcpprobe_fops = { .owner = THIS_MODULE, .open = tcpprobe_open, .read = tcpprobe_read, + .llseek = noop_llseek, }; static __init int tcpprobe_init(void) diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 808bb920c9f..74a6aa00365 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -66,18 +66,18 @@ static void tcp_write_err(struct sock *sk) static int tcp_out_of_resources(struct sock *sk, int do_reset) { struct tcp_sock *tp = tcp_sk(sk); - int orphans = percpu_counter_read_positive(&tcp_orphan_count); + int shift = 0; /* If peer does not open window for long time, or did not transmit * anything for long time, penalize it. */ if ((s32)(tcp_time_stamp - tp->lsndtime) > 2*TCP_RTO_MAX || !do_reset) - orphans <<= 1; + shift++; /* If some dubious ICMP arrived, penalize even more. */ if (sk->sk_err_soft) - orphans <<= 1; + shift++; - if (tcp_too_many_orphans(sk, orphans)) { + if (tcp_too_many_orphans(sk, shift)) { if (net_ratelimit()) printk(KERN_INFO "Out of socket memory\n"); @@ -135,13 +135,16 @@ static void tcp_mtu_probing(struct inet_connection_sock *icsk, struct sock *sk) /* This function calculates a "timeout" which is equivalent to the timeout of a * TCP connection after "boundary" unsuccessful, exponentially backed-off - * retransmissions with an initial RTO of TCP_RTO_MIN. + * retransmissions with an initial RTO of TCP_RTO_MIN or TCP_TIMEOUT_INIT if + * syn_set flag is set. */ static bool retransmits_timed_out(struct sock *sk, - unsigned int boundary) + unsigned int boundary, + unsigned int timeout, + bool syn_set) { - unsigned int timeout, linear_backoff_thresh; - unsigned int start_ts; + unsigned int linear_backoff_thresh, start_ts; + unsigned int rto_base = syn_set ? TCP_TIMEOUT_INIT : TCP_RTO_MIN; if (!inet_csk(sk)->icsk_retransmits) return false; @@ -151,14 +154,15 @@ static bool retransmits_timed_out(struct sock *sk, else start_ts = tcp_sk(sk)->retrans_stamp; - linear_backoff_thresh = ilog2(TCP_RTO_MAX/TCP_RTO_MIN); - - if (boundary <= linear_backoff_thresh) - timeout = ((2 << boundary) - 1) * TCP_RTO_MIN; - else - timeout = ((2 << linear_backoff_thresh) - 1) * TCP_RTO_MIN + - (boundary - linear_backoff_thresh) * TCP_RTO_MAX; + if (likely(timeout == 0)) { + linear_backoff_thresh = ilog2(TCP_RTO_MAX/rto_base); + if (boundary <= linear_backoff_thresh) + timeout = ((2 << boundary) - 1) * rto_base; + else + timeout = ((2 << linear_backoff_thresh) - 1) * rto_base + + (boundary - linear_backoff_thresh) * TCP_RTO_MAX; + } return (tcp_time_stamp - start_ts) >= timeout; } @@ -167,14 +171,15 @@ static int tcp_write_timeout(struct sock *sk) { struct inet_connection_sock *icsk = inet_csk(sk); int retry_until; - bool do_reset; + bool do_reset, syn_set = 0; if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) { if (icsk->icsk_retransmits) dst_negative_advice(sk); retry_until = icsk->icsk_syn_retries ? : sysctl_tcp_syn_retries; + syn_set = 1; } else { - if (retransmits_timed_out(sk, sysctl_tcp_retries1)) { + if (retransmits_timed_out(sk, sysctl_tcp_retries1, 0, 0)) { /* Black hole detection */ tcp_mtu_probing(icsk, sk); @@ -187,14 +192,15 @@ static int tcp_write_timeout(struct sock *sk) retry_until = tcp_orphan_retries(sk, alive); do_reset = alive || - !retransmits_timed_out(sk, retry_until); + !retransmits_timed_out(sk, retry_until, 0, 0); if (tcp_out_of_resources(sk, do_reset)) return 1; } } - if (retransmits_timed_out(sk, retry_until)) { + if (retransmits_timed_out(sk, retry_until, + syn_set ? 0 : icsk->icsk_user_timeout, syn_set)) { /* Has it gone just too far? */ tcp_write_err(sk); return 1; @@ -361,18 +367,19 @@ void tcp_retransmit_timer(struct sock *sk) if (icsk->icsk_retransmits == 0) { int mib_idx; - if (icsk->icsk_ca_state == TCP_CA_Disorder) { - if (tcp_is_sack(tp)) - mib_idx = LINUX_MIB_TCPSACKFAILURES; - else - mib_idx = LINUX_MIB_TCPRENOFAILURES; - } else if (icsk->icsk_ca_state == TCP_CA_Recovery) { + if (icsk->icsk_ca_state == TCP_CA_Recovery) { if (tcp_is_sack(tp)) mib_idx = LINUX_MIB_TCPSACKRECOVERYFAIL; else mib_idx = LINUX_MIB_TCPRENORECOVERYFAIL; } else if (icsk->icsk_ca_state == TCP_CA_Loss) { mib_idx = LINUX_MIB_TCPLOSSFAILURES; + } else if ((icsk->icsk_ca_state == TCP_CA_Disorder) || + tp->sacked_out) { + if (tcp_is_sack(tp)) + mib_idx = LINUX_MIB_TCPSACKFAILURES; + else + mib_idx = LINUX_MIB_TCPRENOFAILURES; } else { mib_idx = LINUX_MIB_TCPTIMEOUTS; } @@ -436,7 +443,7 @@ out_reset_timer: icsk->icsk_rto = min(icsk->icsk_rto << 1, TCP_RTO_MAX); } inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, icsk->icsk_rto, TCP_RTO_MAX); - if (retransmits_timed_out(sk, sysctl_tcp_retries1 + 1)) + if (retransmits_timed_out(sk, sysctl_tcp_retries1 + 1, 0, 0)) __sk_dst_reset(sk); out:; @@ -556,7 +563,14 @@ static void tcp_keepalive_timer (unsigned long data) elapsed = keepalive_time_elapsed(tp); if (elapsed >= keepalive_time_when(tp)) { - if (icsk->icsk_probes_out >= keepalive_probes(tp)) { + /* If the TCP_USER_TIMEOUT option is enabled, use that + * to determine when to timeout instead. + */ + if ((icsk->icsk_user_timeout != 0 && + elapsed >= icsk->icsk_user_timeout && + icsk->icsk_probes_out > 0) || + (icsk->icsk_user_timeout == 0 && + icsk->icsk_probes_out >= keepalive_probes(tp))) { tcp_send_active_reset(sk, GFP_ATOMIC); tcp_write_err(sk); goto out; diff --git a/net/ipv4/tcp_westwood.c b/net/ipv4/tcp_westwood.c index 20151d6a624..a534dda5456 100644 --- a/net/ipv4/tcp_westwood.c +++ b/net/ipv4/tcp_westwood.c @@ -80,7 +80,7 @@ static void tcp_westwood_init(struct sock *sk) */ static inline u32 westwood_do_filter(u32 a, u32 b) { - return (((7 * a) + b) >> 3); + return ((7 * a) + b) >> 3; } static void westwood_filter(struct westwood *w, u32 delta) diff --git a/net/ipv4/tunnel4.c b/net/ipv4/tunnel4.c index 59186ca7808..9a17bd2a0a3 100644 --- a/net/ipv4/tunnel4.c +++ b/net/ipv4/tunnel4.c @@ -14,8 +14,8 @@ #include <net/protocol.h> #include <net/xfrm.h> -static struct xfrm_tunnel *tunnel4_handlers; -static struct xfrm_tunnel *tunnel64_handlers; +static struct xfrm_tunnel *tunnel4_handlers __read_mostly; +static struct xfrm_tunnel *tunnel64_handlers __read_mostly; static DEFINE_MUTEX(tunnel4_mutex); static inline struct xfrm_tunnel **fam_handlers(unsigned short family) @@ -39,7 +39,7 @@ int xfrm4_tunnel_register(struct xfrm_tunnel *handler, unsigned short family) } handler->next = *pprev; - *pprev = handler; + rcu_assign_pointer(*pprev, handler); ret = 0; @@ -73,6 +73,11 @@ int xfrm4_tunnel_deregister(struct xfrm_tunnel *handler, unsigned short family) } EXPORT_SYMBOL(xfrm4_tunnel_deregister); +#define for_each_tunnel_rcu(head, handler) \ + for (handler = rcu_dereference(head); \ + handler != NULL; \ + handler = rcu_dereference(handler->next)) \ + static int tunnel4_rcv(struct sk_buff *skb) { struct xfrm_tunnel *handler; @@ -80,7 +85,7 @@ static int tunnel4_rcv(struct sk_buff *skb) if (!pskb_may_pull(skb, sizeof(struct iphdr))) goto drop; - for (handler = tunnel4_handlers; handler; handler = handler->next) + for_each_tunnel_rcu(tunnel4_handlers, handler) if (!handler->handler(skb)) return 0; @@ -99,7 +104,7 @@ static int tunnel64_rcv(struct sk_buff *skb) if (!pskb_may_pull(skb, sizeof(struct ipv6hdr))) goto drop; - for (handler = tunnel64_handlers; handler; handler = handler->next) + for_each_tunnel_rcu(tunnel64_handlers, handler) if (!handler->handler(skb)) return 0; @@ -115,7 +120,7 @@ static void tunnel4_err(struct sk_buff *skb, u32 info) { struct xfrm_tunnel *handler; - for (handler = tunnel4_handlers; handler; handler = handler->next) + for_each_tunnel_rcu(tunnel4_handlers, handler) if (!handler->err_handler(skb, info)) break; } @@ -125,7 +130,7 @@ static void tunnel64_err(struct sk_buff *skb, u32 info) { struct xfrm_tunnel *handler; - for (handler = tunnel64_handlers; handler; handler = handler->next) + for_each_tunnel_rcu(tunnel64_handlers, handler) if (!handler->err_handler(skb, info)) break; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 32e0bef60d0..b3f7e8cf18a 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -797,7 +797,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, return -EOPNOTSUPP; ipc.opt = NULL; - ipc.shtx.flags = 0; + ipc.tx_flags = 0; if (up->pending) { /* @@ -845,7 +845,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, ipc.addr = inet->inet_saddr; ipc.oif = sk->sk_bound_dev_if; - err = sock_tx_timestamp(msg, sk, &ipc.shtx); + err = sock_tx_timestamp(sk, &ipc.tx_flags); if (err) return err; if (msg->msg_controllen) { @@ -1260,6 +1260,49 @@ void udp_lib_unhash(struct sock *sk) } EXPORT_SYMBOL(udp_lib_unhash); +/* + * inet_rcv_saddr was changed, we must rehash secondary hash + */ +void udp_lib_rehash(struct sock *sk, u16 newhash) +{ + if (sk_hashed(sk)) { + struct udp_table *udptable = sk->sk_prot->h.udp_table; + struct udp_hslot *hslot, *hslot2, *nhslot2; + + hslot2 = udp_hashslot2(udptable, udp_sk(sk)->udp_portaddr_hash); + nhslot2 = udp_hashslot2(udptable, newhash); + udp_sk(sk)->udp_portaddr_hash = newhash; + if (hslot2 != nhslot2) { + hslot = udp_hashslot(udptable, sock_net(sk), + udp_sk(sk)->udp_port_hash); + /* we must lock primary chain too */ + spin_lock_bh(&hslot->lock); + + spin_lock(&hslot2->lock); + hlist_nulls_del_init_rcu(&udp_sk(sk)->udp_portaddr_node); + hslot2->count--; + spin_unlock(&hslot2->lock); + + spin_lock(&nhslot2->lock); + hlist_nulls_add_head_rcu(&udp_sk(sk)->udp_portaddr_node, + &nhslot2->head); + nhslot2->count++; + spin_unlock(&nhslot2->lock); + + spin_unlock_bh(&hslot->lock); + } + } +} +EXPORT_SYMBOL(udp_lib_rehash); + +static void udp_v4_rehash(struct sock *sk) +{ + u16 new_hash = udp4_portaddr_hash(sock_net(sk), + inet_sk(sk)->inet_rcv_saddr, + inet_sk(sk)->inet_num); + udp_lib_rehash(sk, new_hash); +} + static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) { int rc; @@ -1843,6 +1886,7 @@ struct proto udp_prot = { .backlog_rcv = __udp_queue_rcv_skb, .hash = udp_lib_hash, .unhash = udp_lib_unhash, + .rehash = udp_v4_rehash, .get_port = udp_v4_get_port, .memory_allocated = &udp_memory_allocated, .sysctl_mem = sysctl_udp_mem, diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 869078d4eeb..4464f3bff6a 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -61,7 +61,7 @@ static int xfrm4_get_saddr(struct net *net, static int xfrm4_get_tos(struct flowi *fl) { - return fl->fl4_tos; + return IPTOS_RT_MASK & fl->fl4_tos; /* Strip ECN bits */ } static int xfrm4_init_path(struct xfrm_dst *path, struct dst_entry *dst, @@ -174,7 +174,7 @@ static inline int xfrm4_garbage_collect(struct dst_ops *ops) struct net *net = container_of(ops, struct net, xfrm.xfrm4_dst_ops); xfrm4_policy_afinfo.garbage_collect(net); - return (atomic_read(&ops->entries) > ops->gc_thresh * 2); + return (dst_entries_get_slow(ops) > ops->gc_thresh * 2); } static void xfrm4_update_pmtu(struct dst_entry *dst, u32 mtu) @@ -232,7 +232,6 @@ static struct dst_ops xfrm4_dst_ops = { .ifdown = xfrm4_dst_ifdown, .local_out = __ip_local_out, .gc_thresh = 1024, - .entries = ATOMIC_INIT(0), }; static struct xfrm_policy_afinfo xfrm4_policy_afinfo = { @@ -288,6 +287,7 @@ void __init xfrm4_init(int rt_max_size) * and start cleaning when were 1/2 full */ xfrm4_dst_ops.gc_thresh = rt_max_size/2; + dst_entries_init(&xfrm4_dst_ops); xfrm4_state_init(); xfrm4_policy_init(); diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c index 1ef1366a0a0..47947624ecc 100644 --- a/net/ipv4/xfrm4_state.c +++ b/net/ipv4/xfrm4_state.c @@ -21,21 +21,25 @@ static int xfrm4_init_flags(struct xfrm_state *x) } static void -__xfrm4_init_tempsel(struct xfrm_state *x, struct flowi *fl, - struct xfrm_tmpl *tmpl, - xfrm_address_t *daddr, xfrm_address_t *saddr) +__xfrm4_init_tempsel(struct xfrm_selector *sel, struct flowi *fl) +{ + sel->daddr.a4 = fl->fl4_dst; + sel->saddr.a4 = fl->fl4_src; + sel->dport = xfrm_flowi_dport(fl); + sel->dport_mask = htons(0xffff); + sel->sport = xfrm_flowi_sport(fl); + sel->sport_mask = htons(0xffff); + sel->family = AF_INET; + sel->prefixlen_d = 32; + sel->prefixlen_s = 32; + sel->proto = fl->proto; + sel->ifindex = fl->oif; +} + +static void +xfrm4_init_temprop(struct xfrm_state *x, struct xfrm_tmpl *tmpl, + xfrm_address_t *daddr, xfrm_address_t *saddr) { - x->sel.daddr.a4 = fl->fl4_dst; - x->sel.saddr.a4 = fl->fl4_src; - x->sel.dport = xfrm_flowi_dport(fl); - x->sel.dport_mask = htons(0xffff); - x->sel.sport = xfrm_flowi_sport(fl); - x->sel.sport_mask = htons(0xffff); - x->sel.family = AF_INET; - x->sel.prefixlen_d = 32; - x->sel.prefixlen_s = 32; - x->sel.proto = fl->proto; - x->sel.ifindex = fl->oif; x->id = tmpl->id; if (x->id.daddr.a4 == 0) x->id.daddr.a4 = daddr->a4; @@ -70,6 +74,7 @@ static struct xfrm_state_afinfo xfrm4_state_afinfo = { .owner = THIS_MODULE, .init_flags = xfrm4_init_flags, .init_tempsel = __xfrm4_init_tempsel, + .init_temprop = xfrm4_init_temprop, .output = xfrm4_output, .extract_input = xfrm4_extract_input, .extract_output = xfrm4_extract_output, diff --git a/net/ipv4/xfrm4_tunnel.c b/net/ipv4/xfrm4_tunnel.c index 41f5982d208..82806455e85 100644 --- a/net/ipv4/xfrm4_tunnel.c +++ b/net/ipv4/xfrm4_tunnel.c @@ -58,14 +58,14 @@ static int xfrm_tunnel_err(struct sk_buff *skb, u32 info) return -ENOENT; } -static struct xfrm_tunnel xfrm_tunnel_handler = { +static struct xfrm_tunnel xfrm_tunnel_handler __read_mostly = { .handler = xfrm_tunnel_rcv, .err_handler = xfrm_tunnel_err, .priority = 2, }; #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -static struct xfrm_tunnel xfrm64_tunnel_handler = { +static struct xfrm_tunnel xfrm64_tunnel_handler __read_mostly = { .handler = xfrm_tunnel_rcv, .err_handler = xfrm_tunnel_err, .priority = 2, diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index ab70a3fbcaf..ec7a91d9e86 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -243,7 +243,7 @@ static inline bool addrconf_qdisc_ok(const struct net_device *dev) /* Check if a route is valid prefix route */ static inline int addrconf_is_prefix_route(const struct rt6_info *rt) { - return ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0); + return (rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0; } static void addrconf_del_timer(struct inet6_ifaddr *ifp) @@ -1544,7 +1544,7 @@ static int addrconf_ifid_infiniband(u8 *eui, struct net_device *dev) return 0; } -int __ipv6_isatap_ifid(u8 *eui, __be32 addr) +static int __ipv6_isatap_ifid(u8 *eui, __be32 addr) { if (addr == 0) return -1; @@ -1560,7 +1560,6 @@ int __ipv6_isatap_ifid(u8 *eui, __be32 addr) memcpy(eui + 4, &addr, 4); return 0; } -EXPORT_SYMBOL(__ipv6_isatap_ifid); static int addrconf_ifid_sit(u8 *eui, struct net_device *dev) { @@ -2964,7 +2963,8 @@ static void addrconf_dad_completed(struct inet6_ifaddr *ifp) start sending router solicitations. */ - if (ifp->idev->cnf.forwarding == 0 && + if ((ifp->idev->cnf.forwarding == 0 || + ifp->idev->cnf.forwarding == 2) && ifp->idev->cnf.rtr_solicits > 0 && (dev->flags&IFF_LOOPBACK) == 0 && (ipv6_addr_type(&ifp->addr) & IPV6_ADDR_LINKLOCAL)) { @@ -4637,10 +4637,12 @@ int __init addrconf_init(void) if (err < 0) { printk(KERN_CRIT "IPv6 Addrconf:" " cannot initialize default policy table: %d.\n", err); - return err; + goto out; } - register_pernet_subsys(&addrconf_ops); + err = register_pernet_subsys(&addrconf_ops); + if (err < 0) + goto out_addrlabel; /* The addrconf netdev notifier requires that loopback_dev * has it's ipv6 private information allocated and setup @@ -4692,7 +4694,9 @@ errout: unregister_netdevice_notifier(&ipv6_dev_notf); errlo: unregister_pernet_subsys(&addrconf_ops); - +out_addrlabel: + ipv6_addr_label_cleanup(); +out: return err; } @@ -4703,6 +4707,7 @@ void addrconf_cleanup(void) unregister_netdevice_notifier(&ipv6_dev_notf); unregister_pernet_subsys(&addrconf_ops); + ipv6_addr_label_cleanup(); rtnl_lock(); diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c index f0e774cea38..c8993e5a337 100644 --- a/net/ipv6/addrlabel.c +++ b/net/ipv6/addrlabel.c @@ -393,6 +393,11 @@ int __init ipv6_addr_label_init(void) return register_pernet_subsys(&ipv6_addr_label_ops); } +void ipv6_addr_label_cleanup(void) +{ + unregister_pernet_subsys(&ipv6_addr_label_ops); +} + static const struct nla_policy ifal_policy[IFAL_MAX+1] = { [IFAL_ADDRESS] = { .len = sizeof(struct in6_addr), }, [IFAL_LABEL] = { .len = sizeof(u32), }, @@ -513,10 +518,9 @@ static int ip6addrlbl_dump(struct sk_buff *skb, struct netlink_callback *cb) static inline int ip6addrlbl_msgsize(void) { - return (NLMSG_ALIGN(sizeof(struct ifaddrlblmsg)) + return NLMSG_ALIGN(sizeof(struct ifaddrlblmsg)) + nla_total_size(16) /* IFAL_ADDRESS */ - + nla_total_size(4) /* IFAL_LABEL */ - ); + + nla_total_size(4); /* IFAL_LABEL */ } static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr* nlh, diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 56b9bf2516f..54e8e42f7a8 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -343,7 +343,8 @@ int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) */ v4addr = LOOPBACK4_IPV6; if (!(addr_type & IPV6_ADDR_MULTICAST)) { - if (!ipv6_chk_addr(net, &addr->sin6_addr, + if (!inet->transparent && + !ipv6_chk_addr(net, &addr->sin6_addr, dev, 0)) { err = -EADDRNOTAVAIL; goto out_unlock; @@ -467,7 +468,7 @@ int inet6_getname(struct socket *sock, struct sockaddr *uaddr, if (ipv6_addr_type(&sin->sin6_addr) & IPV6_ADDR_LINKLOCAL) sin->sin6_scope_id = sk->sk_bound_dev_if; *uaddr_len = sizeof(*sin); - return(0); + return 0; } EXPORT_SYMBOL(inet6_getname); @@ -488,7 +489,7 @@ int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) case SIOCADDRT: case SIOCDELRT: - return(ipv6_route_ioctl(net, cmd, (void __user *)arg)); + return ipv6_route_ioctl(net, cmd, (void __user *)arg); case SIOCSIFADDR: return addrconf_add_ifaddr(net, (void __user *) arg); @@ -502,7 +503,7 @@ int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) return sk->sk_prot->ioctl(sk, cmd, arg); } /*NOTREACHED*/ - return(0); + return 0; } EXPORT_SYMBOL(inet6_ioctl); diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c index 7d929a22cbc..320bdb877ee 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -105,9 +105,12 @@ ipv4_connected: if (ipv6_addr_any(&np->saddr)) ipv6_addr_set_v4mapped(inet->inet_saddr, &np->saddr); - if (ipv6_addr_any(&np->rcv_saddr)) + if (ipv6_addr_any(&np->rcv_saddr)) { ipv6_addr_set_v4mapped(inet->inet_rcv_saddr, &np->rcv_saddr); + if (sk->sk_prot->rehash) + sk->sk_prot->rehash(sk); + } goto out; } @@ -181,6 +184,8 @@ ipv4_connected: if (ipv6_addr_any(&np->rcv_saddr)) { ipv6_addr_copy(&np->rcv_saddr, &fl.fl6_src); inet->inet_rcv_saddr = LOOPBACK4_IPV6; + if (sk->sk_prot->rehash) + sk->sk_prot->rehash(sk); } ip6_dst_store(sk, dst, @@ -572,6 +577,25 @@ int datagram_recv_ctl(struct sock *sk, struct msghdr *msg, struct sk_buff *skb) u8 *ptr = nh + opt->dst1; put_cmsg(msg, SOL_IPV6, IPV6_2292DSTOPTS, (ptr[1]+1)<<3, ptr); } + if (np->rxopt.bits.rxorigdstaddr) { + struct sockaddr_in6 sin6; + u16 *ports = (u16 *) skb_transport_header(skb); + + if (skb_transport_offset(skb) + 4 <= skb->len) { + /* All current transport protocols have the port numbers in the + * first four bytes of the transport header and this function is + * written with this assumption in mind. + */ + + sin6.sin6_family = AF_INET6; + ipv6_addr_copy(&sin6.sin6_addr, &ipv6_hdr(skb)->daddr); + sin6.sin6_port = ports[1]; + sin6.sin6_flowinfo = 0; + sin6.sin6_scope_id = 0; + + put_cmsg(msg, SOL_IPV6, IPV6_ORIGDSTADDR, sizeof(sin6), &sin6); + } + } return 0; } diff --git a/net/ipv6/exthdrs_core.c b/net/ipv6/exthdrs_core.c index e1caa5d526c..14ed0a955b5 100644 --- a/net/ipv6/exthdrs_core.c +++ b/net/ipv6/exthdrs_core.c @@ -13,12 +13,12 @@ int ipv6_ext_hdr(u8 nexthdr) /* * find out if nexthdr is an extension header or a protocol */ - return ( (nexthdr == NEXTHDR_HOP) || + return (nexthdr == NEXTHDR_HOP) || (nexthdr == NEXTHDR_ROUTING) || (nexthdr == NEXTHDR_FRAGMENT) || (nexthdr == NEXTHDR_AUTH) || (nexthdr == NEXTHDR_NONE) || - (nexthdr == NEXTHDR_DEST) ); + (nexthdr == NEXTHDR_DEST); } /* diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index b1108ede18e..d829874d894 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -34,11 +34,10 @@ struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl, { struct fib_lookup_arg arg = { .lookup_ptr = lookup, + .flags = FIB_LOOKUP_NOREF, }; fib_rules_lookup(net->ipv6.fib6_rules_ops, fl, flags, &arg); - if (arg.rule) - fib_rule_put(arg.rule); if (arg.result) return arg.result; diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index b6a585909d3..de382114609 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -1500,15 +1500,18 @@ static void fib6_gc_timer_cb(unsigned long arg) static int __net_init fib6_net_init(struct net *net) { + size_t size = sizeof(struct hlist_head) * FIB6_TABLE_HASHSZ; + setup_timer(&net->ipv6.ip6_fib_timer, fib6_gc_timer_cb, (unsigned long)net); net->ipv6.rt6_stats = kzalloc(sizeof(*net->ipv6.rt6_stats), GFP_KERNEL); if (!net->ipv6.rt6_stats) goto out_timer; - net->ipv6.fib_table_hash = kcalloc(FIB6_TABLE_HASHSZ, - sizeof(*net->ipv6.fib_table_hash), - GFP_KERNEL); + /* Avoid false sharing : Use at least a full cache line */ + size = max_t(size_t, size, L1_CACHE_BYTES); + + net->ipv6.fib_table_hash = kzalloc(size, GFP_KERNEL); if (!net->ipv6.fib_table_hash) goto out_rt6_stats; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index d40b330c0ee..99157b4cd56 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -637,9 +637,9 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) } mtu -= hlen + sizeof(struct frag_hdr); - if (skb_has_frags(skb)) { + if (skb_has_frag_list(skb)) { int first_len = skb_pagelen(skb); - int truesizes = 0; + struct sk_buff *frag2; if (first_len - hlen > mtu || ((first_len - hlen) & 7) || @@ -651,18 +651,18 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) if (frag->len > mtu || ((frag->len & 7) && frag->next) || skb_headroom(frag) < hlen) - goto slow_path; + goto slow_path_clean; /* Partially cloned skb? */ if (skb_shared(frag)) - goto slow_path; + goto slow_path_clean; BUG_ON(frag->sk); if (skb->sk) { frag->sk = skb->sk; frag->destructor = sock_wfree; - truesizes += frag->truesize; } + skb->truesize -= frag->truesize; } err = 0; @@ -693,7 +693,6 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) first_len = skb_pagelen(skb); skb->data_len = first_len - skb_headlen(skb); - skb->truesize -= truesizes; skb->len = first_len; ipv6_hdr(skb)->payload_len = htons(first_len - sizeof(struct ipv6hdr)); @@ -756,6 +755,15 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) IPSTATS_MIB_FRAGFAILS); dst_release(&rt->dst); return err; + +slow_path_clean: + skb_walk_frags(skb, frag2) { + if (frag2 == frag) + break; + frag2->sk = NULL; + frag2->destructor = NULL; + skb->truesize += frag2->truesize; + } } slow_path: @@ -870,8 +878,8 @@ static inline int ip6_rt_check(struct rt6key *rt_key, struct in6_addr *fl_addr, struct in6_addr *addr_cache) { - return ((rt_key->plen != 128 || !ipv6_addr_equal(fl_addr, &rt_key->addr)) && - (addr_cache == NULL || !ipv6_addr_equal(fl_addr, addr_cache))); + return (rt_key->plen != 128 || !ipv6_addr_equal(fl_addr, &rt_key->addr)) && + (addr_cache == NULL || !ipv6_addr_equal(fl_addr, addr_cache)); } static struct dst_entry *ip6_sk_dst_check(struct sock *sk, diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 0fd027f3f47..c2c0f89397b 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -75,7 +75,7 @@ MODULE_LICENSE("GPL"); (addr)->s6_addr32[2] ^ (addr)->s6_addr32[3]) & \ (HASH_SIZE - 1)) -static void ip6_tnl_dev_init(struct net_device *dev); +static int ip6_tnl_dev_init(struct net_device *dev); static void ip6_tnl_dev_setup(struct net_device *dev); static int ip6_tnl_net_id __read_mostly; @@ -83,15 +83,42 @@ struct ip6_tnl_net { /* the IPv6 tunnel fallback device */ struct net_device *fb_tnl_dev; /* lists for storing tunnels in use */ - struct ip6_tnl *tnls_r_l[HASH_SIZE]; - struct ip6_tnl *tnls_wc[1]; - struct ip6_tnl **tnls[2]; + struct ip6_tnl __rcu *tnls_r_l[HASH_SIZE]; + struct ip6_tnl __rcu *tnls_wc[1]; + struct ip6_tnl __rcu **tnls[2]; }; +/* often modified stats are per cpu, other are shared (netdev->stats) */ +struct pcpu_tstats { + unsigned long rx_packets; + unsigned long rx_bytes; + unsigned long tx_packets; + unsigned long tx_bytes; +}; + +static struct net_device_stats *ip6_get_stats(struct net_device *dev) +{ + struct pcpu_tstats sum = { 0 }; + int i; + + for_each_possible_cpu(i) { + const struct pcpu_tstats *tstats = per_cpu_ptr(dev->tstats, i); + + sum.rx_packets += tstats->rx_packets; + sum.rx_bytes += tstats->rx_bytes; + sum.tx_packets += tstats->tx_packets; + sum.tx_bytes += tstats->tx_bytes; + } + dev->stats.rx_packets = sum.rx_packets; + dev->stats.rx_bytes = sum.rx_bytes; + dev->stats.tx_packets = sum.tx_packets; + dev->stats.tx_bytes = sum.tx_bytes; + return &dev->stats; +} + /* - * Locking : hash tables are protected by RCU and a spinlock + * Locking : hash tables are protected by RCU and RTNL */ -static DEFINE_SPINLOCK(ip6_tnl_lock); static inline struct dst_entry *ip6_tnl_dst_check(struct ip6_tnl *t) { @@ -138,8 +165,8 @@ static inline void ip6_tnl_dst_store(struct ip6_tnl *t, struct dst_entry *dst) static struct ip6_tnl * ip6_tnl_lookup(struct net *net, struct in6_addr *remote, struct in6_addr *local) { - unsigned h0 = HASH(remote); - unsigned h1 = HASH(local); + unsigned int h0 = HASH(remote); + unsigned int h1 = HASH(local); struct ip6_tnl *t; struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id); @@ -167,7 +194,7 @@ ip6_tnl_lookup(struct net *net, struct in6_addr *remote, struct in6_addr *local) * Return: head of IPv6 tunnel list **/ -static struct ip6_tnl ** +static struct ip6_tnl __rcu ** ip6_tnl_bucket(struct ip6_tnl_net *ip6n, struct ip6_tnl_parm *p) { struct in6_addr *remote = &p->raddr; @@ -190,12 +217,10 @@ ip6_tnl_bucket(struct ip6_tnl_net *ip6n, struct ip6_tnl_parm *p) static void ip6_tnl_link(struct ip6_tnl_net *ip6n, struct ip6_tnl *t) { - struct ip6_tnl **tp = ip6_tnl_bucket(ip6n, &t->parms); + struct ip6_tnl __rcu **tp = ip6_tnl_bucket(ip6n, &t->parms); - spin_lock_bh(&ip6_tnl_lock); - t->next = *tp; + rcu_assign_pointer(t->next , rtnl_dereference(*tp)); rcu_assign_pointer(*tp, t); - spin_unlock_bh(&ip6_tnl_lock); } /** @@ -206,18 +231,25 @@ ip6_tnl_link(struct ip6_tnl_net *ip6n, struct ip6_tnl *t) static void ip6_tnl_unlink(struct ip6_tnl_net *ip6n, struct ip6_tnl *t) { - struct ip6_tnl **tp; - - for (tp = ip6_tnl_bucket(ip6n, &t->parms); *tp; tp = &(*tp)->next) { - if (t == *tp) { - spin_lock_bh(&ip6_tnl_lock); - *tp = t->next; - spin_unlock_bh(&ip6_tnl_lock); + struct ip6_tnl __rcu **tp; + struct ip6_tnl *iter; + + for (tp = ip6_tnl_bucket(ip6n, &t->parms); + (iter = rtnl_dereference(*tp)) != NULL; + tp = &iter->next) { + if (t == iter) { + rcu_assign_pointer(*tp, t->next); break; } } } +static void ip6_dev_free(struct net_device *dev) +{ + free_percpu(dev->tstats); + free_netdev(dev); +} + /** * ip6_tnl_create() - create a new tunnel * @p: tunnel parameters @@ -256,7 +288,9 @@ static struct ip6_tnl *ip6_tnl_create(struct net *net, struct ip6_tnl_parm *p) t = netdev_priv(dev); t->parms = *p; - ip6_tnl_dev_init(dev); + err = ip6_tnl_dev_init(dev); + if (err < 0) + goto failed_free; if ((err = register_netdevice(dev)) < 0) goto failed_free; @@ -266,7 +300,7 @@ static struct ip6_tnl *ip6_tnl_create(struct net *net, struct ip6_tnl_parm *p) return t; failed_free: - free_netdev(dev); + ip6_dev_free(dev); failed: return NULL; } @@ -290,10 +324,13 @@ static struct ip6_tnl *ip6_tnl_locate(struct net *net, { struct in6_addr *remote = &p->raddr; struct in6_addr *local = &p->laddr; + struct ip6_tnl __rcu **tp; struct ip6_tnl *t; struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id); - for (t = *ip6_tnl_bucket(ip6n, p); t; t = t->next) { + for (tp = ip6_tnl_bucket(ip6n, p); + (t = rtnl_dereference(*tp)) != NULL; + tp = &t->next) { if (ipv6_addr_equal(local, &t->parms.laddr) && ipv6_addr_equal(remote, &t->parms.raddr)) return t; @@ -318,13 +355,10 @@ ip6_tnl_dev_uninit(struct net_device *dev) struct net *net = dev_net(dev); struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id); - if (dev == ip6n->fb_tnl_dev) { - spin_lock_bh(&ip6_tnl_lock); - ip6n->tnls_wc[0] = NULL; - spin_unlock_bh(&ip6_tnl_lock); - } else { + if (dev == ip6n->fb_tnl_dev) + rcu_assign_pointer(ip6n->tnls_wc[0], NULL); + else ip6_tnl_unlink(ip6n, t); - } ip6_tnl_dst_reset(t); dev_put(dev); } @@ -702,6 +736,8 @@ static int ip6_tnl_rcv(struct sk_buff *skb, __u16 protocol, if ((t = ip6_tnl_lookup(dev_net(skb->dev), &ipv6h->saddr, &ipv6h->daddr)) != NULL) { + struct pcpu_tstats *tstats; + if (t->parms.proto != ipproto && t->parms.proto != 0) { rcu_read_unlock(); goto discard; @@ -724,10 +760,16 @@ static int ip6_tnl_rcv(struct sk_buff *skb, __u16 protocol, skb->pkt_type = PACKET_HOST; memset(skb->cb, 0, sizeof(struct inet6_skb_parm)); - skb_tunnel_rx(skb, t->dev); + tstats = this_cpu_ptr(t->dev->tstats); + tstats->rx_packets++; + tstats->rx_bytes += skb->len; + + __skb_tunnel_rx(skb, t->dev); dscp_ecn_decapsulate(t, ipv6h, skb); + netif_rx(skb); + rcu_read_unlock(); return 0; } @@ -934,8 +976,10 @@ static int ip6_tnl_xmit2(struct sk_buff *skb, err = ip6_local_out(skb); if (net_xmit_eval(err) == 0) { - stats->tx_bytes += pkt_len; - stats->tx_packets++; + struct pcpu_tstats *tstats = this_cpu_ptr(t->dev->tstats); + + tstats->tx_bytes += pkt_len; + tstats->tx_packets++; } else { stats->tx_errors++; stats->tx_aborted_errors++; @@ -1300,12 +1344,14 @@ ip6_tnl_change_mtu(struct net_device *dev, int new_mtu) static const struct net_device_ops ip6_tnl_netdev_ops = { - .ndo_uninit = ip6_tnl_dev_uninit, + .ndo_uninit = ip6_tnl_dev_uninit, .ndo_start_xmit = ip6_tnl_xmit, - .ndo_do_ioctl = ip6_tnl_ioctl, + .ndo_do_ioctl = ip6_tnl_ioctl, .ndo_change_mtu = ip6_tnl_change_mtu, + .ndo_get_stats = ip6_get_stats, }; + /** * ip6_tnl_dev_setup - setup virtual tunnel device * @dev: virtual device associated with tunnel @@ -1317,7 +1363,7 @@ static const struct net_device_ops ip6_tnl_netdev_ops = { static void ip6_tnl_dev_setup(struct net_device *dev) { dev->netdev_ops = &ip6_tnl_netdev_ops; - dev->destructor = free_netdev; + dev->destructor = ip6_dev_free; dev->type = ARPHRD_TUNNEL6; dev->hard_header_len = LL_MAX_HEADER + sizeof (struct ipv6hdr); @@ -1333,12 +1379,17 @@ static void ip6_tnl_dev_setup(struct net_device *dev) * @dev: virtual device associated with tunnel **/ -static inline void +static inline int ip6_tnl_dev_init_gen(struct net_device *dev) { struct ip6_tnl *t = netdev_priv(dev); + t->dev = dev; strcpy(t->parms.name, dev->name); + dev->tstats = alloc_percpu(struct pcpu_tstats); + if (!dev->tstats) + return -ENOMEM; + return 0; } /** @@ -1346,11 +1397,15 @@ ip6_tnl_dev_init_gen(struct net_device *dev) * @dev: virtual device associated with tunnel **/ -static void ip6_tnl_dev_init(struct net_device *dev) +static int ip6_tnl_dev_init(struct net_device *dev) { struct ip6_tnl *t = netdev_priv(dev); - ip6_tnl_dev_init_gen(dev); + int err = ip6_tnl_dev_init_gen(dev); + + if (err) + return err; ip6_tnl_link_config(t); + return 0; } /** @@ -1360,25 +1415,29 @@ static void ip6_tnl_dev_init(struct net_device *dev) * Return: 0 **/ -static void __net_init ip6_fb_tnl_dev_init(struct net_device *dev) +static int __net_init ip6_fb_tnl_dev_init(struct net_device *dev) { struct ip6_tnl *t = netdev_priv(dev); struct net *net = dev_net(dev); struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id); + int err = ip6_tnl_dev_init_gen(dev); + + if (err) + return err; - ip6_tnl_dev_init_gen(dev); t->parms.proto = IPPROTO_IPV6; dev_hold(dev); - ip6n->tnls_wc[0] = t; + rcu_assign_pointer(ip6n->tnls_wc[0], t); + return 0; } -static struct xfrm6_tunnel ip4ip6_handler = { +static struct xfrm6_tunnel ip4ip6_handler __read_mostly = { .handler = ip4ip6_rcv, .err_handler = ip4ip6_err, .priority = 1, }; -static struct xfrm6_tunnel ip6ip6_handler = { +static struct xfrm6_tunnel ip6ip6_handler __read_mostly = { .handler = ip6ip6_rcv, .err_handler = ip6ip6_err, .priority = 1, @@ -1391,14 +1450,14 @@ static void __net_exit ip6_tnl_destroy_tunnels(struct ip6_tnl_net *ip6n) LIST_HEAD(list); for (h = 0; h < HASH_SIZE; h++) { - t = ip6n->tnls_r_l[h]; + t = rtnl_dereference(ip6n->tnls_r_l[h]); while (t != NULL) { unregister_netdevice_queue(t->dev, &list); - t = t->next; + t = rtnl_dereference(t->next); } } - t = ip6n->tnls_wc[0]; + t = rtnl_dereference(ip6n->tnls_wc[0]); unregister_netdevice_queue(t->dev, &list); unregister_netdevice_many(&list); } @@ -1419,7 +1478,9 @@ static int __net_init ip6_tnl_init_net(struct net *net) goto err_alloc_dev; dev_net_set(ip6n->fb_tnl_dev, net); - ip6_fb_tnl_dev_init(ip6n->fb_tnl_dev); + err = ip6_fb_tnl_dev_init(ip6n->fb_tnl_dev); + if (err < 0) + goto err_register; err = register_netdev(ip6n->fb_tnl_dev); if (err < 0) @@ -1427,7 +1488,7 @@ static int __net_init ip6_tnl_init_net(struct net *net) return 0; err_register: - free_netdev(ip6n->fb_tnl_dev); + ip6_dev_free(ip6n->fb_tnl_dev); err_alloc_dev: return err; } diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index 66078dad7fe..6f32ffce702 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -667,6 +667,7 @@ static int pim6_rcv(struct sk_buff *skb) skb_tunnel_rx(skb, reg_dev); netif_rx(skb); + dev_put(reg_dev); return 0; drop: diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index a7f66bc8f0b..0553867a317 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -342,6 +342,21 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, retv = 0; break; + case IPV6_TRANSPARENT: + if (optlen < sizeof(int)) + goto e_inval; + /* we don't have a separate transparent bit for IPV6 we use the one in the IPv4 socket */ + inet_sk(sk)->transparent = valbool; + retv = 0; + break; + + case IPV6_RECVORIGDSTADDR: + if (optlen < sizeof(int)) + goto e_inval; + np->rxopt.bits.rxorigdstaddr = valbool; + retv = 0; + break; + case IPV6_HOPOPTS: case IPV6_RTHDRDSTOPTS: case IPV6_RTHDR: @@ -1104,6 +1119,14 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname, break; } + case IPV6_TRANSPARENT: + val = inet_sk(sk)->transparent; + break; + + case IPV6_RECVORIGDSTADDR: + val = np->rxopt.bits.rxorigdstaddr; + break; + case IPV6_UNICAST_HOPS: case IPV6_MULTICAST_HOPS: { diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 58841c4ae94..998d6d27e7c 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -91,7 +91,9 @@ #include <linux/netfilter.h> #include <linux/netfilter_ipv6.h> -static u32 ndisc_hash(const void *pkey, const struct net_device *dev); +static u32 ndisc_hash(const void *pkey, + const struct net_device *dev, + __u32 rnd); static int ndisc_constructor(struct neighbour *neigh); static void ndisc_solicit(struct neighbour *neigh, struct sk_buff *skb); static void ndisc_error_report(struct neighbour *neigh, struct sk_buff *skb); @@ -228,12 +230,12 @@ static struct nd_opt_hdr *ndisc_next_option(struct nd_opt_hdr *cur, do { cur = ((void *)cur) + (cur->nd_opt_len << 3); } while(cur < end && cur->nd_opt_type != type); - return (cur <= end && cur->nd_opt_type == type ? cur : NULL); + return cur <= end && cur->nd_opt_type == type ? cur : NULL; } static inline int ndisc_is_useropt(struct nd_opt_hdr *opt) { - return (opt->nd_opt_type == ND_OPT_RDNSS); + return opt->nd_opt_type == ND_OPT_RDNSS; } static struct nd_opt_hdr *ndisc_next_useropt(struct nd_opt_hdr *cur, @@ -244,7 +246,7 @@ static struct nd_opt_hdr *ndisc_next_useropt(struct nd_opt_hdr *cur, do { cur = ((void *)cur) + (cur->nd_opt_len << 3); } while(cur < end && !ndisc_is_useropt(cur)); - return (cur <= end && ndisc_is_useropt(cur) ? cur : NULL); + return cur <= end && ndisc_is_useropt(cur) ? cur : NULL; } static struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len, @@ -319,7 +321,7 @@ static inline u8 *ndisc_opt_addr_data(struct nd_opt_hdr *p, int prepad = ndisc_addr_option_pad(dev->type); if (lladdrlen != NDISC_OPT_SPACE(dev->addr_len + prepad)) return NULL; - return (lladdr + prepad); + return lladdr + prepad; } int ndisc_mc_map(struct in6_addr *addr, char *buf, struct net_device *dev, int dir) @@ -350,7 +352,9 @@ int ndisc_mc_map(struct in6_addr *addr, char *buf, struct net_device *dev, int d EXPORT_SYMBOL(ndisc_mc_map); -static u32 ndisc_hash(const void *pkey, const struct net_device *dev) +static u32 ndisc_hash(const void *pkey, + const struct net_device *dev, + __u32 hash_rnd) { const u32 *p32 = pkey; u32 addr_hash, i; @@ -359,7 +363,7 @@ static u32 ndisc_hash(const void *pkey, const struct net_device *dev) for (i = 0; i < (sizeof(struct in6_addr) / sizeof(u32)); i++) addr_hash ^= *p32++; - return jhash_2words(addr_hash, dev->ifindex, nd_tbl.hash_rnd); + return jhash_2words(addr_hash, dev->ifindex, hash_rnd); } static int ndisc_constructor(struct neighbour *neigh) @@ -1105,6 +1109,18 @@ errout: rtnl_set_sk_err(net, RTNLGRP_ND_USEROPT, err); } +static inline int accept_ra(struct inet6_dev *in6_dev) +{ + /* + * If forwarding is enabled, RA are not accepted unless the special + * hybrid mode (accept_ra=2) is enabled. + */ + if (in6_dev->cnf.forwarding && in6_dev->cnf.accept_ra < 2) + return 0; + + return in6_dev->cnf.accept_ra; +} + static void ndisc_router_discovery(struct sk_buff *skb) { struct ra_msg *ra_msg = (struct ra_msg *)skb_transport_header(skb); @@ -1158,8 +1174,7 @@ static void ndisc_router_discovery(struct sk_buff *skb) return; } - /* skip route and link configuration on routers */ - if (in6_dev->cnf.forwarding || !in6_dev->cnf.accept_ra) + if (!accept_ra(in6_dev)) goto skip_linkparms; #ifdef CONFIG_IPV6_NDISC_NODETYPE @@ -1309,8 +1324,7 @@ skip_linkparms: NEIGH_UPDATE_F_ISROUTER); } - /* skip route and link configuration on routers */ - if (in6_dev->cnf.forwarding || !in6_dev->cnf.accept_ra) + if (!accept_ra(in6_dev)) goto out; #ifdef CONFIG_IPV6_ROUTE_INFO diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig index 29d643bcafa..44d2eeac089 100644 --- a/net/ipv6/netfilter/Kconfig +++ b/net/ipv6/netfilter/Kconfig @@ -132,10 +132,10 @@ config IP6_NF_MATCH_RT # The targets config IP6_NF_TARGET_HL tristate '"HL" hoplimit target support' - depends on NETFILTER_ADVANCED + depends on NETFILTER_ADVANCED && IP6_NF_MANGLE select NETFILTER_XT_TARGET_HL ---help--- - This is a backwards-compat option for the user's convenience + This is a backwards-compatible option for the user's convenience (e.g. when running oldconfig). It selects CONFIG_NETFILTER_XT_TARGET_HL. diff --git a/net/ipv6/netfilter/Makefile b/net/ipv6/netfilter/Makefile index aafbba30c89..3f8e4a3d83c 100644 --- a/net/ipv6/netfilter/Makefile +++ b/net/ipv6/netfilter/Makefile @@ -11,10 +11,11 @@ obj-$(CONFIG_IP6_NF_RAW) += ip6table_raw.o obj-$(CONFIG_IP6_NF_SECURITY) += ip6table_security.o # objects for l3 independent conntrack -nf_conntrack_ipv6-objs := nf_conntrack_l3proto_ipv6.o nf_conntrack_proto_icmpv6.o nf_conntrack_reasm.o +nf_conntrack_ipv6-objs := nf_conntrack_l3proto_ipv6.o nf_conntrack_proto_icmpv6.o +nf_defrag_ipv6-objs := nf_defrag_ipv6_hooks.o nf_conntrack_reasm.o # l3 independent conntrack -obj-$(CONFIG_NF_CONNTRACK_IPV6) += nf_conntrack_ipv6.o +obj-$(CONFIG_NF_CONNTRACK_IPV6) += nf_conntrack_ipv6.o nf_defrag_ipv6.o # matches obj-$(CONFIG_IP6_NF_MATCH_AH) += ip6t_ah.o diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 5359ef4daac..51df035897e 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -82,13 +82,13 @@ EXPORT_SYMBOL_GPL(ip6t_alloc_initial_table); int ip6t_ext_hdr(u8 nexthdr) { - return ( (nexthdr == IPPROTO_HOPOPTS) || - (nexthdr == IPPROTO_ROUTING) || - (nexthdr == IPPROTO_FRAGMENT) || - (nexthdr == IPPROTO_ESP) || - (nexthdr == IPPROTO_AH) || - (nexthdr == IPPROTO_NONE) || - (nexthdr == IPPROTO_DSTOPTS) ); + return (nexthdr == IPPROTO_HOPOPTS) || + (nexthdr == IPPROTO_ROUTING) || + (nexthdr == IPPROTO_FRAGMENT) || + (nexthdr == IPPROTO_ESP) || + (nexthdr == IPPROTO_AH) || + (nexthdr == IPPROTO_NONE) || + (nexthdr == IPPROTO_DSTOPTS); } /* Returns whether matches rule or not. */ @@ -215,7 +215,7 @@ static inline bool unconditional(const struct ip6t_ip6 *ipv6) return memcmp(ipv6, &uncond, sizeof(uncond)) == 0; } -static inline const struct ip6t_entry_target * +static inline const struct xt_entry_target * ip6t_get_target_c(const struct ip6t_entry *e) { return ip6t_get_target((struct ip6t_entry *)e); @@ -260,9 +260,9 @@ get_chainname_rulenum(const struct ip6t_entry *s, const struct ip6t_entry *e, const char *hookname, const char **chainname, const char **comment, unsigned int *rulenum) { - const struct ip6t_standard_target *t = (void *)ip6t_get_target_c(s); + const struct xt_standard_target *t = (void *)ip6t_get_target_c(s); - if (strcmp(t->target.u.kernel.target->name, IP6T_ERROR_TARGET) == 0) { + if (strcmp(t->target.u.kernel.target->name, XT_ERROR_TARGET) == 0) { /* Head of user chain: ERROR target with chainname */ *chainname = t->target.data; (*rulenum) = 0; @@ -271,7 +271,7 @@ get_chainname_rulenum(const struct ip6t_entry *s, const struct ip6t_entry *e, if (s->target_offset == sizeof(struct ip6t_entry) && strcmp(t->target.u.kernel.target->name, - IP6T_STANDARD_TARGET) == 0 && + XT_STANDARD_TARGET) == 0 && t->verdict < 0 && unconditional(&s->ipv6)) { /* Tail of chains: STANDARD target (return/policy) */ @@ -369,7 +369,7 @@ ip6t_do_table(struct sk_buff *skb, e = get_entry(table_base, private->hook_entry[hook]); do { - const struct ip6t_entry_target *t; + const struct xt_entry_target *t; const struct xt_entry_match *ematch; IP_NF_ASSERT(e); @@ -403,10 +403,10 @@ ip6t_do_table(struct sk_buff *skb, if (!t->u.kernel.target->target) { int v; - v = ((struct ip6t_standard_target *)t)->verdict; + v = ((struct xt_standard_target *)t)->verdict; if (v < 0) { /* Pop from stack? */ - if (v != IP6T_RETURN) { + if (v != XT_RETURN) { verdict = (unsigned)(-v) - 1; break; } @@ -434,7 +434,7 @@ ip6t_do_table(struct sk_buff *skb, acpar.targinfo = t->data; verdict = t->u.kernel.target->target(skb, &acpar); - if (verdict == IP6T_CONTINUE) + if (verdict == XT_CONTINUE) e = ip6t_next_entry(e); else /* Verdict */ @@ -474,7 +474,7 @@ mark_source_chains(const struct xt_table_info *newinfo, e->counters.pcnt = pos; for (;;) { - const struct ip6t_standard_target *t + const struct xt_standard_target *t = (void *)ip6t_get_target_c(e); int visited = e->comefrom & (1 << hook); @@ -488,13 +488,13 @@ mark_source_chains(const struct xt_table_info *newinfo, /* Unconditional return/END. */ if ((e->target_offset == sizeof(struct ip6t_entry) && (strcmp(t->target.u.user.name, - IP6T_STANDARD_TARGET) == 0) && + XT_STANDARD_TARGET) == 0) && t->verdict < 0 && unconditional(&e->ipv6)) || visited) { unsigned int oldpos, size; if ((strcmp(t->target.u.user.name, - IP6T_STANDARD_TARGET) == 0) && + XT_STANDARD_TARGET) == 0) && t->verdict < -NF_MAX_VERDICT - 1) { duprintf("mark_source_chains: bad " "negative verdict (%i)\n", @@ -537,7 +537,7 @@ mark_source_chains(const struct xt_table_info *newinfo, int newpos = t->verdict; if (strcmp(t->target.u.user.name, - IP6T_STANDARD_TARGET) == 0 && + XT_STANDARD_TARGET) == 0 && newpos >= 0) { if (newpos > newinfo->size - sizeof(struct ip6t_entry)) { @@ -565,7 +565,7 @@ mark_source_chains(const struct xt_table_info *newinfo, return 1; } -static void cleanup_match(struct ip6t_entry_match *m, struct net *net) +static void cleanup_match(struct xt_entry_match *m, struct net *net) { struct xt_mtdtor_param par; @@ -581,14 +581,14 @@ static void cleanup_match(struct ip6t_entry_match *m, struct net *net) static int check_entry(const struct ip6t_entry *e, const char *name) { - const struct ip6t_entry_target *t; + const struct xt_entry_target *t; if (!ip6_checkentry(&e->ipv6)) { duprintf("ip_tables: ip check failed %p %s.\n", e, name); return -EINVAL; } - if (e->target_offset + sizeof(struct ip6t_entry_target) > + if (e->target_offset + sizeof(struct xt_entry_target) > e->next_offset) return -EINVAL; @@ -599,7 +599,7 @@ check_entry(const struct ip6t_entry *e, const char *name) return 0; } -static int check_match(struct ip6t_entry_match *m, struct xt_mtchk_param *par) +static int check_match(struct xt_entry_match *m, struct xt_mtchk_param *par) { const struct ip6t_ip6 *ipv6 = par->entryinfo; int ret; @@ -618,7 +618,7 @@ static int check_match(struct ip6t_entry_match *m, struct xt_mtchk_param *par) } static int -find_check_match(struct ip6t_entry_match *m, struct xt_mtchk_param *par) +find_check_match(struct xt_entry_match *m, struct xt_mtchk_param *par) { struct xt_match *match; int ret; @@ -643,7 +643,7 @@ err: static int check_target(struct ip6t_entry *e, struct net *net, const char *name) { - struct ip6t_entry_target *t = ip6t_get_target(e); + struct xt_entry_target *t = ip6t_get_target(e); struct xt_tgchk_param par = { .net = net, .table = name, @@ -670,7 +670,7 @@ static int find_check_entry(struct ip6t_entry *e, struct net *net, const char *name, unsigned int size) { - struct ip6t_entry_target *t; + struct xt_entry_target *t; struct xt_target *target; int ret; unsigned int j; @@ -721,7 +721,7 @@ find_check_entry(struct ip6t_entry *e, struct net *net, const char *name, static bool check_underflow(const struct ip6t_entry *e) { - const struct ip6t_entry_target *t; + const struct xt_entry_target *t; unsigned int verdict; if (!unconditional(&e->ipv6)) @@ -729,7 +729,7 @@ static bool check_underflow(const struct ip6t_entry *e) t = ip6t_get_target_c(e); if (strcmp(t->u.user.name, XT_STANDARD_TARGET) != 0) return false; - verdict = ((struct ip6t_standard_target *)t)->verdict; + verdict = ((struct xt_standard_target *)t)->verdict; verdict = -verdict - 1; return verdict == NF_DROP || verdict == NF_ACCEPT; } @@ -752,7 +752,7 @@ check_entry_size_and_hooks(struct ip6t_entry *e, } if (e->next_offset - < sizeof(struct ip6t_entry) + sizeof(struct ip6t_entry_target)) { + < sizeof(struct ip6t_entry) + sizeof(struct xt_entry_target)) { duprintf("checking: element %p size %u\n", e, e->next_offset); return -EINVAL; @@ -784,7 +784,7 @@ check_entry_size_and_hooks(struct ip6t_entry *e, static void cleanup_entry(struct ip6t_entry *e, struct net *net) { struct xt_tgdtor_param par; - struct ip6t_entry_target *t; + struct xt_entry_target *t; struct xt_entry_match *ematch; /* Cleanup all matches */ @@ -922,6 +922,7 @@ get_counters(const struct xt_table_info *t, if (cpu == curcpu) continue; i = 0; + local_bh_disable(); xt_info_wrlock(cpu); xt_entry_foreach(iter, t->entries[cpu], t->size) { ADD_COUNTER(counters[i], iter->counters.bcnt, @@ -929,6 +930,7 @@ get_counters(const struct xt_table_info *t, ++i; } xt_info_wrunlock(cpu); + local_bh_enable(); } put_cpu(); } @@ -983,8 +985,8 @@ copy_entries_to_user(unsigned int total_size, /* ... then go back and fix counters and names */ for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){ unsigned int i; - const struct ip6t_entry_match *m; - const struct ip6t_entry_target *t; + const struct xt_entry_match *m; + const struct xt_entry_target *t; e = (struct ip6t_entry *)(loc_cpu_entry + off); if (copy_to_user(userptr + off @@ -1001,7 +1003,7 @@ copy_entries_to_user(unsigned int total_size, m = (void *)e + i; if (copy_to_user(userptr + off + i - + offsetof(struct ip6t_entry_match, + + offsetof(struct xt_entry_match, u.user.name), m->u.kernel.match->name, strlen(m->u.kernel.match->name)+1) @@ -1013,7 +1015,7 @@ copy_entries_to_user(unsigned int total_size, t = ip6t_get_target_c(e); if (copy_to_user(userptr + off + e->target_offset - + offsetof(struct ip6t_entry_target, + + offsetof(struct xt_entry_target, u.user.name), t->u.kernel.target->name, strlen(t->u.kernel.target->name)+1) != 0) { @@ -1051,7 +1053,7 @@ static int compat_calc_entry(const struct ip6t_entry *e, const void *base, struct xt_table_info *newinfo) { const struct xt_entry_match *ematch; - const struct ip6t_entry_target *t; + const struct xt_entry_target *t; unsigned int entry_offset; int off, i, ret; @@ -1103,7 +1105,7 @@ static int compat_table_info(const struct xt_table_info *info, static int get_info(struct net *net, void __user *user, const int *len, int compat) { - char name[IP6T_TABLE_MAXNAMELEN]; + char name[XT_TABLE_MAXNAMELEN]; struct xt_table *t; int ret; @@ -1116,7 +1118,7 @@ static int get_info(struct net *net, void __user *user, if (copy_from_user(name, user, sizeof(name)) != 0) return -EFAULT; - name[IP6T_TABLE_MAXNAMELEN-1] = '\0'; + name[XT_TABLE_MAXNAMELEN-1] = '\0'; #ifdef CONFIG_COMPAT if (compat) xt_compat_lock(AF_INET6); @@ -1413,14 +1415,14 @@ do_add_counters(struct net *net, const void __user *user, unsigned int len, #ifdef CONFIG_COMPAT struct compat_ip6t_replace { - char name[IP6T_TABLE_MAXNAMELEN]; + char name[XT_TABLE_MAXNAMELEN]; u32 valid_hooks; u32 num_entries; u32 size; u32 hook_entry[NF_INET_NUMHOOKS]; u32 underflow[NF_INET_NUMHOOKS]; u32 num_counters; - compat_uptr_t counters; /* struct ip6t_counters * */ + compat_uptr_t counters; /* struct xt_counters * */ struct compat_ip6t_entry entries[0]; }; @@ -1429,7 +1431,7 @@ compat_copy_entry_to_user(struct ip6t_entry *e, void __user **dstptr, unsigned int *size, struct xt_counters *counters, unsigned int i) { - struct ip6t_entry_target *t; + struct xt_entry_target *t; struct compat_ip6t_entry __user *ce; u_int16_t target_offset, next_offset; compat_uint_t origsize; @@ -1464,7 +1466,7 @@ compat_copy_entry_to_user(struct ip6t_entry *e, void __user **dstptr, } static int -compat_find_calc_match(struct ip6t_entry_match *m, +compat_find_calc_match(struct xt_entry_match *m, const char *name, const struct ip6t_ip6 *ipv6, unsigned int hookmask, @@ -1486,7 +1488,7 @@ compat_find_calc_match(struct ip6t_entry_match *m, static void compat_release_entry(struct compat_ip6t_entry *e) { - struct ip6t_entry_target *t; + struct xt_entry_target *t; struct xt_entry_match *ematch; /* Cleanup all matches */ @@ -1507,7 +1509,7 @@ check_compat_entry_size_and_hooks(struct compat_ip6t_entry *e, const char *name) { struct xt_entry_match *ematch; - struct ip6t_entry_target *t; + struct xt_entry_target *t; struct xt_target *target; unsigned int entry_offset; unsigned int j; @@ -1589,7 +1591,7 @@ compat_copy_entry_from_user(struct compat_ip6t_entry *e, void **dstptr, unsigned int *size, const char *name, struct xt_table_info *newinfo, unsigned char *base) { - struct ip6t_entry_target *t; + struct xt_entry_target *t; struct xt_target *target; struct ip6t_entry *de; unsigned int origsize; @@ -1764,6 +1766,9 @@ translate_compat_table(struct net *net, if (ret != 0) break; ++i; + if (strcmp(ip6t_get_target(iter1)->u.user.name, + XT_ERROR_TARGET) == 0) + ++newinfo->stacksize; } if (ret) { /* @@ -1894,7 +1899,7 @@ compat_do_ip6t_set_ctl(struct sock *sk, int cmd, void __user *user, } struct compat_ip6t_get_entries { - char name[IP6T_TABLE_MAXNAMELEN]; + char name[XT_TABLE_MAXNAMELEN]; compat_uint_t size; struct compat_ip6t_entry entrytable[0]; }; @@ -2049,7 +2054,7 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) case IP6T_SO_GET_REVISION_MATCH: case IP6T_SO_GET_REVISION_TARGET: { - struct ip6t_get_revision rev; + struct xt_get_revision rev; int target; if (*len != sizeof(rev)) { @@ -2186,7 +2191,7 @@ static int icmp6_checkentry(const struct xt_mtchk_param *par) /* The built-in targets: standard (NULL) and error. */ static struct xt_target ip6t_builtin_tg[] __read_mostly = { { - .name = IP6T_STANDARD_TARGET, + .name = XT_STANDARD_TARGET, .targetsize = sizeof(int), .family = NFPROTO_IPV6, #ifdef CONFIG_COMPAT @@ -2196,9 +2201,9 @@ static struct xt_target ip6t_builtin_tg[] __read_mostly = { #endif }, { - .name = IP6T_ERROR_TARGET, + .name = XT_ERROR_TARGET, .target = ip6t_error, - .targetsize = IP6T_FUNCTION_MAXNAMELEN, + .targetsize = XT_FUNCTION_MAXNAMELEN, .family = NFPROTO_IPV6, }, }; diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c index 0a07ae7b933..09c88891a75 100644 --- a/net/ipv6/netfilter/ip6t_LOG.c +++ b/net/ipv6/netfilter/ip6t_LOG.c @@ -23,6 +23,7 @@ #include <linux/netfilter/x_tables.h> #include <linux/netfilter_ipv6/ip6_tables.h> #include <net/netfilter/nf_log.h> +#include <net/netfilter/xt_log.h> MODULE_AUTHOR("Jan Rekorajski <baggins@pld.org.pl>"); MODULE_DESCRIPTION("Xtables: IPv6 packet logging to syslog"); @@ -32,11 +33,9 @@ struct in_device; #include <net/route.h> #include <linux/netfilter_ipv6/ip6t_LOG.h> -/* Use lock to serialize, so printks don't overlap */ -static DEFINE_SPINLOCK(log_lock); - /* One level of recursion won't kill us */ -static void dump_packet(const struct nf_loginfo *info, +static void dump_packet(struct sbuff *m, + const struct nf_loginfo *info, const struct sk_buff *skb, unsigned int ip6hoff, int recurse) { @@ -55,15 +54,15 @@ static void dump_packet(const struct nf_loginfo *info, ih = skb_header_pointer(skb, ip6hoff, sizeof(_ip6h), &_ip6h); if (ih == NULL) { - printk("TRUNCATED"); + sb_add(m, "TRUNCATED"); return; } /* Max length: 88 "SRC=0000.0000.0000.0000.0000.0000.0000.0000 DST=0000.0000.0000.0000.0000.0000.0000.0000 " */ - printk("SRC=%pI6 DST=%pI6 ", &ih->saddr, &ih->daddr); + sb_add(m, "SRC=%pI6 DST=%pI6 ", &ih->saddr, &ih->daddr); /* Max length: 44 "LEN=65535 TC=255 HOPLIMIT=255 FLOWLBL=FFFFF " */ - printk("LEN=%Zu TC=%u HOPLIMIT=%u FLOWLBL=%u ", + sb_add(m, "LEN=%Zu TC=%u HOPLIMIT=%u FLOWLBL=%u ", ntohs(ih->payload_len) + sizeof(struct ipv6hdr), (ntohl(*(__be32 *)ih) & 0x0ff00000) >> 20, ih->hop_limit, @@ -78,35 +77,35 @@ static void dump_packet(const struct nf_loginfo *info, hp = skb_header_pointer(skb, ptr, sizeof(_hdr), &_hdr); if (hp == NULL) { - printk("TRUNCATED"); + sb_add(m, "TRUNCATED"); return; } /* Max length: 48 "OPT (...) " */ if (logflags & IP6T_LOG_IPOPT) - printk("OPT ( "); + sb_add(m, "OPT ( "); switch (currenthdr) { case IPPROTO_FRAGMENT: { struct frag_hdr _fhdr; const struct frag_hdr *fh; - printk("FRAG:"); + sb_add(m, "FRAG:"); fh = skb_header_pointer(skb, ptr, sizeof(_fhdr), &_fhdr); if (fh == NULL) { - printk("TRUNCATED "); + sb_add(m, "TRUNCATED "); return; } /* Max length: 6 "65535 " */ - printk("%u ", ntohs(fh->frag_off) & 0xFFF8); + sb_add(m, "%u ", ntohs(fh->frag_off) & 0xFFF8); /* Max length: 11 "INCOMPLETE " */ if (fh->frag_off & htons(0x0001)) - printk("INCOMPLETE "); + sb_add(m, "INCOMPLETE "); - printk("ID:%08x ", ntohl(fh->identification)); + sb_add(m, "ID:%08x ", ntohl(fh->identification)); if (ntohs(fh->frag_off) & 0xFFF8) fragment = 1; @@ -120,7 +119,7 @@ static void dump_packet(const struct nf_loginfo *info, case IPPROTO_HOPOPTS: if (fragment) { if (logflags & IP6T_LOG_IPOPT) - printk(")"); + sb_add(m, ")"); return; } hdrlen = ipv6_optlen(hp); @@ -132,10 +131,10 @@ static void dump_packet(const struct nf_loginfo *info, const struct ip_auth_hdr *ah; /* Max length: 3 "AH " */ - printk("AH "); + sb_add(m, "AH "); if (fragment) { - printk(")"); + sb_add(m, ")"); return; } @@ -146,13 +145,13 @@ static void dump_packet(const struct nf_loginfo *info, * Max length: 26 "INCOMPLETE [65535 * bytes] )" */ - printk("INCOMPLETE [%u bytes] )", + sb_add(m, "INCOMPLETE [%u bytes] )", skb->len - ptr); return; } /* Length: 15 "SPI=0xF1234567 */ - printk("SPI=0x%x ", ntohl(ah->spi)); + sb_add(m, "SPI=0x%x ", ntohl(ah->spi)); } @@ -164,10 +163,10 @@ static void dump_packet(const struct nf_loginfo *info, const struct ip_esp_hdr *eh; /* Max length: 4 "ESP " */ - printk("ESP "); + sb_add(m, "ESP "); if (fragment) { - printk(")"); + sb_add(m, ")"); return; } @@ -177,23 +176,23 @@ static void dump_packet(const struct nf_loginfo *info, eh = skb_header_pointer(skb, ptr, sizeof(_esph), &_esph); if (eh == NULL) { - printk("INCOMPLETE [%u bytes] )", + sb_add(m, "INCOMPLETE [%u bytes] )", skb->len - ptr); return; } /* Length: 16 "SPI=0xF1234567 )" */ - printk("SPI=0x%x )", ntohl(eh->spi) ); + sb_add(m, "SPI=0x%x )", ntohl(eh->spi) ); } return; default: /* Max length: 20 "Unknown Ext Hdr 255" */ - printk("Unknown Ext Hdr %u", currenthdr); + sb_add(m, "Unknown Ext Hdr %u", currenthdr); return; } if (logflags & IP6T_LOG_IPOPT) - printk(") "); + sb_add(m, ") "); currenthdr = hp->nexthdr; ptr += hdrlen; @@ -205,7 +204,7 @@ static void dump_packet(const struct nf_loginfo *info, const struct tcphdr *th; /* Max length: 10 "PROTO=TCP " */ - printk("PROTO=TCP "); + sb_add(m, "PROTO=TCP "); if (fragment) break; @@ -213,40 +212,40 @@ static void dump_packet(const struct nf_loginfo *info, /* Max length: 25 "INCOMPLETE [65535 bytes] " */ th = skb_header_pointer(skb, ptr, sizeof(_tcph), &_tcph); if (th == NULL) { - printk("INCOMPLETE [%u bytes] ", skb->len - ptr); + sb_add(m, "INCOMPLETE [%u bytes] ", skb->len - ptr); return; } /* Max length: 20 "SPT=65535 DPT=65535 " */ - printk("SPT=%u DPT=%u ", + sb_add(m, "SPT=%u DPT=%u ", ntohs(th->source), ntohs(th->dest)); /* Max length: 30 "SEQ=4294967295 ACK=4294967295 " */ if (logflags & IP6T_LOG_TCPSEQ) - printk("SEQ=%u ACK=%u ", + sb_add(m, "SEQ=%u ACK=%u ", ntohl(th->seq), ntohl(th->ack_seq)); /* Max length: 13 "WINDOW=65535 " */ - printk("WINDOW=%u ", ntohs(th->window)); + sb_add(m, "WINDOW=%u ", ntohs(th->window)); /* Max length: 9 "RES=0x3C " */ - printk("RES=0x%02x ", (u_int8_t)(ntohl(tcp_flag_word(th) & TCP_RESERVED_BITS) >> 22)); + sb_add(m, "RES=0x%02x ", (u_int8_t)(ntohl(tcp_flag_word(th) & TCP_RESERVED_BITS) >> 22)); /* Max length: 32 "CWR ECE URG ACK PSH RST SYN FIN " */ if (th->cwr) - printk("CWR "); + sb_add(m, "CWR "); if (th->ece) - printk("ECE "); + sb_add(m, "ECE "); if (th->urg) - printk("URG "); + sb_add(m, "URG "); if (th->ack) - printk("ACK "); + sb_add(m, "ACK "); if (th->psh) - printk("PSH "); + sb_add(m, "PSH "); if (th->rst) - printk("RST "); + sb_add(m, "RST "); if (th->syn) - printk("SYN "); + sb_add(m, "SYN "); if (th->fin) - printk("FIN "); + sb_add(m, "FIN "); /* Max length: 11 "URGP=65535 " */ - printk("URGP=%u ", ntohs(th->urg_ptr)); + sb_add(m, "URGP=%u ", ntohs(th->urg_ptr)); if ((logflags & IP6T_LOG_TCPOPT) && th->doff * 4 > sizeof(struct tcphdr)) { @@ -260,15 +259,15 @@ static void dump_packet(const struct nf_loginfo *info, ptr + sizeof(struct tcphdr), optsize, _opt); if (op == NULL) { - printk("OPT (TRUNCATED)"); + sb_add(m, "OPT (TRUNCATED)"); return; } /* Max length: 127 "OPT (" 15*4*2chars ") " */ - printk("OPT ("); + sb_add(m, "OPT ("); for (i =0; i < optsize; i++) - printk("%02X", op[i]); - printk(") "); + sb_add(m, "%02X", op[i]); + sb_add(m, ") "); } break; } @@ -279,9 +278,9 @@ static void dump_packet(const struct nf_loginfo *info, if (currenthdr == IPPROTO_UDP) /* Max length: 10 "PROTO=UDP " */ - printk("PROTO=UDP " ); + sb_add(m, "PROTO=UDP " ); else /* Max length: 14 "PROTO=UDPLITE " */ - printk("PROTO=UDPLITE "); + sb_add(m, "PROTO=UDPLITE "); if (fragment) break; @@ -289,12 +288,12 @@ static void dump_packet(const struct nf_loginfo *info, /* Max length: 25 "INCOMPLETE [65535 bytes] " */ uh = skb_header_pointer(skb, ptr, sizeof(_udph), &_udph); if (uh == NULL) { - printk("INCOMPLETE [%u bytes] ", skb->len - ptr); + sb_add(m, "INCOMPLETE [%u bytes] ", skb->len - ptr); return; } /* Max length: 20 "SPT=65535 DPT=65535 " */ - printk("SPT=%u DPT=%u LEN=%u ", + sb_add(m, "SPT=%u DPT=%u LEN=%u ", ntohs(uh->source), ntohs(uh->dest), ntohs(uh->len)); break; @@ -304,7 +303,7 @@ static void dump_packet(const struct nf_loginfo *info, const struct icmp6hdr *ic; /* Max length: 13 "PROTO=ICMPv6 " */ - printk("PROTO=ICMPv6 "); + sb_add(m, "PROTO=ICMPv6 "); if (fragment) break; @@ -312,18 +311,18 @@ static void dump_packet(const struct nf_loginfo *info, /* Max length: 25 "INCOMPLETE [65535 bytes] " */ ic = skb_header_pointer(skb, ptr, sizeof(_icmp6h), &_icmp6h); if (ic == NULL) { - printk("INCOMPLETE [%u bytes] ", skb->len - ptr); + sb_add(m, "INCOMPLETE [%u bytes] ", skb->len - ptr); return; } /* Max length: 18 "TYPE=255 CODE=255 " */ - printk("TYPE=%u CODE=%u ", ic->icmp6_type, ic->icmp6_code); + sb_add(m, "TYPE=%u CODE=%u ", ic->icmp6_type, ic->icmp6_code); switch (ic->icmp6_type) { case ICMPV6_ECHO_REQUEST: case ICMPV6_ECHO_REPLY: /* Max length: 19 "ID=65535 SEQ=65535 " */ - printk("ID=%u SEQ=%u ", + sb_add(m, "ID=%u SEQ=%u ", ntohs(ic->icmp6_identifier), ntohs(ic->icmp6_sequence)); break; @@ -334,35 +333,35 @@ static void dump_packet(const struct nf_loginfo *info, case ICMPV6_PARAMPROB: /* Max length: 17 "POINTER=ffffffff " */ - printk("POINTER=%08x ", ntohl(ic->icmp6_pointer)); + sb_add(m, "POINTER=%08x ", ntohl(ic->icmp6_pointer)); /* Fall through */ case ICMPV6_DEST_UNREACH: case ICMPV6_PKT_TOOBIG: case ICMPV6_TIME_EXCEED: /* Max length: 3+maxlen */ if (recurse) { - printk("["); - dump_packet(info, skb, ptr + sizeof(_icmp6h), - 0); - printk("] "); + sb_add(m, "["); + dump_packet(m, info, skb, + ptr + sizeof(_icmp6h), 0); + sb_add(m, "] "); } /* Max length: 10 "MTU=65535 " */ if (ic->icmp6_type == ICMPV6_PKT_TOOBIG) - printk("MTU=%u ", ntohl(ic->icmp6_mtu)); + sb_add(m, "MTU=%u ", ntohl(ic->icmp6_mtu)); } break; } /* Max length: 10 "PROTO=255 " */ default: - printk("PROTO=%u ", currenthdr); + sb_add(m, "PROTO=%u ", currenthdr); } /* Max length: 15 "UID=4294967295 " */ if ((logflags & IP6T_LOG_UID) && recurse && skb->sk) { read_lock_bh(&skb->sk->sk_callback_lock); if (skb->sk->sk_socket && skb->sk->sk_socket->file) - printk("UID=%u GID=%u ", + sb_add(m, "UID=%u GID=%u ", skb->sk->sk_socket->file->f_cred->fsuid, skb->sk->sk_socket->file->f_cred->fsgid); read_unlock_bh(&skb->sk->sk_callback_lock); @@ -370,10 +369,11 @@ static void dump_packet(const struct nf_loginfo *info, /* Max length: 16 "MARK=0xFFFFFFFF " */ if (!recurse && skb->mark) - printk("MARK=0x%x ", skb->mark); + sb_add(m, "MARK=0x%x ", skb->mark); } -static void dump_mac_header(const struct nf_loginfo *info, +static void dump_mac_header(struct sbuff *m, + const struct nf_loginfo *info, const struct sk_buff *skb) { struct net_device *dev = skb->dev; @@ -387,7 +387,7 @@ static void dump_mac_header(const struct nf_loginfo *info, switch (dev->type) { case ARPHRD_ETHER: - printk("MACSRC=%pM MACDST=%pM MACPROTO=%04x ", + sb_add(m, "MACSRC=%pM MACDST=%pM MACPROTO=%04x ", eth_hdr(skb)->h_source, eth_hdr(skb)->h_dest, ntohs(eth_hdr(skb)->h_proto)); return; @@ -396,7 +396,7 @@ static void dump_mac_header(const struct nf_loginfo *info, } fallback: - printk("MAC="); + sb_add(m, "MAC="); if (dev->hard_header_len && skb->mac_header != skb->network_header) { const unsigned char *p = skb_mac_header(skb); @@ -408,19 +408,19 @@ fallback: p = NULL; if (p != NULL) { - printk("%02x", *p++); + sb_add(m, "%02x", *p++); for (i = 1; i < len; i++) - printk(":%02x", p[i]); + sb_add(m, ":%02x", p[i]); } - printk(" "); + sb_add(m, " "); if (dev->type == ARPHRD_SIT) { const struct iphdr *iph = (struct iphdr *)skb_mac_header(skb); - printk("TUNNEL=%pI4->%pI4 ", &iph->saddr, &iph->daddr); + sb_add(m, "TUNNEL=%pI4->%pI4 ", &iph->saddr, &iph->daddr); } } else - printk(" "); + sb_add(m, " "); } static struct nf_loginfo default_loginfo = { @@ -442,22 +442,23 @@ ip6t_log_packet(u_int8_t pf, const struct nf_loginfo *loginfo, const char *prefix) { + struct sbuff *m = sb_open(); + if (!loginfo) loginfo = &default_loginfo; - spin_lock_bh(&log_lock); - printk("<%d>%sIN=%s OUT=%s ", loginfo->u.log.level, - prefix, - in ? in->name : "", - out ? out->name : ""); + sb_add(m, "<%d>%sIN=%s OUT=%s ", loginfo->u.log.level, + prefix, + in ? in->name : "", + out ? out->name : ""); /* MAC logging for input path only. */ if (in && !out) - dump_mac_header(loginfo, skb); + dump_mac_header(m, loginfo, skb); + + dump_packet(m, loginfo, skb, skb_network_offset(skb), 1); - dump_packet(loginfo, skb, skb_network_offset(skb), 1); - printk("\n"); - spin_unlock_bh(&log_lock); + sb_close(m); } static unsigned int diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c index ff43461704b..c8af58b2256 100644 --- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -16,7 +16,6 @@ #include <linux/module.h> #include <linux/skbuff.h> #include <linux/icmp.h> -#include <linux/sysctl.h> #include <net/ipv6.h> #include <net/inet_frag.h> @@ -29,6 +28,7 @@ #include <net/netfilter/nf_conntrack_core.h> #include <net/netfilter/nf_conntrack_zones.h> #include <net/netfilter/ipv6/nf_conntrack_ipv6.h> +#include <net/netfilter/ipv6/nf_defrag_ipv6.h> #include <net/netfilter/nf_log.h> static bool ipv6_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, @@ -189,53 +189,6 @@ out: return nf_conntrack_confirm(skb); } -static enum ip6_defrag_users nf_ct6_defrag_user(unsigned int hooknum, - struct sk_buff *skb) -{ - u16 zone = NF_CT_DEFAULT_ZONE; - - if (skb->nfct) - zone = nf_ct_zone((struct nf_conn *)skb->nfct); - -#ifdef CONFIG_BRIDGE_NETFILTER - if (skb->nf_bridge && - skb->nf_bridge->mask & BRNF_NF_BRIDGE_PREROUTING) - return IP6_DEFRAG_CONNTRACK_BRIDGE_IN + zone; -#endif - if (hooknum == NF_INET_PRE_ROUTING) - return IP6_DEFRAG_CONNTRACK_IN + zone; - else - return IP6_DEFRAG_CONNTRACK_OUT + zone; - -} - -static unsigned int ipv6_defrag(unsigned int hooknum, - struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) -{ - struct sk_buff *reasm; - - /* Previously seen (loopback)? */ - if (skb->nfct && !nf_ct_is_template((struct nf_conn *)skb->nfct)) - return NF_ACCEPT; - - reasm = nf_ct_frag6_gather(skb, nf_ct6_defrag_user(hooknum, skb)); - /* queued */ - if (reasm == NULL) - return NF_STOLEN; - - /* error occured or not fragmented */ - if (reasm == skb) - return NF_ACCEPT; - - nf_ct_frag6_output(hooknum, reasm, (struct net_device *)in, - (struct net_device *)out, okfn); - - return NF_STOLEN; -} - static unsigned int __ipv6_conntrack_in(struct net *net, unsigned int hooknum, struct sk_buff *skb, @@ -288,13 +241,6 @@ static unsigned int ipv6_conntrack_local(unsigned int hooknum, static struct nf_hook_ops ipv6_conntrack_ops[] __read_mostly = { { - .hook = ipv6_defrag, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_PRE_ROUTING, - .priority = NF_IP6_PRI_CONNTRACK_DEFRAG, - }, - { .hook = ipv6_conntrack_in, .owner = THIS_MODULE, .pf = NFPROTO_IPV6, @@ -309,13 +255,6 @@ static struct nf_hook_ops ipv6_conntrack_ops[] __read_mostly = { .priority = NF_IP6_PRI_CONNTRACK, }, { - .hook = ipv6_defrag, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_LOCAL_OUT, - .priority = NF_IP6_PRI_CONNTRACK_DEFRAG, - }, - { .hook = ipv6_confirm, .owner = THIS_MODULE, .pf = NFPROTO_IPV6, @@ -387,10 +326,6 @@ struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6 __read_mostly = { .nlattr_to_tuple = ipv6_nlattr_to_tuple, .nla_policy = ipv6_nla_policy, #endif -#ifdef CONFIG_SYSCTL - .ctl_table_path = nf_net_netfilter_sysctl_path, - .ctl_table = nf_ct_ipv6_sysctl_table, -#endif .me = THIS_MODULE, }; @@ -403,16 +338,12 @@ static int __init nf_conntrack_l3proto_ipv6_init(void) int ret = 0; need_conntrack(); + nf_defrag_ipv6_enable(); - ret = nf_ct_frag6_init(); - if (ret < 0) { - pr_err("nf_conntrack_ipv6: can't initialize frag6.\n"); - return ret; - } ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_tcp6); if (ret < 0) { pr_err("nf_conntrack_ipv6: can't register tcp.\n"); - goto cleanup_frag6; + return ret; } ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_udp6); @@ -450,8 +381,6 @@ static int __init nf_conntrack_l3proto_ipv6_init(void) nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp6); cleanup_tcp: nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp6); - cleanup_frag6: - nf_ct_frag6_cleanup(); return ret; } @@ -463,7 +392,6 @@ static void __exit nf_conntrack_l3proto_ipv6_fini(void) nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmpv6); nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp6); nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp6); - nf_ct_frag6_cleanup(); } module_init(nf_conntrack_l3proto_ipv6_init); diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index 13ef5bc05cf..489d71b844a 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -73,7 +73,7 @@ static struct inet_frags nf_frags; static struct netns_frags nf_init_frags; #ifdef CONFIG_SYSCTL -struct ctl_table nf_ct_ipv6_sysctl_table[] = { +struct ctl_table nf_ct_frag6_sysctl_table[] = { { .procname = "nf_conntrack_frag6_timeout", .data = &nf_init_frags.timeout, @@ -97,6 +97,8 @@ struct ctl_table nf_ct_ipv6_sysctl_table[] = { }, { } }; + +static struct ctl_table_header *nf_ct_frag6_sysctl_header; #endif static unsigned int nf_hashfn(struct inet_frag_queue *q) @@ -113,14 +115,6 @@ static void nf_skb_free(struct sk_buff *skb) kfree_skb(NFCT_FRAG6_CB(skb)->orig); } -/* Memory Tracking Functions. */ -static void frag_kfree_skb(struct sk_buff *skb) -{ - atomic_sub(skb->truesize, &nf_init_frags.mem); - nf_skb_free(skb); - kfree_skb(skb); -} - /* Destruction primitives. */ static __inline__ void fq_put(struct nf_ct_frag6_queue *fq) @@ -282,66 +276,22 @@ static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb, } found: - /* We found where to put this one. Check for overlap with - * preceding fragment, and, if needed, align things so that - * any overlaps are eliminated. - */ - if (prev) { - int i = (NFCT_FRAG6_CB(prev)->offset + prev->len) - offset; - - if (i > 0) { - offset += i; - if (end <= offset) { - pr_debug("overlap\n"); - goto err; - } - if (!pskb_pull(skb, i)) { - pr_debug("Can't pull\n"); - goto err; - } - if (skb->ip_summed != CHECKSUM_UNNECESSARY) - skb->ip_summed = CHECKSUM_NONE; - } - } - - /* Look for overlap with succeeding segments. - * If we can merge fragments, do it. + /* RFC5722, Section 4: + * When reassembling an IPv6 datagram, if + * one or more its constituent fragments is determined to be an + * overlapping fragment, the entire datagram (and any constituent + * fragments, including those not yet received) MUST be silently + * discarded. */ - while (next && NFCT_FRAG6_CB(next)->offset < end) { - /* overlap is 'i' bytes */ - int i = end - NFCT_FRAG6_CB(next)->offset; - - if (i < next->len) { - /* Eat head of the next overlapped fragment - * and leave the loop. The next ones cannot overlap. - */ - pr_debug("Eat head of the overlapped parts.: %d", i); - if (!pskb_pull(next, i)) - goto err; - - /* next fragment */ - NFCT_FRAG6_CB(next)->offset += i; - fq->q.meat -= i; - if (next->ip_summed != CHECKSUM_UNNECESSARY) - next->ip_summed = CHECKSUM_NONE; - break; - } else { - struct sk_buff *free_it = next; - - /* Old fragmnet is completely overridden with - * new one drop it. - */ - next = next->next; - if (prev) - prev->next = next; - else - fq->q.fragments = next; + /* Check for overlap with preceding fragment. */ + if (prev && + (NFCT_FRAG6_CB(prev)->offset + prev->len) - offset > 0) + goto discard_fq; - fq->q.meat -= free_it->len; - frag_kfree_skb(free_it); - } - } + /* Look for overlap with succeeding segment. */ + if (next && NFCT_FRAG6_CB(next)->offset < end) + goto discard_fq; NFCT_FRAG6_CB(skb)->offset = offset; @@ -371,6 +321,8 @@ found: write_unlock(&nf_frags.lock); return 0; +discard_fq: + fq_kill(fq); err: return -1; } @@ -413,7 +365,7 @@ nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev) /* If the first fragment is fragmented itself, we split * it to two chunks: the first with data and paged part * and the second, holding only fragments. */ - if (skb_has_frags(head)) { + if (skb_has_frag_list(head)) { struct sk_buff *clone; int i, plen = 0; @@ -673,11 +625,21 @@ int nf_ct_frag6_init(void) inet_frags_init_net(&nf_init_frags); inet_frags_init(&nf_frags); + nf_ct_frag6_sysctl_header = register_sysctl_paths(nf_net_netfilter_sysctl_path, + nf_ct_frag6_sysctl_table); + if (!nf_ct_frag6_sysctl_header) { + inet_frags_fini(&nf_frags); + return -ENOMEM; + } + return 0; } void nf_ct_frag6_cleanup(void) { + unregister_sysctl_table(nf_ct_frag6_sysctl_header); + nf_ct_frag6_sysctl_header = NULL; + inet_frags_fini(&nf_frags); nf_init_frags.low_thresh = 0; diff --git a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c new file mode 100644 index 00000000000..99abfb53bab --- /dev/null +++ b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c @@ -0,0 +1,131 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/types.h> +#include <linux/ipv6.h> +#include <linux/in6.h> +#include <linux/netfilter.h> +#include <linux/module.h> +#include <linux/skbuff.h> +#include <linux/icmp.h> +#include <linux/sysctl.h> +#include <net/ipv6.h> +#include <net/inet_frag.h> + +#include <linux/netfilter_ipv6.h> +#include <linux/netfilter_bridge.h> +#include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_helper.h> +#include <net/netfilter/nf_conntrack_l4proto.h> +#include <net/netfilter/nf_conntrack_l3proto.h> +#include <net/netfilter/nf_conntrack_core.h> +#include <net/netfilter/nf_conntrack_zones.h> +#include <net/netfilter/ipv6/nf_conntrack_ipv6.h> +#include <net/netfilter/ipv6/nf_defrag_ipv6.h> + +static enum ip6_defrag_users nf_ct6_defrag_user(unsigned int hooknum, + struct sk_buff *skb) +{ + u16 zone = NF_CT_DEFAULT_ZONE; + + if (skb->nfct) + zone = nf_ct_zone((struct nf_conn *)skb->nfct); + +#ifdef CONFIG_BRIDGE_NETFILTER + if (skb->nf_bridge && + skb->nf_bridge->mask & BRNF_NF_BRIDGE_PREROUTING) + return IP6_DEFRAG_CONNTRACK_BRIDGE_IN + zone; +#endif + if (hooknum == NF_INET_PRE_ROUTING) + return IP6_DEFRAG_CONNTRACK_IN + zone; + else + return IP6_DEFRAG_CONNTRACK_OUT + zone; + +} + +static unsigned int ipv6_defrag(unsigned int hooknum, + struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct sk_buff *reasm; + + /* Previously seen (loopback)? */ + if (skb->nfct && !nf_ct_is_template((struct nf_conn *)skb->nfct)) + return NF_ACCEPT; + + reasm = nf_ct_frag6_gather(skb, nf_ct6_defrag_user(hooknum, skb)); + /* queued */ + if (reasm == NULL) + return NF_STOLEN; + + /* error occured or not fragmented */ + if (reasm == skb) + return NF_ACCEPT; + + nf_ct_frag6_output(hooknum, reasm, (struct net_device *)in, + (struct net_device *)out, okfn); + + return NF_STOLEN; +} + +static struct nf_hook_ops ipv6_defrag_ops[] = { + { + .hook = ipv6_defrag, + .owner = THIS_MODULE, + .pf = NFPROTO_IPV6, + .hooknum = NF_INET_PRE_ROUTING, + .priority = NF_IP6_PRI_CONNTRACK_DEFRAG, + }, + { + .hook = ipv6_defrag, + .owner = THIS_MODULE, + .pf = NFPROTO_IPV6, + .hooknum = NF_INET_LOCAL_OUT, + .priority = NF_IP6_PRI_CONNTRACK_DEFRAG, + }, +}; + +static int __init nf_defrag_init(void) +{ + int ret = 0; + + ret = nf_ct_frag6_init(); + if (ret < 0) { + pr_err("nf_defrag_ipv6: can't initialize frag6.\n"); + return ret; + } + ret = nf_register_hooks(ipv6_defrag_ops, ARRAY_SIZE(ipv6_defrag_ops)); + if (ret < 0) { + pr_err("nf_defrag_ipv6: can't register hooks\n"); + goto cleanup_frag6; + } + return ret; + +cleanup_frag6: + nf_ct_frag6_cleanup(); + return ret; + +} + +static void __exit nf_defrag_fini(void) +{ + nf_unregister_hooks(ipv6_defrag_ops, ARRAY_SIZE(ipv6_defrag_ops)); + nf_ct_frag6_cleanup(); +} + +void nf_defrag_ipv6_enable(void) +{ +} +EXPORT_SYMBOL_GPL(nf_defrag_ipv6_enable); + +module_init(nf_defrag_init); +module_exit(nf_defrag_fini); + +MODULE_LICENSE("GPL"); diff --git a/net/ipv6/protocol.c b/net/ipv6/protocol.c index 1fa3468f0f3..9bb936ae245 100644 --- a/net/ipv6/protocol.c +++ b/net/ipv6/protocol.c @@ -25,28 +25,14 @@ #include <linux/spinlock.h> #include <net/protocol.h> -const struct inet6_protocol *inet6_protos[MAX_INET_PROTOS]; -static DEFINE_SPINLOCK(inet6_proto_lock); - +const struct inet6_protocol *inet6_protos[MAX_INET_PROTOS] __read_mostly; int inet6_add_protocol(const struct inet6_protocol *prot, unsigned char protocol) { - int ret, hash = protocol & (MAX_INET_PROTOS - 1); - - spin_lock_bh(&inet6_proto_lock); - - if (inet6_protos[hash]) { - ret = -1; - } else { - inet6_protos[hash] = prot; - ret = 0; - } - - spin_unlock_bh(&inet6_proto_lock); + int hash = protocol & (MAX_INET_PROTOS - 1); - return ret; + return !cmpxchg(&inet6_protos[hash], NULL, prot) ? 0 : -1; } - EXPORT_SYMBOL(inet6_add_protocol); /* @@ -57,20 +43,10 @@ int inet6_del_protocol(const struct inet6_protocol *prot, unsigned char protocol { int ret, hash = protocol & (MAX_INET_PROTOS - 1); - spin_lock_bh(&inet6_proto_lock); - - if (inet6_protos[hash] != prot) { - ret = -1; - } else { - inet6_protos[hash] = NULL; - ret = 0; - } - - spin_unlock_bh(&inet6_proto_lock); + ret = (cmpxchg(&inet6_protos[hash], prot, NULL) == prot) ? 0 : -1; synchronize_net(); return ret; } - EXPORT_SYMBOL(inet6_del_protocol); diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index e677937a07f..45e6efb7f17 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -764,7 +764,7 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk, return -EINVAL; if (sin6->sin6_family && sin6->sin6_family != AF_INET6) - return(-EAFNOSUPPORT); + return -EAFNOSUPPORT; /* port is the proto value [0..255] carried in nexthdr */ proto = ntohs(sin6->sin6_port); @@ -772,10 +772,10 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk, if (!proto) proto = inet->inet_num; else if (proto != inet->inet_num) - return(-EINVAL); + return -EINVAL; if (proto > 255) - return(-EINVAL); + return -EINVAL; daddr = &sin6->sin6_addr; if (np->sndflow) { @@ -985,7 +985,7 @@ static int do_rawv6_setsockopt(struct sock *sk, int level, int optname, /* You may get strange result with a positive odd offset; RFC2292bis agrees with me. */ if (val > 0 && (val&1)) - return(-EINVAL); + return -EINVAL; if (val < 0) { rp->checksum = 0; } else { @@ -997,7 +997,7 @@ static int do_rawv6_setsockopt(struct sock *sk, int level, int optname, break; default: - return(-ENOPROTOOPT); + return -ENOPROTOOPT; } } @@ -1190,7 +1190,7 @@ static int rawv6_init_sk(struct sock *sk) default: break; } - return(0); + return 0; } struct proto rawv6_prot = { diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index 545c4141b75..c7ba3149633 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -149,13 +149,6 @@ int ip6_frag_match(struct inet_frag_queue *q, void *a) } EXPORT_SYMBOL(ip6_frag_match); -/* Memory Tracking Functions. */ -static void frag_kfree_skb(struct netns_frags *nf, struct sk_buff *skb) -{ - atomic_sub(skb->truesize, &nf->mem); - kfree_skb(skb); -} - void ip6_frag_init(struct inet_frag_queue *q, void *a) { struct frag_queue *fq = container_of(q, struct frag_queue, q); @@ -346,58 +339,22 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, } found: - /* We found where to put this one. Check for overlap with - * preceding fragment, and, if needed, align things so that - * any overlaps are eliminated. + /* RFC5722, Section 4: + * When reassembling an IPv6 datagram, if + * one or more its constituent fragments is determined to be an + * overlapping fragment, the entire datagram (and any constituent + * fragments, including those not yet received) MUST be silently + * discarded. */ - if (prev) { - int i = (FRAG6_CB(prev)->offset + prev->len) - offset; - if (i > 0) { - offset += i; - if (end <= offset) - goto err; - if (!pskb_pull(skb, i)) - goto err; - if (skb->ip_summed != CHECKSUM_UNNECESSARY) - skb->ip_summed = CHECKSUM_NONE; - } - } + /* Check for overlap with preceding fragment. */ + if (prev && + (FRAG6_CB(prev)->offset + prev->len) - offset > 0) + goto discard_fq; - /* Look for overlap with succeeding segments. - * If we can merge fragments, do it. - */ - while (next && FRAG6_CB(next)->offset < end) { - int i = end - FRAG6_CB(next)->offset; /* overlap is 'i' bytes */ - - if (i < next->len) { - /* Eat head of the next overlapped fragment - * and leave the loop. The next ones cannot overlap. - */ - if (!pskb_pull(next, i)) - goto err; - FRAG6_CB(next)->offset += i; /* next fragment */ - fq->q.meat -= i; - if (next->ip_summed != CHECKSUM_UNNECESSARY) - next->ip_summed = CHECKSUM_NONE; - break; - } else { - struct sk_buff *free_it = next; - - /* Old fragment is completely overridden with - * new one drop it. - */ - next = next->next; - - if (prev) - prev->next = next; - else - fq->q.fragments = next; - - fq->q.meat -= free_it->len; - frag_kfree_skb(fq->q.net, free_it); - } - } + /* Look for overlap with succeeding segment. */ + if (next && FRAG6_CB(next)->offset < end) + goto discard_fq; FRAG6_CB(skb)->offset = offset; @@ -436,6 +393,8 @@ found: write_unlock(&ip6_frags.lock); return -1; +discard_fq: + fq_kill(fq); err: IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_REASMFAILS); @@ -499,7 +458,7 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *prev, /* If the first fragment is fragmented itself, we split * it to two chunks: the first with data and paged part * and the second, holding only fragments. */ - if (skb_has_frags(head)) { + if (skb_has_frag_list(head)) { struct sk_buff *clone; int i, plen = 0; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 8f2d0400cf8..25661f968f3 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -109,7 +109,6 @@ static struct dst_ops ip6_dst_ops_template = { .link_failure = ip6_link_failure, .update_pmtu = ip6_rt_update_pmtu, .local_out = __ip6_local_out, - .entries = ATOMIC_INIT(0), }; static void ip6_rt_blackhole_update_pmtu(struct dst_entry *dst, u32 mtu) @@ -122,7 +121,6 @@ static struct dst_ops ip6_dst_blackhole_ops = { .destroy = ip6_dst_destroy, .check = ip6_dst_check, .update_pmtu = ip6_rt_blackhole_update_pmtu, - .entries = ATOMIC_INIT(0), }; static struct rt6_info ip6_null_entry_template = { @@ -217,14 +215,14 @@ static void ip6_dst_ifdown(struct dst_entry *dst, struct net_device *dev, static __inline__ int rt6_check_expired(const struct rt6_info *rt) { - return (rt->rt6i_flags & RTF_EXPIRES && - time_after(jiffies, rt->rt6i_expires)); + return (rt->rt6i_flags & RTF_EXPIRES) && + time_after(jiffies, rt->rt6i_expires); } static inline int rt6_need_strict(struct in6_addr *daddr) { - return (ipv6_addr_type(daddr) & - (IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL | IPV6_ADDR_LOOPBACK)); + return ipv6_addr_type(daddr) & + (IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL | IPV6_ADDR_LOOPBACK); } /* @@ -440,7 +438,7 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict) __func__, match); net = dev_net(rt0->rt6i_dev); - return (match ? match : net->ipv6.ip6_null_entry); + return match ? match : net->ipv6.ip6_null_entry; } #ifdef CONFIG_IPV6_ROUTE_INFO @@ -670,7 +668,7 @@ static struct rt6_info *rt6_alloc_cow(struct rt6_info *ort, struct in6_addr *dad if (net_ratelimit()) printk(KERN_WARNING - "Neighbour table overflow.\n"); + "ipv6: Neighbour table overflow.\n"); dst_free(&rt->dst); return NULL; } @@ -859,7 +857,7 @@ int ip6_dst_blackhole(struct sock *sk, struct dst_entry **dstp, struct flowi *fl dst_release(*dstp); *dstp = new; - return (new ? 0 : -ENOMEM); + return new ? 0 : -ENOMEM; } EXPORT_SYMBOL_GPL(ip6_dst_blackhole); @@ -1058,19 +1056,22 @@ static int ip6_dst_gc(struct dst_ops *ops) int rt_elasticity = net->ipv6.sysctl.ip6_rt_gc_elasticity; int rt_gc_timeout = net->ipv6.sysctl.ip6_rt_gc_timeout; unsigned long rt_last_gc = net->ipv6.ip6_rt_last_gc; + int entries; + entries = dst_entries_get_fast(ops); if (time_after(rt_last_gc + rt_min_interval, now) && - atomic_read(&ops->entries) <= rt_max_size) + entries <= rt_max_size) goto out; net->ipv6.ip6_rt_gc_expire++; fib6_run_gc(net->ipv6.ip6_rt_gc_expire, net); net->ipv6.ip6_rt_last_gc = now; - if (atomic_read(&ops->entries) < ops->gc_thresh) + entries = dst_entries_get_slow(ops); + if (entries < ops->gc_thresh) net->ipv6.ip6_rt_gc_expire = rt_gc_timeout>>1; out: net->ipv6.ip6_rt_gc_expire -= net->ipv6.ip6_rt_gc_expire>>rt_elasticity; - return (atomic_read(&ops->entries) > rt_max_size); + return entries > rt_max_size; } /* Clean host part of a prefix. Not necessary in radix tree, @@ -1169,6 +1170,8 @@ int ip6_route_add(struct fib6_config *cfg) if (addr_type & IPV6_ADDR_MULTICAST) rt->dst.input = ip6_mc_input; + else if (cfg->fc_flags & RTF_LOCAL) + rt->dst.input = ip6_input; else rt->dst.input = ip6_forward; @@ -1190,7 +1193,8 @@ int ip6_route_add(struct fib6_config *cfg) they would result in kernel looping; promote them to reject routes */ if ((cfg->fc_flags & RTF_REJECT) || - (dev && (dev->flags&IFF_LOOPBACK) && !(addr_type&IPV6_ADDR_LOOPBACK))) { + (dev && (dev->flags&IFF_LOOPBACK) && !(addr_type&IPV6_ADDR_LOOPBACK) + && !(cfg->fc_flags&RTF_LOCAL))) { /* hold loopback dev/idev if we haven't done so. */ if (dev != net->loopback_dev) { if (dev) { @@ -1556,14 +1560,13 @@ out: * i.e. Path MTU discovery */ -void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr, - struct net_device *dev, u32 pmtu) +static void rt6_do_pmtu_disc(struct in6_addr *daddr, struct in6_addr *saddr, + struct net *net, u32 pmtu, int ifindex) { struct rt6_info *rt, *nrt; - struct net *net = dev_net(dev); int allfrag = 0; - rt = rt6_lookup(net, daddr, saddr, dev->ifindex, 0); + rt = rt6_lookup(net, daddr, saddr, ifindex, 0); if (rt == NULL) return; @@ -1631,6 +1634,27 @@ out: dst_release(&rt->dst); } +void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr, + struct net_device *dev, u32 pmtu) +{ + struct net *net = dev_net(dev); + + /* + * RFC 1981 states that a node "MUST reduce the size of the packets it + * is sending along the path" that caused the Packet Too Big message. + * Since it's not possible in the general case to determine which + * interface was used to send the original packet, we update the MTU + * on the interface that will be used to send future packets. We also + * update the MTU on the interface that received the Packet Too Big in + * case the original packet was forced out that interface with + * SO_BINDTODEVICE or similar. This is the next best thing to the + * correct behaviour, which would be to update the MTU on all + * interfaces. + */ + rt6_do_pmtu_disc(daddr, saddr, net, pmtu, 0); + rt6_do_pmtu_disc(daddr, saddr, net, pmtu, dev->ifindex); +} + /* * Misc support functions */ @@ -2082,6 +2106,9 @@ static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh, if (rtm->rtm_type == RTN_UNREACHABLE) cfg->fc_flags |= RTF_REJECT; + if (rtm->rtm_type == RTN_LOCAL) + cfg->fc_flags |= RTF_LOCAL; + cfg->fc_nlinfo.pid = NETLINK_CB(skb).pid; cfg->fc_nlinfo.nlh = nlh; cfg->fc_nlinfo.nl_net = sock_net(skb->sk); @@ -2202,6 +2229,8 @@ static int rt6_fill_node(struct net *net, NLA_PUT_U32(skb, RTA_TABLE, table); if (rt->rt6i_flags&RTF_REJECT) rtm->rtm_type = RTN_UNREACHABLE; + else if (rt->rt6i_flags&RTF_LOCAL) + rtm->rtm_type = RTN_LOCAL; else if (rt->rt6i_dev && (rt->rt6i_dev->flags&IFF_LOOPBACK)) rtm->rtm_type = RTN_LOCAL; else @@ -2496,7 +2525,7 @@ static int rt6_stats_seq_show(struct seq_file *seq, void *v) net->ipv6.rt6_stats->fib_rt_alloc, net->ipv6.rt6_stats->fib_rt_entries, net->ipv6.rt6_stats->fib_rt_cache, - atomic_read(&net->ipv6.ip6_dst_ops.entries), + dst_entries_get_slow(&net->ipv6.ip6_dst_ops), net->ipv6.rt6_stats->fib_discarded_routes); return 0; @@ -2580,7 +2609,7 @@ ctl_table ipv6_route_table_template[] = { .data = &init_net.ipv6.sysctl.ip6_rt_gc_elasticity, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = proc_dointvec_jiffies, + .proc_handler = proc_dointvec, }, { .procname = "mtu_expires", @@ -2594,7 +2623,7 @@ ctl_table ipv6_route_table_template[] = { .data = &init_net.ipv6.sysctl.ip6_rt_min_advmss, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = proc_dointvec_jiffies, + .proc_handler = proc_dointvec, }, { .procname = "gc_min_interval_ms", @@ -2638,11 +2667,14 @@ static int __net_init ip6_route_net_init(struct net *net) memcpy(&net->ipv6.ip6_dst_ops, &ip6_dst_ops_template, sizeof(net->ipv6.ip6_dst_ops)); + if (dst_entries_init(&net->ipv6.ip6_dst_ops) < 0) + goto out_ip6_dst_ops; + net->ipv6.ip6_null_entry = kmemdup(&ip6_null_entry_template, sizeof(*net->ipv6.ip6_null_entry), GFP_KERNEL); if (!net->ipv6.ip6_null_entry) - goto out_ip6_dst_ops; + goto out_ip6_dst_entries; net->ipv6.ip6_null_entry->dst.path = (struct dst_entry *)net->ipv6.ip6_null_entry; net->ipv6.ip6_null_entry->dst.ops = &net->ipv6.ip6_dst_ops; @@ -2692,6 +2724,8 @@ out_ip6_prohibit_entry: out_ip6_null_entry: kfree(net->ipv6.ip6_null_entry); #endif +out_ip6_dst_entries: + dst_entries_destroy(&net->ipv6.ip6_dst_ops); out_ip6_dst_ops: goto out; } @@ -2730,10 +2764,14 @@ int __init ip6_route_init(void) if (!ip6_dst_ops_template.kmem_cachep) goto out; - ret = register_pernet_subsys(&ip6_route_net_ops); + ret = dst_entries_init(&ip6_dst_blackhole_ops); if (ret) goto out_kmem_cache; + ret = register_pernet_subsys(&ip6_route_net_ops); + if (ret) + goto out_dst_entries; + ip6_dst_blackhole_ops.kmem_cachep = ip6_dst_ops_template.kmem_cachep; /* Registering of the loopback is done before this portion of code, @@ -2780,6 +2818,8 @@ out_fib6_init: fib6_gc_cleanup(); out_register_subsys: unregister_pernet_subsys(&ip6_route_net_ops); +out_dst_entries: + dst_entries_destroy(&ip6_dst_blackhole_ops); out_kmem_cache: kmem_cache_destroy(ip6_dst_ops_template.kmem_cachep); goto out; diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 4699cd3c311..367a6cc584c 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -63,36 +63,63 @@ #define HASH_SIZE 16 #define HASH(addr) (((__force u32)addr^((__force u32)addr>>4))&0xF) -static void ipip6_tunnel_init(struct net_device *dev); +static int ipip6_tunnel_init(struct net_device *dev); static void ipip6_tunnel_setup(struct net_device *dev); +static void ipip6_dev_free(struct net_device *dev); static int sit_net_id __read_mostly; struct sit_net { - struct ip_tunnel *tunnels_r_l[HASH_SIZE]; - struct ip_tunnel *tunnels_r[HASH_SIZE]; - struct ip_tunnel *tunnels_l[HASH_SIZE]; - struct ip_tunnel *tunnels_wc[1]; - struct ip_tunnel **tunnels[4]; + struct ip_tunnel __rcu *tunnels_r_l[HASH_SIZE]; + struct ip_tunnel __rcu *tunnels_r[HASH_SIZE]; + struct ip_tunnel __rcu *tunnels_l[HASH_SIZE]; + struct ip_tunnel __rcu *tunnels_wc[1]; + struct ip_tunnel __rcu **tunnels[4]; struct net_device *fb_tunnel_dev; }; /* - * Locking : hash tables are protected by RCU and a spinlock + * Locking : hash tables are protected by RCU and RTNL */ -static DEFINE_SPINLOCK(ipip6_lock); #define for_each_ip_tunnel_rcu(start) \ for (t = rcu_dereference(start); t; t = rcu_dereference(t->next)) +/* often modified stats are per cpu, other are shared (netdev->stats) */ +struct pcpu_tstats { + unsigned long rx_packets; + unsigned long rx_bytes; + unsigned long tx_packets; + unsigned long tx_bytes; +}; + +static struct net_device_stats *ipip6_get_stats(struct net_device *dev) +{ + struct pcpu_tstats sum = { 0 }; + int i; + + for_each_possible_cpu(i) { + const struct pcpu_tstats *tstats = per_cpu_ptr(dev->tstats, i); + + sum.rx_packets += tstats->rx_packets; + sum.rx_bytes += tstats->rx_bytes; + sum.tx_packets += tstats->tx_packets; + sum.tx_bytes += tstats->tx_bytes; + } + dev->stats.rx_packets = sum.rx_packets; + dev->stats.rx_bytes = sum.rx_bytes; + dev->stats.tx_packets = sum.tx_packets; + dev->stats.tx_bytes = sum.tx_bytes; + return &dev->stats; +} /* * Must be invoked with rcu_read_lock */ static struct ip_tunnel * ipip6_tunnel_lookup(struct net *net, struct net_device *dev, __be32 remote, __be32 local) { - unsigned h0 = HASH(remote); - unsigned h1 = HASH(local); + unsigned int h0 = HASH(remote); + unsigned int h1 = HASH(local); struct ip_tunnel *t; struct sit_net *sitn = net_generic(net, sit_net_id); @@ -121,12 +148,12 @@ static struct ip_tunnel * ipip6_tunnel_lookup(struct net *net, return NULL; } -static struct ip_tunnel **__ipip6_bucket(struct sit_net *sitn, +static struct ip_tunnel __rcu **__ipip6_bucket(struct sit_net *sitn, struct ip_tunnel_parm *parms) { __be32 remote = parms->iph.daddr; __be32 local = parms->iph.saddr; - unsigned h = 0; + unsigned int h = 0; int prio = 0; if (remote) { @@ -140,7 +167,7 @@ static struct ip_tunnel **__ipip6_bucket(struct sit_net *sitn, return &sitn->tunnels[prio][h]; } -static inline struct ip_tunnel **ipip6_bucket(struct sit_net *sitn, +static inline struct ip_tunnel __rcu **ipip6_bucket(struct sit_net *sitn, struct ip_tunnel *t) { return __ipip6_bucket(sitn, &t->parms); @@ -148,13 +175,14 @@ static inline struct ip_tunnel **ipip6_bucket(struct sit_net *sitn, static void ipip6_tunnel_unlink(struct sit_net *sitn, struct ip_tunnel *t) { - struct ip_tunnel **tp; - - for (tp = ipip6_bucket(sitn, t); *tp; tp = &(*tp)->next) { - if (t == *tp) { - spin_lock_bh(&ipip6_lock); - *tp = t->next; - spin_unlock_bh(&ipip6_lock); + struct ip_tunnel __rcu **tp; + struct ip_tunnel *iter; + + for (tp = ipip6_bucket(sitn, t); + (iter = rtnl_dereference(*tp)) != NULL; + tp = &iter->next) { + if (t == iter) { + rcu_assign_pointer(*tp, t->next); break; } } @@ -162,12 +190,10 @@ static void ipip6_tunnel_unlink(struct sit_net *sitn, struct ip_tunnel *t) static void ipip6_tunnel_link(struct sit_net *sitn, struct ip_tunnel *t) { - struct ip_tunnel **tp = ipip6_bucket(sitn, t); + struct ip_tunnel __rcu **tp = ipip6_bucket(sitn, t); - spin_lock_bh(&ipip6_lock); - t->next = *tp; + rcu_assign_pointer(t->next, rtnl_dereference(*tp)); rcu_assign_pointer(*tp, t); - spin_unlock_bh(&ipip6_lock); } static void ipip6_tunnel_clone_6rd(struct net_device *dev, struct sit_net *sitn) @@ -187,17 +213,20 @@ static void ipip6_tunnel_clone_6rd(struct net_device *dev, struct sit_net *sitn) #endif } -static struct ip_tunnel * ipip6_tunnel_locate(struct net *net, +static struct ip_tunnel *ipip6_tunnel_locate(struct net *net, struct ip_tunnel_parm *parms, int create) { __be32 remote = parms->iph.daddr; __be32 local = parms->iph.saddr; - struct ip_tunnel *t, **tp, *nt; + struct ip_tunnel *t, *nt; + struct ip_tunnel __rcu **tp; struct net_device *dev; char name[IFNAMSIZ]; struct sit_net *sitn = net_generic(net, sit_net_id); - for (tp = __ipip6_bucket(sitn, parms); (t = *tp) != NULL; tp = &t->next) { + for (tp = __ipip6_bucket(sitn, parms); + (t = rtnl_dereference(*tp)) != NULL; + tp = &t->next) { if (local == t->parms.iph.saddr && remote == t->parms.iph.daddr && parms->link == t->parms.link) { @@ -213,7 +242,7 @@ static struct ip_tunnel * ipip6_tunnel_locate(struct net *net, if (parms->name[0]) strlcpy(name, parms->name, IFNAMSIZ); else - sprintf(name, "sit%%d"); + strcpy(name, "sit%d"); dev = alloc_netdev(sizeof(*t), name, ipip6_tunnel_setup); if (dev == NULL) @@ -229,7 +258,8 @@ static struct ip_tunnel * ipip6_tunnel_locate(struct net *net, nt = netdev_priv(dev); nt->parms = *parms; - ipip6_tunnel_init(dev); + if (ipip6_tunnel_init(dev) < 0) + goto failed_free; ipip6_tunnel_clone_6rd(dev, sitn); if (parms->i_flags & SIT_ISATAP) @@ -244,7 +274,7 @@ static struct ip_tunnel * ipip6_tunnel_locate(struct net *net, return nt; failed_free: - free_netdev(dev); + ipip6_dev_free(dev); failed: return NULL; } @@ -340,7 +370,7 @@ ipip6_tunnel_add_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a, int chg) ASSERT_RTNL(); - for (p = t->prl; p; p = p->next) { + for (p = rtnl_dereference(t->prl); p; p = rtnl_dereference(p->next)) { if (p->addr == a->addr) { if (chg) { p->flags = a->flags; @@ -451,15 +481,12 @@ static void ipip6_tunnel_uninit(struct net_device *dev) struct sit_net *sitn = net_generic(net, sit_net_id); if (dev == sitn->fb_tunnel_dev) { - spin_lock_bh(&ipip6_lock); - sitn->tunnels_wc[0] = NULL; - spin_unlock_bh(&ipip6_lock); - dev_put(dev); + rcu_assign_pointer(sitn->tunnels_wc[0], NULL); } else { ipip6_tunnel_unlink(sitn, netdev_priv(dev)); ipip6_tunnel_del_prl(netdev_priv(dev), NULL); - dev_put(dev); } + dev_put(dev); } @@ -548,6 +575,8 @@ static int ipip6_rcv(struct sk_buff *skb) tunnel = ipip6_tunnel_lookup(dev_net(skb->dev), skb->dev, iph->saddr, iph->daddr); if (tunnel != NULL) { + struct pcpu_tstats *tstats; + secpath_reset(skb); skb->mac_header = skb->network_header; skb_reset_network_header(skb); @@ -563,10 +592,16 @@ static int ipip6_rcv(struct sk_buff *skb) return 0; } - skb_tunnel_rx(skb, tunnel->dev); + tstats = this_cpu_ptr(tunnel->dev->tstats); + tstats->rx_packets++; + tstats->rx_bytes += skb->len; + + __skb_tunnel_rx(skb, tunnel->dev); ipip6_ecn_decapsulate(iph, skb); + netif_rx(skb); + rcu_read_unlock(); return 0; } @@ -590,7 +625,7 @@ __be32 try_6rd(struct in6_addr *v6dst, struct ip_tunnel *tunnel) #ifdef CONFIG_IPV6_SIT_6RD if (ipv6_prefix_equal(v6dst, &tunnel->ip6rd.prefix, tunnel->ip6rd.prefixlen)) { - unsigned pbw0, pbi0; + unsigned int pbw0, pbi0; int pbi1; u32 d; @@ -625,14 +660,13 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); - struct net_device_stats *stats = &dev->stats; - struct netdev_queue *txq = netdev_get_tx_queue(dev, 0); + struct pcpu_tstats *tstats; struct iphdr *tiph = &tunnel->parms.iph; struct ipv6hdr *iph6 = ipv6_hdr(skb); u8 tos = tunnel->parms.iph.tos; __be16 df = tiph->frag_off; struct rtable *rt; /* Route to the other host */ - struct net_device *tdev; /* Device to other host */ + struct net_device *tdev; /* Device to other host */ struct iphdr *iph; /* Our new IP header */ unsigned int max_headroom; /* The extra header space needed */ __be32 dst = tiph->daddr; @@ -703,20 +737,20 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb, .oif = tunnel->parms.link, .proto = IPPROTO_IPV6 }; if (ip_route_output_key(dev_net(dev), &rt, &fl)) { - stats->tx_carrier_errors++; + dev->stats.tx_carrier_errors++; goto tx_error_icmp; } } if (rt->rt_type != RTN_UNICAST) { ip_rt_put(rt); - stats->tx_carrier_errors++; + dev->stats.tx_carrier_errors++; goto tx_error_icmp; } tdev = rt->dst.dev; if (tdev == dev) { ip_rt_put(rt); - stats->collisions++; + dev->stats.collisions++; goto tx_error; } @@ -724,7 +758,7 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb, mtu = dst_mtu(&rt->dst) - sizeof(struct iphdr); if (mtu < 68) { - stats->collisions++; + dev->stats.collisions++; ip_rt_put(rt); goto tx_error; } @@ -763,7 +797,7 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb, struct sk_buff *new_skb = skb_realloc_headroom(skb, max_headroom); if (!new_skb) { ip_rt_put(rt); - txq->tx_dropped++; + dev->stats.tx_dropped++; dev_kfree_skb(skb); return NETDEV_TX_OK; } @@ -799,14 +833,14 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb, iph->ttl = iph6->hop_limit; nf_reset(skb); - - IPTUNNEL_XMIT(); + tstats = this_cpu_ptr(dev->tstats); + __IPTUNNEL_XMIT(tstats, &dev->stats); return NETDEV_TX_OK; tx_error_icmp: dst_link_failure(skb); tx_error: - stats->tx_errors++; + dev->stats.tx_errors++; dev_kfree_skb(skb); return NETDEV_TX_OK; } @@ -1083,12 +1117,19 @@ static const struct net_device_ops ipip6_netdev_ops = { .ndo_start_xmit = ipip6_tunnel_xmit, .ndo_do_ioctl = ipip6_tunnel_ioctl, .ndo_change_mtu = ipip6_tunnel_change_mtu, + .ndo_get_stats = ipip6_get_stats, }; +static void ipip6_dev_free(struct net_device *dev) +{ + free_percpu(dev->tstats); + free_netdev(dev); +} + static void ipip6_tunnel_setup(struct net_device *dev) { dev->netdev_ops = &ipip6_netdev_ops; - dev->destructor = free_netdev; + dev->destructor = ipip6_dev_free; dev->type = ARPHRD_SIT; dev->hard_header_len = LL_MAX_HEADER + sizeof(struct iphdr); @@ -1098,9 +1139,10 @@ static void ipip6_tunnel_setup(struct net_device *dev) dev->iflink = 0; dev->addr_len = 4; dev->features |= NETIF_F_NETNS_LOCAL; + dev->features |= NETIF_F_LLTX; } -static void ipip6_tunnel_init(struct net_device *dev) +static int ipip6_tunnel_init(struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); @@ -1111,9 +1153,14 @@ static void ipip6_tunnel_init(struct net_device *dev) memcpy(dev->broadcast, &tunnel->parms.iph.daddr, 4); ipip6_tunnel_bind_dev(dev); + dev->tstats = alloc_percpu(struct pcpu_tstats); + if (!dev->tstats) + return -ENOMEM; + + return 0; } -static void __net_init ipip6_fb_tunnel_init(struct net_device *dev) +static int __net_init ipip6_fb_tunnel_init(struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); struct iphdr *iph = &tunnel->parms.iph; @@ -1128,11 +1175,15 @@ static void __net_init ipip6_fb_tunnel_init(struct net_device *dev) iph->ihl = 5; iph->ttl = 64; + dev->tstats = alloc_percpu(struct pcpu_tstats); + if (!dev->tstats) + return -ENOMEM; dev_hold(dev); sitn->tunnels_wc[0] = tunnel; + return 0; } -static struct xfrm_tunnel sit_handler = { +static struct xfrm_tunnel sit_handler __read_mostly = { .handler = ipip6_rcv, .err_handler = ipip6_err, .priority = 1, @@ -1173,7 +1224,10 @@ static int __net_init sit_init_net(struct net *net) } dev_net_set(sitn->fb_tunnel_dev, net); - ipip6_fb_tunnel_init(sitn->fb_tunnel_dev); + err = ipip6_fb_tunnel_init(sitn->fb_tunnel_dev); + if (err) + goto err_dev_free; + ipip6_tunnel_clone_6rd(sitn->fb_tunnel_dev, sitn); if ((err = register_netdev(sitn->fb_tunnel_dev))) @@ -1183,7 +1237,8 @@ static int __net_init sit_init_net(struct net *net) err_reg_dev: dev_put(sitn->fb_tunnel_dev); - free_netdev(sitn->fb_tunnel_dev); +err_dev_free: + ipip6_dev_free(sitn->fb_tunnel_dev); err_alloc_dev: return err; } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index fe6d40418c0..7e41e2cbb85 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -139,7 +139,7 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr, return -EINVAL; if (usin->sin6_family != AF_INET6) - return(-EAFNOSUPPORT); + return -EAFNOSUPPORT; memset(&fl, 0, sizeof(fl)); @@ -1409,7 +1409,7 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, newsk = tcp_create_openreq_child(sk, req, skb); if (newsk == NULL) - goto out; + goto out_nonewsk; /* * No need to charge this sock to the relevant IPv6 refcnt debug socks @@ -1497,18 +1497,22 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, } #endif + if (__inet_inherit_port(sk, newsk) < 0) { + sock_put(newsk); + goto out; + } __inet6_hash(newsk, NULL); - __inet_inherit_port(sk, newsk); return newsk; out_overflow: NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENOVERFLOWS); -out: - NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS); +out_nonewsk: if (opt && opt != np->opt) sock_kfree_s(sk, opt, opt->tot_len); dst_release(dst); +out: + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS); return NULL; } diff --git a/net/ipv6/tunnel6.c b/net/ipv6/tunnel6.c index fc3c86a4745..d9864725d0c 100644 --- a/net/ipv6/tunnel6.c +++ b/net/ipv6/tunnel6.c @@ -30,8 +30,8 @@ #include <net/protocol.h> #include <net/xfrm.h> -static struct xfrm6_tunnel *tunnel6_handlers; -static struct xfrm6_tunnel *tunnel46_handlers; +static struct xfrm6_tunnel *tunnel6_handlers __read_mostly; +static struct xfrm6_tunnel *tunnel46_handlers __read_mostly; static DEFINE_MUTEX(tunnel6_mutex); int xfrm6_tunnel_register(struct xfrm6_tunnel *handler, unsigned short family) @@ -51,7 +51,7 @@ int xfrm6_tunnel_register(struct xfrm6_tunnel *handler, unsigned short family) } handler->next = *pprev; - *pprev = handler; + rcu_assign_pointer(*pprev, handler); ret = 0; @@ -88,6 +88,11 @@ int xfrm6_tunnel_deregister(struct xfrm6_tunnel *handler, unsigned short family) EXPORT_SYMBOL(xfrm6_tunnel_deregister); +#define for_each_tunnel_rcu(head, handler) \ + for (handler = rcu_dereference(head); \ + handler != NULL; \ + handler = rcu_dereference(handler->next)) \ + static int tunnel6_rcv(struct sk_buff *skb) { struct xfrm6_tunnel *handler; @@ -95,7 +100,7 @@ static int tunnel6_rcv(struct sk_buff *skb) if (!pskb_may_pull(skb, sizeof(struct ipv6hdr))) goto drop; - for (handler = tunnel6_handlers; handler; handler = handler->next) + for_each_tunnel_rcu(tunnel6_handlers, handler) if (!handler->handler(skb)) return 0; @@ -113,7 +118,7 @@ static int tunnel46_rcv(struct sk_buff *skb) if (!pskb_may_pull(skb, sizeof(struct iphdr))) goto drop; - for (handler = tunnel46_handlers; handler; handler = handler->next) + for_each_tunnel_rcu(tunnel46_handlers, handler) if (!handler->handler(skb)) return 0; @@ -129,7 +134,7 @@ static void tunnel6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, { struct xfrm6_tunnel *handler; - for (handler = tunnel6_handlers; handler; handler = handler->next) + for_each_tunnel_rcu(tunnel6_handlers, handler) if (!handler->err_handler(skb, opt, type, code, offset, info)) break; } diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 1dd1affdead..c84dad43211 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -111,10 +111,19 @@ int udp_v6_get_port(struct sock *sk, unsigned short snum) return udp_lib_get_port(sk, snum, ipv6_rcv_saddr_equal, hash2_nulladdr); } +static void udp_v6_rehash(struct sock *sk) +{ + u16 new_hash = udp6_portaddr_hash(sock_net(sk), + &inet6_sk(sk)->rcv_saddr, + inet_sk(sk)->inet_num); + + udp_lib_rehash(sk, new_hash); +} + static inline int compute_score(struct sock *sk, struct net *net, unsigned short hnum, - struct in6_addr *saddr, __be16 sport, - struct in6_addr *daddr, __be16 dport, + const struct in6_addr *saddr, __be16 sport, + const struct in6_addr *daddr, __be16 dport, int dif) { int score = -1; @@ -230,8 +239,8 @@ exact_match: } static struct sock *__udp6_lib_lookup(struct net *net, - struct in6_addr *saddr, __be16 sport, - struct in6_addr *daddr, __be16 dport, + const struct in6_addr *saddr, __be16 sport, + const struct in6_addr *daddr, __be16 dport, int dif, struct udp_table *udptable) { struct sock *sk, *result; @@ -311,6 +320,14 @@ static struct sock *__udp6_lib_lookup_skb(struct sk_buff *skb, udptable); } +struct sock *udp6_lib_lookup(struct net *net, const struct in6_addr *saddr, __be16 sport, + const struct in6_addr *daddr, __be16 dport, int dif) +{ + return __udp6_lib_lookup(net, saddr, sport, daddr, dport, dif, &udp_table); +} +EXPORT_SYMBOL_GPL(udp6_lib_lookup); + + /* * This should be easy, if there is something there we * return it, otherwise we block. @@ -1447,6 +1464,7 @@ struct proto udpv6_prot = { .backlog_rcv = udpv6_queue_rcv_skb, .hash = udp_lib_hash, .unhash = udp_lib_unhash, + .rehash = udp_v6_rehash, .get_port = udp_v6_get_port, .memory_allocated = &udp_memory_allocated, .sysctl_mem = sysctl_udp_mem, diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 6baeabbbca8..7e74023ea6e 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -199,7 +199,7 @@ static inline int xfrm6_garbage_collect(struct dst_ops *ops) struct net *net = container_of(ops, struct net, xfrm.xfrm6_dst_ops); xfrm6_policy_afinfo.garbage_collect(net); - return (atomic_read(&ops->entries) > ops->gc_thresh * 2); + return dst_entries_get_fast(ops) > ops->gc_thresh * 2; } static void xfrm6_update_pmtu(struct dst_entry *dst, u32 mtu) @@ -255,7 +255,6 @@ static struct dst_ops xfrm6_dst_ops = { .ifdown = xfrm6_dst_ifdown, .local_out = __ip6_local_out, .gc_thresh = 1024, - .entries = ATOMIC_INIT(0), }; static struct xfrm_policy_afinfo xfrm6_policy_afinfo = { @@ -312,11 +311,13 @@ int __init xfrm6_init(void) */ gc_thresh = FIB6_TABLE_HASHSZ * 8; xfrm6_dst_ops.gc_thresh = (gc_thresh < 1024) ? 1024 : gc_thresh; + dst_entries_init(&xfrm6_dst_ops); ret = xfrm6_policy_init(); - if (ret) + if (ret) { + dst_entries_destroy(&xfrm6_dst_ops); goto out; - + } ret = xfrm6_state_init(); if (ret) goto out_policy; @@ -341,4 +342,5 @@ void xfrm6_fini(void) //xfrm6_input_fini(); xfrm6_policy_fini(); xfrm6_state_fini(); + dst_entries_destroy(&xfrm6_dst_ops); } diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c index f417b77fa0e..a67575d472a 100644 --- a/net/ipv6/xfrm6_state.c +++ b/net/ipv6/xfrm6_state.c @@ -20,23 +20,27 @@ #include <net/addrconf.h> static void -__xfrm6_init_tempsel(struct xfrm_state *x, struct flowi *fl, - struct xfrm_tmpl *tmpl, - xfrm_address_t *daddr, xfrm_address_t *saddr) +__xfrm6_init_tempsel(struct xfrm_selector *sel, struct flowi *fl) { /* Initialize temporary selector matching only * to current session. */ - ipv6_addr_copy((struct in6_addr *)&x->sel.daddr, &fl->fl6_dst); - ipv6_addr_copy((struct in6_addr *)&x->sel.saddr, &fl->fl6_src); - x->sel.dport = xfrm_flowi_dport(fl); - x->sel.dport_mask = htons(0xffff); - x->sel.sport = xfrm_flowi_sport(fl); - x->sel.sport_mask = htons(0xffff); - x->sel.family = AF_INET6; - x->sel.prefixlen_d = 128; - x->sel.prefixlen_s = 128; - x->sel.proto = fl->proto; - x->sel.ifindex = fl->oif; + ipv6_addr_copy((struct in6_addr *)&sel->daddr, &fl->fl6_dst); + ipv6_addr_copy((struct in6_addr *)&sel->saddr, &fl->fl6_src); + sel->dport = xfrm_flowi_dport(fl); + sel->dport_mask = htons(0xffff); + sel->sport = xfrm_flowi_sport(fl); + sel->sport_mask = htons(0xffff); + sel->family = AF_INET6; + sel->prefixlen_d = 128; + sel->prefixlen_s = 128; + sel->proto = fl->proto; + sel->ifindex = fl->oif; +} + +static void +xfrm6_init_temprop(struct xfrm_state *x, struct xfrm_tmpl *tmpl, + xfrm_address_t *daddr, xfrm_address_t *saddr) +{ x->id = tmpl->id; if (ipv6_addr_any((struct in6_addr*)&x->id.daddr)) memcpy(&x->id.daddr, daddr, sizeof(x->sel.daddr)); @@ -168,6 +172,7 @@ static struct xfrm_state_afinfo xfrm6_state_afinfo = { .eth_proto = htons(ETH_P_IPV6), .owner = THIS_MODULE, .init_tempsel = __xfrm6_init_tempsel, + .init_temprop = xfrm6_init_temprop, .tmpl_sort = __xfrm6_tmpl_sort, .state_sort = __xfrm6_state_sort, .output = xfrm6_output, diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c index 2ce3a8278f2..2969cad408d 100644 --- a/net/ipv6/xfrm6_tunnel.c +++ b/net/ipv6/xfrm6_tunnel.c @@ -199,7 +199,7 @@ static void x6spi_destroy_rcu(struct rcu_head *head) container_of(head, struct xfrm6_tunnel_spi, rcu_head)); } -void xfrm6_tunnel_free_spi(struct net *net, xfrm_address_t *saddr) +static void xfrm6_tunnel_free_spi(struct net *net, xfrm_address_t *saddr) { struct xfrm6_tunnel_net *xfrm6_tn = xfrm6_tunnel_pernet(net); struct xfrm6_tunnel_spi *x6spi; @@ -223,8 +223,6 @@ void xfrm6_tunnel_free_spi(struct net *net, xfrm_address_t *saddr) spin_unlock_bh(&xfrm6_tunnel_spi_lock); } -EXPORT_SYMBOL(xfrm6_tunnel_free_spi); - static int xfrm6_tunnel_output(struct xfrm_state *x, struct sk_buff *skb) { skb_push(skb, -skb_network_offset(skb)); @@ -317,13 +315,13 @@ static const struct xfrm_type xfrm6_tunnel_type = { .output = xfrm6_tunnel_output, }; -static struct xfrm6_tunnel xfrm6_tunnel_handler = { +static struct xfrm6_tunnel xfrm6_tunnel_handler __read_mostly = { .handler = xfrm6_tunnel_rcv, .err_handler = xfrm6_tunnel_err, .priority = 2, }; -static struct xfrm6_tunnel xfrm46_tunnel_handler = { +static struct xfrm6_tunnel xfrm46_tunnel_handler __read_mostly = { .handler = xfrm6_tunnel_rcv, .err_handler = xfrm6_tunnel_err, .priority = 2, diff --git a/net/ipx/Kconfig b/net/ipx/Kconfig index e9ad0062fbb..02549cb2c32 100644 --- a/net/ipx/Kconfig +++ b/net/ipx/Kconfig @@ -3,6 +3,7 @@ # config IPX tristate "The IPX protocol" + depends on BKL # should be fixable select LLC ---help--- This is support for the Novell networking protocol, IPX, commonly diff --git a/net/irda/af_irda.c b/net/irda/af_irda.c index 79986a674f6..7f097989cde 100644 --- a/net/irda/af_irda.c +++ b/net/irda/af_irda.c @@ -573,9 +573,9 @@ static int irda_find_lsap_sel(struct irda_sock *self, char *name) /* Requested object/attribute doesn't exist */ if((self->errno == IAS_CLASS_UNKNOWN) || (self->errno == IAS_ATTRIB_UNKNOWN)) - return (-EADDRNOTAVAIL); + return -EADDRNOTAVAIL; else - return (-EHOSTUNREACH); + return -EHOSTUNREACH; } /* Get the remote TSAP selector */ @@ -663,7 +663,7 @@ static int irda_discover_daddr_and_lsap_sel(struct irda_sock *self, char *name) __func__, name); self->daddr = DEV_ADDR_ANY; kfree(discoveries); - return(-ENOTUNIQ); + return -ENOTUNIQ; } /* First time we found that one, save it ! */ daddr = self->daddr; @@ -677,7 +677,7 @@ static int irda_discover_daddr_and_lsap_sel(struct irda_sock *self, char *name) IRDA_DEBUG(0, "%s(), unexpected IAS query failure\n", __func__); self->daddr = DEV_ADDR_ANY; kfree(discoveries); - return(-EHOSTUNREACH); + return -EHOSTUNREACH; break; } } @@ -689,7 +689,7 @@ static int irda_discover_daddr_and_lsap_sel(struct irda_sock *self, char *name) IRDA_DEBUG(1, "%s(), cannot discover service ''%s'' in any device !!!\n", __func__, name); self->daddr = DEV_ADDR_ANY; - return(-EADDRNOTAVAIL); + return -EADDRNOTAVAIL; } /* Revert back to discovered device & service */ @@ -715,14 +715,11 @@ static int irda_getname(struct socket *sock, struct sockaddr *uaddr, struct sockaddr_irda saddr; struct sock *sk = sock->sk; struct irda_sock *self = irda_sk(sk); - int err; - lock_kernel(); memset(&saddr, 0, sizeof(saddr)); if (peer) { - err = -ENOTCONN; if (sk->sk_state != TCP_ESTABLISHED) - goto out; + return -ENOTCONN; saddr.sir_family = AF_IRDA; saddr.sir_lsap_sel = self->dtsap_sel; @@ -739,10 +736,8 @@ static int irda_getname(struct socket *sock, struct sockaddr *uaddr, /* uaddr_len come to us uninitialised */ *uaddr_len = sizeof (struct sockaddr_irda); memcpy(uaddr, &saddr, *uaddr_len); - err = 0; -out: - unlock_kernel(); - return err; + + return 0; } /* @@ -758,7 +753,8 @@ static int irda_listen(struct socket *sock, int backlog) IRDA_DEBUG(2, "%s()\n", __func__); - lock_kernel(); + lock_sock(sk); + if ((sk->sk_type != SOCK_STREAM) && (sk->sk_type != SOCK_SEQPACKET) && (sk->sk_type != SOCK_DGRAM)) goto out; @@ -770,7 +766,7 @@ static int irda_listen(struct socket *sock, int backlog) err = 0; } out: - unlock_kernel(); + release_sock(sk); return err; } @@ -793,7 +789,7 @@ static int irda_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) if (addr_len != sizeof(struct sockaddr_irda)) return -EINVAL; - lock_kernel(); + lock_sock(sk); #ifdef CONFIG_IRDA_ULTRA /* Special care for Ultra sockets */ if ((sk->sk_type == SOCK_DGRAM) && @@ -824,8 +820,8 @@ static int irda_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) err = irda_open_tsap(self, addr->sir_lsap_sel, addr->sir_name); if (err < 0) { - kfree(self->ias_obj->name); - kfree(self->ias_obj); + irias_delete_object(self->ias_obj); + self->ias_obj = NULL; goto out; } @@ -836,7 +832,7 @@ static int irda_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) err = 0; out: - unlock_kernel(); + release_sock(sk); return err; } @@ -856,12 +852,13 @@ static int irda_accept(struct socket *sock, struct socket *newsock, int flags) IRDA_DEBUG(2, "%s()\n", __func__); - lock_kernel(); err = irda_create(sock_net(sk), newsock, sk->sk_protocol, 0); if (err) - goto out; + return err; err = -EINVAL; + + lock_sock(sk); if (sock->state != SS_UNCONNECTED) goto out; @@ -947,7 +944,7 @@ static int irda_accept(struct socket *sock, struct socket *newsock, int flags) irda_connect_response(new); err = 0; out: - unlock_kernel(); + release_sock(sk); return err; } @@ -981,7 +978,7 @@ static int irda_connect(struct socket *sock, struct sockaddr *uaddr, IRDA_DEBUG(2, "%s(%p)\n", __func__, self); - lock_kernel(); + lock_sock(sk); /* Don't allow connect for Ultra sockets */ err = -ESOCKTNOSUPPORT; if ((sk->sk_type == SOCK_DGRAM) && (sk->sk_protocol == IRDAPROTO_ULTRA)) @@ -1072,6 +1069,8 @@ static int irda_connect(struct socket *sock, struct sockaddr *uaddr, if (sk->sk_state != TCP_ESTABLISHED) { sock->state = SS_UNCONNECTED; + if (sk->sk_prot->disconnect(sk, flags)) + sock->state = SS_DISCONNECTING; err = sock_error(sk); if (!err) err = -ECONNRESET; @@ -1084,7 +1083,7 @@ static int irda_connect(struct socket *sock, struct sockaddr *uaddr, self->saddr = irttp_get_saddr(self->tsap); err = 0; out: - unlock_kernel(); + release_sock(sk); return err; } @@ -1231,7 +1230,6 @@ static int irda_release(struct socket *sock) if (sk == NULL) return 0; - lock_kernel(); lock_sock(sk); sk->sk_state = TCP_CLOSE; sk->sk_shutdown |= SEND_SHUTDOWN; @@ -1250,7 +1248,6 @@ static int irda_release(struct socket *sock) /* Destroy networking socket if we are the last reference on it, * i.e. if(sk->sk_refcnt == 0) -> sk_free(sk) */ sock_put(sk); - unlock_kernel(); /* Notes on socket locking and deallocation... - Jean II * In theory we should put pairs of sock_hold() / sock_put() to @@ -1298,7 +1295,6 @@ static int irda_sendmsg(struct kiocb *iocb, struct socket *sock, IRDA_DEBUG(4, "%s(), len=%zd\n", __func__, len); - lock_kernel(); /* Note : socket.c set MSG_EOR on SEQPACKET sockets */ if (msg->msg_flags & ~(MSG_DONTWAIT | MSG_EOR | MSG_CMSG_COMPAT | MSG_NOSIGNAL)) { @@ -1306,6 +1302,8 @@ static int irda_sendmsg(struct kiocb *iocb, struct socket *sock, goto out; } + lock_sock(sk); + if (sk->sk_shutdown & SEND_SHUTDOWN) goto out_err; @@ -1361,14 +1359,14 @@ static int irda_sendmsg(struct kiocb *iocb, struct socket *sock, goto out_err; } - unlock_kernel(); + release_sock(sk); /* Tell client how much data we actually sent */ return len; out_err: err = sk_stream_error(sk, msg->msg_flags, err); out: - unlock_kernel(); + release_sock(sk); return err; } @@ -1390,14 +1388,10 @@ static int irda_recvmsg_dgram(struct kiocb *iocb, struct socket *sock, IRDA_DEBUG(4, "%s()\n", __func__); - lock_kernel(); - if ((err = sock_error(sk)) < 0) - goto out; - skb = skb_recv_datagram(sk, flags & ~MSG_DONTWAIT, flags & MSG_DONTWAIT, &err); if (!skb) - goto out; + return err; skb_reset_transport_header(skb); copied = skb->len; @@ -1425,12 +1419,8 @@ static int irda_recvmsg_dgram(struct kiocb *iocb, struct socket *sock, irttp_flow_request(self->tsap, FLOW_START); } } - unlock_kernel(); - return copied; -out: - unlock_kernel(); - return err; + return copied; } /* @@ -1448,17 +1438,15 @@ static int irda_recvmsg_stream(struct kiocb *iocb, struct socket *sock, IRDA_DEBUG(3, "%s()\n", __func__); - lock_kernel(); if ((err = sock_error(sk)) < 0) - goto out; + return err; - err = -EINVAL; if (sock->flags & __SO_ACCEPTCON) - goto out; + return -EINVAL; err =-EOPNOTSUPP; if (flags & MSG_OOB) - goto out; + return -EOPNOTSUPP; err = 0; target = sock_rcvlowat(sk, flags & MSG_WAITALL, size); @@ -1500,7 +1488,7 @@ static int irda_recvmsg_stream(struct kiocb *iocb, struct socket *sock, finish_wait(sk_sleep(sk), &wait); if (err) - goto out; + return err; if (sk->sk_shutdown & RCV_SHUTDOWN) break; @@ -1553,9 +1541,7 @@ static int irda_recvmsg_stream(struct kiocb *iocb, struct socket *sock, } } -out: - unlock_kernel(); - return err ? : copied; + return copied; } /* @@ -1573,13 +1559,12 @@ static int irda_sendmsg_dgram(struct kiocb *iocb, struct socket *sock, struct sk_buff *skb; int err; - lock_kernel(); - IRDA_DEBUG(4, "%s(), len=%zd\n", __func__, len); - err = -EINVAL; if (msg->msg_flags & ~(MSG_DONTWAIT|MSG_CMSG_COMPAT)) - goto out; + return -EINVAL; + + lock_sock(sk); if (sk->sk_shutdown & SEND_SHUTDOWN) { send_sig(SIGPIPE, current, 0); @@ -1630,10 +1615,12 @@ static int irda_sendmsg_dgram(struct kiocb *iocb, struct socket *sock, IRDA_DEBUG(0, "%s(), err=%d\n", __func__, err); goto out; } - unlock_kernel(); + + release_sock(sk); return len; + out: - unlock_kernel(); + release_sock(sk); return err; } @@ -1656,10 +1643,11 @@ static int irda_sendmsg_ultra(struct kiocb *iocb, struct socket *sock, IRDA_DEBUG(4, "%s(), len=%zd\n", __func__, len); - lock_kernel(); err = -EINVAL; if (msg->msg_flags & ~(MSG_DONTWAIT|MSG_CMSG_COMPAT)) - goto out; + return -EINVAL; + + lock_sock(sk); err = -EPIPE; if (sk->sk_shutdown & SEND_SHUTDOWN) { @@ -1732,7 +1720,7 @@ static int irda_sendmsg_ultra(struct kiocb *iocb, struct socket *sock, if (err) IRDA_DEBUG(0, "%s(), err=%d\n", __func__, err); out: - unlock_kernel(); + release_sock(sk); return err ? : len; } #endif /* CONFIG_IRDA_ULTRA */ @@ -1747,7 +1735,7 @@ static int irda_shutdown(struct socket *sock, int how) IRDA_DEBUG(1, "%s(%p)\n", __func__, self); - lock_kernel(); + lock_sock(sk); sk->sk_state = TCP_CLOSE; sk->sk_shutdown |= SEND_SHUTDOWN; @@ -1769,7 +1757,7 @@ static int irda_shutdown(struct socket *sock, int how) self->daddr = DEV_ADDR_ANY; /* Until we get re-connected */ self->saddr = 0x0; /* so IrLMP assign us any link */ - unlock_kernel(); + release_sock(sk); return 0; } @@ -1786,7 +1774,6 @@ static unsigned int irda_poll(struct file * file, struct socket *sock, IRDA_DEBUG(4, "%s()\n", __func__); - lock_kernel(); poll_wait(file, sk_sleep(sk), wait); mask = 0; @@ -1834,20 +1821,8 @@ static unsigned int irda_poll(struct file * file, struct socket *sock, default: break; } - unlock_kernel(); - return mask; -} - -static unsigned int irda_datagram_poll(struct file *file, struct socket *sock, - poll_table *wait) -{ - int err; - - lock_kernel(); - err = datagram_poll(file, sock, wait); - unlock_kernel(); - return err; + return mask; } /* @@ -1860,7 +1835,6 @@ static int irda_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) IRDA_DEBUG(4, "%s(), cmd=%#x\n", __func__, cmd); - lock_kernel(); err = -EINVAL; switch (cmd) { case TIOCOUTQ: { @@ -1903,7 +1877,6 @@ static int irda_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) IRDA_DEBUG(1, "%s(), doing device ioctl!\n", __func__); err = -ENOIOCTLCMD; } - unlock_kernel(); return err; } @@ -1927,7 +1900,7 @@ static int irda_compat_ioctl(struct socket *sock, unsigned int cmd, unsigned lon * Set some options for the socket * */ -static int __irda_setsockopt(struct socket *sock, int level, int optname, +static int irda_setsockopt(struct socket *sock, int level, int optname, char __user *optval, unsigned int optlen) { struct sock *sk = sock->sk; @@ -1935,13 +1908,15 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, struct irda_ias_set *ias_opt; struct ias_object *ias_obj; struct ias_attrib * ias_attr; /* Attribute in IAS object */ - int opt, free_ias = 0; + int opt, free_ias = 0, err = 0; IRDA_DEBUG(2, "%s(%p)\n", __func__, self); if (level != SOL_IRLMP) return -ENOPROTOOPT; + lock_sock(sk); + switch (optname) { case IRLMP_IAS_SET: /* The user want to add an attribute to an existing IAS object @@ -1951,17 +1926,22 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, * create the right attribute... */ - if (optlen != sizeof(struct irda_ias_set)) - return -EINVAL; + if (optlen != sizeof(struct irda_ias_set)) { + err = -EINVAL; + goto out; + } ias_opt = kmalloc(sizeof(struct irda_ias_set), GFP_ATOMIC); - if (ias_opt == NULL) - return -ENOMEM; + if (ias_opt == NULL) { + err = -ENOMEM; + goto out; + } /* Copy query to the driver. */ if (copy_from_user(ias_opt, optval, optlen)) { kfree(ias_opt); - return -EFAULT; + err = -EFAULT; + goto out; } /* Find the object we target. @@ -1971,7 +1951,8 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, if(ias_opt->irda_class_name[0] == '\0') { if(self->ias_obj == NULL) { kfree(ias_opt); - return -EINVAL; + err = -EINVAL; + goto out; } ias_obj = self->ias_obj; } else @@ -1983,7 +1964,8 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, if((!capable(CAP_NET_ADMIN)) && ((ias_obj == NULL) || (ias_obj != self->ias_obj))) { kfree(ias_opt); - return -EPERM; + err = -EPERM; + goto out; } /* If the object doesn't exist, create it */ @@ -1993,7 +1975,8 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, jiffies); if (ias_obj == NULL) { kfree(ias_opt); - return -ENOMEM; + err = -ENOMEM; + goto out; } free_ias = 1; } @@ -2005,7 +1988,8 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, kfree(ias_obj->name); kfree(ias_obj); } - return -EINVAL; + err = -EINVAL; + goto out; } /* Look at the type */ @@ -2028,7 +2012,8 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, kfree(ias_obj); } - return -EINVAL; + err = -EINVAL; + goto out; } /* Add an octet sequence attribute */ irias_add_octseq_attrib( @@ -2060,7 +2045,8 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, kfree(ias_obj->name); kfree(ias_obj); } - return -EINVAL; + err = -EINVAL; + goto out; } irias_insert_object(ias_obj); kfree(ias_opt); @@ -2071,17 +2057,22 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, * object is not owned by the kernel and delete it. */ - if (optlen != sizeof(struct irda_ias_set)) - return -EINVAL; + if (optlen != sizeof(struct irda_ias_set)) { + err = -EINVAL; + goto out; + } ias_opt = kmalloc(sizeof(struct irda_ias_set), GFP_ATOMIC); - if (ias_opt == NULL) - return -ENOMEM; + if (ias_opt == NULL) { + err = -ENOMEM; + goto out; + } /* Copy query to the driver. */ if (copy_from_user(ias_opt, optval, optlen)) { kfree(ias_opt); - return -EFAULT; + err = -EFAULT; + goto out; } /* Find the object we target. @@ -2094,7 +2085,8 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, ias_obj = irias_find_object(ias_opt->irda_class_name); if(ias_obj == (struct ias_object *) NULL) { kfree(ias_opt); - return -EINVAL; + err = -EINVAL; + goto out; } /* Only ROOT can mess with the global IAS database. @@ -2103,7 +2095,8 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, if((!capable(CAP_NET_ADMIN)) && ((ias_obj == NULL) || (ias_obj != self->ias_obj))) { kfree(ias_opt); - return -EPERM; + err = -EPERM; + goto out; } /* Find the attribute (in the object) we target */ @@ -2111,14 +2104,16 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, ias_opt->irda_attrib_name); if(ias_attr == (struct ias_attrib *) NULL) { kfree(ias_opt); - return -EINVAL; + err = -EINVAL; + goto out; } /* Check is the user space own the object */ if(ias_attr->value->owner != IAS_USER_ATTR) { IRDA_DEBUG(1, "%s(), attempting to delete a kernel attribute\n", __func__); kfree(ias_opt); - return -EPERM; + err = -EPERM; + goto out; } /* Remove the attribute (and maybe the object) */ @@ -2126,11 +2121,15 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, kfree(ias_opt); break; case IRLMP_MAX_SDU_SIZE: - if (optlen < sizeof(int)) - return -EINVAL; + if (optlen < sizeof(int)) { + err = -EINVAL; + goto out; + } - if (get_user(opt, (int __user *)optval)) - return -EFAULT; + if (get_user(opt, (int __user *)optval)) { + err = -EFAULT; + goto out; + } /* Only possible for a seqpacket service (TTP with SAR) */ if (sk->sk_type != SOCK_SEQPACKET) { @@ -2140,16 +2139,21 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, } else { IRDA_WARNING("%s: not allowed to set MAXSDUSIZE for this socket type!\n", __func__); - return -ENOPROTOOPT; + err = -ENOPROTOOPT; + goto out; } break; case IRLMP_HINTS_SET: - if (optlen < sizeof(int)) - return -EINVAL; + if (optlen < sizeof(int)) { + err = -EINVAL; + goto out; + } /* The input is really a (__u8 hints[2]), easier as an int */ - if (get_user(opt, (int __user *)optval)) - return -EFAULT; + if (get_user(opt, (int __user *)optval)) { + err = -EFAULT; + goto out; + } /* Unregister any old registration */ if (self->skey) @@ -2163,12 +2167,16 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, * making a discovery (nodes which don't match any hint * bit in the mask are not reported). */ - if (optlen < sizeof(int)) - return -EINVAL; + if (optlen < sizeof(int)) { + err = -EINVAL; + goto out; + } /* The input is really a (__u8 hints[2]), easier as an int */ - if (get_user(opt, (int __user *)optval)) - return -EFAULT; + if (get_user(opt, (int __user *)optval)) { + err = -EFAULT; + goto out; + } /* Set the new hint mask */ self->mask.word = (__u16) opt; @@ -2180,19 +2188,12 @@ static int __irda_setsockopt(struct socket *sock, int level, int optname, break; default: - return -ENOPROTOOPT; + err = -ENOPROTOOPT; + break; } - return 0; -} - -static int irda_setsockopt(struct socket *sock, int level, int optname, - char __user *optval, unsigned int optlen) -{ - int err; - lock_kernel(); - err = __irda_setsockopt(sock, level, optname, optval, optlen); - unlock_kernel(); +out: + release_sock(sk); return err; } @@ -2249,7 +2250,7 @@ static int irda_extract_ias_value(struct irda_ias_set *ias_opt, /* * Function irda_getsockopt (sock, level, optname, optval, optlen) */ -static int __irda_getsockopt(struct socket *sock, int level, int optname, +static int irda_getsockopt(struct socket *sock, int level, int optname, char __user *optval, int __user *optlen) { struct sock *sk = sock->sk; @@ -2262,7 +2263,7 @@ static int __irda_getsockopt(struct socket *sock, int level, int optname, int daddr = DEV_ADDR_ANY; /* Dest address for IAS queries */ int val = 0; int len = 0; - int err; + int err = 0; int offset, total; IRDA_DEBUG(2, "%s(%p)\n", __func__, self); @@ -2276,15 +2277,18 @@ static int __irda_getsockopt(struct socket *sock, int level, int optname, if(len < 0) return -EINVAL; + lock_sock(sk); + switch (optname) { case IRLMP_ENUMDEVICES: /* Ask lmp for the current discovery log */ discoveries = irlmp_get_discoveries(&list.len, self->mask.word, self->nslots); /* Check if the we got some results */ - if (discoveries == NULL) - return -EAGAIN; /* Didn't find any devices */ - err = 0; + if (discoveries == NULL) { + err = -EAGAIN; + goto out; /* Didn't find any devices */ + } /* Write total list length back to client */ if (copy_to_user(optval, &list, @@ -2297,8 +2301,7 @@ static int __irda_getsockopt(struct socket *sock, int level, int optname, sizeof(struct irda_device_info); /* Copy the list itself - watch for overflow */ - if(list.len > 2048) - { + if (list.len > 2048) { err = -EINVAL; goto bed; } @@ -2314,17 +2317,20 @@ static int __irda_getsockopt(struct socket *sock, int level, int optname, bed: /* Free up our buffer */ kfree(discoveries); - if (err) - return err; break; case IRLMP_MAX_SDU_SIZE: val = self->max_data_size; len = sizeof(int); - if (put_user(len, optlen)) - return -EFAULT; + if (put_user(len, optlen)) { + err = -EFAULT; + goto out; + } + + if (copy_to_user(optval, &val, len)) { + err = -EFAULT; + goto out; + } - if (copy_to_user(optval, &val, len)) - return -EFAULT; break; case IRLMP_IAS_GET: /* The user want an object from our local IAS database. @@ -2332,17 +2338,22 @@ bed: * that we found */ /* Check that the user has allocated the right space for us */ - if (len != sizeof(struct irda_ias_set)) - return -EINVAL; + if (len != sizeof(struct irda_ias_set)) { + err = -EINVAL; + goto out; + } ias_opt = kmalloc(sizeof(struct irda_ias_set), GFP_ATOMIC); - if (ias_opt == NULL) - return -ENOMEM; + if (ias_opt == NULL) { + err = -ENOMEM; + goto out; + } /* Copy query to the driver. */ if (copy_from_user(ias_opt, optval, len)) { kfree(ias_opt); - return -EFAULT; + err = -EFAULT; + goto out; } /* Find the object we target. @@ -2355,7 +2366,8 @@ bed: ias_obj = irias_find_object(ias_opt->irda_class_name); if(ias_obj == (struct ias_object *) NULL) { kfree(ias_opt); - return -EINVAL; + err = -EINVAL; + goto out; } /* Find the attribute (in the object) we target */ @@ -2363,21 +2375,23 @@ bed: ias_opt->irda_attrib_name); if(ias_attr == (struct ias_attrib *) NULL) { kfree(ias_opt); - return -EINVAL; + err = -EINVAL; + goto out; } /* Translate from internal to user structure */ err = irda_extract_ias_value(ias_opt, ias_attr->value); if(err) { kfree(ias_opt); - return err; + goto out; } /* Copy reply to the user */ if (copy_to_user(optval, ias_opt, sizeof(struct irda_ias_set))) { kfree(ias_opt); - return -EFAULT; + err = -EFAULT; + goto out; } /* Note : don't need to put optlen, we checked it */ kfree(ias_opt); @@ -2388,17 +2402,22 @@ bed: * then wait for the answer to come back. */ /* Check that the user has allocated the right space for us */ - if (len != sizeof(struct irda_ias_set)) - return -EINVAL; + if (len != sizeof(struct irda_ias_set)) { + err = -EINVAL; + goto out; + } ias_opt = kmalloc(sizeof(struct irda_ias_set), GFP_ATOMIC); - if (ias_opt == NULL) - return -ENOMEM; + if (ias_opt == NULL) { + err = -ENOMEM; + goto out; + } /* Copy query to the driver. */ if (copy_from_user(ias_opt, optval, len)) { kfree(ias_opt); - return -EFAULT; + err = -EFAULT; + goto out; } /* At this point, there are two cases... @@ -2419,7 +2438,8 @@ bed: daddr = ias_opt->daddr; if((!daddr) || (daddr == DEV_ADDR_ANY)) { kfree(ias_opt); - return -EINVAL; + err = -EINVAL; + goto out; } } @@ -2428,7 +2448,8 @@ bed: IRDA_WARNING("%s: busy with a previous query\n", __func__); kfree(ias_opt); - return -EBUSY; + err = -EBUSY; + goto out; } self->iriap = iriap_open(LSAP_ANY, IAS_CLIENT, self, @@ -2436,7 +2457,8 @@ bed: if (self->iriap == NULL) { kfree(ias_opt); - return -ENOMEM; + err = -ENOMEM; + goto out; } /* Treat unexpected wakeup as disconnect */ @@ -2455,7 +2477,8 @@ bed: * we can free it regardless! */ kfree(ias_opt); /* Treat signals as disconnect */ - return -EHOSTUNREACH; + err = -EHOSTUNREACH; + goto out; } /* Check what happened */ @@ -2465,9 +2488,11 @@ bed: /* Requested object/attribute doesn't exist */ if((self->errno == IAS_CLASS_UNKNOWN) || (self->errno == IAS_ATTRIB_UNKNOWN)) - return (-EADDRNOTAVAIL); + err = -EADDRNOTAVAIL; else - return (-EHOSTUNREACH); + err = -EHOSTUNREACH; + + goto out; } /* Translate from internal to user structure */ @@ -2476,14 +2501,15 @@ bed: irias_delete_value(self->ias_result); if (err) { kfree(ias_opt); - return err; + goto out; } /* Copy reply to the user */ if (copy_to_user(optval, ias_opt, sizeof(struct irda_ias_set))) { kfree(ias_opt); - return -EFAULT; + err = -EFAULT; + goto out; } /* Note : don't need to put optlen, we checked it */ kfree(ias_opt); @@ -2504,11 +2530,15 @@ bed: */ /* Check that the user is passing us an int */ - if (len != sizeof(int)) - return -EINVAL; + if (len != sizeof(int)) { + err = -EINVAL; + goto out; + } /* Get timeout in ms (max time we block the caller) */ - if (get_user(val, (int __user *)optval)) - return -EFAULT; + if (get_user(val, (int __user *)optval)) { + err = -EFAULT; + goto out; + } /* Tell IrLMP we want to be notified */ irlmp_update_client(self->ckey, self->mask.word, @@ -2520,8 +2550,6 @@ bed: /* Wait until a node is discovered */ if (!self->cachedaddr) { - int ret = 0; - IRDA_DEBUG(1, "%s(), nothing discovered yet, going to sleep...\n", __func__); /* Set watchdog timer to expire in <val> ms. */ @@ -2534,7 +2562,7 @@ bed: /* Wait for IR-LMP to call us back */ __wait_event_interruptible(self->query_wait, (self->cachedaddr != 0 || self->errno == -ETIME), - ret); + err); /* If watchdog is still activated, kill it! */ if(timer_pending(&(self->watchdog))) @@ -2542,8 +2570,8 @@ bed: IRDA_DEBUG(1, "%s(), ...waking up !\n", __func__); - if (ret != 0) - return ret; + if (err != 0) + goto out; } else IRDA_DEBUG(1, "%s(), found immediately !\n", @@ -2566,25 +2594,19 @@ bed: * If the user want more details, he should query * the whole discovery log and pick one device... */ - if (put_user(daddr, (int __user *)optval)) - return -EFAULT; + if (put_user(daddr, (int __user *)optval)) { + err = -EFAULT; + goto out; + } break; default: - return -ENOPROTOOPT; + err = -ENOPROTOOPT; } - return 0; -} - -static int irda_getsockopt(struct socket *sock, int level, int optname, - char __user *optval, int __user *optlen) -{ - int err; +out: - lock_kernel(); - err = __irda_getsockopt(sock, level, optname, optval, optlen); - unlock_kernel(); + release_sock(sk); return err; } @@ -2628,7 +2650,7 @@ static const struct proto_ops irda_seqpacket_ops = { .socketpair = sock_no_socketpair, .accept = irda_accept, .getname = irda_getname, - .poll = irda_datagram_poll, + .poll = datagram_poll, .ioctl = irda_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = irda_compat_ioctl, @@ -2652,7 +2674,7 @@ static const struct proto_ops irda_dgram_ops = { .socketpair = sock_no_socketpair, .accept = irda_accept, .getname = irda_getname, - .poll = irda_datagram_poll, + .poll = datagram_poll, .ioctl = irda_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = irda_compat_ioctl, @@ -2677,7 +2699,7 @@ static const struct proto_ops irda_ultra_ops = { .socketpair = sock_no_socketpair, .accept = sock_no_accept, .getname = irda_getname, - .poll = irda_datagram_poll, + .poll = datagram_poll, .ioctl = irda_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = irda_compat_ioctl, diff --git a/net/irda/discovery.c b/net/irda/discovery.c index c1c8ae93912..36c3f037f17 100644 --- a/net/irda/discovery.c +++ b/net/irda/discovery.c @@ -315,7 +315,7 @@ struct irda_device_info *irlmp_copy_discoveries(hashbin_t *log, int *pn, /* Get the actual number of device in the buffer and return */ *pn = i; - return(buffer); + return buffer; } #ifdef CONFIG_PROC_FS diff --git a/net/irda/ircomm/ircomm_tty.c b/net/irda/ircomm/ircomm_tty.c index faa82ca2dfd..a39cca8331d 100644 --- a/net/irda/ircomm/ircomm_tty.c +++ b/net/irda/ircomm/ircomm_tty.c @@ -449,8 +449,8 @@ static int ircomm_tty_open(struct tty_struct *tty, struct file *filp) } #ifdef SERIAL_DO_RESTART - return ((self->flags & ASYNC_HUP_NOTIFY) ? - -EAGAIN : -ERESTARTSYS); + return (self->flags & ASYNC_HUP_NOTIFY) ? + -EAGAIN : -ERESTARTSYS; #else return -EAGAIN; #endif diff --git a/net/irda/iriap.c b/net/irda/iriap.c index fce364c6c71..5b743bdd89b 100644 --- a/net/irda/iriap.c +++ b/net/irda/iriap.c @@ -502,7 +502,8 @@ static void iriap_getvaluebyclass_confirm(struct iriap_cb *self, IRDA_DEBUG(4, "%s(), strlen=%d\n", __func__, value_len); /* Make sure the string is null-terminated */ - fp[n+value_len] = 0x00; + if (n + value_len < skb->len) + fp[n + value_len] = 0x00; IRDA_DEBUG(4, "Got string %s\n", fp+n); /* Will truncate to IAS_MAX_STRING bytes */ diff --git a/net/irda/irlan/irlan_common.c b/net/irda/irlan/irlan_common.c index a788f9e9427..6130f9d9dbe 100644 --- a/net/irda/irlan/irlan_common.c +++ b/net/irda/irlan/irlan_common.c @@ -1102,7 +1102,7 @@ int irlan_extract_param(__u8 *buf, char *name, char *value, __u16 *len) memcpy(&val_len, buf+n, 2); /* To avoid alignment problems */ le16_to_cpus(&val_len); n+=2; - if (val_len > 1016) { + if (val_len >= 1016) { IRDA_DEBUG(2, "%s(), parameter length to long\n", __func__ ); return -RSP_INVALID_COMMAND_FORMAT; } diff --git a/net/irda/irlan/irlan_eth.c b/net/irda/irlan/irlan_eth.c index 9616c32d107..8ee1ff6c742 100644 --- a/net/irda/irlan/irlan_eth.c +++ b/net/irda/irlan/irlan_eth.c @@ -45,13 +45,11 @@ static int irlan_eth_close(struct net_device *dev); static netdev_tx_t irlan_eth_xmit(struct sk_buff *skb, struct net_device *dev); static void irlan_eth_set_multicast_list( struct net_device *dev); -static struct net_device_stats *irlan_eth_get_stats(struct net_device *dev); static const struct net_device_ops irlan_eth_netdev_ops = { .ndo_open = irlan_eth_open, .ndo_stop = irlan_eth_close, .ndo_start_xmit = irlan_eth_xmit, - .ndo_get_stats = irlan_eth_get_stats, .ndo_set_multicast_list = irlan_eth_set_multicast_list, .ndo_change_mtu = eth_change_mtu, .ndo_validate_addr = eth_validate_addr, @@ -169,6 +167,7 @@ static netdev_tx_t irlan_eth_xmit(struct sk_buff *skb, { struct irlan_cb *self = netdev_priv(dev); int ret; + unsigned int len; /* skb headroom large enough to contain all IrDA-headers? */ if ((skb_headroom(skb) < self->max_header_size) || (skb_shared(skb))) { @@ -188,6 +187,7 @@ static netdev_tx_t irlan_eth_xmit(struct sk_buff *skb, dev->trans_start = jiffies; + len = skb->len; /* Now queue the packet in the transport layer */ if (self->use_udata) ret = irttp_udata_request(self->tsap_data, skb); @@ -206,10 +206,10 @@ static netdev_tx_t irlan_eth_xmit(struct sk_buff *skb, * tried :-) DB */ /* irttp_data_request already free the packet */ - self->stats.tx_dropped++; + dev->stats.tx_dropped++; } else { - self->stats.tx_packets++; - self->stats.tx_bytes += skb->len; + dev->stats.tx_packets++; + dev->stats.tx_bytes += len; } return NETDEV_TX_OK; @@ -224,15 +224,16 @@ static netdev_tx_t irlan_eth_xmit(struct sk_buff *skb, int irlan_eth_receive(void *instance, void *sap, struct sk_buff *skb) { struct irlan_cb *self = instance; + struct net_device *dev = self->dev; if (skb == NULL) { - ++self->stats.rx_dropped; + dev->stats.rx_dropped++; return 0; } if (skb->len < ETH_HLEN) { IRDA_DEBUG(0, "%s() : IrLAN frame too short (%d)\n", __func__, skb->len); - ++self->stats.rx_dropped; + dev->stats.rx_dropped++; dev_kfree_skb(skb); return 0; } @@ -242,10 +243,10 @@ int irlan_eth_receive(void *instance, void *sap, struct sk_buff *skb) * might have been previously set by the low level IrDA network * device driver */ - skb->protocol = eth_type_trans(skb, self->dev); /* Remove eth header */ + skb->protocol = eth_type_trans(skb, dev); /* Remove eth header */ - self->stats.rx_packets++; - self->stats.rx_bytes += skb->len; + dev->stats.rx_packets++; + dev->stats.rx_bytes += skb->len; netif_rx(skb); /* Eat it! */ @@ -346,16 +347,3 @@ static void irlan_eth_set_multicast_list(struct net_device *dev) else irlan_set_broadcast_filter(self, FALSE); } - -/* - * Function irlan_get_stats (dev) - * - * Get the current statistics for this device - * - */ -static struct net_device_stats *irlan_eth_get_stats(struct net_device *dev) -{ - struct irlan_cb *self = netdev_priv(dev); - - return &self->stats; -} diff --git a/net/irda/irlan/irlan_event.c b/net/irda/irlan/irlan_event.c index cbcb4eb5403..43f16040a6f 100644 --- a/net/irda/irlan/irlan_event.c +++ b/net/irda/irlan/irlan_event.c @@ -24,7 +24,7 @@ #include <net/irda/irlan_event.h> -char *irlan_state[] = { +const char * const irlan_state[] = { "IRLAN_IDLE", "IRLAN_QUERY", "IRLAN_CONN", diff --git a/net/irda/irlmp.c b/net/irda/irlmp.c index 0e7d8bde145..6115a44c0a2 100644 --- a/net/irda/irlmp.c +++ b/net/irda/irlmp.c @@ -939,7 +939,7 @@ struct irda_device_info *irlmp_get_discoveries(int *pn, __u16 mask, int nslots) } /* Return current cached discovery log */ - return(irlmp_copy_discoveries(irlmp->cachelog, pn, mask, TRUE)); + return irlmp_copy_discoveries(irlmp->cachelog, pn, mask, TRUE); } EXPORT_SYMBOL(irlmp_get_discoveries); diff --git a/net/irda/irlmp_frame.c b/net/irda/irlmp_frame.c index 3750884094d..062e63b1c5c 100644 --- a/net/irda/irlmp_frame.c +++ b/net/irda/irlmp_frame.c @@ -448,7 +448,7 @@ static struct lsap_cb *irlmp_find_lsap(struct lap_cb *self, __u8 dlsap_sel, (self->cache.slsap_sel == slsap_sel) && (self->cache.dlsap_sel == dlsap_sel)) { - return (self->cache.lsap); + return self->cache.lsap; } #endif diff --git a/net/irda/irnet/irnet.h b/net/irda/irnet/irnet.h index 4300df35d37..0d82ff5aeff 100644 --- a/net/irda/irnet/irnet.h +++ b/net/irda/irnet/irnet.h @@ -458,6 +458,8 @@ typedef struct irnet_socket int disco_index; /* Last read in the discovery log */ int disco_number; /* Size of the discovery log */ + struct mutex lock; + } irnet_socket; /* diff --git a/net/irda/irnet/irnet_irda.c b/net/irda/irnet/irnet_irda.c index e98e40d76f4..7f17a8020e8 100644 --- a/net/irda/irnet/irnet_irda.c +++ b/net/irda/irnet/irnet_irda.c @@ -238,7 +238,7 @@ irnet_ias_to_tsap(irnet_socket * self, DEXIT(IRDA_SR_TRACE, "\n"); /* Return the TSAP */ - return(dtsap_sel); + return dtsap_sel; } /*------------------------------------------------------------------*/ @@ -301,7 +301,7 @@ irnet_connect_tsap(irnet_socket * self) { clear_bit(0, &self->ttp_connect); DERROR(IRDA_SR_ERROR, "connect aborted!\n"); - return(err); + return err; } /* Connect to remote device */ @@ -312,7 +312,7 @@ irnet_connect_tsap(irnet_socket * self) { clear_bit(0, &self->ttp_connect); DERROR(IRDA_SR_ERROR, "connect aborted!\n"); - return(err); + return err; } /* The above call is non-blocking. @@ -321,7 +321,7 @@ irnet_connect_tsap(irnet_socket * self) * See you there ;-) */ DEXIT(IRDA_SR_TRACE, "\n"); - return(err); + return err; } /*------------------------------------------------------------------*/ @@ -362,10 +362,10 @@ irnet_discover_next_daddr(irnet_socket * self) /* The above request is non-blocking. * After a while, IrDA will call us back in irnet_discovervalue_confirm() * We will then call irnet_ias_to_tsap() and come back here again... */ - return(0); + return 0; } else - return(1); + return 1; } /*------------------------------------------------------------------*/ @@ -436,7 +436,7 @@ irnet_discover_daddr_and_lsap_sel(irnet_socket * self) /* Follow me in irnet_discovervalue_confirm() */ DEXIT(IRDA_SR_TRACE, "\n"); - return(0); + return 0; } /*------------------------------------------------------------------*/ @@ -485,7 +485,7 @@ irnet_dname_to_daddr(irnet_socket * self) /* No luck ! */ DEBUG(IRDA_SR_INFO, "cannot discover device ``%s'' !!!\n", self->rname); kfree(discoveries); - return(-EADDRNOTAVAIL); + return -EADDRNOTAVAIL; } @@ -527,7 +527,7 @@ irda_irnet_create(irnet_socket * self) INIT_WORK(&self->disconnect_work, irnet_ppp_disconnect); DEXIT(IRDA_SOCK_TRACE, "\n"); - return(0); + return 0; } /*------------------------------------------------------------------*/ @@ -601,7 +601,7 @@ irda_irnet_connect(irnet_socket * self) * We will finish the connection procedure in irnet_connect_tsap(). */ DEXIT(IRDA_SOCK_TRACE, "\n"); - return(0); + return 0; } /*------------------------------------------------------------------*/ @@ -733,7 +733,7 @@ irnet_daddr_to_dname(irnet_socket * self) /* No luck ! */ DEXIT(IRDA_SERV_INFO, ": cannot discover device 0x%08x !!!\n", self->daddr); kfree(discoveries); - return(-EADDRNOTAVAIL); + return -EADDRNOTAVAIL; } /*------------------------------------------------------------------*/ diff --git a/net/irda/irnet/irnet_ppp.c b/net/irda/irnet/irnet_ppp.c index 36f43cac8ea..7fa86373de4 100644 --- a/net/irda/irnet/irnet_ppp.c +++ b/net/irda/irnet/irnet_ppp.c @@ -166,7 +166,7 @@ irnet_ctrl_write(irnet_socket * ap, } /* Success : we have parsed all commands successfully */ - return(count); + return count; } #ifdef INITIAL_DISCOVERY @@ -300,7 +300,7 @@ irnet_ctrl_read(irnet_socket * ap, } DEXIT(CTRL_TRACE, "\n"); - return(strlen(event)); + return strlen(event); } #endif /* INITIAL_DISCOVERY */ @@ -409,7 +409,7 @@ irnet_ctrl_read(irnet_socket * ap, } DEXIT(CTRL_TRACE, "\n"); - return(strlen(event)); + return strlen(event); } /*------------------------------------------------------------------*/ @@ -480,7 +480,6 @@ dev_irnet_open(struct inode * inode, ap = kzalloc(sizeof(*ap), GFP_KERNEL); DABORT(ap == NULL, -ENOMEM, FS_ERROR, "Can't allocate struct irnet...\n"); - lock_kernel(); /* initialize the irnet structure */ ap->file = file; @@ -502,18 +501,20 @@ dev_irnet_open(struct inode * inode, { DERROR(FS_ERROR, "Can't setup IrDA link...\n"); kfree(ap); - unlock_kernel(); + return err; } /* For the control channel */ ap->event_index = irnet_events.index; /* Cancel all past events */ + mutex_init(&ap->lock); + /* Put our stuff where we will be able to find it later */ file->private_data = ap; DEXIT(FS_TRACE, " - ap=0x%p\n", ap); - unlock_kernel(); + return 0; } @@ -623,7 +624,7 @@ dev_irnet_poll(struct file * file, mask |= irnet_ctrl_poll(ap, file, wait); DEXIT(FS_TRACE, " - mask=0x%X\n", mask); - return(mask); + return mask; } /*------------------------------------------------------------------*/ @@ -664,7 +665,9 @@ dev_irnet_ioctl( { DEBUG(FS_INFO, "Entering PPP discipline.\n"); /* PPP channel setup (ap->chan in configured in dev_irnet_open())*/ - lock_kernel(); + if (mutex_lock_interruptible(&ap->lock)) + return -EINTR; + err = ppp_register_channel(&ap->chan); if(err == 0) { @@ -677,14 +680,17 @@ dev_irnet_ioctl( } else DERROR(FS_ERROR, "Can't setup PPP channel...\n"); - unlock_kernel(); + + mutex_unlock(&ap->lock); } else { /* In theory, should be N_TTY */ DEBUG(FS_INFO, "Exiting PPP discipline.\n"); /* Disconnect from the generic PPP layer */ - lock_kernel(); + if (mutex_lock_interruptible(&ap->lock)) + return -EINTR; + if(ap->ppp_open) { ap->ppp_open = 0; @@ -693,24 +699,31 @@ dev_irnet_ioctl( else DERROR(FS_ERROR, "Channel not registered !\n"); err = 0; - unlock_kernel(); + + mutex_unlock(&ap->lock); } break; /* Query PPP channel and unit number */ case PPPIOCGCHAN: - lock_kernel(); + if (mutex_lock_interruptible(&ap->lock)) + return -EINTR; + if(ap->ppp_open && !put_user(ppp_channel_index(&ap->chan), (int __user *)argp)) err = 0; - unlock_kernel(); + + mutex_unlock(&ap->lock); break; case PPPIOCGUNIT: - lock_kernel(); + if (mutex_lock_interruptible(&ap->lock)) + return -EINTR; + if(ap->ppp_open && !put_user(ppp_unit_number(&ap->chan), (int __user *)argp)) err = 0; - unlock_kernel(); + + mutex_unlock(&ap->lock); break; /* All these ioctls can be passed both directly and from ppp_generic, @@ -730,9 +743,12 @@ dev_irnet_ioctl( if(!capable(CAP_NET_ADMIN)) err = -EPERM; else { - lock_kernel(); + if (mutex_lock_interruptible(&ap->lock)) + return -EINTR; + err = ppp_irnet_ioctl(&ap->chan, cmd, arg); - unlock_kernel(); + + mutex_unlock(&ap->lock); } break; @@ -740,7 +756,9 @@ dev_irnet_ioctl( /* Get termios */ case TCGETS: DEBUG(FS_INFO, "Get termios.\n"); - lock_kernel(); + if (mutex_lock_interruptible(&ap->lock)) + return -EINTR; + #ifndef TCGETS2 if(!kernel_termios_to_user_termios((struct termios __user *)argp, &ap->termios)) err = 0; @@ -748,12 +766,15 @@ dev_irnet_ioctl( if(kernel_termios_to_user_termios_1((struct termios __user *)argp, &ap->termios)) err = 0; #endif - unlock_kernel(); + + mutex_unlock(&ap->lock); break; /* Set termios */ case TCSETSF: DEBUG(FS_INFO, "Set termios.\n"); - lock_kernel(); + if (mutex_lock_interruptible(&ap->lock)) + return -EINTR; + #ifndef TCGETS2 if(!user_termios_to_kernel_termios(&ap->termios, (struct termios __user *)argp)) err = 0; @@ -761,7 +782,8 @@ dev_irnet_ioctl( if(!user_termios_to_kernel_termios_1(&ap->termios, (struct termios __user *)argp)) err = 0; #endif - unlock_kernel(); + + mutex_unlock(&ap->lock); break; /* Set DTR/RTS */ @@ -784,9 +806,10 @@ dev_irnet_ioctl( * We should also worry that we don't accept junk here and that * we get rid of our own buffers */ #ifdef FLUSH_TO_PPP - lock_kernel(); + if (mutex_lock_interruptible(&ap->lock)) + return -EINTR; ppp_output_wakeup(&ap->chan); - unlock_kernel(); + mutex_unlock(&ap->lock); #endif /* FLUSH_TO_PPP */ err = 0; break; diff --git a/net/irda/irnet/irnet_ppp.h b/net/irda/irnet/irnet_ppp.h index b5df2418f90..940225866da 100644 --- a/net/irda/irnet/irnet_ppp.h +++ b/net/irda/irnet/irnet_ppp.h @@ -103,7 +103,8 @@ static const struct file_operations irnet_device_fops = .poll = dev_irnet_poll, .unlocked_ioctl = dev_irnet_ioctl, .open = dev_irnet_open, - .release = dev_irnet_close + .release = dev_irnet_close, + .llseek = noop_llseek, /* Also : llseek, readdir, mmap, flush, fsync, fasync, lock, readv, writev */ }; diff --git a/net/irda/parameters.c b/net/irda/parameters.c index fc1a20565e2..71cd38c1a67 100644 --- a/net/irda/parameters.c +++ b/net/irda/parameters.c @@ -298,6 +298,8 @@ static int irda_extract_string(void *self, __u8 *buf, int len, __u8 pi, p.pi = pi; /* In case handler needs to know */ p.pl = buf[1]; /* Extract length of value */ + if (p.pl > 32) + p.pl = 32; IRDA_DEBUG(2, "%s(), pi=%#x, pl=%d\n", __func__, p.pi, p.pl); @@ -318,7 +320,7 @@ static int irda_extract_string(void *self, __u8 *buf, int len, __u8 pi, (__u8) str[0], (__u8) str[1]); /* Null terminate string */ - str[p.pl+1] = '\0'; + str[p.pl] = '\0'; p.pv.c = str; /* Handler will need to take a copy */ diff --git a/net/key/af_key.c b/net/key/af_key.c index 43040e97c47..d87c22df6f1 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -565,12 +565,12 @@ pfkey_proto2satype(uint16_t proto) static uint8_t pfkey_proto_to_xfrm(uint8_t proto) { - return (proto == IPSEC_PROTO_ANY ? 0 : proto); + return proto == IPSEC_PROTO_ANY ? 0 : proto; } static uint8_t pfkey_proto_from_xfrm(uint8_t proto) { - return (proto ? proto : IPSEC_PROTO_ANY); + return proto ? proto : IPSEC_PROTO_ANY; } static inline int pfkey_sockaddr_len(sa_family_t family) diff --git a/net/l2tp/l2tp_eth.c b/net/l2tp/l2tp_eth.c index 58c6c4cda73..8d9ce0accc9 100644 --- a/net/l2tp/l2tp_eth.c +++ b/net/l2tp/l2tp_eth.c @@ -132,7 +132,7 @@ static void l2tp_eth_dev_recv(struct l2tp_session *session, struct sk_buff *skb, printk("\n"); } - if (data_len < ETH_HLEN) + if (!pskb_may_pull(skb, sizeof(ETH_HLEN))) goto error; secpath_reset(skb); @@ -144,7 +144,6 @@ static void l2tp_eth_dev_recv(struct l2tp_session *session, struct sk_buff *skb, nf_reset(skb); if (dev_forward_skb(dev, skb) == NET_RX_SUCCESS) { - dev->last_rx = jiffies; dev->stats.rx_packets++; dev->stats.rx_bytes += data_len; } else diff --git a/net/l2tp/l2tp_ip.c b/net/l2tp/l2tp_ip.c index 226a0ae3bcf..1c770c0644d 100644 --- a/net/l2tp/l2tp_ip.c +++ b/net/l2tp/l2tp_ip.c @@ -65,9 +65,7 @@ static struct sock *__l2tp_ip_bind_lookup(struct net *net, __be32 laddr, int dif continue; if ((l2tp->conn_id == tunnel_id) && -#ifdef CONFIG_NET_NS - (sk->sk_net == net) && -#endif + net_eq(sock_net(sk), net) && !(inet->inet_rcv_saddr && inet->inet_rcv_saddr != laddr) && !(sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif)) goto found; diff --git a/net/l2tp/l2tp_ppp.c b/net/l2tp/l2tp_ppp.c index ff954b3e94b..39a21d0c61c 100644 --- a/net/l2tp/l2tp_ppp.c +++ b/net/l2tp/l2tp_ppp.c @@ -1768,7 +1768,7 @@ static const struct proto_ops pppol2tp_ops = { .ioctl = pppox_ioctl, }; -static struct pppox_proto pppol2tp_proto = { +static const struct pppox_proto pppol2tp_proto = { .create = pppol2tp_create, .ioctl = pppol2tp_ioctl }; diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c index 023ba820236..58261299821 100644 --- a/net/llc/af_llc.c +++ b/net/llc/af_llc.c @@ -1024,7 +1024,8 @@ static int llc_ui_setsockopt(struct socket *sock, int level, int optname, { struct sock *sk = sock->sk; struct llc_sock *llc = llc_sk(sk); - int rc = -EINVAL, opt; + unsigned int opt; + int rc = -EINVAL; lock_sock(sk); if (unlikely(level != SOL_LLC || optlen != sizeof(int))) diff --git a/net/llc/llc_station.c b/net/llc/llc_station.c index e4dae0244d7..cf4aea3ba30 100644 --- a/net/llc/llc_station.c +++ b/net/llc/llc_station.c @@ -689,7 +689,7 @@ static void llc_station_rcv(struct sk_buff *skb) int __init llc_station_init(void) { - u16 rc = -ENOBUFS; + int rc = -ENOBUFS; struct sk_buff *skb; struct llc_station_state_ev *ev; diff --git a/net/mac80211/aes_ccm.c b/net/mac80211/aes_ccm.c index a87cb3ba2df..d2b03e0851e 100644 --- a/net/mac80211/aes_ccm.c +++ b/net/mac80211/aes_ccm.c @@ -138,10 +138,8 @@ struct crypto_cipher *ieee80211_aes_key_setup_encrypt(const u8 key[]) struct crypto_cipher *tfm; tfm = crypto_alloc_cipher("aes", 0, CRYPTO_ALG_ASYNC); - if (IS_ERR(tfm)) - return NULL; - - crypto_cipher_setkey(tfm, key, ALG_CCMP_KEY_LEN); + if (!IS_ERR(tfm)) + crypto_cipher_setkey(tfm, key, ALG_CCMP_KEY_LEN); return tfm; } diff --git a/net/mac80211/aes_cmac.c b/net/mac80211/aes_cmac.c index 3d097b3d7b6..b4d66cca76d 100644 --- a/net/mac80211/aes_cmac.c +++ b/net/mac80211/aes_cmac.c @@ -119,10 +119,8 @@ struct crypto_cipher * ieee80211_aes_cmac_key_setup(const u8 key[]) struct crypto_cipher *tfm; tfm = crypto_alloc_cipher("aes", 0, CRYPTO_ALG_ASYNC); - if (IS_ERR(tfm)) - return NULL; - - crypto_cipher_setkey(tfm, key, AES_CMAC_KEY_LEN); + if (!IS_ERR(tfm)) + crypto_cipher_setkey(tfm, key, AES_CMAC_KEY_LEN); return tfm; } diff --git a/net/mac80211/agg-rx.c b/net/mac80211/agg-rx.c index 965b272499f..720b7a84af5 100644 --- a/net/mac80211/agg-rx.c +++ b/net/mac80211/agg-rx.c @@ -56,7 +56,7 @@ static void ieee80211_free_tid_rx(struct rcu_head *h) } void ___ieee80211_stop_rx_ba_session(struct sta_info *sta, u16 tid, - u16 initiator, u16 reason) + u16 initiator, u16 reason, bool tx) { struct ieee80211_local *local = sta->local; struct tid_ampdu_rx *tid_rx; @@ -81,20 +81,21 @@ void ___ieee80211_stop_rx_ba_session(struct sta_info *sta, u16 tid, "aggregation for tid %d\n", tid); /* check if this is a self generated aggregation halt */ - if (initiator == WLAN_BACK_RECIPIENT) + if (initiator == WLAN_BACK_RECIPIENT && tx) ieee80211_send_delba(sta->sdata, sta->sta.addr, tid, 0, reason); del_timer_sync(&tid_rx->session_timer); + del_timer_sync(&tid_rx->reorder_timer); call_rcu(&tid_rx->rcu_head, ieee80211_free_tid_rx); } void __ieee80211_stop_rx_ba_session(struct sta_info *sta, u16 tid, - u16 initiator, u16 reason) + u16 initiator, u16 reason, bool tx) { mutex_lock(&sta->ampdu_mlme.mtx); - ___ieee80211_stop_rx_ba_session(sta, tid, initiator, reason); + ___ieee80211_stop_rx_ba_session(sta, tid, initiator, reason, tx); mutex_unlock(&sta->ampdu_mlme.mtx); } @@ -120,6 +121,20 @@ static void sta_rx_agg_session_timer_expired(unsigned long data) ieee80211_queue_work(&sta->local->hw, &sta->ampdu_mlme.work); } +static void sta_rx_agg_reorder_timer_expired(unsigned long data) +{ + u8 *ptid = (u8 *)data; + u8 *timer_to_id = ptid - *ptid; + struct sta_info *sta = container_of(timer_to_id, struct sta_info, + timer_to_tid[0]); + + rcu_read_lock(); + spin_lock(&sta->lock); + ieee80211_release_reorder_timeout(sta, *ptid); + spin_unlock(&sta->lock); + rcu_read_unlock(); +} + static void ieee80211_send_addba_resp(struct ieee80211_sub_if_data *sdata, u8 *da, u16 tid, u8 dialog_token, u16 status, u16 policy, u16 buf_size, u16 timeout) @@ -251,11 +266,18 @@ void ieee80211_process_addba_request(struct ieee80211_local *local, goto end; } + spin_lock_init(&tid_agg_rx->reorder_lock); + /* rx timer */ tid_agg_rx->session_timer.function = sta_rx_agg_session_timer_expired; tid_agg_rx->session_timer.data = (unsigned long)&sta->timer_to_tid[tid]; init_timer(&tid_agg_rx->session_timer); + /* rx reorder timer */ + tid_agg_rx->reorder_timer.function = sta_rx_agg_reorder_timer_expired; + tid_agg_rx->reorder_timer.data = (unsigned long)&sta->timer_to_tid[tid]; + init_timer(&tid_agg_rx->reorder_timer); + /* prepare reordering buffer */ tid_agg_rx->reorder_buf = kcalloc(buf_size, sizeof(struct sk_buff *), GFP_ATOMIC); diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c index c893f236ace..d4679b265ba 100644 --- a/net/mac80211/agg-tx.c +++ b/net/mac80211/agg-tx.c @@ -145,7 +145,8 @@ static void kfree_tid_tx(struct rcu_head *rcu_head) } int ___ieee80211_stop_tx_ba_session(struct sta_info *sta, u16 tid, - enum ieee80211_back_parties initiator) + enum ieee80211_back_parties initiator, + bool tx) { struct ieee80211_local *local = sta->local; struct tid_ampdu_tx *tid_tx = sta->ampdu_mlme.tid_tx[tid]; @@ -175,6 +176,8 @@ int ___ieee80211_stop_tx_ba_session(struct sta_info *sta, u16 tid, set_bit(HT_AGG_STATE_STOPPING, &tid_tx->state); + del_timer_sync(&tid_tx->addba_resp_timer); + /* * After this packets are no longer handed right through * to the driver but are put onto tid_tx->pending instead, @@ -183,6 +186,7 @@ int ___ieee80211_stop_tx_ba_session(struct sta_info *sta, u16 tid, clear_bit(HT_AGG_STATE_OPERATIONAL, &tid_tx->state); tid_tx->stop_initiator = initiator; + tid_tx->tx_stop = tx; ret = drv_ampdu_action(local, sta->sdata, IEEE80211_AMPDU_TX_STOP, @@ -575,13 +579,14 @@ void ieee80211_start_tx_ba_cb_irqsafe(struct ieee80211_vif *vif, EXPORT_SYMBOL(ieee80211_start_tx_ba_cb_irqsafe); int __ieee80211_stop_tx_ba_session(struct sta_info *sta, u16 tid, - enum ieee80211_back_parties initiator) + enum ieee80211_back_parties initiator, + bool tx) { int ret; mutex_lock(&sta->ampdu_mlme.mtx); - ret = ___ieee80211_stop_tx_ba_session(sta, tid, initiator); + ret = ___ieee80211_stop_tx_ba_session(sta, tid, initiator, tx); mutex_unlock(&sta->ampdu_mlme.mtx); @@ -670,7 +675,7 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_vif *vif, u8 *ra, u8 tid) goto unlock_sta; } - if (tid_tx->stop_initiator == WLAN_BACK_INITIATOR) + if (tid_tx->stop_initiator == WLAN_BACK_INITIATOR && tid_tx->tx_stop) ieee80211_send_delba(sta->sdata, ra, tid, WLAN_BACK_INITIATOR, WLAN_REASON_QSTA_NOT_USE); @@ -770,7 +775,8 @@ void ieee80211_process_addba_resp(struct ieee80211_local *local, sta->ampdu_mlme.addba_req_num[tid] = 0; } else { - ___ieee80211_stop_tx_ba_session(sta, tid, WLAN_BACK_INITIATOR); + ___ieee80211_stop_tx_ba_session(sta, tid, WLAN_BACK_INITIATOR, + true); } out: diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 29ac8e1a509..18bd0e55060 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -19,33 +19,6 @@ #include "rate.h" #include "mesh.h" -static bool nl80211_type_check(enum nl80211_iftype type) -{ - switch (type) { - case NL80211_IFTYPE_ADHOC: - case NL80211_IFTYPE_STATION: - case NL80211_IFTYPE_MONITOR: -#ifdef CONFIG_MAC80211_MESH - case NL80211_IFTYPE_MESH_POINT: -#endif - case NL80211_IFTYPE_AP: - case NL80211_IFTYPE_AP_VLAN: - case NL80211_IFTYPE_WDS: - return true; - default: - return false; - } -} - -static bool nl80211_params_check(enum nl80211_iftype type, - struct vif_params *params) -{ - if (!nl80211_type_check(type)) - return false; - - return true; -} - static int ieee80211_add_iface(struct wiphy *wiphy, char *name, enum nl80211_iftype type, u32 *flags, struct vif_params *params) @@ -55,9 +28,6 @@ static int ieee80211_add_iface(struct wiphy *wiphy, char *name, struct ieee80211_sub_if_data *sdata; int err; - if (!nl80211_params_check(type, params)) - return -EINVAL; - err = ieee80211_if_add(local, name, &dev, type, params); if (err || type != NL80211_IFTYPE_MONITOR || !flags) return err; @@ -82,12 +52,6 @@ static int ieee80211_change_iface(struct wiphy *wiphy, struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); int ret; - if (ieee80211_sdata_running(sdata)) - return -EBUSY; - - if (!nl80211_params_check(type, params)) - return -EINVAL; - ret = ieee80211_if_change_type(sdata, type); if (ret) return ret; @@ -104,54 +68,71 @@ static int ieee80211_change_iface(struct wiphy *wiphy, params && params->use_4addr >= 0) sdata->u.mgd.use_4addr = params->use_4addr; - if (sdata->vif.type == NL80211_IFTYPE_MONITOR && flags) - sdata->u.mntr_flags = *flags; + if (sdata->vif.type == NL80211_IFTYPE_MONITOR && flags) { + struct ieee80211_local *local = sdata->local; + + if (ieee80211_sdata_running(sdata)) { + /* + * Prohibit MONITOR_FLAG_COOK_FRAMES to be + * changed while the interface is up. + * Else we would need to add a lot of cruft + * to update everything: + * cooked_mntrs, monitor and all fif_* counters + * reconfigure hardware + */ + if ((*flags & MONITOR_FLAG_COOK_FRAMES) != + (sdata->u.mntr_flags & MONITOR_FLAG_COOK_FRAMES)) + return -EBUSY; + + ieee80211_adjust_monitor_flags(sdata, -1); + sdata->u.mntr_flags = *flags; + ieee80211_adjust_monitor_flags(sdata, 1); + + ieee80211_configure_filter(local); + } else { + /* + * Because the interface is down, ieee80211_do_stop + * and ieee80211_do_open take care of "everything" + * mentioned in the comment above. + */ + sdata->u.mntr_flags = *flags; + } + } return 0; } static int ieee80211_add_key(struct wiphy *wiphy, struct net_device *dev, - u8 key_idx, const u8 *mac_addr, + u8 key_idx, bool pairwise, const u8 *mac_addr, struct key_params *params) { - struct ieee80211_sub_if_data *sdata; + struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); struct sta_info *sta = NULL; - enum ieee80211_key_alg alg; struct ieee80211_key *key; int err; - if (!netif_running(dev)) + if (!ieee80211_sdata_running(sdata)) return -ENETDOWN; - sdata = IEEE80211_DEV_TO_SUB_IF(dev); - + /* reject WEP and TKIP keys if WEP failed to initialize */ switch (params->cipher) { case WLAN_CIPHER_SUITE_WEP40: - case WLAN_CIPHER_SUITE_WEP104: - alg = ALG_WEP; - break; case WLAN_CIPHER_SUITE_TKIP: - alg = ALG_TKIP; - break; - case WLAN_CIPHER_SUITE_CCMP: - alg = ALG_CCMP; - break; - case WLAN_CIPHER_SUITE_AES_CMAC: - alg = ALG_AES_CMAC; + case WLAN_CIPHER_SUITE_WEP104: + if (IS_ERR(sdata->local->wep_tx_tfm)) + return -EINVAL; break; default: - return -EINVAL; + break; } - /* reject WEP and TKIP keys if WEP failed to initialize */ - if ((alg == ALG_WEP || alg == ALG_TKIP) && - IS_ERR(sdata->local->wep_tx_tfm)) - return -EINVAL; + key = ieee80211_key_alloc(params->cipher, key_idx, params->key_len, + params->key, params->seq_len, params->seq); + if (IS_ERR(key)) + return PTR_ERR(key); - key = ieee80211_key_alloc(alg, key_idx, params->key_len, params->key, - params->seq_len, params->seq); - if (!key) - return -ENOMEM; + if (pairwise) + key->conf.flags |= IEEE80211_KEY_FLAG_PAIRWISE; mutex_lock(&sdata->local->sta_mtx); @@ -164,9 +145,10 @@ static int ieee80211_add_key(struct wiphy *wiphy, struct net_device *dev, } } - ieee80211_key_link(key, sdata, sta); + err = ieee80211_key_link(key, sdata, sta); + if (err) + ieee80211_key_free(sdata->local, key); - err = 0; out_unlock: mutex_unlock(&sdata->local->sta_mtx); @@ -174,7 +156,7 @@ static int ieee80211_add_key(struct wiphy *wiphy, struct net_device *dev, } static int ieee80211_del_key(struct wiphy *wiphy, struct net_device *dev, - u8 key_idx, const u8 *mac_addr) + u8 key_idx, bool pairwise, const u8 *mac_addr) { struct ieee80211_sub_if_data *sdata; struct sta_info *sta; @@ -191,10 +173,17 @@ static int ieee80211_del_key(struct wiphy *wiphy, struct net_device *dev, if (!sta) goto out_unlock; - if (sta->key) { - ieee80211_key_free(sdata->local, sta->key); - WARN_ON(sta->key); - ret = 0; + if (pairwise) { + if (sta->ptk) { + ieee80211_key_free(sdata->local, sta->ptk); + ret = 0; + } + } else { + if (sta->gtk[key_idx]) { + ieee80211_key_free(sdata->local, + sta->gtk[key_idx]); + ret = 0; + } } goto out_unlock; @@ -216,7 +205,8 @@ static int ieee80211_del_key(struct wiphy *wiphy, struct net_device *dev, } static int ieee80211_get_key(struct wiphy *wiphy, struct net_device *dev, - u8 key_idx, const u8 *mac_addr, void *cookie, + u8 key_idx, bool pairwise, const u8 *mac_addr, + void *cookie, void (*callback)(void *cookie, struct key_params *params)) { @@ -224,7 +214,7 @@ static int ieee80211_get_key(struct wiphy *wiphy, struct net_device *dev, struct sta_info *sta = NULL; u8 seq[6] = {0}; struct key_params params; - struct ieee80211_key *key; + struct ieee80211_key *key = NULL; u32 iv32; u16 iv16; int err = -ENOENT; @@ -238,7 +228,10 @@ static int ieee80211_get_key(struct wiphy *wiphy, struct net_device *dev, if (!sta) goto out; - key = sta->key; + if (pairwise) + key = sta->ptk; + else if (key_idx < NUM_DEFAULT_KEYS) + key = sta->gtk[key_idx]; } else key = sdata->keys[key_idx]; @@ -247,10 +240,10 @@ static int ieee80211_get_key(struct wiphy *wiphy, struct net_device *dev, memset(¶ms, 0, sizeof(params)); - switch (key->conf.alg) { - case ALG_TKIP: - params.cipher = WLAN_CIPHER_SUITE_TKIP; + params.cipher = key->conf.cipher; + switch (key->conf.cipher) { + case WLAN_CIPHER_SUITE_TKIP: iv32 = key->u.tkip.tx.iv32; iv16 = key->u.tkip.tx.iv16; @@ -268,8 +261,7 @@ static int ieee80211_get_key(struct wiphy *wiphy, struct net_device *dev, params.seq = seq; params.seq_len = 6; break; - case ALG_CCMP: - params.cipher = WLAN_CIPHER_SUITE_CCMP; + case WLAN_CIPHER_SUITE_CCMP: seq[0] = key->u.ccmp.tx_pn[5]; seq[1] = key->u.ccmp.tx_pn[4]; seq[2] = key->u.ccmp.tx_pn[3]; @@ -279,14 +271,7 @@ static int ieee80211_get_key(struct wiphy *wiphy, struct net_device *dev, params.seq = seq; params.seq_len = 6; break; - case ALG_WEP: - if (key->conf.keylen == 5) - params.cipher = WLAN_CIPHER_SUITE_WEP40; - else - params.cipher = WLAN_CIPHER_SUITE_WEP104; - break; - case ALG_AES_CMAC: - params.cipher = WLAN_CIPHER_SUITE_AES_CMAC; + case WLAN_CIPHER_SUITE_AES_CMAC: seq[0] = key->u.aes_cmac.tx_pn[5]; seq[1] = key->u.aes_cmac.tx_pn[4]; seq[2] = key->u.aes_cmac.tx_pn[3]; @@ -342,13 +327,19 @@ static void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo) STATION_INFO_TX_BYTES | STATION_INFO_RX_PACKETS | STATION_INFO_TX_PACKETS | - STATION_INFO_TX_BITRATE; + STATION_INFO_TX_RETRIES | + STATION_INFO_TX_FAILED | + STATION_INFO_TX_BITRATE | + STATION_INFO_RX_DROP_MISC; sinfo->inactive_time = jiffies_to_msecs(jiffies - sta->last_rx); sinfo->rx_bytes = sta->rx_bytes; sinfo->tx_bytes = sta->tx_bytes; sinfo->rx_packets = sta->rx_packets; sinfo->tx_packets = sta->tx_packets; + sinfo->tx_retries = sta->tx_retry_count; + sinfo->tx_failed = sta->tx_retry_failed; + sinfo->rx_dropped_misc = sta->rx_dropped; if ((sta->local->hw.flags & IEEE80211_HW_SIGNAL_DBM) || (sta->local->hw.flags & IEEE80211_HW_SIGNAL_UNSPEC)) { @@ -634,6 +625,7 @@ static void sta_apply_parameters(struct ieee80211_local *local, struct sta_info *sta, struct station_parameters *params) { + unsigned long flags; u32 rates; int i, j; struct ieee80211_supported_band *sband; @@ -642,7 +634,7 @@ static void sta_apply_parameters(struct ieee80211_local *local, sband = local->hw.wiphy->bands[local->oper_channel->band]; - spin_lock_bh(&sta->lock); + spin_lock_irqsave(&sta->flaglock, flags); mask = params->sta_flags_mask; set = params->sta_flags_set; @@ -669,7 +661,7 @@ static void sta_apply_parameters(struct ieee80211_local *local, if (set & BIT(NL80211_STA_FLAG_MFP)) sta->flags |= WLAN_STA_MFP; } - spin_unlock_bh(&sta->lock); + spin_unlock_irqrestore(&sta->flaglock, flags); /* * cfg80211 validates this (1-2007) and allows setting the AID @@ -1143,9 +1135,9 @@ static int ieee80211_set_txq_params(struct wiphy *wiphy, p.uapsd = false; if (drv_conf_tx(local, params->queue, &p)) { - printk(KERN_DEBUG "%s: failed to set TX queue " - "parameters for queue %d\n", - wiphy_name(local->hw.wiphy), params->queue); + wiphy_debug(local->hw.wiphy, + "failed to set TX queue parameters for queue %d\n", + params->queue); return -EINVAL; } @@ -1207,15 +1199,26 @@ static int ieee80211_scan(struct wiphy *wiphy, struct net_device *dev, struct cfg80211_scan_request *req) { - struct ieee80211_sub_if_data *sdata; - - sdata = IEEE80211_DEV_TO_SUB_IF(dev); + struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); - if (sdata->vif.type != NL80211_IFTYPE_STATION && - sdata->vif.type != NL80211_IFTYPE_ADHOC && - sdata->vif.type != NL80211_IFTYPE_MESH_POINT && - (sdata->vif.type != NL80211_IFTYPE_AP || sdata->u.ap.beacon)) + switch (ieee80211_vif_type_p2p(&sdata->vif)) { + case NL80211_IFTYPE_STATION: + case NL80211_IFTYPE_ADHOC: + case NL80211_IFTYPE_MESH_POINT: + case NL80211_IFTYPE_P2P_CLIENT: + break; + case NL80211_IFTYPE_P2P_GO: + if (sdata->local->ops->hw_scan) + break; + /* FIXME: implement NoA while scanning in software */ + return -EOPNOTSUPP; + case NL80211_IFTYPE_AP: + if (sdata->u.ap.beacon) + return -EOPNOTSUPP; + break; + default: return -EOPNOTSUPP; + } return ieee80211_request_scan(sdata, req); } @@ -1362,7 +1365,7 @@ static int ieee80211_get_tx_power(struct wiphy *wiphy, int *dbm) } static int ieee80211_set_wds_peer(struct wiphy *wiphy, struct net_device *dev, - u8 *addr) + const u8 *addr) { struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); @@ -1411,7 +1414,7 @@ int __ieee80211_request_smps(struct ieee80211_sub_if_data *sdata, if (!sdata->u.mgd.associated || sdata->vif.bss_conf.channel_type == NL80211_CHAN_NO_HT) { mutex_lock(&sdata->local->iflist_mtx); - ieee80211_recalc_smps(sdata->local, sdata); + ieee80211_recalc_smps(sdata->local); mutex_unlock(&sdata->local->iflist_mtx); return 0; } @@ -1541,11 +1544,11 @@ static int ieee80211_cancel_remain_on_channel(struct wiphy *wiphy, return ieee80211_wk_cancel_remain_on_channel(sdata, cookie); } -static int ieee80211_action(struct wiphy *wiphy, struct net_device *dev, - struct ieee80211_channel *chan, - enum nl80211_channel_type channel_type, - bool channel_type_valid, - const u8 *buf, size_t len, u64 *cookie) +static int ieee80211_mgmt_tx(struct wiphy *wiphy, struct net_device *dev, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + bool channel_type_valid, + const u8 *buf, size_t len, u64 *cookie) { struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); struct ieee80211_local *local = sdata->local; @@ -1566,7 +1569,11 @@ static int ieee80211_action(struct wiphy *wiphy, struct net_device *dev, switch (sdata->vif.type) { case NL80211_IFTYPE_ADHOC: - if (mgmt->u.action.category == WLAN_CATEGORY_PUBLIC) + case NL80211_IFTYPE_AP: + case NL80211_IFTYPE_AP_VLAN: + case NL80211_IFTYPE_P2P_GO: + if (!ieee80211_is_action(mgmt->frame_control) || + mgmt->u.action.category == WLAN_CATEGORY_PUBLIC) break; rcu_read_lock(); sta = sta_info_get(sdata, mgmt->da); @@ -1575,8 +1582,7 @@ static int ieee80211_action(struct wiphy *wiphy, struct net_device *dev, return -ENOLINK; break; case NL80211_IFTYPE_STATION: - if (!(sdata->u.mgd.flags & IEEE80211_STA_MFP_ENABLED)) - flags |= IEEE80211_TX_INTFL_DONT_ENCRYPT; + case NL80211_IFTYPE_P2P_CLIENT: break; default: return -EOPNOTSUPP; @@ -1598,6 +1604,23 @@ static int ieee80211_action(struct wiphy *wiphy, struct net_device *dev, return 0; } +static void ieee80211_mgmt_frame_register(struct wiphy *wiphy, + struct net_device *dev, + u16 frame_type, bool reg) +{ + struct ieee80211_local *local = wiphy_priv(wiphy); + + if (frame_type != (IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_PROBE_REQ)) + return; + + if (reg) + local->probe_req_reg++; + else + local->probe_req_reg--; + + ieee80211_queue_work(&local->hw, &local->reconfig_filter); +} + struct cfg80211_ops mac80211_config_ops = { .add_virtual_intf = ieee80211_add_iface, .del_virtual_intf = ieee80211_del_iface, @@ -1647,6 +1670,7 @@ struct cfg80211_ops mac80211_config_ops = { .set_bitrate_mask = ieee80211_set_bitrate_mask, .remain_on_channel = ieee80211_remain_on_channel, .cancel_remain_on_channel = ieee80211_cancel_remain_on_channel, - .action = ieee80211_action, + .mgmt_tx = ieee80211_mgmt_tx, .set_cqm_rssi_config = ieee80211_set_cqm_rssi_config, + .mgmt_frame_register = ieee80211_mgmt_frame_register, }; diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c index 32be11e4c4d..5b24740fc0b 100644 --- a/net/mac80211/chan.c +++ b/net/mac80211/chan.c @@ -11,7 +11,7 @@ __ieee80211_get_channel_mode(struct ieee80211_local *local, { struct ieee80211_sub_if_data *sdata; - WARN_ON(!mutex_is_locked(&local->iflist_mtx)); + lockdep_assert_held(&local->iflist_mtx); list_for_each_entry(sdata, &local->interfaces, list) { if (sdata == ignore) diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c index a694c593ff6..18260aa99c5 100644 --- a/net/mac80211/debugfs.c +++ b/net/mac80211/debugfs.c @@ -36,6 +36,7 @@ static ssize_t name## _read(struct file *file, char __user *userbuf, \ static const struct file_operations name## _ops = { \ .read = name## _read, \ .open = mac80211_open_file_generic, \ + .llseek = generic_file_llseek, \ }; #define DEBUGFS_ADD(name) \ @@ -85,13 +86,15 @@ static ssize_t tsf_write(struct file *file, if (strncmp(buf, "reset", 5) == 0) { if (local->ops->reset_tsf) { drv_reset_tsf(local); - printk(KERN_INFO "%s: debugfs reset TSF\n", wiphy_name(local->hw.wiphy)); + wiphy_info(local->hw.wiphy, "debugfs reset TSF\n"); } } else { tsf = simple_strtoul(buf, NULL, 0); if (local->ops->set_tsf) { drv_set_tsf(local, tsf); - printk(KERN_INFO "%s: debugfs set TSF to %#018llx\n", wiphy_name(local->hw.wiphy), tsf); + wiphy_info(local->hw.wiphy, + "debugfs set TSF to %#018llx\n", tsf); + } } @@ -101,7 +104,8 @@ static ssize_t tsf_write(struct file *file, static const struct file_operations tsf_ops = { .read = tsf_read, .write = tsf_write, - .open = mac80211_open_file_generic + .open = mac80211_open_file_generic, + .llseek = default_llseek, }; static ssize_t reset_write(struct file *file, const char __user *user_buf, @@ -120,6 +124,7 @@ static ssize_t reset_write(struct file *file, const char __user *user_buf, static const struct file_operations reset_ops = { .write = reset_write, .open = mac80211_open_file_generic, + .llseek = noop_llseek, }; static ssize_t noack_read(struct file *file, char __user *user_buf, @@ -155,7 +160,8 @@ static ssize_t noack_write(struct file *file, static const struct file_operations noack_ops = { .read = noack_read, .write = noack_write, - .open = mac80211_open_file_generic + .open = mac80211_open_file_generic, + .llseek = default_llseek, }; static ssize_t uapsd_queues_read(struct file *file, char __user *user_buf, @@ -201,7 +207,8 @@ static ssize_t uapsd_queues_write(struct file *file, static const struct file_operations uapsd_queues_ops = { .read = uapsd_queues_read, .write = uapsd_queues_write, - .open = mac80211_open_file_generic + .open = mac80211_open_file_generic, + .llseek = default_llseek, }; static ssize_t uapsd_max_sp_len_read(struct file *file, char __user *user_buf, @@ -247,7 +254,8 @@ static ssize_t uapsd_max_sp_len_write(struct file *file, static const struct file_operations uapsd_max_sp_len_ops = { .read = uapsd_max_sp_len_read, .write = uapsd_max_sp_len_write, - .open = mac80211_open_file_generic + .open = mac80211_open_file_generic, + .llseek = default_llseek, }; static ssize_t channel_type_read(struct file *file, char __user *user_buf, @@ -279,7 +287,8 @@ static ssize_t channel_type_read(struct file *file, char __user *user_buf, static const struct file_operations channel_type_ops = { .read = channel_type_read, - .open = mac80211_open_file_generic + .open = mac80211_open_file_generic, + .llseek = default_llseek, }; static ssize_t queues_read(struct file *file, char __user *user_buf, @@ -302,7 +311,8 @@ static ssize_t queues_read(struct file *file, char __user *user_buf, static const struct file_operations queues_ops = { .read = queues_read, - .open = mac80211_open_file_generic + .open = mac80211_open_file_generic, + .llseek = default_llseek, }; /* statistics stuff */ @@ -346,6 +356,7 @@ static ssize_t stats_ ##name## _read(struct file *file, \ static const struct file_operations stats_ ##name## _ops = { \ .read = stats_ ##name## _read, \ .open = mac80211_open_file_generic, \ + .llseek = generic_file_llseek, \ }; #define DEBUGFS_STATS_ADD(name, field) \ @@ -366,7 +377,6 @@ void debugfs_hw_add(struct ieee80211_local *local) if (!phyd) return; - local->debugfs.stations = debugfs_create_dir("stations", phyd); local->debugfs.keys = debugfs_create_dir("keys", phyd); DEBUGFS_ADD(frequency); diff --git a/net/mac80211/debugfs_key.c b/net/mac80211/debugfs_key.c index fa5e76e658e..4aa47d074a7 100644 --- a/net/mac80211/debugfs_key.c +++ b/net/mac80211/debugfs_key.c @@ -32,6 +32,7 @@ static ssize_t key_##name##_read(struct file *file, \ static const struct file_operations key_ ##name## _ops = { \ .read = key_##name##_read, \ .open = mac80211_open_file_generic, \ + .llseek = generic_file_llseek, \ } #define KEY_FILE(name, format) \ @@ -46,6 +47,7 @@ static const struct file_operations key_ ##name## _ops = { \ static const struct file_operations key_ ##name## _ops = { \ .read = key_conf_##name##_read, \ .open = mac80211_open_file_generic, \ + .llseek = generic_file_llseek, \ } #define KEY_CONF_FILE(name, format) \ @@ -64,26 +66,13 @@ static ssize_t key_algorithm_read(struct file *file, char __user *userbuf, size_t count, loff_t *ppos) { - char *alg; + char buf[15]; struct ieee80211_key *key = file->private_data; + u32 c = key->conf.cipher; - switch (key->conf.alg) { - case ALG_WEP: - alg = "WEP\n"; - break; - case ALG_TKIP: - alg = "TKIP\n"; - break; - case ALG_CCMP: - alg = "CCMP\n"; - break; - case ALG_AES_CMAC: - alg = "AES-128-CMAC\n"; - break; - default: - return 0; - } - return simple_read_from_buffer(userbuf, count, ppos, alg, strlen(alg)); + sprintf(buf, "%.2x-%.2x-%.2x:%d\n", + c >> 24, (c >> 16) & 0xff, (c >> 8) & 0xff, c & 0xff); + return simple_read_from_buffer(userbuf, count, ppos, buf, strlen(buf)); } KEY_OPS(algorithm); @@ -95,21 +84,22 @@ static ssize_t key_tx_spec_read(struct file *file, char __user *userbuf, int len; struct ieee80211_key *key = file->private_data; - switch (key->conf.alg) { - case ALG_WEP: + switch (key->conf.cipher) { + case WLAN_CIPHER_SUITE_WEP40: + case WLAN_CIPHER_SUITE_WEP104: len = scnprintf(buf, sizeof(buf), "\n"); break; - case ALG_TKIP: + case WLAN_CIPHER_SUITE_TKIP: len = scnprintf(buf, sizeof(buf), "%08x %04x\n", key->u.tkip.tx.iv32, key->u.tkip.tx.iv16); break; - case ALG_CCMP: + case WLAN_CIPHER_SUITE_CCMP: tpn = key->u.ccmp.tx_pn; len = scnprintf(buf, sizeof(buf), "%02x%02x%02x%02x%02x%02x\n", tpn[0], tpn[1], tpn[2], tpn[3], tpn[4], tpn[5]); break; - case ALG_AES_CMAC: + case WLAN_CIPHER_SUITE_AES_CMAC: tpn = key->u.aes_cmac.tx_pn; len = scnprintf(buf, sizeof(buf), "%02x%02x%02x%02x%02x%02x\n", tpn[0], tpn[1], tpn[2], tpn[3], tpn[4], @@ -130,11 +120,12 @@ static ssize_t key_rx_spec_read(struct file *file, char __user *userbuf, int i, len; const u8 *rpn; - switch (key->conf.alg) { - case ALG_WEP: + switch (key->conf.cipher) { + case WLAN_CIPHER_SUITE_WEP40: + case WLAN_CIPHER_SUITE_WEP104: len = scnprintf(buf, sizeof(buf), "\n"); break; - case ALG_TKIP: + case WLAN_CIPHER_SUITE_TKIP: for (i = 0; i < NUM_RX_DATA_QUEUES; i++) p += scnprintf(p, sizeof(buf)+buf-p, "%08x %04x\n", @@ -142,7 +133,7 @@ static ssize_t key_rx_spec_read(struct file *file, char __user *userbuf, key->u.tkip.rx[i].iv16); len = p - buf; break; - case ALG_CCMP: + case WLAN_CIPHER_SUITE_CCMP: for (i = 0; i < NUM_RX_DATA_QUEUES + 1; i++) { rpn = key->u.ccmp.rx_pn[i]; p += scnprintf(p, sizeof(buf)+buf-p, @@ -152,7 +143,7 @@ static ssize_t key_rx_spec_read(struct file *file, char __user *userbuf, } len = p - buf; break; - case ALG_AES_CMAC: + case WLAN_CIPHER_SUITE_AES_CMAC: rpn = key->u.aes_cmac.rx_pn; p += scnprintf(p, sizeof(buf)+buf-p, "%02x%02x%02x%02x%02x%02x\n", @@ -174,11 +165,11 @@ static ssize_t key_replays_read(struct file *file, char __user *userbuf, char buf[20]; int len; - switch (key->conf.alg) { - case ALG_CCMP: + switch (key->conf.cipher) { + case WLAN_CIPHER_SUITE_CCMP: len = scnprintf(buf, sizeof(buf), "%u\n", key->u.ccmp.replays); break; - case ALG_AES_CMAC: + case WLAN_CIPHER_SUITE_AES_CMAC: len = scnprintf(buf, sizeof(buf), "%u\n", key->u.aes_cmac.replays); break; @@ -196,8 +187,8 @@ static ssize_t key_icverrors_read(struct file *file, char __user *userbuf, char buf[20]; int len; - switch (key->conf.alg) { - case ALG_AES_CMAC: + switch (key->conf.cipher) { + case WLAN_CIPHER_SUITE_AES_CMAC: len = scnprintf(buf, sizeof(buf), "%u\n", key->u.aes_cmac.icverrors); break; diff --git a/net/mac80211/debugfs_netdev.c b/net/mac80211/debugfs_netdev.c index 20b2998fa0e..cbdf36d7841 100644 --- a/net/mac80211/debugfs_netdev.c +++ b/net/mac80211/debugfs_netdev.c @@ -121,6 +121,7 @@ static const struct file_operations name##_ops = { \ .read = ieee80211_if_read_##name, \ .write = (_write), \ .open = mac80211_open_file_generic, \ + .llseek = generic_file_llseek, \ } #define __IEEE80211_IF_FILE_W(name) \ @@ -409,6 +410,9 @@ void ieee80211_debugfs_add_netdev(struct ieee80211_sub_if_data *sdata) sprintf(buf, "netdev:%s", sdata->name); sdata->debugfs.dir = debugfs_create_dir(buf, sdata->local->hw.wiphy->debugfsdir); + if (sdata->debugfs.dir) + sdata->debugfs.subdir_stations = debugfs_create_dir("stations", + sdata->debugfs.dir); add_files(sdata); } diff --git a/net/mac80211/debugfs_sta.c b/net/mac80211/debugfs_sta.c index 76839d4dfaa..4601fea1784 100644 --- a/net/mac80211/debugfs_sta.c +++ b/net/mac80211/debugfs_sta.c @@ -36,6 +36,7 @@ static ssize_t sta_ ##name## _read(struct file *file, \ static const struct file_operations sta_ ##name## _ops = { \ .read = sta_##name##_read, \ .open = mac80211_open_file_generic, \ + .llseek = generic_file_llseek, \ } #define STA_OPS_RW(name) \ @@ -43,6 +44,7 @@ static const struct file_operations sta_ ##name## _ops = { \ .read = sta_##name##_read, \ .write = sta_##name##_write, \ .open = mac80211_open_file_generic, \ + .llseek = generic_file_llseek, \ } #define STA_FILE(name, field, format) \ @@ -196,7 +198,8 @@ static ssize_t sta_agg_status_write(struct file *file, const char __user *userbu else ret = ieee80211_stop_tx_ba_session(&sta->sta, tid); } else { - __ieee80211_stop_rx_ba_session(sta, tid, WLAN_BACK_RECIPIENT, 3); + __ieee80211_stop_rx_ba_session(sta, tid, WLAN_BACK_RECIPIENT, + 3, true); ret = 0; } @@ -300,7 +303,7 @@ STA_OPS(ht_capa); void ieee80211_sta_debugfs_add(struct sta_info *sta) { - struct dentry *stations_dir = sta->local->debugfs.stations; + struct dentry *stations_dir = sta->sdata->debugfs.subdir_stations; u8 mac[3*ETH_ALEN]; sta->debugfs.add_has_run = true; diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index 14123dce544..16983825f8e 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -54,6 +54,20 @@ static inline int drv_add_interface(struct ieee80211_local *local, return ret; } +static inline int drv_change_interface(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, + enum nl80211_iftype type, bool p2p) +{ + int ret; + + might_sleep(); + + trace_drv_change_interface(local, sdata, type, p2p); + ret = local->ops->change_interface(&local->hw, &sdata->vif, type, p2p); + trace_drv_return_int(local, ret); + return ret; +} + static inline void drv_remove_interface(struct ieee80211_local *local, struct ieee80211_vif *vif) { diff --git a/net/mac80211/driver-trace.h b/net/mac80211/driver-trace.h index 5d5d2a97466..6831fb1641c 100644 --- a/net/mac80211/driver-trace.h +++ b/net/mac80211/driver-trace.h @@ -25,12 +25,14 @@ static inline void trace_ ## name(proto) {} #define STA_PR_FMT " sta:%pM" #define STA_PR_ARG __entry->sta_addr -#define VIF_ENTRY __field(enum nl80211_iftype, vif_type) __field(void *, sdata) \ +#define VIF_ENTRY __field(enum nl80211_iftype, vif_type) __field(void *, sdata) \ + __field(bool, p2p) \ __string(vif_name, sdata->dev ? sdata->dev->name : "<nodev>") -#define VIF_ASSIGN __entry->vif_type = sdata->vif.type; __entry->sdata = sdata; \ +#define VIF_ASSIGN __entry->vif_type = sdata->vif.type; __entry->sdata = sdata; \ + __entry->p2p = sdata->vif.p2p; \ __assign_str(vif_name, sdata->dev ? sdata->dev->name : "<nodev>") -#define VIF_PR_FMT " vif:%s(%d)" -#define VIF_PR_ARG __get_str(vif_name), __entry->vif_type +#define VIF_PR_FMT " vif:%s(%d%s)" +#define VIF_PR_ARG __get_str(vif_name), __entry->vif_type, __entry->p2p ? "/p2p" : "" /* * Tracing for driver callbacks. @@ -136,6 +138,34 @@ TRACE_EVENT(drv_add_interface, ) ); +TRACE_EVENT(drv_change_interface, + TP_PROTO(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, + enum nl80211_iftype type, bool p2p), + + TP_ARGS(local, sdata, type, p2p), + + TP_STRUCT__entry( + LOCAL_ENTRY + VIF_ENTRY + __field(u32, new_type) + __field(bool, new_p2p) + ), + + TP_fast_assign( + LOCAL_ASSIGN; + VIF_ASSIGN; + __entry->new_type = type; + __entry->new_p2p = p2p; + ), + + TP_printk( + LOCAL_PR_FMT VIF_PR_FMT " new type:%d%s", + LOCAL_PR_ARG, VIF_PR_ARG, __entry->new_type, + __entry->new_p2p ? "/p2p" : "" + ) +); + TRACE_EVENT(drv_remove_interface, TP_PROTO(struct ieee80211_local *local, struct ieee80211_sub_if_data *sdata), @@ -336,7 +366,7 @@ TRACE_EVENT(drv_set_key, LOCAL_ENTRY VIF_ENTRY STA_ENTRY - __field(enum ieee80211_key_alg, alg) + __field(u32, cipher) __field(u8, hw_key_idx) __field(u8, flags) __field(s8, keyidx) @@ -346,7 +376,7 @@ TRACE_EVENT(drv_set_key, LOCAL_ASSIGN; VIF_ASSIGN; STA_ASSIGN; - __entry->alg = key->alg; + __entry->cipher = key->cipher; __entry->flags = key->flags; __entry->keyidx = key->keyidx; __entry->hw_key_idx = key->hw_key_idx; diff --git a/net/mac80211/ht.c b/net/mac80211/ht.c index 9d101fb3386..75d679d75e6 100644 --- a/net/mac80211/ht.c +++ b/net/mac80211/ht.c @@ -101,16 +101,16 @@ void ieee80211_ht_cap_ie_to_sta_ht_cap(struct ieee80211_supported_band *sband, ht_cap->mcs.rx_mask[32/8] |= 1; } -void ieee80211_sta_tear_down_BA_sessions(struct sta_info *sta) +void ieee80211_sta_tear_down_BA_sessions(struct sta_info *sta, bool tx) { int i; cancel_work_sync(&sta->ampdu_mlme.work); for (i = 0; i < STA_TID_NUM; i++) { - __ieee80211_stop_tx_ba_session(sta, i, WLAN_BACK_INITIATOR); + __ieee80211_stop_tx_ba_session(sta, i, WLAN_BACK_INITIATOR, tx); __ieee80211_stop_rx_ba_session(sta, i, WLAN_BACK_RECIPIENT, - WLAN_REASON_QSTA_LEAVE_QBSS); + WLAN_REASON_QSTA_LEAVE_QBSS, tx); } } @@ -135,7 +135,7 @@ void ieee80211_ba_session_work(struct work_struct *work) if (test_and_clear_bit(tid, sta->ampdu_mlme.tid_rx_timer_expired)) ___ieee80211_stop_rx_ba_session( sta, tid, WLAN_BACK_RECIPIENT, - WLAN_REASON_QSTA_TIMEOUT); + WLAN_REASON_QSTA_TIMEOUT, true); tid_tx = sta->ampdu_mlme.tid_tx[tid]; if (!tid_tx) @@ -146,7 +146,8 @@ void ieee80211_ba_session_work(struct work_struct *work) else if (test_and_clear_bit(HT_AGG_STATE_WANT_STOP, &tid_tx->state)) ___ieee80211_stop_tx_ba_session(sta, tid, - WLAN_BACK_INITIATOR); + WLAN_BACK_INITIATOR, + true); } mutex_unlock(&sta->ampdu_mlme.mtx); } @@ -214,9 +215,11 @@ void ieee80211_process_delba(struct ieee80211_sub_if_data *sdata, #endif /* CONFIG_MAC80211_HT_DEBUG */ if (initiator == WLAN_BACK_INITIATOR) - __ieee80211_stop_rx_ba_session(sta, tid, WLAN_BACK_INITIATOR, 0); + __ieee80211_stop_rx_ba_session(sta, tid, WLAN_BACK_INITIATOR, 0, + true); else - __ieee80211_stop_tx_ba_session(sta, tid, WLAN_BACK_RECIPIENT); + __ieee80211_stop_tx_ba_session(sta, tid, WLAN_BACK_RECIPIENT, + true); } int ieee80211_send_smps_action(struct ieee80211_sub_if_data *sdata, @@ -265,3 +268,33 @@ int ieee80211_send_smps_action(struct ieee80211_sub_if_data *sdata, return 0; } + +void ieee80211_request_smps_work(struct work_struct *work) +{ + struct ieee80211_sub_if_data *sdata = + container_of(work, struct ieee80211_sub_if_data, + u.mgd.request_smps_work); + + mutex_lock(&sdata->u.mgd.mtx); + __ieee80211_request_smps(sdata, sdata->u.mgd.driver_smps_mode); + mutex_unlock(&sdata->u.mgd.mtx); +} + +void ieee80211_request_smps(struct ieee80211_vif *vif, + enum ieee80211_smps_mode smps_mode) +{ + struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif); + + if (WARN_ON(vif->type != NL80211_IFTYPE_STATION)) + return; + + if (WARN_ON(smps_mode == IEEE80211_SMPS_OFF)) + smps_mode = IEEE80211_SMPS_AUTOMATIC; + + sdata->u.mgd.driver_smps_mode = smps_mode; + + ieee80211_queue_work(&sdata->local->hw, + &sdata->u.mgd.request_smps_work); +} +/* this might change ... don't want non-open drivers using it */ +EXPORT_SYMBOL_GPL(ieee80211_request_smps); diff --git a/net/mac80211/ibss.c b/net/mac80211/ibss.c index c691780725a..ff60c022f51 100644 --- a/net/mac80211/ibss.c +++ b/net/mac80211/ibss.c @@ -173,6 +173,19 @@ static void __ieee80211_sta_join_ibss(struct ieee80211_sub_if_data *sdata, memcpy(skb_put(skb, ifibss->ie_len), ifibss->ie, ifibss->ie_len); + if (local->hw.queues >= 4) { + pos = skb_put(skb, 9); + *pos++ = WLAN_EID_VENDOR_SPECIFIC; + *pos++ = 7; /* len */ + *pos++ = 0x00; /* Microsoft OUI 00:50:F2 */ + *pos++ = 0x50; + *pos++ = 0xf2; + *pos++ = 2; /* WME */ + *pos++ = 0; /* WME info */ + *pos++ = 1; /* WME ver */ + *pos++ = 0; /* U-APSD no in use */ + } + rcu_assign_pointer(ifibss->presp, skb); sdata->vif.bss_conf.beacon_int = beacon_int; @@ -266,37 +279,45 @@ static void ieee80211_rx_bss_info(struct ieee80211_sub_if_data *sdata, if (!channel || channel->flags & IEEE80211_CHAN_DISABLED) return; - if (sdata->vif.type == NL80211_IFTYPE_ADHOC && elems->supp_rates && + if (sdata->vif.type == NL80211_IFTYPE_ADHOC && memcmp(mgmt->bssid, sdata->u.ibss.bssid, ETH_ALEN) == 0) { - supp_rates = ieee80211_sta_get_rates(local, elems, band); rcu_read_lock(); - sta = sta_info_get(sdata, mgmt->sa); - if (sta) { - u32 prev_rates; - prev_rates = sta->sta.supp_rates[band]; - /* make sure mandatory rates are always added */ - sta->sta.supp_rates[band] = supp_rates | - ieee80211_mandatory_rates(local, band); + if (elems->supp_rates) { + supp_rates = ieee80211_sta_get_rates(local, elems, + band); + if (sta) { + u32 prev_rates; + + prev_rates = sta->sta.supp_rates[band]; + /* make sure mandatory rates are always added */ + sta->sta.supp_rates[band] = supp_rates | + ieee80211_mandatory_rates(local, band); - if (sta->sta.supp_rates[band] != prev_rates) { + if (sta->sta.supp_rates[band] != prev_rates) { #ifdef CONFIG_MAC80211_IBSS_DEBUG - printk(KERN_DEBUG "%s: updated supp_rates set " - "for %pM based on beacon/probe_response " - "(0x%x -> 0x%x)\n", - sdata->name, sta->sta.addr, - prev_rates, sta->sta.supp_rates[band]); + printk(KERN_DEBUG + "%s: updated supp_rates set " + "for %pM based on beacon" + "/probe_resp (0x%x -> 0x%x)\n", + sdata->name, sta->sta.addr, + prev_rates, + sta->sta.supp_rates[band]); #endif - rate_control_rate_init(sta); - } - rcu_read_unlock(); - } else { - rcu_read_unlock(); - ieee80211_ibss_add_sta(sdata, mgmt->bssid, mgmt->sa, - supp_rates, GFP_KERNEL); + rate_control_rate_init(sta); + } + } else + sta = ieee80211_ibss_add_sta(sdata, mgmt->bssid, + mgmt->sa, supp_rates, + GFP_ATOMIC); } + + if (sta && elems->wmm_info) + set_sta_flags(sta, WLAN_STA_WME); + + rcu_read_unlock(); } bss = ieee80211_bss_info_update(local, rx_status, mgmt, len, elems, @@ -427,8 +448,8 @@ struct sta_info *ieee80211_ibss_add_sta(struct ieee80211_sub_if_data *sdata, return NULL; #ifdef CONFIG_MAC80211_VERBOSE_DEBUG - printk(KERN_DEBUG "%s: Adding new IBSS station %pM (dev=%s)\n", - wiphy_name(local->hw.wiphy), addr, sdata->name); + wiphy_debug(local->hw.wiphy, "Adding new IBSS station %pM (dev=%s)\n", + addr, sdata->name); #endif sta = sta_info_alloc(sdata, addr, gfp); @@ -920,12 +941,14 @@ int ieee80211_ibss_join(struct ieee80211_sub_if_data *sdata, memcpy(sdata->u.ibss.ssid, params->ssid, IEEE80211_MAX_SSID_LEN); sdata->u.ibss.ssid_len = params->ssid_len; + mutex_unlock(&sdata->u.ibss.mtx); + + mutex_lock(&sdata->local->mtx); ieee80211_recalc_idle(sdata->local); + mutex_unlock(&sdata->local->mtx); ieee80211_queue_work(&sdata->local->hw, &sdata->work); - mutex_unlock(&sdata->u.ibss.mtx); - return 0; } @@ -980,7 +1003,9 @@ int ieee80211_ibss_leave(struct ieee80211_sub_if_data *sdata) mutex_unlock(&sdata->u.ibss.mtx); + mutex_lock(&local->mtx); ieee80211_recalc_idle(sdata->local); + mutex_unlock(&local->mtx); return 0; } diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 65e0ed6c297..b80c3868992 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -50,12 +50,6 @@ struct ieee80211_local; * increased memory use (about 2 kB of RAM per entry). */ #define IEEE80211_FRAGMENT_MAX 4 -/* - * Time after which we ignore scan results and no longer report/use - * them in any way. - */ -#define IEEE80211_SCAN_RESULT_EXPIRE (10 * HZ) - #define TU_TO_EXP_TIME(x) (jiffies + usecs_to_jiffies((x) * 1024)) #define IEEE80211_DEFAULT_UAPSD_QUEUES \ @@ -165,12 +159,37 @@ typedef unsigned __bitwise__ ieee80211_rx_result; #define RX_DROP_MONITOR ((__force ieee80211_rx_result) 2u) #define RX_QUEUED ((__force ieee80211_rx_result) 3u) -#define IEEE80211_RX_IN_SCAN BIT(0) -/* frame is destined to interface currently processed (incl. multicast frames) */ -#define IEEE80211_RX_RA_MATCH BIT(1) -#define IEEE80211_RX_AMSDU BIT(2) -#define IEEE80211_RX_FRAGMENTED BIT(3) -/* only add flags here that do not change with subframes of an aMPDU */ +/** + * enum ieee80211_packet_rx_flags - packet RX flags + * @IEEE80211_RX_RA_MATCH: frame is destined to interface currently processed + * (incl. multicast frames) + * @IEEE80211_RX_IN_SCAN: received while scanning + * @IEEE80211_RX_FRAGMENTED: fragmented frame + * @IEEE80211_RX_AMSDU: a-MSDU packet + * @IEEE80211_RX_MALFORMED_ACTION_FRM: action frame is malformed + * + * These are per-frame flags that are attached to a frame in the + * @rx_flags field of &struct ieee80211_rx_status. + */ +enum ieee80211_packet_rx_flags { + IEEE80211_RX_IN_SCAN = BIT(0), + IEEE80211_RX_RA_MATCH = BIT(1), + IEEE80211_RX_FRAGMENTED = BIT(2), + IEEE80211_RX_AMSDU = BIT(3), + IEEE80211_RX_MALFORMED_ACTION_FRM = BIT(4), +}; + +/** + * enum ieee80211_rx_flags - RX data flags + * + * @IEEE80211_RX_CMNTR: received on cooked monitor already + * + * These flags are used across handling multiple interfaces + * for a single frame. + */ +enum ieee80211_rx_flags { + IEEE80211_RX_CMNTR = BIT(0), +}; struct ieee80211_rx_data { struct sk_buff *skb; @@ -343,10 +362,14 @@ struct ieee80211_if_managed { unsigned long timers_running; /* used for quiesce/restart */ bool powersave; /* powersave requested for this iface */ enum ieee80211_smps_mode req_smps, /* requested smps mode */ - ap_smps; /* smps mode AP thinks we're in */ + ap_smps, /* smps mode AP thinks we're in */ + driver_smps_mode; /* smps mode request */ + + struct work_struct request_smps_work; unsigned int flags; + bool beacon_crc_valid; u32 beacon_crc; enum { @@ -371,6 +394,13 @@ struct ieee80211_if_managed { int ave_beacon_signal; /* + * Number of Beacon frames used in ave_beacon_signal. This can be used + * to avoid generating less reliable cqm events that would be based + * only on couple of received frames. + */ + unsigned int count_beacon_signal; + + /* * Last Beacon frame signal strength average (ave_beacon_signal / 16) * that triggered a cqm event. 0 indicates that no event has been * generated for the current association. @@ -474,6 +504,19 @@ enum ieee80211_sub_if_data_flags { IEEE80211_SDATA_DONT_BRIDGE_PACKETS = BIT(3), }; +/** + * enum ieee80211_sdata_state_bits - virtual interface state bits + * @SDATA_STATE_RUNNING: virtual interface is up & running; this + * mirrors netif_running() but is separate for interface type + * change handling while the interface is up + * @SDATA_STATE_OFFCHANNEL: This interface is currently in offchannel + * mode, so queues are stopped + */ +enum ieee80211_sdata_state_bits { + SDATA_STATE_RUNNING, + SDATA_STATE_OFFCHANNEL, +}; + struct ieee80211_sub_if_data { struct list_head list; @@ -487,6 +530,8 @@ struct ieee80211_sub_if_data { unsigned int flags; + unsigned long state; + int drop_unencrypted; char name[IFNAMSIZ]; @@ -497,17 +542,20 @@ struct ieee80211_sub_if_data { */ bool ht_opmode_valid; + /* to detect idle changes */ + bool old_idle; + /* Fragment table for host-based reassembly */ struct ieee80211_fragment_entry fragments[IEEE80211_FRAGMENT_MAX]; unsigned int fragment_next; -#define NUM_DEFAULT_KEYS 4 -#define NUM_DEFAULT_MGMT_KEYS 2 struct ieee80211_key *keys[NUM_DEFAULT_KEYS + NUM_DEFAULT_MGMT_KEYS]; struct ieee80211_key *default_key; struct ieee80211_key *default_mgmt_key; u16 sequence_number; + __be16 control_port_protocol; + bool control_port_no_encrypt; struct work_struct work; struct sk_buff_head skb_queue; @@ -539,6 +587,7 @@ struct ieee80211_sub_if_data { #ifdef CONFIG_MAC80211_DEBUGFS struct { struct dentry *dir; + struct dentry *subdir_stations; struct dentry *default_key; struct dentry *default_mgmt_key; } debugfs; @@ -595,11 +644,17 @@ enum queue_stop_reason { * determine if we are on the operating channel or not * @SCAN_OFF_CHANNEL: We're off our operating channel for scanning, * gets only set in conjunction with SCAN_SW_SCANNING + * @SCAN_COMPLETED: Set for our scan work function when the driver reported + * that the scan completed. + * @SCAN_ABORTED: Set for our scan work function when the driver reported + * a scan complete for an aborted scan. */ enum { SCAN_SW_SCANNING, SCAN_HW_SCANNING, SCAN_OFF_CHANNEL, + SCAN_COMPLETED, + SCAN_ABORTED, }; /** @@ -634,7 +689,6 @@ struct ieee80211_local { /* * work stuff, potentially off-channel (in the future) */ - struct mutex work_mtx; struct list_head work_list; struct timer_list work_timer; struct work_struct work_work; @@ -653,9 +707,13 @@ struct ieee80211_local { int open_count; int monitors, cooked_mntrs; /* number of interfaces with corresponding FIF_ flags */ - int fif_fcsfail, fif_plcpfail, fif_control, fif_other_bss, fif_pspoll; + int fif_fcsfail, fif_plcpfail, fif_control, fif_other_bss, fif_pspoll, + fif_probe_req; + int probe_req_reg; unsigned int filter_flags; /* FIF_* */ + bool wiphy_ciphers_allocated; + /* protects the aggregated multicast list and filter calls */ spinlock_t filter_lock; @@ -746,9 +804,10 @@ struct ieee80211_local { */ struct mutex key_mtx; + /* mutex for scan and work locking */ + struct mutex mtx; /* Scanning and BSS list */ - struct mutex scan_mtx; unsigned long scanning; struct cfg80211_ssid scan_ssid; struct cfg80211_scan_request *int_scan_req; @@ -866,10 +925,14 @@ struct ieee80211_local { #ifdef CONFIG_MAC80211_DEBUGFS struct local_debugfsdentries { struct dentry *rcdir; - struct dentry *stations; struct dentry *keys; } debugfs; #endif + + /* dummy netdev for use w/ NAPI */ + struct net_device napi_dev; + + struct napi_struct napi; }; static inline struct ieee80211_sub_if_data * @@ -1003,6 +1066,8 @@ void ieee80211_sta_restart(struct ieee80211_sub_if_data *sdata); void ieee80211_sta_work(struct ieee80211_sub_if_data *sdata); void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb); +void ieee80211_sta_reset_beacon_monitor(struct ieee80211_sub_if_data *sdata); +void ieee80211_sta_reset_conn_monitor(struct ieee80211_sub_if_data *sdata); /* IBSS code */ void ieee80211_ibss_notify_scan_completed(struct ieee80211_local *local); @@ -1068,10 +1133,12 @@ void ieee80211_if_remove(struct ieee80211_sub_if_data *sdata); void ieee80211_remove_interfaces(struct ieee80211_local *local); u32 __ieee80211_recalc_idle(struct ieee80211_local *local); void ieee80211_recalc_idle(struct ieee80211_local *local); +void ieee80211_adjust_monitor_flags(struct ieee80211_sub_if_data *sdata, + const int offset); static inline bool ieee80211_sdata_running(struct ieee80211_sub_if_data *sdata) { - return netif_running(sdata->dev); + return test_bit(SDATA_STATE_RUNNING, &sdata->state); } /* tx handling */ @@ -1105,12 +1172,13 @@ void ieee80211_send_delba(struct ieee80211_sub_if_data *sdata, int ieee80211_send_smps_action(struct ieee80211_sub_if_data *sdata, enum ieee80211_smps_mode smps, const u8 *da, const u8 *bssid); +void ieee80211_request_smps_work(struct work_struct *work); void ___ieee80211_stop_rx_ba_session(struct sta_info *sta, u16 tid, - u16 initiator, u16 reason); + u16 initiator, u16 reason, bool stop); void __ieee80211_stop_rx_ba_session(struct sta_info *sta, u16 tid, - u16 initiator, u16 reason); -void ieee80211_sta_tear_down_BA_sessions(struct sta_info *sta); + u16 initiator, u16 reason, bool stop); +void ieee80211_sta_tear_down_BA_sessions(struct sta_info *sta, bool tx); void ieee80211_process_delba(struct ieee80211_sub_if_data *sdata, struct sta_info *sta, struct ieee80211_mgmt *mgmt, size_t len); @@ -1124,13 +1192,16 @@ void ieee80211_process_addba_request(struct ieee80211_local *local, size_t len); int __ieee80211_stop_tx_ba_session(struct sta_info *sta, u16 tid, - enum ieee80211_back_parties initiator); + enum ieee80211_back_parties initiator, + bool tx); int ___ieee80211_stop_tx_ba_session(struct sta_info *sta, u16 tid, - enum ieee80211_back_parties initiator); + enum ieee80211_back_parties initiator, + bool tx); void ieee80211_start_tx_ba_cb(struct ieee80211_vif *vif, u8 *ra, u16 tid); void ieee80211_stop_tx_ba_cb(struct ieee80211_vif *vif, u8 *ra, u8 tid); void ieee80211_ba_session_work(struct work_struct *work); void ieee80211_tx_ba_session_handle_start(struct sta_info *sta, int tid); +void ieee80211_release_reorder_timeout(struct sta_info *sta, int tid); /* Spectrum management */ void ieee80211_process_measurement_req(struct ieee80211_sub_if_data *sdata, @@ -1146,6 +1217,12 @@ int __ieee80211_suspend(struct ieee80211_hw *hw); static inline int __ieee80211_resume(struct ieee80211_hw *hw) { + struct ieee80211_local *local = hw_to_local(hw); + + WARN(test_bit(SCAN_HW_SCANNING, &local->scanning), + "%s: resume with hardware scan still in progress\n", + wiphy_name(hw->wiphy)); + return ieee80211_reconfig(hw_to_local(hw)); } #else @@ -1208,7 +1285,8 @@ void ieee80211_send_auth(struct ieee80211_sub_if_data *sdata, const u8 *key, u8 key_len, u8 key_idx); int ieee80211_build_preq_ies(struct ieee80211_local *local, u8 *buffer, const u8 *ie, size_t ie_len, - enum ieee80211_band band); + enum ieee80211_band band, u32 rate_mask, + u8 channel); void ieee80211_send_probe_req(struct ieee80211_sub_if_data *sdata, u8 *dst, const u8 *ssid, size_t ssid_len, const u8 *ie, size_t ie_len); @@ -1221,8 +1299,7 @@ u32 ieee80211_sta_get_rates(struct ieee80211_local *local, enum ieee80211_band band); int __ieee80211_request_smps(struct ieee80211_sub_if_data *sdata, enum ieee80211_smps_mode smps_mode); -void ieee80211_recalc_smps(struct ieee80211_local *local, - struct ieee80211_sub_if_data *forsdata); +void ieee80211_recalc_smps(struct ieee80211_local *local); size_t ieee80211_ie_split(const u8 *ies, size_t ielen, const u8 *ids, int n_ids, size_t offset); diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index ebbe264e2b0..f9163b12c7f 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -24,6 +24,7 @@ #include "led.h" #include "driver-ops.h" #include "wme.h" +#include "rate.h" /** * DOC: Interface list locking @@ -94,21 +95,14 @@ static inline int identical_mac_addr_allowed(int type1, int type2) type2 == NL80211_IFTYPE_AP_VLAN)); } -static int ieee80211_open(struct net_device *dev) +static int ieee80211_check_concurrent_iface(struct ieee80211_sub_if_data *sdata, + enum nl80211_iftype iftype) { - struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); - struct ieee80211_sub_if_data *nsdata; struct ieee80211_local *local = sdata->local; - struct sta_info *sta; - u32 changed = 0; - int res; - u32 hw_reconf_flags = 0; - u8 null_addr[ETH_ALEN] = {0}; + struct ieee80211_sub_if_data *nsdata; + struct net_device *dev = sdata->dev; - /* fail early if user set an invalid address */ - if (compare_ether_addr(dev->dev_addr, null_addr) && - !is_valid_ether_addr(dev->dev_addr)) - return -EADDRNOTAVAIL; + ASSERT_RTNL(); /* we hold the RTNL here so can safely walk the list */ list_for_each_entry(nsdata, &local->interfaces, list) { @@ -125,7 +119,7 @@ static int ieee80211_open(struct net_device *dev) * belonging to the same hardware. Then, however, we're * faced with having to adopt two different TSF timers... */ - if (sdata->vif.type == NL80211_IFTYPE_ADHOC && + if (iftype == NL80211_IFTYPE_ADHOC && nsdata->vif.type == NL80211_IFTYPE_ADHOC) return -EBUSY; @@ -139,19 +133,56 @@ static int ieee80211_open(struct net_device *dev) /* * check whether it may have the same address */ - if (!identical_mac_addr_allowed(sdata->vif.type, + if (!identical_mac_addr_allowed(iftype, nsdata->vif.type)) return -ENOTUNIQ; /* * can only add VLANs to enabled APs */ - if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN && + if (iftype == NL80211_IFTYPE_AP_VLAN && nsdata->vif.type == NL80211_IFTYPE_AP) sdata->bss = &nsdata->u.ap; } } + return 0; +} + +void ieee80211_adjust_monitor_flags(struct ieee80211_sub_if_data *sdata, + const int offset) +{ + struct ieee80211_local *local = sdata->local; + u32 flags = sdata->u.mntr_flags; + +#define ADJUST(_f, _s) do { \ + if (flags & MONITOR_FLAG_##_f) \ + local->fif_##_s += offset; \ + } while (0) + + ADJUST(FCSFAIL, fcsfail); + ADJUST(PLCPFAIL, plcpfail); + ADJUST(CONTROL, control); + ADJUST(CONTROL, pspoll); + ADJUST(OTHER_BSS, other_bss); + +#undef ADJUST +} + +/* + * NOTE: Be very careful when changing this function, it must NOT return + * an error on interface type changes that have been pre-checked, so most + * checks should be in ieee80211_check_concurrent_iface. + */ +static int ieee80211_do_open(struct net_device *dev, bool coming_up) +{ + struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); + struct ieee80211_local *local = sdata->local; + struct sta_info *sta; + u32 changed = 0; + int res; + u32 hw_reconf_flags = 0; + switch (sdata->vif.type) { case NL80211_IFTYPE_WDS: if (!is_valid_ether_addr(sdata->u.wds.remote_addr)) @@ -177,7 +208,9 @@ static int ieee80211_open(struct net_device *dev) /* no special treatment */ break; case NL80211_IFTYPE_UNSPECIFIED: - case __NL80211_IFTYPE_AFTER_LAST: + case NUM_NL80211_IFTYPES: + case NL80211_IFTYPE_P2P_CLIENT: + case NL80211_IFTYPE_P2P_GO: /* cannot happen */ WARN_ON(1); break; @@ -187,39 +220,30 @@ static int ieee80211_open(struct net_device *dev) res = drv_start(local); if (res) goto err_del_bss; + if (local->ops->napi_poll) + napi_enable(&local->napi); /* we're brought up, everything changes */ hw_reconf_flags = ~0; ieee80211_led_radio(local, true); } /* - * Check all interfaces and copy the hopefully now-present - * MAC address to those that have the special null one. + * Copy the hopefully now-present MAC address to + * this interface, if it has the special null one. */ - list_for_each_entry(nsdata, &local->interfaces, list) { - struct net_device *ndev = nsdata->dev; - - /* - * No need to check running since we do not allow - * it to start up with this invalid address. - */ - if (compare_ether_addr(null_addr, ndev->dev_addr) == 0) { - memcpy(ndev->dev_addr, - local->hw.wiphy->perm_addr, - ETH_ALEN); - memcpy(ndev->perm_addr, ndev->dev_addr, ETH_ALEN); + if (is_zero_ether_addr(dev->dev_addr)) { + memcpy(dev->dev_addr, + local->hw.wiphy->perm_addr, + ETH_ALEN); + memcpy(dev->perm_addr, dev->dev_addr, ETH_ALEN); + + if (!is_valid_ether_addr(dev->dev_addr)) { + if (!local->open_count) + drv_stop(local); + return -EADDRNOTAVAIL; } } - /* - * Validate the MAC address for this device. - */ - if (!is_valid_ether_addr(dev->dev_addr)) { - if (!local->open_count) - drv_stop(local); - return -EADDRNOTAVAIL; - } - switch (sdata->vif.type) { case NL80211_IFTYPE_AP_VLAN: /* no need to tell driver */ @@ -237,25 +261,17 @@ static int ieee80211_open(struct net_device *dev) hw_reconf_flags |= IEEE80211_CONF_CHANGE_MONITOR; } - if (sdata->u.mntr_flags & MONITOR_FLAG_FCSFAIL) - local->fif_fcsfail++; - if (sdata->u.mntr_flags & MONITOR_FLAG_PLCPFAIL) - local->fif_plcpfail++; - if (sdata->u.mntr_flags & MONITOR_FLAG_CONTROL) { - local->fif_control++; - local->fif_pspoll++; - } - if (sdata->u.mntr_flags & MONITOR_FLAG_OTHER_BSS) - local->fif_other_bss++; - + ieee80211_adjust_monitor_flags(sdata, 1); ieee80211_configure_filter(local); netif_carrier_on(dev); break; default: - res = drv_add_interface(local, &sdata->vif); - if (res) - goto err_stop; + if (coming_up) { + res = drv_add_interface(local, &sdata->vif); + if (res) + goto err_stop; + } if (ieee80211_vif_is_mesh(&sdata->vif)) { local->fif_other_bss++; @@ -264,8 +280,11 @@ static int ieee80211_open(struct net_device *dev) ieee80211_start_mesh(sdata); } else if (sdata->vif.type == NL80211_IFTYPE_AP) { local->fif_pspoll++; + local->fif_probe_req++; ieee80211_configure_filter(local); + } else if (sdata->vif.type == NL80211_IFTYPE_ADHOC) { + local->fif_probe_req++; } changed |= ieee80211_reset_erp_info(sdata); @@ -277,6 +296,8 @@ static int ieee80211_open(struct net_device *dev) netif_carrier_on(dev); } + set_bit(SDATA_STATE_RUNNING, &sdata->state); + if (sdata->vif.type == NL80211_IFTYPE_WDS) { /* Create STA entry for the WDS peer */ sta = sta_info_alloc(sdata, sdata->u.wds.remote_addr, @@ -294,6 +315,8 @@ static int ieee80211_open(struct net_device *dev) /* STA has been freed */ goto err_del_interface; } + + rate_control_rate_init(sta); } /* @@ -307,9 +330,13 @@ static int ieee80211_open(struct net_device *dev) if (sdata->flags & IEEE80211_SDATA_PROMISC) atomic_inc(&local->iff_promiscs); + mutex_lock(&local->mtx); hw_reconf_flags |= __ieee80211_recalc_idle(local); + mutex_unlock(&local->mtx); + + if (coming_up) + local->open_count++; - local->open_count++; if (hw_reconf_flags) { ieee80211_hw_config(local, hw_reconf_flags); /* @@ -334,22 +361,42 @@ static int ieee80211_open(struct net_device *dev) sdata->bss = NULL; if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN) list_del(&sdata->u.vlan.list); + clear_bit(SDATA_STATE_RUNNING, &sdata->state); return res; } -static int ieee80211_stop(struct net_device *dev) +static int ieee80211_open(struct net_device *dev) { struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); + int err; + + /* fail early if user set an invalid address */ + if (!is_zero_ether_addr(dev->dev_addr) && + !is_valid_ether_addr(dev->dev_addr)) + return -EADDRNOTAVAIL; + + err = ieee80211_check_concurrent_iface(sdata, sdata->vif.type); + if (err) + return err; + + return ieee80211_do_open(dev, true); +} + +static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, + bool going_down) +{ struct ieee80211_local *local = sdata->local; unsigned long flags; struct sk_buff *skb, *tmp; u32 hw_reconf_flags = 0; int i; + clear_bit(SDATA_STATE_RUNNING, &sdata->state); + /* * Stop TX on this interface first. */ - netif_tx_stop_all_queues(dev); + netif_tx_stop_all_queues(sdata->dev); /* * Purge work for this interface. @@ -366,12 +413,9 @@ static int ieee80211_stop(struct net_device *dev) * (because if we remove a STA after ops->remove_interface() * the driver will have removed the vif info already!) * - * We could relax this and only unlink the stations from the - * hash table and list but keep them on a per-sdata list that - * will be inserted back again when the interface is brought - * up again, but I don't currently see a use case for that, - * except with WDS which gets a STA entry created when it is - * brought up. + * This is relevant only in AP, WDS and mesh modes, since in + * all other modes we've already removed all stations when + * disconnecting etc. */ sta_info_flush(local, sdata); @@ -387,14 +431,19 @@ static int ieee80211_stop(struct net_device *dev) if (sdata->flags & IEEE80211_SDATA_PROMISC) atomic_dec(&local->iff_promiscs); - if (sdata->vif.type == NL80211_IFTYPE_AP) + if (sdata->vif.type == NL80211_IFTYPE_AP) { local->fif_pspoll--; + local->fif_probe_req--; + } else if (sdata->vif.type == NL80211_IFTYPE_ADHOC) { + local->fif_probe_req--; + } - netif_addr_lock_bh(dev); + netif_addr_lock_bh(sdata->dev); spin_lock_bh(&local->filter_lock); - __hw_addr_unsync(&local->mc_list, &dev->mc, dev->addr_len); + __hw_addr_unsync(&local->mc_list, &sdata->dev->mc, + sdata->dev->addr_len); spin_unlock_bh(&local->filter_lock); - netif_addr_unlock_bh(dev); + netif_addr_unlock_bh(sdata->dev); ieee80211_configure_filter(local); @@ -406,11 +455,21 @@ static int ieee80211_stop(struct net_device *dev) struct ieee80211_sub_if_data *vlan, *tmpsdata; struct beacon_data *old_beacon = sdata->u.ap.beacon; + /* sdata_running will return false, so this will disable */ + ieee80211_bss_info_change_notify(sdata, + BSS_CHANGED_BEACON_ENABLED); + /* remove beacon */ rcu_assign_pointer(sdata->u.ap.beacon, NULL); synchronize_rcu(); kfree(old_beacon); + /* free all potentially still buffered bcast frames */ + while ((skb = skb_dequeue(&sdata->u.ap.ps_bc_buf))) { + local->total_ps_buffered--; + dev_kfree_skb(skb); + } + /* down all dependent devices, that is VLANs */ list_for_each_entry_safe(vlan, tmpsdata, &sdata->u.ap.vlans, u.vlan.list) @@ -418,7 +477,8 @@ static int ieee80211_stop(struct net_device *dev) WARN_ON(!list_empty(&sdata->u.ap.vlans)); } - local->open_count--; + if (going_down) + local->open_count--; switch (sdata->vif.type) { case NL80211_IFTYPE_AP_VLAN: @@ -437,40 +497,9 @@ static int ieee80211_stop(struct net_device *dev) hw_reconf_flags |= IEEE80211_CONF_CHANGE_MONITOR; } - if (sdata->u.mntr_flags & MONITOR_FLAG_FCSFAIL) - local->fif_fcsfail--; - if (sdata->u.mntr_flags & MONITOR_FLAG_PLCPFAIL) - local->fif_plcpfail--; - if (sdata->u.mntr_flags & MONITOR_FLAG_CONTROL) { - local->fif_pspoll--; - local->fif_control--; - } - if (sdata->u.mntr_flags & MONITOR_FLAG_OTHER_BSS) - local->fif_other_bss--; - + ieee80211_adjust_monitor_flags(sdata, -1); ieee80211_configure_filter(local); break; - case NL80211_IFTYPE_STATION: - del_timer_sync(&sdata->u.mgd.chswitch_timer); - del_timer_sync(&sdata->u.mgd.timer); - del_timer_sync(&sdata->u.mgd.conn_mon_timer); - del_timer_sync(&sdata->u.mgd.bcn_mon_timer); - /* - * If any of the timers fired while we waited for it, it will - * have queued its work. Now the work will be running again - * but will not rearm the timer again because it checks - * whether the interface is running, which, at this point, - * it no longer is. - */ - cancel_work_sync(&sdata->u.mgd.chswitch_work); - cancel_work_sync(&sdata->u.mgd.monitor_work); - cancel_work_sync(&sdata->u.mgd.beacon_connection_loss_work); - - /* fall through */ - case NL80211_IFTYPE_ADHOC: - if (sdata->vif.type == NL80211_IFTYPE_ADHOC) - del_timer_sync(&sdata->u.ibss.timer); - /* fall through */ case NL80211_IFTYPE_MESH_POINT: if (ieee80211_vif_is_mesh(&sdata->vif)) { /* other_bss and allmulti are always set on mesh @@ -498,27 +527,34 @@ static int ieee80211_stop(struct net_device *dev) ieee80211_scan_cancel(local); /* - * Disable beaconing for AP and mesh, IBSS can't - * still be joined to a network at this point. + * Disable beaconing here for mesh only, AP and IBSS + * are already taken care of. */ - if (sdata->vif.type == NL80211_IFTYPE_AP || - sdata->vif.type == NL80211_IFTYPE_MESH_POINT) { + if (sdata->vif.type == NL80211_IFTYPE_MESH_POINT) ieee80211_bss_info_change_notify(sdata, BSS_CHANGED_BEACON_ENABLED); - } - /* free all remaining keys, there shouldn't be any */ + /* + * Free all remaining keys, there shouldn't be any, + * except maybe group keys in AP more or WDS? + */ ieee80211_free_keys(sdata); - drv_remove_interface(local, &sdata->vif); + + if (going_down) + drv_remove_interface(local, &sdata->vif); } sdata->bss = NULL; + mutex_lock(&local->mtx); hw_reconf_flags |= __ieee80211_recalc_idle(local); + mutex_unlock(&local->mtx); ieee80211_recalc_ps(local, -1); if (local->open_count == 0) { + if (local->ops->napi_poll) + napi_disable(&local->napi); ieee80211_clear_tx_pending(local); ieee80211_stop_device(local); @@ -541,6 +577,13 @@ static int ieee80211_stop(struct net_device *dev) } } spin_unlock_irqrestore(&local->queue_stop_reason_lock, flags); +} + +static int ieee80211_stop(struct net_device *dev) +{ + struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); + + ieee80211_do_stop(sdata, true); return 0; } @@ -585,8 +628,6 @@ static void ieee80211_teardown_sdata(struct net_device *dev) { struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); struct ieee80211_local *local = sdata->local; - struct beacon_data *beacon; - struct sk_buff *skb; int flushed; int i; @@ -599,37 +640,8 @@ static void ieee80211_teardown_sdata(struct net_device *dev) __skb_queue_purge(&sdata->fragments[i].skb_list); sdata->fragment_next = 0; - switch (sdata->vif.type) { - case NL80211_IFTYPE_AP: - beacon = sdata->u.ap.beacon; - rcu_assign_pointer(sdata->u.ap.beacon, NULL); - synchronize_rcu(); - kfree(beacon); - - while ((skb = skb_dequeue(&sdata->u.ap.ps_bc_buf))) { - local->total_ps_buffered--; - dev_kfree_skb(skb); - } - - break; - case NL80211_IFTYPE_MESH_POINT: - if (ieee80211_vif_is_mesh(&sdata->vif)) - mesh_rmc_free(sdata); - break; - case NL80211_IFTYPE_ADHOC: - if (WARN_ON(sdata->u.ibss.presp)) - kfree_skb(sdata->u.ibss.presp); - break; - case NL80211_IFTYPE_STATION: - case NL80211_IFTYPE_WDS: - case NL80211_IFTYPE_AP_VLAN: - case NL80211_IFTYPE_MONITOR: - break; - case NL80211_IFTYPE_UNSPECIFIED: - case __NL80211_IFTYPE_AFTER_LAST: - BUG(); - break; - } + if (ieee80211_vif_is_mesh(&sdata->vif)) + mesh_rmc_free(sdata); flushed = sta_info_flush(local, sdata); WARN_ON(flushed); @@ -791,7 +803,8 @@ static void ieee80211_iface_work(struct work_struct *work) __ieee80211_stop_rx_ba_session( sta, tid, WLAN_BACK_RECIPIENT, - WLAN_REASON_QSTA_REQUIRE_SETUP); + WLAN_REASON_QSTA_REQUIRE_SETUP, + true); } mutex_unlock(&local->sta_mtx); } else switch (sdata->vif.type) { @@ -844,9 +857,13 @@ static void ieee80211_setup_sdata(struct ieee80211_sub_if_data *sdata, /* and set some type-dependent values */ sdata->vif.type = type; + sdata->vif.p2p = false; sdata->dev->netdev_ops = &ieee80211_dataif_ops; sdata->wdev.iftype = type; + sdata->control_port_protocol = cpu_to_be16(ETH_P_PAE); + sdata->control_port_no_encrypt = false; + /* only monitor differs */ sdata->dev->type = ARPHRD_ETHER; @@ -854,10 +871,20 @@ static void ieee80211_setup_sdata(struct ieee80211_sub_if_data *sdata, INIT_WORK(&sdata->work, ieee80211_iface_work); switch (type) { + case NL80211_IFTYPE_P2P_GO: + type = NL80211_IFTYPE_AP; + sdata->vif.type = type; + sdata->vif.p2p = true; + /* fall through */ case NL80211_IFTYPE_AP: skb_queue_head_init(&sdata->u.ap.ps_bc_buf); INIT_LIST_HEAD(&sdata->u.ap.vlans); break; + case NL80211_IFTYPE_P2P_CLIENT: + type = NL80211_IFTYPE_STATION; + sdata->vif.type = type; + sdata->vif.p2p = true; + /* fall through */ case NL80211_IFTYPE_STATION: ieee80211_sta_setup_sdata(sdata); break; @@ -878,7 +905,7 @@ static void ieee80211_setup_sdata(struct ieee80211_sub_if_data *sdata, case NL80211_IFTYPE_AP_VLAN: break; case NL80211_IFTYPE_UNSPECIFIED: - case __NL80211_IFTYPE_AFTER_LAST: + case NUM_NL80211_IFTYPES: BUG(); break; } @@ -886,12 +913,85 @@ static void ieee80211_setup_sdata(struct ieee80211_sub_if_data *sdata, ieee80211_debugfs_add_netdev(sdata); } +static int ieee80211_runtime_change_iftype(struct ieee80211_sub_if_data *sdata, + enum nl80211_iftype type) +{ + struct ieee80211_local *local = sdata->local; + int ret, err; + enum nl80211_iftype internal_type = type; + bool p2p = false; + + ASSERT_RTNL(); + + if (!local->ops->change_interface) + return -EBUSY; + + switch (sdata->vif.type) { + case NL80211_IFTYPE_AP: + case NL80211_IFTYPE_STATION: + case NL80211_IFTYPE_ADHOC: + /* + * Could maybe also all others here? + * Just not sure how that interacts + * with the RX/config path e.g. for + * mesh. + */ + break; + default: + return -EBUSY; + } + + switch (type) { + case NL80211_IFTYPE_AP: + case NL80211_IFTYPE_STATION: + case NL80211_IFTYPE_ADHOC: + /* + * Could probably support everything + * but WDS here (WDS do_open can fail + * under memory pressure, which this + * code isn't prepared to handle). + */ + break; + case NL80211_IFTYPE_P2P_CLIENT: + p2p = true; + internal_type = NL80211_IFTYPE_STATION; + break; + case NL80211_IFTYPE_P2P_GO: + p2p = true; + internal_type = NL80211_IFTYPE_AP; + break; + default: + return -EBUSY; + } + + ret = ieee80211_check_concurrent_iface(sdata, internal_type); + if (ret) + return ret; + + ieee80211_do_stop(sdata, false); + + ieee80211_teardown_sdata(sdata->dev); + + ret = drv_change_interface(local, sdata, internal_type, p2p); + if (ret) + type = sdata->vif.type; + + ieee80211_setup_sdata(sdata, type); + + err = ieee80211_do_open(sdata->dev, false); + WARN(err, "type change: do_open returned %d", err); + + return ret; +} + int ieee80211_if_change_type(struct ieee80211_sub_if_data *sdata, enum nl80211_iftype type) { + int ret; + ASSERT_RTNL(); - if (type == sdata->vif.type) + if (type == ieee80211_vif_type_p2p(&sdata->vif)) return 0; /* Setting ad-hoc mode on non-IBSS channel is not supported. */ @@ -899,18 +999,15 @@ int ieee80211_if_change_type(struct ieee80211_sub_if_data *sdata, type == NL80211_IFTYPE_ADHOC) return -EOPNOTSUPP; - /* - * We could, here, on changes between IBSS/STA/MESH modes, - * invoke an MLME function instead that disassociates etc. - * and goes into the requested mode. - */ - - if (ieee80211_sdata_running(sdata)) - return -EBUSY; - - /* Purge and reset type-dependent state. */ - ieee80211_teardown_sdata(sdata->dev); - ieee80211_setup_sdata(sdata, type); + if (ieee80211_sdata_running(sdata)) { + ret = ieee80211_runtime_change_iftype(sdata, type); + if (ret) + return ret; + } else { + /* Purge and reset type-dependent state. */ + ieee80211_teardown_sdata(sdata->dev); + ieee80211_setup_sdata(sdata, type); + } /* reset some values that shouldn't be kept across type changes */ sdata->vif.bss_conf.basic_rates = @@ -1167,8 +1264,7 @@ static u32 ieee80211_idle_off(struct ieee80211_local *local, return 0; #ifdef CONFIG_MAC80211_VERBOSE_DEBUG - printk(KERN_DEBUG "%s: device no longer idle - %s\n", - wiphy_name(local->hw.wiphy), reason); + wiphy_debug(local->hw.wiphy, "device no longer idle - %s\n", reason); #endif local->hw.conf.flags &= ~IEEE80211_CONF_IDLE; @@ -1181,8 +1277,7 @@ static u32 ieee80211_idle_on(struct ieee80211_local *local) return 0; #ifdef CONFIG_MAC80211_VERBOSE_DEBUG - printk(KERN_DEBUG "%s: device now idle\n", - wiphy_name(local->hw.wiphy)); + wiphy_debug(local->hw.wiphy, "device now idle\n"); #endif drv_flush(local, false); @@ -1195,28 +1290,61 @@ u32 __ieee80211_recalc_idle(struct ieee80211_local *local) { struct ieee80211_sub_if_data *sdata; int count = 0; + bool working = false, scanning = false; + struct ieee80211_work *wk; - if (!list_empty(&local->work_list)) - return ieee80211_idle_off(local, "working"); - - if (local->scanning) - return ieee80211_idle_off(local, "scanning"); +#ifdef CONFIG_PROVE_LOCKING + WARN_ON(debug_locks && !lockdep_rtnl_is_held() && + !lockdep_is_held(&local->iflist_mtx)); +#endif + lockdep_assert_held(&local->mtx); list_for_each_entry(sdata, &local->interfaces, list) { - if (!ieee80211_sdata_running(sdata)) + if (!ieee80211_sdata_running(sdata)) { + sdata->vif.bss_conf.idle = true; continue; + } + + sdata->old_idle = sdata->vif.bss_conf.idle; + /* do not count disabled managed interfaces */ if (sdata->vif.type == NL80211_IFTYPE_STATION && - !sdata->u.mgd.associated) + !sdata->u.mgd.associated) { + sdata->vif.bss_conf.idle = true; continue; + } /* do not count unused IBSS interfaces */ if (sdata->vif.type == NL80211_IFTYPE_ADHOC && - !sdata->u.ibss.ssid_len) + !sdata->u.ibss.ssid_len) { + sdata->vif.bss_conf.idle = true; continue; + } /* count everything else */ count++; } + list_for_each_entry(wk, &local->work_list, list) { + working = true; + wk->sdata->vif.bss_conf.idle = false; + } + + if (local->scan_sdata) { + scanning = true; + local->scan_sdata->vif.bss_conf.idle = false; + } + + list_for_each_entry(sdata, &local->interfaces, list) { + if (sdata->old_idle == sdata->vif.bss_conf.idle) + continue; + if (!ieee80211_sdata_running(sdata)) + continue; + ieee80211_bss_info_change_notify(sdata, BSS_CHANGED_IDLE); + } + + if (working) + return ieee80211_idle_off(local, "working"); + if (scanning) + return ieee80211_idle_off(local, "scanning"); if (!count) return ieee80211_idle_on(local); else diff --git a/net/mac80211/key.c b/net/mac80211/key.c index 1b9d87ed143..ccd676b2f59 100644 --- a/net/mac80211/key.c +++ b/net/mac80211/key.c @@ -49,7 +49,7 @@ static const u8 bcast_addr[ETH_ALEN] = { 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF }; static void assert_key_lock(struct ieee80211_local *local) { - WARN_ON(!mutex_is_locked(&local->key_mtx)); + lockdep_assert_held(&local->key_mtx); } static struct ieee80211_sta *get_sta_for_key(struct ieee80211_key *key) @@ -60,7 +60,7 @@ static struct ieee80211_sta *get_sta_for_key(struct ieee80211_key *key) return NULL; } -static void ieee80211_key_enable_hw_accel(struct ieee80211_key *key) +static int ieee80211_key_enable_hw_accel(struct ieee80211_key *key) { struct ieee80211_sub_if_data *sdata; struct ieee80211_sta *sta; @@ -69,12 +69,20 @@ static void ieee80211_key_enable_hw_accel(struct ieee80211_key *key) might_sleep(); if (!key->local->ops->set_key) - return; + goto out_unsupported; assert_key_lock(key->local); sta = get_sta_for_key(key); + /* + * If this is a per-STA GTK, check if it + * is supported; if not, return. + */ + if (sta && !(key->conf.flags & IEEE80211_KEY_FLAG_PAIRWISE) && + !(key->local->hw.flags & IEEE80211_HW_SUPPORTS_PER_STA_GTK)) + goto out_unsupported; + sdata = key->sdata; if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN) sdata = container_of(sdata->bss, @@ -83,14 +91,28 @@ static void ieee80211_key_enable_hw_accel(struct ieee80211_key *key) ret = drv_set_key(key->local, SET_KEY, sdata, sta, &key->conf); - if (!ret) + if (!ret) { key->flags |= KEY_FLAG_UPLOADED_TO_HARDWARE; + return 0; + } - if (ret && ret != -ENOSPC && ret != -EOPNOTSUPP) - printk(KERN_ERR "mac80211-%s: failed to set key " - "(%d, %pM) to hardware (%d)\n", - wiphy_name(key->local->hw.wiphy), - key->conf.keyidx, sta ? sta->addr : bcast_addr, ret); + if (ret != -ENOSPC && ret != -EOPNOTSUPP) + wiphy_err(key->local->hw.wiphy, + "failed to set key (%d, %pM) to hardware (%d)\n", + key->conf.keyidx, sta ? sta->addr : bcast_addr, ret); + + out_unsupported: + switch (key->conf.cipher) { + case WLAN_CIPHER_SUITE_WEP40: + case WLAN_CIPHER_SUITE_WEP104: + case WLAN_CIPHER_SUITE_TKIP: + case WLAN_CIPHER_SUITE_CCMP: + case WLAN_CIPHER_SUITE_AES_CMAC: + /* all of these we can do in software */ + return 0; + default: + return -EINVAL; + } } static void ieee80211_key_disable_hw_accel(struct ieee80211_key *key) @@ -121,14 +143,33 @@ static void ieee80211_key_disable_hw_accel(struct ieee80211_key *key) sta, &key->conf); if (ret) - printk(KERN_ERR "mac80211-%s: failed to remove key " - "(%d, %pM) from hardware (%d)\n", - wiphy_name(key->local->hw.wiphy), - key->conf.keyidx, sta ? sta->addr : bcast_addr, ret); + wiphy_err(key->local->hw.wiphy, + "failed to remove key (%d, %pM) from hardware (%d)\n", + key->conf.keyidx, sta ? sta->addr : bcast_addr, ret); key->flags &= ~KEY_FLAG_UPLOADED_TO_HARDWARE; } +void ieee80211_key_removed(struct ieee80211_key_conf *key_conf) +{ + struct ieee80211_key *key; + + key = container_of(key_conf, struct ieee80211_key, conf); + + might_sleep(); + assert_key_lock(key->local); + + key->flags &= ~KEY_FLAG_UPLOADED_TO_HARDWARE; + + /* + * Flush TX path to avoid attempts to use this key + * after this function returns. Until then, drivers + * must be prepared to handle the key. + */ + synchronize_rcu(); +} +EXPORT_SYMBOL_GPL(ieee80211_key_removed); + static void __ieee80211_set_default_key(struct ieee80211_sub_if_data *sdata, int idx) { @@ -184,6 +225,7 @@ void ieee80211_set_default_mgmt_key(struct ieee80211_sub_if_data *sdata, static void __ieee80211_key_replace(struct ieee80211_sub_if_data *sdata, struct sta_info *sta, + bool pairwise, struct ieee80211_key *old, struct ieee80211_key *new) { @@ -192,8 +234,14 @@ static void __ieee80211_key_replace(struct ieee80211_sub_if_data *sdata, if (new) list_add(&new->list, &sdata->key_list); - if (sta) { - rcu_assign_pointer(sta->key, new); + if (sta && pairwise) { + rcu_assign_pointer(sta->ptk, new); + } else if (sta) { + if (old) + idx = old->conf.keyidx; + else + idx = new->conf.keyidx; + rcu_assign_pointer(sta->gtk[idx], new); } else { WARN_ON(new && old && new->conf.keyidx != old->conf.keyidx); @@ -227,20 +275,18 @@ static void __ieee80211_key_replace(struct ieee80211_sub_if_data *sdata, } } -struct ieee80211_key *ieee80211_key_alloc(enum ieee80211_key_alg alg, - int idx, - size_t key_len, +struct ieee80211_key *ieee80211_key_alloc(u32 cipher, int idx, size_t key_len, const u8 *key_data, size_t seq_len, const u8 *seq) { struct ieee80211_key *key; - int i, j; + int i, j, err; BUG_ON(idx < 0 || idx >= NUM_DEFAULT_KEYS + NUM_DEFAULT_MGMT_KEYS); key = kzalloc(sizeof(struct ieee80211_key) + key_len, GFP_KERNEL); if (!key) - return NULL; + return ERR_PTR(-ENOMEM); /* * Default to software encryption; we'll later upload the @@ -249,15 +295,16 @@ struct ieee80211_key *ieee80211_key_alloc(enum ieee80211_key_alg alg, key->conf.flags = 0; key->flags = 0; - key->conf.alg = alg; + key->conf.cipher = cipher; key->conf.keyidx = idx; key->conf.keylen = key_len; - switch (alg) { - case ALG_WEP: + switch (cipher) { + case WLAN_CIPHER_SUITE_WEP40: + case WLAN_CIPHER_SUITE_WEP104: key->conf.iv_len = WEP_IV_LEN; key->conf.icv_len = WEP_ICV_LEN; break; - case ALG_TKIP: + case WLAN_CIPHER_SUITE_TKIP: key->conf.iv_len = TKIP_IV_LEN; key->conf.icv_len = TKIP_ICV_LEN; if (seq) { @@ -269,7 +316,7 @@ struct ieee80211_key *ieee80211_key_alloc(enum ieee80211_key_alg alg, } } break; - case ALG_CCMP: + case WLAN_CIPHER_SUITE_CCMP: key->conf.iv_len = CCMP_HDR_LEN; key->conf.icv_len = CCMP_MIC_LEN; if (seq) { @@ -278,42 +325,38 @@ struct ieee80211_key *ieee80211_key_alloc(enum ieee80211_key_alg alg, key->u.ccmp.rx_pn[i][j] = seq[CCMP_PN_LEN - j - 1]; } - break; - case ALG_AES_CMAC: - key->conf.iv_len = 0; - key->conf.icv_len = sizeof(struct ieee80211_mmie); - if (seq) - for (j = 0; j < 6; j++) - key->u.aes_cmac.rx_pn[j] = seq[6 - j - 1]; - break; - } - memcpy(key->conf.key, key_data, key_len); - INIT_LIST_HEAD(&key->list); - - if (alg == ALG_CCMP) { /* * Initialize AES key state here as an optimization so that * it does not need to be initialized for every packet. */ key->u.ccmp.tfm = ieee80211_aes_key_setup_encrypt(key_data); - if (!key->u.ccmp.tfm) { + if (IS_ERR(key->u.ccmp.tfm)) { + err = PTR_ERR(key->u.ccmp.tfm); kfree(key); - return NULL; + key = ERR_PTR(err); } - } - - if (alg == ALG_AES_CMAC) { + break; + case WLAN_CIPHER_SUITE_AES_CMAC: + key->conf.iv_len = 0; + key->conf.icv_len = sizeof(struct ieee80211_mmie); + if (seq) + for (j = 0; j < 6; j++) + key->u.aes_cmac.rx_pn[j] = seq[6 - j - 1]; /* * Initialize AES key state here as an optimization so that * it does not need to be initialized for every packet. */ key->u.aes_cmac.tfm = ieee80211_aes_cmac_key_setup(key_data); - if (!key->u.aes_cmac.tfm) { + if (IS_ERR(key->u.aes_cmac.tfm)) { + err = PTR_ERR(key->u.aes_cmac.tfm); kfree(key); - return NULL; + key = ERR_PTR(err); } + break; } + memcpy(key->conf.key, key_data, key_len); + INIT_LIST_HEAD(&key->list); return key; } @@ -326,9 +369,9 @@ static void __ieee80211_key_destroy(struct ieee80211_key *key) if (key->local) ieee80211_key_disable_hw_accel(key); - if (key->conf.alg == ALG_CCMP) + if (key->conf.cipher == WLAN_CIPHER_SUITE_CCMP) ieee80211_aes_key_free(key->u.ccmp.tfm); - if (key->conf.alg == ALG_AES_CMAC) + if (key->conf.cipher == WLAN_CIPHER_SUITE_AES_CMAC) ieee80211_aes_cmac_key_free(key->u.aes_cmac.tfm); if (key->local) ieee80211_debugfs_key_remove(key); @@ -336,12 +379,13 @@ static void __ieee80211_key_destroy(struct ieee80211_key *key) kfree(key); } -void ieee80211_key_link(struct ieee80211_key *key, - struct ieee80211_sub_if_data *sdata, - struct sta_info *sta) +int ieee80211_key_link(struct ieee80211_key *key, + struct ieee80211_sub_if_data *sdata, + struct sta_info *sta) { struct ieee80211_key *old_key; - int idx; + int idx, ret; + bool pairwise = key->conf.flags & IEEE80211_KEY_FLAG_PAIRWISE; BUG_ON(!sdata); BUG_ON(!key); @@ -358,13 +402,6 @@ void ieee80211_key_link(struct ieee80211_key *key, */ if (test_sta_flags(sta, WLAN_STA_WME)) key->conf.flags |= IEEE80211_KEY_FLAG_WMM_STA; - - /* - * This key is for a specific sta interface, - * inform the driver that it should try to store - * this key as pairwise key. - */ - key->conf.flags |= IEEE80211_KEY_FLAG_PAIRWISE; } else { if (sdata->vif.type == NL80211_IFTYPE_STATION) { struct sta_info *ap; @@ -386,19 +423,23 @@ void ieee80211_key_link(struct ieee80211_key *key, mutex_lock(&sdata->local->key_mtx); - if (sta) - old_key = sta->key; + if (sta && pairwise) + old_key = sta->ptk; + else if (sta) + old_key = sta->gtk[idx]; else old_key = sdata->keys[idx]; - __ieee80211_key_replace(sdata, sta, old_key, key); + __ieee80211_key_replace(sdata, sta, pairwise, old_key, key); __ieee80211_key_destroy(old_key); ieee80211_debugfs_key_add(key); - ieee80211_key_enable_hw_accel(key); + ret = ieee80211_key_enable_hw_accel(key); mutex_unlock(&sdata->local->key_mtx); + + return ret; } static void __ieee80211_key_free(struct ieee80211_key *key) @@ -408,7 +449,8 @@ static void __ieee80211_key_free(struct ieee80211_key *key) */ if (key->sdata) __ieee80211_key_replace(key->sdata, key->sta, - key, NULL); + key->conf.flags & IEEE80211_KEY_FLAG_PAIRWISE, + key, NULL); __ieee80211_key_destroy(key); } diff --git a/net/mac80211/key.h b/net/mac80211/key.h index b665bbb7a47..0db1c0f5f69 100644 --- a/net/mac80211/key.h +++ b/net/mac80211/key.h @@ -16,6 +16,9 @@ #include <linux/rcupdate.h> #include <net/mac80211.h> +#define NUM_DEFAULT_KEYS 4 +#define NUM_DEFAULT_MGMT_KEYS 2 + #define WEP_IV_LEN 4 #define WEP_ICV_LEN 4 #define ALG_TKIP_KEY_LEN 32 @@ -123,18 +126,16 @@ struct ieee80211_key { struct ieee80211_key_conf conf; }; -struct ieee80211_key *ieee80211_key_alloc(enum ieee80211_key_alg alg, - int idx, - size_t key_len, +struct ieee80211_key *ieee80211_key_alloc(u32 cipher, int idx, size_t key_len, const u8 *key_data, size_t seq_len, const u8 *seq); /* * Insert a key into data structures (sdata, sta if necessary) * to make it used, free old key. */ -void ieee80211_key_link(struct ieee80211_key *key, - struct ieee80211_sub_if_data *sdata, - struct sta_info *sta); +int __must_check ieee80211_key_link(struct ieee80211_key *key, + struct ieee80211_sub_if_data *sdata, + struct sta_info *sta); void ieee80211_key_free(struct ieee80211_local *local, struct ieee80211_key *key); void ieee80211_set_default_key(struct ieee80211_sub_if_data *sdata, int idx); diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 798a91b100c..22bc42b1899 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -54,6 +54,9 @@ void ieee80211_configure_filter(struct ieee80211_local *local) if (local->monitors || local->scanning) new_flags |= FIF_BCN_PRBRESP_PROMISC; + if (local->fif_probe_req || local->probe_req_reg) + new_flags |= FIF_PROBE_REQ; + if (local->fif_fcsfail) new_flags |= FIF_FCSFAIL; @@ -99,16 +102,19 @@ int ieee80211_hw_config(struct ieee80211_local *local, u32 changed) int ret = 0; int power; enum nl80211_channel_type channel_type; + u32 offchannel_flag; might_sleep(); scan_chan = local->scan_channel; + offchannel_flag = local->hw.conf.flags & IEEE80211_CONF_OFFCHANNEL; if (scan_chan) { chan = scan_chan; channel_type = NL80211_CHAN_NO_HT; local->hw.conf.flags |= IEEE80211_CONF_OFFCHANNEL; - } else if (local->tmp_channel) { + } else if (local->tmp_channel && + local->oper_channel != local->tmp_channel) { chan = scan_chan = local->tmp_channel; channel_type = local->tmp_channel_type; local->hw.conf.flags |= IEEE80211_CONF_OFFCHANNEL; @@ -117,8 +123,9 @@ int ieee80211_hw_config(struct ieee80211_local *local, u32 changed) channel_type = local->_oper_channel_type; local->hw.conf.flags &= ~IEEE80211_CONF_OFFCHANNEL; } + offchannel_flag ^= local->hw.conf.flags & IEEE80211_CONF_OFFCHANNEL; - if (chan != local->hw.conf.channel || + if (offchannel_flag || chan != local->hw.conf.channel || channel_type != local->hw.conf.channel_type) { local->hw.conf.channel = chan; local->hw.conf.channel_type = channel_type; @@ -197,6 +204,8 @@ void ieee80211_bss_info_change_notify(struct ieee80211_sub_if_data *sdata, sdata->vif.bss_conf.bssid = sdata->u.ibss.bssid; else if (sdata->vif.type == NL80211_IFTYPE_AP) sdata->vif.bss_conf.bssid = sdata->vif.addr; + else if (sdata->vif.type == NL80211_IFTYPE_WDS) + sdata->vif.bss_conf.bssid = NULL; else if (ieee80211_vif_is_mesh(&sdata->vif)) { sdata->vif.bss_conf.bssid = zero; } else { @@ -207,6 +216,7 @@ void ieee80211_bss_info_change_notify(struct ieee80211_sub_if_data *sdata, switch (sdata->vif.type) { case NL80211_IFTYPE_AP: case NL80211_IFTYPE_ADHOC: + case NL80211_IFTYPE_WDS: case NL80211_IFTYPE_MESH_POINT: break; default: @@ -291,7 +301,16 @@ static void ieee80211_restart_work(struct work_struct *work) struct ieee80211_local *local = container_of(work, struct ieee80211_local, restart_work); + /* wait for scan work complete */ + flush_workqueue(local->workqueue); + + mutex_lock(&local->mtx); + WARN(test_bit(SCAN_HW_SCANNING, &local->scanning), + "%s called with hardware scan in progress\n", __func__); + mutex_unlock(&local->mtx); + rtnl_lock(); + ieee80211_scan_cancel(local); ieee80211_reconfig(local); rtnl_unlock(); } @@ -302,7 +321,7 @@ void ieee80211_restart_hw(struct ieee80211_hw *hw) trace_api_restart_hw(local); - /* use this reason, __ieee80211_resume will unblock it */ + /* use this reason, ieee80211_reconfig will unblock it */ ieee80211_stop_queues_by_reason(hw, IEEE80211_QUEUE_STOP_REASON_SUSPEND); @@ -316,7 +335,7 @@ static void ieee80211_recalc_smps_work(struct work_struct *work) container_of(work, struct ieee80211_local, recalc_smps); mutex_lock(&local->iflist_mtx); - ieee80211_recalc_smps(local, NULL); + ieee80211_recalc_smps(local); mutex_unlock(&local->iflist_mtx); } @@ -336,9 +355,6 @@ static int ieee80211_ifa_changed(struct notifier_block *nb, struct ieee80211_if_managed *ifmgd; int c = 0; - if (!netif_running(ndev)) - return NOTIFY_DONE; - /* Make sure it's our interface that got changed */ if (!wdev) return NOTIFY_DONE; @@ -349,11 +365,14 @@ static int ieee80211_ifa_changed(struct notifier_block *nb, sdata = IEEE80211_DEV_TO_SUB_IF(ndev); bss_conf = &sdata->vif.bss_conf; + if (!ieee80211_sdata_running(sdata)) + return NOTIFY_DONE; + /* ARP filtering is only supported in managed mode */ if (sdata->vif.type != NL80211_IFTYPE_STATION) return NOTIFY_DONE; - idev = sdata->dev->ip_ptr; + idev = __in_dev_get_rtnl(sdata->dev); if (!idev) return NOTIFY_DONE; @@ -390,6 +409,80 @@ static int ieee80211_ifa_changed(struct notifier_block *nb, } #endif +static int ieee80211_napi_poll(struct napi_struct *napi, int budget) +{ + struct ieee80211_local *local = + container_of(napi, struct ieee80211_local, napi); + + return local->ops->napi_poll(&local->hw, budget); +} + +void ieee80211_napi_schedule(struct ieee80211_hw *hw) +{ + struct ieee80211_local *local = hw_to_local(hw); + + napi_schedule(&local->napi); +} +EXPORT_SYMBOL(ieee80211_napi_schedule); + +void ieee80211_napi_complete(struct ieee80211_hw *hw) +{ + struct ieee80211_local *local = hw_to_local(hw); + + napi_complete(&local->napi); +} +EXPORT_SYMBOL(ieee80211_napi_complete); + +/* There isn't a lot of sense in it, but you can transmit anything you like */ +static const struct ieee80211_txrx_stypes +ieee80211_default_mgmt_stypes[NUM_NL80211_IFTYPES] = { + [NL80211_IFTYPE_ADHOC] = { + .tx = 0xffff, + .rx = BIT(IEEE80211_STYPE_ACTION >> 4), + }, + [NL80211_IFTYPE_STATION] = { + .tx = 0xffff, + .rx = BIT(IEEE80211_STYPE_ACTION >> 4) | + BIT(IEEE80211_STYPE_PROBE_REQ >> 4), + }, + [NL80211_IFTYPE_AP] = { + .tx = 0xffff, + .rx = BIT(IEEE80211_STYPE_ASSOC_REQ >> 4) | + BIT(IEEE80211_STYPE_REASSOC_REQ >> 4) | + BIT(IEEE80211_STYPE_PROBE_REQ >> 4) | + BIT(IEEE80211_STYPE_DISASSOC >> 4) | + BIT(IEEE80211_STYPE_AUTH >> 4) | + BIT(IEEE80211_STYPE_DEAUTH >> 4) | + BIT(IEEE80211_STYPE_ACTION >> 4), + }, + [NL80211_IFTYPE_AP_VLAN] = { + /* copy AP */ + .tx = 0xffff, + .rx = BIT(IEEE80211_STYPE_ASSOC_REQ >> 4) | + BIT(IEEE80211_STYPE_REASSOC_REQ >> 4) | + BIT(IEEE80211_STYPE_PROBE_REQ >> 4) | + BIT(IEEE80211_STYPE_DISASSOC >> 4) | + BIT(IEEE80211_STYPE_AUTH >> 4) | + BIT(IEEE80211_STYPE_DEAUTH >> 4) | + BIT(IEEE80211_STYPE_ACTION >> 4), + }, + [NL80211_IFTYPE_P2P_CLIENT] = { + .tx = 0xffff, + .rx = BIT(IEEE80211_STYPE_ACTION >> 4) | + BIT(IEEE80211_STYPE_PROBE_REQ >> 4), + }, + [NL80211_IFTYPE_P2P_GO] = { + .tx = 0xffff, + .rx = BIT(IEEE80211_STYPE_ASSOC_REQ >> 4) | + BIT(IEEE80211_STYPE_REASSOC_REQ >> 4) | + BIT(IEEE80211_STYPE_PROBE_REQ >> 4) | + BIT(IEEE80211_STYPE_DISASSOC >> 4) | + BIT(IEEE80211_STYPE_AUTH >> 4) | + BIT(IEEE80211_STYPE_DEAUTH >> 4) | + BIT(IEEE80211_STYPE_ACTION >> 4), + }, +}; + struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len, const struct ieee80211_ops *ops) { @@ -419,6 +512,8 @@ struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len, if (!wiphy) return NULL; + wiphy->mgmt_stypes = ieee80211_default_mgmt_stypes; + wiphy->flags |= WIPHY_FLAG_NETNS_OK | WIPHY_FLAG_4ADDR_AP | WIPHY_FLAG_4ADDR_STATION; @@ -444,6 +539,7 @@ struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len, /* set up some defaults */ local->hw.queues = 1; local->hw.max_rates = 1; + local->hw.max_report_rates = 0; local->hw.conf.long_frame_max_tx_count = wiphy->retry_long; local->hw.conf.short_frame_max_tx_count = wiphy->retry_short; local->user_power_level = -1; @@ -455,7 +551,7 @@ struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len, __hw_addr_init(&local->mc_list); mutex_init(&local->iflist_mtx); - mutex_init(&local->scan_mtx); + mutex_init(&local->mtx); mutex_init(&local->key_mtx); spin_lock_init(&local->filter_lock); @@ -494,6 +590,9 @@ struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len, skb_queue_head_init(&local->skb_queue); skb_queue_head_init(&local->skb_queue_unreliable); + /* init dummy netdev for use w/ NAPI */ + init_dummy_netdev(&local->napi_dev); + return local_to_hw(local); } EXPORT_SYMBOL(ieee80211_alloc_hw); @@ -506,6 +605,7 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) int channels, max_bitrates; bool supp_ht; static const u32 cipher_suites[] = { + /* keep WEP first, it may be removed below */ WLAN_CIPHER_SUITE_WEP40, WLAN_CIPHER_SUITE_WEP104, WLAN_CIPHER_SUITE_TKIP, @@ -515,6 +615,9 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) WLAN_CIPHER_SUITE_AES_CMAC }; + if (hw->max_report_rates == 0) + hw->max_report_rates = hw->max_rates; + /* * generic code guarantees at least one band, * set this very early because much code assumes @@ -554,6 +657,14 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) /* mac80211 always supports monitor */ local->hw.wiphy->interface_modes |= BIT(NL80211_IFTYPE_MONITOR); +#ifndef CONFIG_MAC80211_MESH + /* mesh depends on Kconfig, but drivers should set it if they want */ + local->hw.wiphy->interface_modes &= ~BIT(NL80211_IFTYPE_MESH_POINT); +#endif + + /* mac80211 supports control port protocol changing */ + local->hw.wiphy->flags |= WIPHY_FLAG_CONTROL_PORT_PROTOCOL; + if (local->hw.flags & IEEE80211_HW_SIGNAL_DBM) local->hw.wiphy->signal_type = CFG80211_SIGNAL_TYPE_MBM; else if (local->hw.flags & IEEE80211_HW_SIGNAL_UNSPEC) @@ -589,10 +700,41 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) if (local->hw.wiphy->max_scan_ie_len) local->hw.wiphy->max_scan_ie_len -= local->scan_ies_len; - local->hw.wiphy->cipher_suites = cipher_suites; - local->hw.wiphy->n_cipher_suites = ARRAY_SIZE(cipher_suites); - if (!(local->hw.flags & IEEE80211_HW_MFP_CAPABLE)) - local->hw.wiphy->n_cipher_suites--; + /* Set up cipher suites unless driver already did */ + if (!local->hw.wiphy->cipher_suites) { + local->hw.wiphy->cipher_suites = cipher_suites; + local->hw.wiphy->n_cipher_suites = ARRAY_SIZE(cipher_suites); + if (!(local->hw.flags & IEEE80211_HW_MFP_CAPABLE)) + local->hw.wiphy->n_cipher_suites--; + } + if (IS_ERR(local->wep_tx_tfm) || IS_ERR(local->wep_rx_tfm)) { + if (local->hw.wiphy->cipher_suites == cipher_suites) { + local->hw.wiphy->cipher_suites += 2; + local->hw.wiphy->n_cipher_suites -= 2; + } else { + u32 *suites; + int r, w = 0; + + /* Filter out WEP */ + + suites = kmemdup( + local->hw.wiphy->cipher_suites, + sizeof(u32) * local->hw.wiphy->n_cipher_suites, + GFP_KERNEL); + if (!suites) + return -ENOMEM; + for (r = 0; r < local->hw.wiphy->n_cipher_suites; r++) { + u32 suite = local->hw.wiphy->cipher_suites[r]; + if (suite == WLAN_CIPHER_SUITE_WEP40 || + suite == WLAN_CIPHER_SUITE_WEP104) + continue; + suites[w++] = suite; + } + local->hw.wiphy->cipher_suites = suites; + local->hw.wiphy->n_cipher_suites = w; + local->wiphy_ciphers_allocated = true; + } + } result = wiphy_register(local->hw.wiphy); if (result < 0) @@ -641,16 +783,16 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) result = ieee80211_wep_init(local); if (result < 0) - printk(KERN_DEBUG "%s: Failed to initialize wep: %d\n", - wiphy_name(local->hw.wiphy), result); + wiphy_debug(local->hw.wiphy, "Failed to initialize wep: %d\n", + result); rtnl_lock(); result = ieee80211_init_rate_ctrl_alg(local, hw->rate_control_algorithm); if (result < 0) { - printk(KERN_DEBUG "%s: Failed to initialize rate control " - "algorithm\n", wiphy_name(local->hw.wiphy)); + wiphy_debug(local->hw.wiphy, + "Failed to initialize rate control algorithm\n"); goto fail_rate; } @@ -659,8 +801,8 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) result = ieee80211_if_add(local, "wlan%d", NULL, NL80211_IFTYPE_STATION, NULL); if (result) - printk(KERN_WARNING "%s: Failed to add default virtual iface\n", - wiphy_name(local->hw.wiphy)); + wiphy_warn(local->hw.wiphy, + "Failed to add default virtual iface\n"); } rtnl_unlock(); @@ -683,6 +825,9 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) goto fail_ifa; #endif + netif_napi_add(&local->napi_dev, &local->napi, ieee80211_napi_poll, + local->hw.napi_weight); + return 0; #ifdef CONFIG_INET @@ -703,6 +848,8 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) fail_workqueue: wiphy_unregister(local->hw.wiphy); fail_wiphy_register: + if (local->wiphy_ciphers_allocated) + kfree(local->hw.wiphy->cipher_suites); kfree(local->int_scan_req); return result; } @@ -732,6 +879,13 @@ void ieee80211_unregister_hw(struct ieee80211_hw *hw) rtnl_unlock(); + /* + * Now all work items will be gone, but the + * timer might still be armed, so delete it + */ + del_timer_sync(&local->work_timer); + + cancel_work_sync(&local->restart_work); cancel_work_sync(&local->reconfig_filter); ieee80211_clear_tx_pending(local); @@ -740,8 +894,7 @@ void ieee80211_unregister_hw(struct ieee80211_hw *hw) if (skb_queue_len(&local->skb_queue) || skb_queue_len(&local->skb_queue_unreliable)) - printk(KERN_WARNING "%s: skb_queue not empty\n", - wiphy_name(local->hw.wiphy)); + wiphy_warn(local->hw.wiphy, "skb_queue not empty\n"); skb_queue_purge(&local->skb_queue); skb_queue_purge(&local->skb_queue_unreliable); @@ -758,7 +911,10 @@ void ieee80211_free_hw(struct ieee80211_hw *hw) struct ieee80211_local *local = hw_to_local(hw); mutex_destroy(&local->iflist_mtx); - mutex_destroy(&local->scan_mtx); + mutex_destroy(&local->mtx); + + if (local->wiphy_ciphers_allocated) + kfree(local->hw.wiphy->cipher_suites); wiphy_free(local->hw.wiphy); } diff --git a/net/mac80211/mesh_plink.c b/net/mac80211/mesh_plink.c index ea13a80a476..1c91f0f3c30 100644 --- a/net/mac80211/mesh_plink.c +++ b/net/mac80211/mesh_plink.c @@ -412,7 +412,7 @@ void mesh_rx_plink_frame(struct ieee80211_sub_if_data *sdata, struct ieee80211_m enum plink_event event; enum plink_frame_type ftype; size_t baselen; - bool deactivated; + bool deactivated, matches_local = true; u8 ie_len; u8 *baseaddr; __le16 plid, llid, reason; @@ -487,6 +487,7 @@ void mesh_rx_plink_frame(struct ieee80211_sub_if_data *sdata, struct ieee80211_m /* Now we will figure out the appropriate event... */ event = PLINK_UNDEFINED; if (ftype != PLINK_CLOSE && (!mesh_matches_local(&elems, sdata))) { + matches_local = false; switch (ftype) { case PLINK_OPEN: event = OPN_RJCT; @@ -498,7 +499,15 @@ void mesh_rx_plink_frame(struct ieee80211_sub_if_data *sdata, struct ieee80211_m /* avoid warning */ break; } - spin_lock_bh(&sta->lock); + } + + if (!sta && !matches_local) { + rcu_read_unlock(); + reason = cpu_to_le16(MESH_CAPABILITY_POLICY_VIOLATION); + llid = 0; + mesh_plink_frame_tx(sdata, PLINK_CLOSE, mgmt->sa, llid, + plid, reason); + return; } else if (!sta) { /* ftype == PLINK_OPEN */ u32 rates; @@ -522,7 +531,7 @@ void mesh_rx_plink_frame(struct ieee80211_sub_if_data *sdata, struct ieee80211_m } event = OPN_ACPT; spin_lock_bh(&sta->lock); - } else { + } else if (matches_local) { spin_lock_bh(&sta->lock); switch (ftype) { case PLINK_OPEN: @@ -564,6 +573,8 @@ void mesh_rx_plink_frame(struct ieee80211_sub_if_data *sdata, struct ieee80211_m rcu_read_unlock(); return; } + } else { + spin_lock_bh(&sta->lock); } mpl_dbg("Mesh plink (peer, state, llid, plid, event): %pM %s %d %d %d\n", diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index b6c163ac22d..a3a9421555a 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -54,6 +54,12 @@ */ #define IEEE80211_SIGNAL_AVE_WEIGHT 3 +/* + * How many Beacon frames need to have been used in average signal strength + * before starting to indicate signal change events. + */ +#define IEEE80211_SIGNAL_AVE_MIN_COUNT 4 + #define TMR_RUNNING_TIMER 0 #define TMR_RUNNING_CHANSW 1 @@ -86,7 +92,7 @@ enum rx_mgmt_action { /* utils */ static inline void ASSERT_MGD_MTX(struct ieee80211_if_managed *ifmgd) { - WARN_ON(!mutex_is_locked(&ifmgd->mtx)); + lockdep_assert_held(&ifmgd->mtx); } /* @@ -109,7 +115,7 @@ static void run_again(struct ieee80211_if_managed *ifmgd, mod_timer(&ifmgd->timer, timeout); } -static void mod_beacon_timer(struct ieee80211_sub_if_data *sdata) +void ieee80211_sta_reset_beacon_monitor(struct ieee80211_sub_if_data *sdata) { if (sdata->local->hw.flags & IEEE80211_HW_BEACON_FILTER) return; @@ -118,6 +124,19 @@ static void mod_beacon_timer(struct ieee80211_sub_if_data *sdata) round_jiffies_up(jiffies + IEEE80211_BEACON_LOSS_TIME)); } +void ieee80211_sta_reset_conn_monitor(struct ieee80211_sub_if_data *sdata) +{ + struct ieee80211_if_managed *ifmgd = &sdata->u.mgd; + + if (sdata->local->hw.flags & IEEE80211_HW_CONNECTION_MONITOR) + return; + + mod_timer(&sdata->u.mgd.conn_mon_timer, + round_jiffies_up(jiffies + IEEE80211_CONNECTION_IDLE_TIME)); + + ifmgd->probe_send_count = 0; +} + static int ecw2cw(int ecw) { return (1 << ecw) - 1; @@ -778,16 +797,17 @@ static void ieee80211_sta_wmm_params(struct ieee80211_local *local, params.uapsd = uapsd; #ifdef CONFIG_MAC80211_VERBOSE_DEBUG - printk(KERN_DEBUG "%s: WMM queue=%d aci=%d acm=%d aifs=%d " - "cWmin=%d cWmax=%d txop=%d uapsd=%d\n", - wiphy_name(local->hw.wiphy), queue, aci, acm, - params.aifs, params.cw_min, params.cw_max, params.txop, - params.uapsd); + wiphy_debug(local->hw.wiphy, + "WMM queue=%d aci=%d acm=%d aifs=%d " + "cWmin=%d cWmax=%d txop=%d uapsd=%d\n", + queue, aci, acm, + params.aifs, params.cw_min, params.cw_max, + params.txop, params.uapsd); #endif if (drv_conf_tx(local, queue, ¶ms)) - printk(KERN_DEBUG "%s: failed to set TX queue " - "parameters for queue %d\n", - wiphy_name(local->hw.wiphy), queue); + wiphy_debug(local->hw.wiphy, + "failed to set TX queue parameters for queue %d\n", + queue); } /* enable WMM or activate new settings */ @@ -860,14 +880,6 @@ static void ieee80211_set_associated(struct ieee80211_sub_if_data *sdata, sdata->u.mgd.flags &= ~(IEEE80211_STA_CONNECTION_POLL | IEEE80211_STA_BEACON_POLL); - /* - * Always handle WMM once after association regardless - * of the first value the AP uses. Setting -1 here has - * that effect because the AP values is an unsigned - * 4-bit value. - */ - sdata->u.mgd.wmm_last_param_set = -1; - ieee80211_led_assoc(local, 1); if (local->hw.flags & IEEE80211_HW_NEED_DTIM_PERIOD) @@ -901,7 +913,7 @@ static void ieee80211_set_associated(struct ieee80211_sub_if_data *sdata, mutex_lock(&local->iflist_mtx); ieee80211_recalc_ps(local, -1); - ieee80211_recalc_smps(local, sdata); + ieee80211_recalc_smps(local); mutex_unlock(&local->iflist_mtx); netif_tx_start_all_queues(sdata->dev); @@ -909,7 +921,7 @@ static void ieee80211_set_associated(struct ieee80211_sub_if_data *sdata, } static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata, - bool remove_sta) + bool remove_sta, bool tx) { struct ieee80211_if_managed *ifmgd = &sdata->u.mgd; struct ieee80211_local *local = sdata->local; @@ -948,7 +960,7 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata, sta = sta_info_get(sdata, bssid); if (sta) { set_sta_flags(sta, WLAN_STA_BLOCK_BA); - ieee80211_sta_tear_down_BA_sessions(sta); + ieee80211_sta_tear_down_BA_sessions(sta, tx); } mutex_unlock(&local->sta_mtx); @@ -990,6 +1002,11 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata, if (remove_sta) sta_info_destroy_addr(sdata, bssid); + + del_timer_sync(&sdata->u.mgd.conn_mon_timer); + del_timer_sync(&sdata->u.mgd.bcn_mon_timer); + del_timer_sync(&sdata->u.mgd.timer); + del_timer_sync(&sdata->u.mgd.chswitch_timer); } void ieee80211_sta_rx_notify(struct ieee80211_sub_if_data *sdata, @@ -1006,21 +1023,26 @@ void ieee80211_sta_rx_notify(struct ieee80211_sub_if_data *sdata, if (is_multicast_ether_addr(hdr->addr1)) return; - if (sdata->local->hw.flags & IEEE80211_HW_CONNECTION_MONITOR) - return; - - mod_timer(&sdata->u.mgd.conn_mon_timer, - round_jiffies_up(jiffies + IEEE80211_CONNECTION_IDLE_TIME)); + ieee80211_sta_reset_conn_monitor(sdata); } static void ieee80211_mgd_probe_ap_send(struct ieee80211_sub_if_data *sdata) { struct ieee80211_if_managed *ifmgd = &sdata->u.mgd; const u8 *ssid; + u8 *dst = ifmgd->associated->bssid; + u8 unicast_limit = max(1, IEEE80211_MAX_PROBE_TRIES - 3); + + /* + * Try sending broadcast probe requests for the last three + * probe requests after the first ones failed since some + * buggy APs only support broadcast probe requests. + */ + if (ifmgd->probe_send_count >= unicast_limit) + dst = NULL; ssid = ieee80211_bss_get_ie(ifmgd->associated, WLAN_EID_SSID); - ieee80211_send_probe_req(sdata, ifmgd->associated->bssid, - ssid + 2, ssid[1], NULL, 0); + ieee80211_send_probe_req(sdata, dst, ssid + 2, ssid[1], NULL, 0); ifmgd->probe_send_count++; ifmgd->probe_timeout = jiffies + IEEE80211_PROBE_WAIT; @@ -1102,9 +1124,12 @@ static void __ieee80211_connection_loss(struct ieee80211_sub_if_data *sdata) printk(KERN_DEBUG "Connection to AP %pM lost.\n", bssid); - ieee80211_set_disassoc(sdata, true); - ieee80211_recalc_idle(local); + ieee80211_set_disassoc(sdata, true, true); mutex_unlock(&ifmgd->mtx); + + mutex_lock(&local->mtx); + ieee80211_recalc_idle(local); + mutex_unlock(&local->mtx); /* * must be outside lock due to cfg80211, * but that's not a problem. @@ -1172,8 +1197,10 @@ ieee80211_rx_mgmt_deauth(struct ieee80211_sub_if_data *sdata, printk(KERN_DEBUG "%s: deauthenticated from %pM (Reason: %u)\n", sdata->name, bssid, reason_code); - ieee80211_set_disassoc(sdata, true); + ieee80211_set_disassoc(sdata, true, false); + mutex_lock(&sdata->local->mtx); ieee80211_recalc_idle(sdata->local); + mutex_unlock(&sdata->local->mtx); return RX_MGMT_CFG80211_DEAUTH; } @@ -1202,8 +1229,10 @@ ieee80211_rx_mgmt_disassoc(struct ieee80211_sub_if_data *sdata, printk(KERN_DEBUG "%s: disassociated from %pM (Reason: %u)\n", sdata->name, mgmt->sa, reason_code); - ieee80211_set_disassoc(sdata, true); + ieee80211_set_disassoc(sdata, true, false); + mutex_lock(&sdata->local->mtx); ieee80211_recalc_idle(sdata->local); + mutex_unlock(&sdata->local->mtx); return RX_MGMT_CFG80211_DISASSOC; } @@ -1262,7 +1291,7 @@ static bool ieee80211_assoc_success(struct ieee80211_work *wk, rates = 0; basic_rates = 0; - sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; + sband = local->hw.wiphy->bands[wk->chan->band]; for (i = 0; i < elems.supp_rates_len; i++) { int rate = (elems.supp_rates[i] & 0x7f) * 5; @@ -1298,11 +1327,11 @@ static bool ieee80211_assoc_success(struct ieee80211_work *wk, } } - sta->sta.supp_rates[local->hw.conf.channel->band] = rates; + sta->sta.supp_rates[wk->chan->band] = rates; sdata->vif.bss_conf.basic_rates = basic_rates; /* cf. IEEE 802.11 9.2.12 */ - if (local->hw.conf.channel->band == IEEE80211_BAND_2GHZ && + if (wk->chan->band == IEEE80211_BAND_2GHZ && have_higher_than_11mbit) sdata->flags |= IEEE80211_SDATA_OPERATING_GMODE; else @@ -1330,6 +1359,14 @@ static bool ieee80211_assoc_success(struct ieee80211_work *wk, return false; } + /* + * Always handle WMM once after association regardless + * of the first value the AP uses. Setting -1 here has + * that effect because the AP values is an unsigned + * 4-bit value. + */ + ifmgd->wmm_last_param_set = -1; + if (elems.wmm_param) ieee80211_sta_wmm_params(local, sdata, elems.wmm_param, elems.wmm_param_len); @@ -1362,7 +1399,7 @@ static bool ieee80211_assoc_success(struct ieee80211_work *wk, * Also start the timer that will detect beacon loss. */ ieee80211_sta_rx_notify(sdata, (struct ieee80211_hdr *)mgmt); - mod_beacon_timer(sdata); + ieee80211_sta_reset_beacon_monitor(sdata); return true; } @@ -1465,7 +1502,7 @@ static void ieee80211_rx_mgmt_probe_resp(struct ieee80211_sub_if_data *sdata, * we have or will be receiving any beacons or data, so let's * schedule the timers again, just in case. */ - mod_beacon_timer(sdata); + ieee80211_sta_reset_beacon_monitor(sdata); mod_timer(&ifmgd->conn_mon_timer, round_jiffies_up(jiffies + @@ -1540,15 +1577,18 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_sub_if_data *sdata, ifmgd->last_beacon_signal = rx_status->signal; if (ifmgd->flags & IEEE80211_STA_RESET_SIGNAL_AVE) { ifmgd->flags &= ~IEEE80211_STA_RESET_SIGNAL_AVE; - ifmgd->ave_beacon_signal = rx_status->signal; + ifmgd->ave_beacon_signal = rx_status->signal * 16; ifmgd->last_cqm_event_signal = 0; + ifmgd->count_beacon_signal = 1; } else { ifmgd->ave_beacon_signal = (IEEE80211_SIGNAL_AVE_WEIGHT * rx_status->signal * 16 + (16 - IEEE80211_SIGNAL_AVE_WEIGHT) * ifmgd->ave_beacon_signal) / 16; + ifmgd->count_beacon_signal++; } if (bss_conf->cqm_rssi_thold && + ifmgd->count_beacon_signal >= IEEE80211_SIGNAL_AVE_MIN_COUNT && !(local->hw.flags & IEEE80211_HW_SUPPORTS_CQM_RSSI)) { int sig = ifmgd->ave_beacon_signal / 16; int last_event = ifmgd->last_cqm_event_signal; @@ -1588,7 +1628,7 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_sub_if_data *sdata, * Push the beacon loss detection into the future since * we are processing a beacon from the AP just now. */ - mod_beacon_timer(sdata); + ieee80211_sta_reset_beacon_monitor(sdata); ncrc = crc32_be(0, (void *)&mgmt->u.beacon.beacon_int, 4); ncrc = ieee802_11_parse_elems_crc(mgmt->u.beacon.variable, @@ -1599,7 +1639,7 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_sub_if_data *sdata, directed_tim = ieee80211_check_tim(elems.tim, elems.tim_len, ifmgd->aid); - if (ncrc != ifmgd->beacon_crc) { + if (ncrc != ifmgd->beacon_crc || !ifmgd->beacon_crc_valid) { ieee80211_rx_bss_info(sdata, mgmt, len, rx_status, &elems, true); @@ -1630,9 +1670,10 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_sub_if_data *sdata, } } - if (ncrc == ifmgd->beacon_crc) + if (ncrc == ifmgd->beacon_crc && ifmgd->beacon_crc_valid) return; ifmgd->beacon_crc = ncrc; + ifmgd->beacon_crc_valid = true; if (elems.erp_info && elems.erp_info_len >= 1) { erp_valid = true; @@ -1751,7 +1792,7 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata, struct ieee80211_local *local = sdata->local; struct ieee80211_work *wk; - mutex_lock(&local->work_mtx); + mutex_lock(&local->mtx); list_for_each_entry(wk, &local->work_list, list) { if (wk->sdata != sdata) continue; @@ -1783,7 +1824,7 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata, free_work(wk); break; } - mutex_unlock(&local->work_mtx); + mutex_unlock(&local->mtx); cfg80211_send_deauth(sdata->dev, (u8 *)mgmt, skb->len); } @@ -1823,10 +1864,12 @@ void ieee80211_sta_work(struct ieee80211_sub_if_data *sdata) else if (ifmgd->probe_send_count < IEEE80211_MAX_PROBE_TRIES) { #ifdef CONFIG_MAC80211_VERBOSE_DEBUG - printk(KERN_DEBUG "No probe response from AP %pM" - " after %dms, try %d\n", bssid, - (1000 * IEEE80211_PROBE_WAIT)/HZ, - ifmgd->probe_send_count); + wiphy_debug(local->hw.wiphy, + "%s: No probe response from AP %pM" + " after %dms, try %d\n", + sdata->name, + bssid, (1000 * IEEE80211_PROBE_WAIT)/HZ, + ifmgd->probe_send_count); #endif ieee80211_mgd_probe_ap_send(sdata); } else { @@ -1836,12 +1879,16 @@ void ieee80211_sta_work(struct ieee80211_sub_if_data *sdata) */ ifmgd->flags &= ~(IEEE80211_STA_CONNECTION_POLL | IEEE80211_STA_BEACON_POLL); - printk(KERN_DEBUG "No probe response from AP %pM" - " after %dms, disconnecting.\n", - bssid, (1000 * IEEE80211_PROBE_WAIT)/HZ); - ieee80211_set_disassoc(sdata, true); - ieee80211_recalc_idle(local); + wiphy_debug(local->hw.wiphy, + "%s: No probe response from AP %pM" + " after %dms, disconnecting.\n", + sdata->name, + bssid, (1000 * IEEE80211_PROBE_WAIT)/HZ); + ieee80211_set_disassoc(sdata, true, true); mutex_unlock(&ifmgd->mtx); + mutex_lock(&local->mtx); + ieee80211_recalc_idle(local); + mutex_unlock(&local->mtx); /* * must be outside lock due to cfg80211, * but that's not a problem. @@ -1917,6 +1964,8 @@ void ieee80211_sta_quiesce(struct ieee80211_sub_if_data *sdata) * time -- the code here is properly synchronised. */ + cancel_work_sync(&ifmgd->request_smps_work); + cancel_work_sync(&ifmgd->beacon_connection_loss_work); if (del_timer_sync(&ifmgd->timer)) set_bit(TMR_RUNNING_TIMER, &ifmgd->timers_running); @@ -1952,6 +2001,7 @@ void ieee80211_sta_setup_sdata(struct ieee80211_sub_if_data *sdata) INIT_WORK(&ifmgd->chswitch_work, ieee80211_chswitch_work); INIT_WORK(&ifmgd->beacon_connection_loss_work, ieee80211_beacon_connection_loss_work); + INIT_WORK(&ifmgd->request_smps_work, ieee80211_request_smps_work); setup_timer(&ifmgd->timer, ieee80211_sta_timer, (unsigned long) sdata); setup_timer(&ifmgd->bcn_mon_timer, ieee80211_sta_bcn_mon_timer, @@ -2158,7 +2208,7 @@ int ieee80211_mgd_assoc(struct ieee80211_sub_if_data *sdata, } /* Trying to reassociate - clear previous association state */ - ieee80211_set_disassoc(sdata, true); + ieee80211_set_disassoc(sdata, true, false); } mutex_unlock(&ifmgd->mtx); @@ -2169,6 +2219,8 @@ int ieee80211_mgd_assoc(struct ieee80211_sub_if_data *sdata, ifmgd->flags &= ~IEEE80211_STA_DISABLE_11N; ifmgd->flags &= ~IEEE80211_STA_NULLFUNC_ACKED; + ifmgd->beacon_crc_valid = false; + for (i = 0; i < req->crypto.n_ciphers_pairwise; i++) if (req->crypto.ciphers_pairwise[i] == WLAN_CIPHER_SUITE_WEP40 || req->crypto.ciphers_pairwise[i] == WLAN_CIPHER_SUITE_TKIP || @@ -2249,6 +2301,9 @@ int ieee80211_mgd_assoc(struct ieee80211_sub_if_data *sdata, else ifmgd->flags &= ~IEEE80211_STA_CONTROL_PORT; + sdata->control_port_protocol = req->crypto.control_port_ethertype; + sdata->control_port_no_encrypt = req->crypto.control_port_no_encrypt; + ieee80211_add_work(wk); return 0; } @@ -2267,7 +2322,7 @@ int ieee80211_mgd_deauth(struct ieee80211_sub_if_data *sdata, memcpy(bssid, req->bss->bssid, ETH_ALEN); if (ifmgd->associated == req->bss) { - ieee80211_set_disassoc(sdata, false); + ieee80211_set_disassoc(sdata, false, true); mutex_unlock(&ifmgd->mtx); assoc_bss = true; } else { @@ -2275,7 +2330,7 @@ int ieee80211_mgd_deauth(struct ieee80211_sub_if_data *sdata, mutex_unlock(&ifmgd->mtx); - mutex_lock(&local->work_mtx); + mutex_lock(&local->mtx); list_for_each_entry(wk, &local->work_list, list) { if (wk->sdata != sdata) continue; @@ -2294,7 +2349,7 @@ int ieee80211_mgd_deauth(struct ieee80211_sub_if_data *sdata, free_work(wk); break; } - mutex_unlock(&local->work_mtx); + mutex_unlock(&local->mtx); /* * If somebody requests authentication and we haven't @@ -2319,7 +2374,9 @@ int ieee80211_mgd_deauth(struct ieee80211_sub_if_data *sdata, if (assoc_bss) sta_info_destroy_addr(sdata, bssid); + mutex_lock(&sdata->local->mtx); ieee80211_recalc_idle(sdata->local); + mutex_unlock(&sdata->local->mtx); return 0; } @@ -2348,7 +2405,7 @@ int ieee80211_mgd_disassoc(struct ieee80211_sub_if_data *sdata, sdata->name, req->bss->bssid, req->reason_code); memcpy(bssid, req->bss->bssid, ETH_ALEN); - ieee80211_set_disassoc(sdata, false); + ieee80211_set_disassoc(sdata, false, true); mutex_unlock(&ifmgd->mtx); @@ -2357,7 +2414,9 @@ int ieee80211_mgd_disassoc(struct ieee80211_sub_if_data *sdata, cookie, !req->local_state_change); sta_info_destroy_addr(sdata, bssid); + mutex_lock(&sdata->local->mtx); ieee80211_recalc_idle(sdata->local); + mutex_unlock(&sdata->local->mtx); return 0; } diff --git a/net/mac80211/offchannel.c b/net/mac80211/offchannel.c index c36b1911987..4b564091e51 100644 --- a/net/mac80211/offchannel.c +++ b/net/mac80211/offchannel.c @@ -22,12 +22,16 @@ static void ieee80211_offchannel_ps_enable(struct ieee80211_sub_if_data *sdata) { struct ieee80211_local *local = sdata->local; + struct ieee80211_if_managed *ifmgd = &sdata->u.mgd; local->offchannel_ps_enabled = false; /* FIXME: what to do when local->pspolling is true? */ del_timer_sync(&local->dynamic_ps_timer); + del_timer_sync(&ifmgd->bcn_mon_timer); + del_timer_sync(&ifmgd->conn_mon_timer); + cancel_work_sync(&local->dynamic_ps_enable_work); if (local->hw.conf.flags & IEEE80211_CONF_PS) { @@ -85,6 +89,9 @@ static void ieee80211_offchannel_ps_disable(struct ieee80211_sub_if_data *sdata) mod_timer(&local->dynamic_ps_timer, jiffies + msecs_to_jiffies(local->hw.conf.dynamic_ps_timeout)); } + + ieee80211_sta_reset_beacon_monitor(sdata); + ieee80211_sta_reset_conn_monitor(sdata); } void ieee80211_offchannel_stop_beaconing(struct ieee80211_local *local) @@ -112,8 +119,10 @@ void ieee80211_offchannel_stop_beaconing(struct ieee80211_local *local) * used from user space controlled off-channel operations. */ if (sdata->vif.type != NL80211_IFTYPE_STATION && - sdata->vif.type != NL80211_IFTYPE_MONITOR) + sdata->vif.type != NL80211_IFTYPE_MONITOR) { + set_bit(SDATA_STATE_OFFCHANNEL, &sdata->state); netif_tx_stop_all_queues(sdata->dev); + } } mutex_unlock(&local->iflist_mtx); } @@ -131,6 +140,7 @@ void ieee80211_offchannel_stop_station(struct ieee80211_local *local) continue; if (sdata->vif.type == NL80211_IFTYPE_STATION) { + set_bit(SDATA_STATE_OFFCHANNEL, &sdata->state); netif_tx_stop_all_queues(sdata->dev); if (sdata->u.mgd.associated) ieee80211_offchannel_ps_enable(sdata); @@ -155,8 +165,20 @@ void ieee80211_offchannel_return(struct ieee80211_local *local, ieee80211_offchannel_ps_disable(sdata); } - if (sdata->vif.type != NL80211_IFTYPE_MONITOR) + if (sdata->vif.type != NL80211_IFTYPE_MONITOR) { + clear_bit(SDATA_STATE_OFFCHANNEL, &sdata->state); + /* + * This may wake up queues even though the driver + * currently has them stopped. This is not very + * likely, since the driver won't have gotten any + * (or hardly any) new packets while we weren't + * on the right channel, and even if it happens + * it will at most lead to queueing up one more + * packet per queue in mac80211 rather than on + * the interface qdisc. + */ netif_tx_wake_all_queues(sdata->dev); + } /* re-enable beaconing */ if (enable_beaconing && diff --git a/net/mac80211/pm.c b/net/mac80211/pm.c index d287fde0431..e37355193ed 100644 --- a/net/mac80211/pm.c +++ b/net/mac80211/pm.c @@ -45,7 +45,7 @@ int __ieee80211_suspend(struct ieee80211_hw *hw) list_for_each_entry(sta, &local->sta_list, list) { if (hw->flags & IEEE80211_HW_AMPDU_AGGREGATION) { set_sta_flags(sta, WLAN_STA_BLOCK_BA); - ieee80211_sta_tear_down_BA_sessions(sta); + ieee80211_sta_tear_down_BA_sessions(sta, true); } if (sta->uploaded) { diff --git a/net/mac80211/rate.c b/net/mac80211/rate.c index 6d0bd198af1..809cf230d25 100644 --- a/net/mac80211/rate.c +++ b/net/mac80211/rate.c @@ -103,6 +103,7 @@ ieee80211_rate_control_ops_get(const char *name) struct rate_control_ops *ops; const char *alg_name; + kparam_block_sysfs_write(ieee80211_default_rc_algo); if (!name) alg_name = ieee80211_default_rc_algo; else @@ -120,6 +121,7 @@ ieee80211_rate_control_ops_get(const char *name) /* try built-in one if specific alg requested but not found */ if (!ops && strlen(CONFIG_MAC80211_RC_DEFAULT)) ops = ieee80211_try_rate_control_ops_get(CONFIG_MAC80211_RC_DEFAULT); + kparam_unblock_sysfs_write(ieee80211_default_rc_algo); return ops; } @@ -143,6 +145,7 @@ static ssize_t rcname_read(struct file *file, char __user *userbuf, static const struct file_operations rcname_ops = { .read = rcname_read, .open = mac80211_open_file_generic, + .llseek = default_llseek, }; #endif @@ -205,7 +208,7 @@ static bool rc_no_data_or_no_ack(struct ieee80211_tx_rate_control *txrc) fc = hdr->frame_control; - return ((info->flags & IEEE80211_TX_CTL_NO_ACK) || !ieee80211_is_data(fc)); + return (info->flags & IEEE80211_TX_CTL_NO_ACK) || !ieee80211_is_data(fc); } static void rc_send_low_broadcast(s8 *idx, u32 basic_rates, u8 max_rate_idx) @@ -366,8 +369,8 @@ int ieee80211_init_rate_ctrl_alg(struct ieee80211_local *local, ref = rate_control_alloc(name, local); if (!ref) { - printk(KERN_WARNING "%s: Failed to select rate control " - "algorithm\n", wiphy_name(local->hw.wiphy)); + wiphy_warn(local->hw.wiphy, + "Failed to select rate control algorithm\n"); return -ENOENT; } @@ -378,9 +381,8 @@ int ieee80211_init_rate_ctrl_alg(struct ieee80211_local *local, sta_info_flush(local, NULL); } - printk(KERN_DEBUG "%s: Selected rate control " - "algorithm '%s'\n", wiphy_name(local->hw.wiphy), - ref->ops->name); + wiphy_debug(local->hw.wiphy, "Selected rate control algorithm '%s'\n", + ref->ops->name); return 0; } diff --git a/net/mac80211/rc80211_minstrel_debugfs.c b/net/mac80211/rc80211_minstrel_debugfs.c index 241e76f3fdf..a290ad231d7 100644 --- a/net/mac80211/rc80211_minstrel_debugfs.c +++ b/net/mac80211/rc80211_minstrel_debugfs.c @@ -122,6 +122,7 @@ static const struct file_operations minstrel_stat_fops = { .open = minstrel_stats_open, .read = minstrel_stats_read, .release = minstrel_stats_release, + .llseek = default_llseek, }; void diff --git a/net/mac80211/rc80211_minstrel_ht.c b/net/mac80211/rc80211_minstrel_ht.c index c5b465904e3..2a18d6602d4 100644 --- a/net/mac80211/rc80211_minstrel_ht.c +++ b/net/mac80211/rc80211_minstrel_ht.c @@ -397,8 +397,9 @@ minstrel_ht_tx_status(void *priv, struct ieee80211_supported_band *sband, !(info->flags & IEEE80211_TX_STAT_AMPDU)) return; - if (!info->status.ampdu_len) { - info->status.ampdu_ack_len = 1; + if (!(info->flags & IEEE80211_TX_STAT_AMPDU)) { + info->status.ampdu_ack_len = + (info->flags & IEEE80211_TX_STAT_ACK ? 1 : 0); info->status.ampdu_len = 1; } @@ -426,7 +427,7 @@ minstrel_ht_tx_status(void *priv, struct ieee80211_supported_band *sband, group = minstrel_ht_get_group_idx(&ar[i]); rate = &mi->groups[group].rates[ar[i].idx % 8]; - if (last && (info->flags & IEEE80211_TX_STAT_ACK)) + if (last) rate->success += info->status.ampdu_ack_len; rate->attempts += ar[i].count * info->status.ampdu_len; diff --git a/net/mac80211/rc80211_minstrel_ht_debugfs.c b/net/mac80211/rc80211_minstrel_ht_debugfs.c index 4a5a4b3e779..cefcb5d2dae 100644 --- a/net/mac80211/rc80211_minstrel_ht_debugfs.c +++ b/net/mac80211/rc80211_minstrel_ht_debugfs.c @@ -90,7 +90,7 @@ minstrel_ht_stats_open(struct inode *inode, struct file *file) MINSTREL_TRUNC(mi->avg_ampdu_len * 10) % 10); ms->len = p - ms->buf; - return 0; + return nonseekable_open(inode, file); } static const struct file_operations minstrel_ht_stat_fops = { @@ -98,6 +98,7 @@ static const struct file_operations minstrel_ht_stat_fops = { .open = minstrel_ht_stats_open, .read = minstrel_stats_read, .release = minstrel_stats_release, + .llseek = no_llseek, }; void diff --git a/net/mac80211/rc80211_pid_debugfs.c b/net/mac80211/rc80211_pid_debugfs.c index 47438b4a9af..4851e9e2dae 100644 --- a/net/mac80211/rc80211_pid_debugfs.c +++ b/net/mac80211/rc80211_pid_debugfs.c @@ -162,7 +162,7 @@ static ssize_t rate_control_pid_events_read(struct file *file, char __user *buf, file_info->next_entry = (file_info->next_entry + 1) % RC_PID_EVENT_RING_SIZE; - /* Print information about the event. Note that userpace needs to + /* Print information about the event. Note that userspace needs to * provide large enough buffers. */ length = length < RC_PID_PRINT_BUF_SIZE ? length : RC_PID_PRINT_BUF_SIZE; @@ -206,6 +206,7 @@ static const struct file_operations rc_pid_fop_events = { .poll = rate_control_pid_events_poll, .open = rate_control_pid_events_open, .release = rate_control_pid_events_release, + .llseek = noop_llseek, }; void rate_control_pid_add_sta_debugfs(void *priv, void *priv_sta, diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index fa0f37e4afe..902b03ee8f6 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -315,6 +315,7 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb, static void ieee80211_parse_qos(struct ieee80211_rx_data *rx) { struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)rx->skb->data; + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb); int tid; /* does the frame have a qos control field? */ @@ -323,9 +324,7 @@ static void ieee80211_parse_qos(struct ieee80211_rx_data *rx) /* frame has qos control */ tid = *qc & IEEE80211_QOS_CTL_TID_MASK; if (*qc & IEEE80211_QOS_CONTROL_A_MSDU_PRESENT) - rx->flags |= IEEE80211_RX_AMSDU; - else - rx->flags &= ~IEEE80211_RX_AMSDU; + status->rx_flags |= IEEE80211_RX_AMSDU; } else { /* * IEEE 802.11-2007, 7.1.3.4.1 ("Sequence Number field"): @@ -387,26 +386,25 @@ static ieee80211_rx_result debug_noinline ieee80211_rx_h_passive_scan(struct ieee80211_rx_data *rx) { struct ieee80211_local *local = rx->local; + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb); struct sk_buff *skb = rx->skb; - if (unlikely(test_bit(SCAN_HW_SCANNING, &local->scanning))) + if (likely(!(status->rx_flags & IEEE80211_RX_IN_SCAN))) + return RX_CONTINUE; + + if (test_bit(SCAN_HW_SCANNING, &local->scanning)) return ieee80211_scan_rx(rx->sdata, skb); - if (unlikely(test_bit(SCAN_SW_SCANNING, &local->scanning) && - (rx->flags & IEEE80211_RX_IN_SCAN))) { + if (test_bit(SCAN_SW_SCANNING, &local->scanning)) { /* drop all the other packets during a software scan anyway */ if (ieee80211_scan_rx(rx->sdata, skb) != RX_QUEUED) dev_kfree_skb(skb); return RX_QUEUED; } - if (unlikely(rx->flags & IEEE80211_RX_IN_SCAN)) { - /* scanning finished during invoking of handlers */ - I802_DEBUG_INC(local->rx_handlers_drop_passive_scan); - return RX_DROP_UNUSABLE; - } - - return RX_CONTINUE; + /* scanning finished during invoking of handlers */ + I802_DEBUG_INC(local->rx_handlers_drop_passive_scan); + return RX_DROP_UNUSABLE; } @@ -538,20 +536,12 @@ static void ieee80211_release_reorder_frame(struct ieee80211_hw *hw, int index, struct sk_buff_head *frames) { - struct ieee80211_supported_band *sband; - struct ieee80211_rate *rate = NULL; struct sk_buff *skb = tid_agg_rx->reorder_buf[index]; - struct ieee80211_rx_status *status; if (!skb) goto no_frame; - status = IEEE80211_SKB_RXCB(skb); - - /* release the reordered frames to stack */ - sband = hw->wiphy->bands[status->band]; - if (!(status->flag & RX_FLAG_HT)) - rate = &sband->bitrates[status->rate_idx]; + /* release the frame from the reorder ring buffer */ tid_agg_rx->stored_mpdu_num--; tid_agg_rx->reorder_buf[index] = NULL; __skb_queue_tail(frames, skb); @@ -580,9 +570,102 @@ static void ieee80211_release_reorder_frames(struct ieee80211_hw *hw, * frames that have not yet been received are assumed to be lost and the skb * can be released for processing. This may also release other skb's from the * reorder buffer if there are no additional gaps between the frames. + * + * Callers must hold tid_agg_rx->reorder_lock. */ #define HT_RX_REORDER_BUF_TIMEOUT (HZ / 10) +static void ieee80211_sta_reorder_release(struct ieee80211_hw *hw, + struct tid_ampdu_rx *tid_agg_rx, + struct sk_buff_head *frames) +{ + int index, j; + + /* release the buffer until next missing frame */ + index = seq_sub(tid_agg_rx->head_seq_num, tid_agg_rx->ssn) % + tid_agg_rx->buf_size; + if (!tid_agg_rx->reorder_buf[index] && + tid_agg_rx->stored_mpdu_num > 1) { + /* + * No buffers ready to be released, but check whether any + * frames in the reorder buffer have timed out. + */ + int skipped = 1; + for (j = (index + 1) % tid_agg_rx->buf_size; j != index; + j = (j + 1) % tid_agg_rx->buf_size) { + if (!tid_agg_rx->reorder_buf[j]) { + skipped++; + continue; + } + if (!time_after(jiffies, tid_agg_rx->reorder_time[j] + + HT_RX_REORDER_BUF_TIMEOUT)) + goto set_release_timer; + +#ifdef CONFIG_MAC80211_HT_DEBUG + if (net_ratelimit()) + wiphy_debug(hw->wiphy, + "release an RX reorder frame due to timeout on earlier frames\n"); +#endif + ieee80211_release_reorder_frame(hw, tid_agg_rx, + j, frames); + + /* + * Increment the head seq# also for the skipped slots. + */ + tid_agg_rx->head_seq_num = + (tid_agg_rx->head_seq_num + skipped) & SEQ_MASK; + skipped = 0; + } + } else while (tid_agg_rx->reorder_buf[index]) { + ieee80211_release_reorder_frame(hw, tid_agg_rx, index, frames); + index = seq_sub(tid_agg_rx->head_seq_num, tid_agg_rx->ssn) % + tid_agg_rx->buf_size; + } + + /* + * Disable the reorder release timer for now. + * + * The current implementation lacks a proper locking scheme + * which would protect vital statistic and debug counters + * from being updated by two different but concurrent BHs. + * + * More information about the topic is available from: + * - thread: http://marc.info/?t=128635927000001 + * + * What was wrong: + * => http://marc.info/?l=linux-wireless&m=128636170811964 + * "Basically the thing is that until your patch, the data + * in the struct didn't actually need locking because it + * was accessed by the RX path only which is not concurrent." + * + * List of what needs to be fixed: + * => http://marc.info/?l=linux-wireless&m=128656352920957 + * + + if (tid_agg_rx->stored_mpdu_num) { + j = index = seq_sub(tid_agg_rx->head_seq_num, + tid_agg_rx->ssn) % tid_agg_rx->buf_size; + + for (; j != (index - 1) % tid_agg_rx->buf_size; + j = (j + 1) % tid_agg_rx->buf_size) { + if (tid_agg_rx->reorder_buf[j]) + break; + } + + set_release_timer: + + mod_timer(&tid_agg_rx->reorder_timer, + tid_agg_rx->reorder_time[j] + + HT_RX_REORDER_BUF_TIMEOUT); + } else { + del_timer(&tid_agg_rx->reorder_timer); + } + */ + +set_release_timer: + return; +} + /* * As this function belongs to the RX path it must be under * rcu_read_lock protection. It returns false if the frame @@ -598,14 +681,16 @@ static bool ieee80211_sta_manage_reorder_buf(struct ieee80211_hw *hw, u16 mpdu_seq_num = (sc & IEEE80211_SCTL_SEQ) >> 4; u16 head_seq_num, buf_size; int index; + bool ret = true; buf_size = tid_agg_rx->buf_size; head_seq_num = tid_agg_rx->head_seq_num; + spin_lock(&tid_agg_rx->reorder_lock); /* frame with out of date sequence number */ if (seq_less(mpdu_seq_num, head_seq_num)) { dev_kfree_skb(skb); - return true; + goto out; } /* @@ -626,7 +711,7 @@ static bool ieee80211_sta_manage_reorder_buf(struct ieee80211_hw *hw, /* check if we already stored this frame */ if (tid_agg_rx->reorder_buf[index]) { dev_kfree_skb(skb); - return true; + goto out; } /* @@ -636,58 +721,19 @@ static bool ieee80211_sta_manage_reorder_buf(struct ieee80211_hw *hw, if (mpdu_seq_num == tid_agg_rx->head_seq_num && tid_agg_rx->stored_mpdu_num == 0) { tid_agg_rx->head_seq_num = seq_inc(tid_agg_rx->head_seq_num); - return false; + ret = false; + goto out; } /* put the frame in the reordering buffer */ tid_agg_rx->reorder_buf[index] = skb; tid_agg_rx->reorder_time[index] = jiffies; tid_agg_rx->stored_mpdu_num++; - /* release the buffer until next missing frame */ - index = seq_sub(tid_agg_rx->head_seq_num, tid_agg_rx->ssn) % - tid_agg_rx->buf_size; - if (!tid_agg_rx->reorder_buf[index] && - tid_agg_rx->stored_mpdu_num > 1) { - /* - * No buffers ready to be released, but check whether any - * frames in the reorder buffer have timed out. - */ - int j; - int skipped = 1; - for (j = (index + 1) % tid_agg_rx->buf_size; j != index; - j = (j + 1) % tid_agg_rx->buf_size) { - if (!tid_agg_rx->reorder_buf[j]) { - skipped++; - continue; - } - if (!time_after(jiffies, tid_agg_rx->reorder_time[j] + - HT_RX_REORDER_BUF_TIMEOUT)) - break; + ieee80211_sta_reorder_release(hw, tid_agg_rx, frames); -#ifdef CONFIG_MAC80211_HT_DEBUG - if (net_ratelimit()) - printk(KERN_DEBUG "%s: release an RX reorder " - "frame due to timeout on earlier " - "frames\n", - wiphy_name(hw->wiphy)); -#endif - ieee80211_release_reorder_frame(hw, tid_agg_rx, - j, frames); - - /* - * Increment the head seq# also for the skipped slots. - */ - tid_agg_rx->head_seq_num = - (tid_agg_rx->head_seq_num + skipped) & SEQ_MASK; - skipped = 0; - } - } else while (tid_agg_rx->reorder_buf[index]) { - ieee80211_release_reorder_frame(hw, tid_agg_rx, index, frames); - index = seq_sub(tid_agg_rx->head_seq_num, tid_agg_rx->ssn) % - tid_agg_rx->buf_size; - } - - return true; + out: + spin_unlock(&tid_agg_rx->reorder_lock); + return ret; } /* @@ -761,13 +807,14 @@ static ieee80211_rx_result debug_noinline ieee80211_rx_h_check(struct ieee80211_rx_data *rx) { struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)rx->skb->data; + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb); /* Drop duplicate 802.11 retransmissions (IEEE 802.11 Chap. 9.2.9) */ if (rx->sta && !is_multicast_ether_addr(hdr->addr1)) { if (unlikely(ieee80211_has_retry(hdr->frame_control) && rx->sta->last_seq_ctrl[rx->queue] == hdr->seq_ctrl)) { - if (rx->flags & IEEE80211_RX_RA_MATCH) { + if (status->rx_flags & IEEE80211_RX_RA_MATCH) { rx->local->dot11FrameDuplicateCount++; rx->sta->num_duplicates++; } @@ -796,11 +843,12 @@ ieee80211_rx_h_check(struct ieee80211_rx_data *rx) if (unlikely((ieee80211_is_data(hdr->frame_control) || ieee80211_is_pspoll(hdr->frame_control)) && rx->sdata->vif.type != NL80211_IFTYPE_ADHOC && + rx->sdata->vif.type != NL80211_IFTYPE_WDS && (!rx->sta || !test_sta_flags(rx->sta, WLAN_STA_ASSOC)))) { if ((!ieee80211_has_fromds(hdr->frame_control) && !ieee80211_has_tods(hdr->frame_control) && ieee80211_is_data(hdr->frame_control)) || - !(rx->flags & IEEE80211_RX_RA_MATCH)) { + !(status->rx_flags & IEEE80211_RX_RA_MATCH)) { /* Drop IBSS frames and frames for other hosts * silently. */ return RX_DROP_MONITOR; @@ -822,7 +870,7 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx) int keyidx; int hdrlen; ieee80211_rx_result result = RX_DROP_UNUSABLE; - struct ieee80211_key *stakey = NULL; + struct ieee80211_key *sta_ptk = NULL; int mmie_keyidx = -1; __le16 fc; @@ -857,22 +905,25 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx) * No point in finding a key and decrypting if the frame is neither * addressed to us nor a multicast frame. */ - if (!(rx->flags & IEEE80211_RX_RA_MATCH)) + if (!(status->rx_flags & IEEE80211_RX_RA_MATCH)) return RX_CONTINUE; /* start without a key */ rx->key = NULL; if (rx->sta) - stakey = rcu_dereference(rx->sta->key); + sta_ptk = rcu_dereference(rx->sta->ptk); fc = hdr->frame_control; if (!ieee80211_has_protected(fc)) mmie_keyidx = ieee80211_get_mmie_keyidx(rx->skb); - if (!is_multicast_ether_addr(hdr->addr1) && stakey) { - rx->key = stakey; + if (!is_multicast_ether_addr(hdr->addr1) && sta_ptk) { + rx->key = sta_ptk; + if ((status->flag & RX_FLAG_DECRYPTED) && + (status->flag & RX_FLAG_IV_STRIPPED)) + return RX_CONTINUE; /* Skip decryption if the frame is not protected. */ if (!ieee80211_has_protected(fc)) return RX_CONTINUE; @@ -885,7 +936,10 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx) if (mmie_keyidx < NUM_DEFAULT_KEYS || mmie_keyidx >= NUM_DEFAULT_KEYS + NUM_DEFAULT_MGMT_KEYS) return RX_DROP_MONITOR; /* unexpected BIP keyidx */ - rx->key = rcu_dereference(rx->sdata->keys[mmie_keyidx]); + if (rx->sta) + rx->key = rcu_dereference(rx->sta->gtk[mmie_keyidx]); + if (!rx->key) + rx->key = rcu_dereference(rx->sdata->keys[mmie_keyidx]); } else if (!ieee80211_has_protected(fc)) { /* * The frame was not protected, so skip decryption. However, we @@ -928,16 +982,25 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx) skb_copy_bits(rx->skb, hdrlen + 3, &keyid, 1); keyidx = keyid >> 6; - rx->key = rcu_dereference(rx->sdata->keys[keyidx]); + /* check per-station GTK first, if multicast packet */ + if (is_multicast_ether_addr(hdr->addr1) && rx->sta) + rx->key = rcu_dereference(rx->sta->gtk[keyidx]); - /* - * RSNA-protected unicast frames should always be sent with - * pairwise or station-to-station keys, but for WEP we allow - * using a key index as well. - */ - if (rx->key && rx->key->conf.alg != ALG_WEP && - !is_multicast_ether_addr(hdr->addr1)) - rx->key = NULL; + /* if not found, try default key */ + if (!rx->key) { + rx->key = rcu_dereference(rx->sdata->keys[keyidx]); + + /* + * RSNA-protected unicast frames should always be + * sent with pairwise or station-to-station keys, + * but for WEP we allow using a key index as well. + */ + if (rx->key && + rx->key->conf.cipher != WLAN_CIPHER_SUITE_WEP40 && + rx->key->conf.cipher != WLAN_CIPHER_SUITE_WEP104 && + !is_multicast_ether_addr(hdr->addr1)) + rx->key = NULL; + } } if (rx->key) { @@ -951,8 +1014,9 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx) return RX_DROP_UNUSABLE; /* the hdr variable is invalid now! */ - switch (rx->key->conf.alg) { - case ALG_WEP: + switch (rx->key->conf.cipher) { + case WLAN_CIPHER_SUITE_WEP40: + case WLAN_CIPHER_SUITE_WEP104: /* Check for weak IVs if possible */ if (rx->sta && ieee80211_is_data(fc) && (!(status->flag & RX_FLAG_IV_STRIPPED) || @@ -962,15 +1026,21 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx) result = ieee80211_crypto_wep_decrypt(rx); break; - case ALG_TKIP: + case WLAN_CIPHER_SUITE_TKIP: result = ieee80211_crypto_tkip_decrypt(rx); break; - case ALG_CCMP: + case WLAN_CIPHER_SUITE_CCMP: result = ieee80211_crypto_ccmp_decrypt(rx); break; - case ALG_AES_CMAC: + case WLAN_CIPHER_SUITE_AES_CMAC: result = ieee80211_crypto_aes_cmac_decrypt(rx); break; + default: + /* + * We can reach here only with HW-only algorithms + * but why didn't it decrypt the frame?! + */ + return RX_DROP_UNUSABLE; } /* either the frame has been decrypted or will be dropped */ @@ -1079,7 +1149,7 @@ ieee80211_rx_h_sta_process(struct ieee80211_rx_data *rx) sta->last_rx = jiffies; } - if (!(rx->flags & IEEE80211_RX_RA_MATCH)) + if (!(status->rx_flags & IEEE80211_RX_RA_MATCH)) return RX_CONTINUE; if (rx->sdata->vif.type == NL80211_IFTYPE_STATION) @@ -1236,6 +1306,7 @@ ieee80211_rx_h_defragment(struct ieee80211_rx_data *rx) unsigned int frag, seq; struct ieee80211_fragment_entry *entry; struct sk_buff *skb; + struct ieee80211_rx_status *status; hdr = (struct ieee80211_hdr *)rx->skb->data; fc = hdr->frame_control; @@ -1265,7 +1336,7 @@ ieee80211_rx_h_defragment(struct ieee80211_rx_data *rx) /* This is the first fragment of a new frame. */ entry = ieee80211_reassemble_add(rx->sdata, frag, seq, rx->queue, &(rx->skb)); - if (rx->key && rx->key->conf.alg == ALG_CCMP && + if (rx->key && rx->key->conf.cipher == WLAN_CIPHER_SUITE_CCMP && ieee80211_has_protected(fc)) { int queue = ieee80211_is_mgmt(fc) ? NUM_RX_DATA_QUEUES : rx->queue; @@ -1294,7 +1365,7 @@ ieee80211_rx_h_defragment(struct ieee80211_rx_data *rx) int i; u8 pn[CCMP_PN_LEN], *rpn; int queue; - if (!rx->key || rx->key->conf.alg != ALG_CCMP) + if (!rx->key || rx->key->conf.cipher != WLAN_CIPHER_SUITE_CCMP) return RX_DROP_UNUSABLE; memcpy(pn, entry->last_pn, CCMP_PN_LEN); for (i = CCMP_PN_LEN - 1; i >= 0; i--) { @@ -1335,7 +1406,8 @@ ieee80211_rx_h_defragment(struct ieee80211_rx_data *rx) } /* Complete frame has been reassembled - process it now */ - rx->flags |= IEEE80211_RX_FRAGMENTED; + status = IEEE80211_SKB_RXCB(rx->skb); + status->rx_flags |= IEEE80211_RX_FRAGMENTED; out: if (rx->sta) @@ -1352,9 +1424,10 @@ ieee80211_rx_h_ps_poll(struct ieee80211_rx_data *rx) { struct ieee80211_sub_if_data *sdata = rx->sdata; __le16 fc = ((struct ieee80211_hdr *)rx->skb->data)->frame_control; + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb); if (likely(!rx->sta || !ieee80211_is_pspoll(fc) || - !(rx->flags & IEEE80211_RX_RA_MATCH))) + !(status->rx_flags & IEEE80211_RX_RA_MATCH))) return RX_CONTINUE; if ((sdata->vif.type != NL80211_IFTYPE_AP) && @@ -1492,7 +1565,7 @@ static bool ieee80211_frame_allowed(struct ieee80211_rx_data *rx, __le16 fc) * Allow EAPOL frames to us/the PAE group address regardless * of whether the frame was encrypted or not. */ - if (ehdr->h_proto == htons(ETH_P_PAE) && + if (ehdr->h_proto == rx->sdata->control_port_protocol && (compare_ether_addr(ehdr->h_dest, rx->sdata->vif.addr) == 0 || compare_ether_addr(ehdr->h_dest, pae_group_addr) == 0)) return true; @@ -1515,6 +1588,7 @@ ieee80211_deliver_skb(struct ieee80211_rx_data *rx) struct sk_buff *skb, *xmit_skb; struct ethhdr *ehdr = (struct ethhdr *) rx->skb->data; struct sta_info *dsta; + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb); skb = rx->skb; xmit_skb = NULL; @@ -1522,7 +1596,7 @@ ieee80211_deliver_skb(struct ieee80211_rx_data *rx) if ((sdata->vif.type == NL80211_IFTYPE_AP || sdata->vif.type == NL80211_IFTYPE_AP_VLAN) && !(sdata->flags & IEEE80211_SDATA_DONT_BRIDGE_PACKETS) && - (rx->flags & IEEE80211_RX_RA_MATCH) && + (status->rx_flags & IEEE80211_RX_RA_MATCH) && (sdata->vif.type != NL80211_IFTYPE_AP_VLAN || !sdata->u.vlan.sta)) { if (is_multicast_ether_addr(ehdr->h_dest)) { /* @@ -1599,6 +1673,7 @@ ieee80211_rx_h_amsdu(struct ieee80211_rx_data *rx) struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data; __le16 fc = hdr->frame_control; struct sk_buff_head frame_list; + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb); if (unlikely(!ieee80211_is_data(fc))) return RX_CONTINUE; @@ -1606,7 +1681,7 @@ ieee80211_rx_h_amsdu(struct ieee80211_rx_data *rx) if (unlikely(!ieee80211_is_data_present(fc))) return RX_DROP_MONITOR; - if (!(rx->flags & IEEE80211_RX_AMSDU)) + if (!(status->rx_flags & IEEE80211_RX_AMSDU)) return RX_CONTINUE; if (ieee80211_has_a4(hdr->frame_control) && @@ -1657,6 +1732,7 @@ ieee80211_rx_h_mesh_fwding(struct ieee80211_rx_data *rx) struct sk_buff *skb = rx->skb, *fwd_skb; struct ieee80211_local *local = rx->local; struct ieee80211_sub_if_data *sdata = rx->sdata; + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb); hdr = (struct ieee80211_hdr *) skb->data; hdrlen = ieee80211_hdrlen(hdr->frame_control); @@ -1702,7 +1778,7 @@ ieee80211_rx_h_mesh_fwding(struct ieee80211_rx_data *rx) mesh_hdr->ttl--; - if (rx->flags & IEEE80211_RX_RA_MATCH) { + if (status->rx_flags & IEEE80211_RX_RA_MATCH) { if (!mesh_hdr->ttl) IEEE80211_IFSTA_MESH_CTR_INC(&rx->sdata->u.mesh, dropped_frames_ttl); @@ -1909,13 +1985,38 @@ static void ieee80211_process_sa_query_req(struct ieee80211_sub_if_data *sdata, } static ieee80211_rx_result debug_noinline +ieee80211_rx_h_mgmt_check(struct ieee80211_rx_data *rx) +{ + struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *) rx->skb->data; + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb); + + /* + * From here on, look only at management frames. + * Data and control frames are already handled, + * and unknown (reserved) frames are useless. + */ + if (rx->skb->len < 24) + return RX_DROP_MONITOR; + + if (!ieee80211_is_mgmt(mgmt->frame_control)) + return RX_DROP_MONITOR; + + if (!(status->rx_flags & IEEE80211_RX_RA_MATCH)) + return RX_DROP_MONITOR; + + if (ieee80211_drop_unencrypted_mgmt(rx)) + return RX_DROP_UNUSABLE; + + return RX_CONTINUE; +} + +static ieee80211_rx_result debug_noinline ieee80211_rx_h_action(struct ieee80211_rx_data *rx) { struct ieee80211_local *local = rx->local; struct ieee80211_sub_if_data *sdata = rx->sdata; struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *) rx->skb->data; - struct sk_buff *nskb; - struct ieee80211_rx_status *status; + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb); int len = rx->skb->len; if (!ieee80211_is_action(mgmt->frame_control)) @@ -1928,10 +2029,7 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx) if (!rx->sta && mgmt->u.action.category != WLAN_CATEGORY_PUBLIC) return RX_DROP_UNUSABLE; - if (!(rx->flags & IEEE80211_RX_RA_MATCH)) - return RX_DROP_UNUSABLE; - - if (ieee80211_drop_unencrypted_mgmt(rx)) + if (!(status->rx_flags & IEEE80211_RX_RA_MATCH)) return RX_DROP_UNUSABLE; switch (mgmt->u.action.category) { @@ -2024,17 +2122,36 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx) goto queue; } + return RX_CONTINUE; + invalid: - /* - * For AP mode, hostapd is responsible for handling any action - * frames that we didn't handle, including returning unknown - * ones. For all other modes we will return them to the sender, - * setting the 0x80 bit in the action category, as required by - * 802.11-2007 7.3.1.11. - */ - if (sdata->vif.type == NL80211_IFTYPE_AP || - sdata->vif.type == NL80211_IFTYPE_AP_VLAN) - return RX_DROP_MONITOR; + status->rx_flags |= IEEE80211_RX_MALFORMED_ACTION_FRM; + /* will return in the next handlers */ + return RX_CONTINUE; + + handled: + if (rx->sta) + rx->sta->rx_packets++; + dev_kfree_skb(rx->skb); + return RX_QUEUED; + + queue: + rx->skb->pkt_type = IEEE80211_SDATA_QUEUE_TYPE_FRAME; + skb_queue_tail(&sdata->skb_queue, rx->skb); + ieee80211_queue_work(&local->hw, &sdata->work); + if (rx->sta) + rx->sta->rx_packets++; + return RX_QUEUED; +} + +static ieee80211_rx_result debug_noinline +ieee80211_rx_h_userspace_mgmt(struct ieee80211_rx_data *rx) +{ + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb); + + /* skip known-bad action frames and return them in the next handler */ + if (status->rx_flags & IEEE80211_RX_MALFORMED_ACTION_FRM) + return RX_CONTINUE; /* * Getting here means the kernel doesn't know how to handle @@ -2042,12 +2159,46 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx) * so userspace can register for those to know whether ones * it transmitted were processed or returned. */ - status = IEEE80211_SKB_RXCB(rx->skb); - if (cfg80211_rx_action(rx->sdata->dev, status->freq, - rx->skb->data, rx->skb->len, - GFP_ATOMIC)) - goto handled; + if (cfg80211_rx_mgmt(rx->sdata->dev, status->freq, + rx->skb->data, rx->skb->len, + GFP_ATOMIC)) { + if (rx->sta) + rx->sta->rx_packets++; + dev_kfree_skb(rx->skb); + return RX_QUEUED; + } + + + return RX_CONTINUE; +} + +static ieee80211_rx_result debug_noinline +ieee80211_rx_h_action_return(struct ieee80211_rx_data *rx) +{ + struct ieee80211_local *local = rx->local; + struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *) rx->skb->data; + struct sk_buff *nskb; + struct ieee80211_sub_if_data *sdata = rx->sdata; + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb); + + if (!ieee80211_is_action(mgmt->frame_control)) + return RX_CONTINUE; + + /* + * For AP mode, hostapd is responsible for handling any action + * frames that we didn't handle, including returning unknown + * ones. For all other modes we will return them to the sender, + * setting the 0x80 bit in the action category, as required by + * 802.11-2007 7.3.1.11. + * Newer versions of hostapd shall also use the management frame + * registration mechanisms, but older ones still use cooked + * monitor interfaces so push all frames there. + */ + if (!(status->rx_flags & IEEE80211_RX_MALFORMED_ACTION_FRM) && + (sdata->vif.type == NL80211_IFTYPE_AP || + sdata->vif.type == NL80211_IFTYPE_AP_VLAN)) + return RX_DROP_MONITOR; /* do not return rejected action frames */ if (mgmt->u.action.category & 0x80) @@ -2066,20 +2217,8 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx) ieee80211_tx_skb(rx->sdata, nskb); } - - handled: - if (rx->sta) - rx->sta->rx_packets++; dev_kfree_skb(rx->skb); return RX_QUEUED; - - queue: - rx->skb->pkt_type = IEEE80211_SDATA_QUEUE_TYPE_FRAME; - skb_queue_tail(&sdata->skb_queue, rx->skb); - ieee80211_queue_work(&local->hw, &sdata->work); - if (rx->sta) - rx->sta->rx_packets++; - return RX_QUEUED; } static ieee80211_rx_result debug_noinline @@ -2090,15 +2229,6 @@ ieee80211_rx_h_mgmt(struct ieee80211_rx_data *rx) struct ieee80211_mgmt *mgmt = (void *)rx->skb->data; __le16 stype; - if (!(rx->flags & IEEE80211_RX_RA_MATCH)) - return RX_DROP_MONITOR; - - if (rx->skb->len < 24) - return RX_DROP_MONITOR; - - if (ieee80211_drop_unencrypted_mgmt(rx)) - return RX_DROP_UNUSABLE; - rxs = ieee80211_work_rx_mgmt(rx->sdata, rx->skb); if (rxs != RX_CONTINUE) return rxs; @@ -2199,8 +2329,13 @@ static void ieee80211_rx_cooked_monitor(struct ieee80211_rx_data *rx, struct net_device *prev_dev = NULL; struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb); - if (status->flag & RX_FLAG_INTERNAL_CMTR) + /* + * If cooked monitor has been processed already, then + * don't do it again. If not, set the flag. + */ + if (rx->flags & IEEE80211_RX_CMNTR) goto out_free_skb; + rx->flags |= IEEE80211_RX_CMNTR; if (skb_headroom(skb) < sizeof(*rthdr) && pskb_expand_head(skb, sizeof(*rthdr), 0, GFP_ATOMIC)) @@ -2256,30 +2391,53 @@ static void ieee80211_rx_cooked_monitor(struct ieee80211_rx_data *rx, if (prev_dev) { skb->dev = prev_dev; netif_receive_skb(skb); - skb = NULL; - } else - goto out_free_skb; - - status->flag |= RX_FLAG_INTERNAL_CMTR; - return; + return; + } out_free_skb: dev_kfree_skb(skb); } +static void ieee80211_rx_handlers_result(struct ieee80211_rx_data *rx, + ieee80211_rx_result res) +{ + switch (res) { + case RX_DROP_MONITOR: + I802_DEBUG_INC(rx->sdata->local->rx_handlers_drop); + if (rx->sta) + rx->sta->rx_dropped++; + /* fall through */ + case RX_CONTINUE: { + struct ieee80211_rate *rate = NULL; + struct ieee80211_supported_band *sband; + struct ieee80211_rx_status *status; + + status = IEEE80211_SKB_RXCB((rx->skb)); + + sband = rx->local->hw.wiphy->bands[status->band]; + if (!(status->flag & RX_FLAG_HT)) + rate = &sband->bitrates[status->rate_idx]; + + ieee80211_rx_cooked_monitor(rx, rate); + break; + } + case RX_DROP_UNUSABLE: + I802_DEBUG_INC(rx->sdata->local->rx_handlers_drop); + if (rx->sta) + rx->sta->rx_dropped++; + dev_kfree_skb(rx->skb); + break; + case RX_QUEUED: + I802_DEBUG_INC(rx->sdata->local->rx_handlers_queued); + break; + } +} -static void ieee80211_invoke_rx_handlers(struct ieee80211_sub_if_data *sdata, - struct ieee80211_rx_data *rx, - struct sk_buff *skb, - struct ieee80211_rate *rate) +static void ieee80211_rx_handlers(struct ieee80211_rx_data *rx, + struct sk_buff_head *frames) { - struct sk_buff_head reorder_release; ieee80211_rx_result res = RX_DROP_MONITOR; - - __skb_queue_head_init(&reorder_release); - - rx->skb = skb; - rx->sdata = sdata; + struct sk_buff *skb; #define CALL_RXH(rxh) \ do { \ @@ -2288,23 +2446,14 @@ static void ieee80211_invoke_rx_handlers(struct ieee80211_sub_if_data *sdata, goto rxh_next; \ } while (0); - /* - * NB: the rxh_next label works even if we jump - * to it from here because then the list will - * be empty, which is a trivial check - */ - CALL_RXH(ieee80211_rx_h_passive_scan) - CALL_RXH(ieee80211_rx_h_check) - - ieee80211_rx_reorder_ampdu(rx, &reorder_release); - - while ((skb = __skb_dequeue(&reorder_release))) { + while ((skb = __skb_dequeue(frames))) { /* * all the other fields are valid across frames * that belong to an aMPDU since they are on the * same TID from the same station */ rx->skb = skb; + rx->flags = 0; CALL_RXH(ieee80211_rx_h_decrypt) CALL_RXH(ieee80211_rx_h_check_more_data) @@ -2316,50 +2465,92 @@ static void ieee80211_invoke_rx_handlers(struct ieee80211_sub_if_data *sdata, CALL_RXH(ieee80211_rx_h_remove_qos_control) CALL_RXH(ieee80211_rx_h_amsdu) #ifdef CONFIG_MAC80211_MESH - if (ieee80211_vif_is_mesh(&sdata->vif)) + if (ieee80211_vif_is_mesh(&rx->sdata->vif)) CALL_RXH(ieee80211_rx_h_mesh_fwding); #endif CALL_RXH(ieee80211_rx_h_data) /* special treatment -- needs the queue */ - res = ieee80211_rx_h_ctrl(rx, &reorder_release); + res = ieee80211_rx_h_ctrl(rx, frames); if (res != RX_CONTINUE) goto rxh_next; + CALL_RXH(ieee80211_rx_h_mgmt_check) CALL_RXH(ieee80211_rx_h_action) + CALL_RXH(ieee80211_rx_h_userspace_mgmt) + CALL_RXH(ieee80211_rx_h_action_return) CALL_RXH(ieee80211_rx_h_mgmt) + rxh_next: + ieee80211_rx_handlers_result(rx, res); + #undef CALL_RXH + } +} + +static void ieee80211_invoke_rx_handlers(struct ieee80211_rx_data *rx) +{ + struct sk_buff_head reorder_release; + ieee80211_rx_result res = RX_DROP_MONITOR; + + __skb_queue_head_init(&reorder_release); + +#define CALL_RXH(rxh) \ + do { \ + res = rxh(rx); \ + if (res != RX_CONTINUE) \ + goto rxh_next; \ + } while (0); + + CALL_RXH(ieee80211_rx_h_passive_scan) + CALL_RXH(ieee80211_rx_h_check) + + ieee80211_rx_reorder_ampdu(rx, &reorder_release); + + ieee80211_rx_handlers(rx, &reorder_release); + return; rxh_next: - switch (res) { - case RX_DROP_MONITOR: - I802_DEBUG_INC(sdata->local->rx_handlers_drop); - if (rx->sta) - rx->sta->rx_dropped++; - /* fall through */ - case RX_CONTINUE: - ieee80211_rx_cooked_monitor(rx, rate); - break; - case RX_DROP_UNUSABLE: - I802_DEBUG_INC(sdata->local->rx_handlers_drop); - if (rx->sta) - rx->sta->rx_dropped++; - dev_kfree_skb(rx->skb); - break; - case RX_QUEUED: - I802_DEBUG_INC(sdata->local->rx_handlers_queued); - break; - } - } + ieee80211_rx_handlers_result(rx, res); + +#undef CALL_RXH +} + +/* + * This function makes calls into the RX path. Therefore the + * caller must hold the sta_info->lock and everything has to + * be under rcu_read_lock protection as well. + */ +void ieee80211_release_reorder_timeout(struct sta_info *sta, int tid) +{ + struct sk_buff_head frames; + struct ieee80211_rx_data rx = { + .sta = sta, + .sdata = sta->sdata, + .local = sta->local, + .queue = tid, + }; + struct tid_ampdu_rx *tid_agg_rx; + + tid_agg_rx = rcu_dereference(sta->ampdu_mlme.tid_rx[tid]); + if (!tid_agg_rx) + return; + + __skb_queue_head_init(&frames); + + spin_lock(&tid_agg_rx->reorder_lock); + ieee80211_sta_reorder_release(&sta->local->hw, tid_agg_rx, &frames); + spin_unlock(&tid_agg_rx->reorder_lock); + + ieee80211_rx_handlers(&rx, &frames); } /* main receive path */ -static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata, - struct ieee80211_rx_data *rx, +static int prepare_for_handlers(struct ieee80211_rx_data *rx, struct ieee80211_hdr *hdr) { + struct ieee80211_sub_if_data *sdata = rx->sdata; struct sk_buff *skb = rx->skb; struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb); u8 *bssid = ieee80211_get_bssid(hdr, skb->len, sdata->vif.type); @@ -2373,7 +2564,7 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata, compare_ether_addr(sdata->vif.addr, hdr->addr1) != 0) { if (!(sdata->dev->flags & IFF_PROMISC)) return 0; - rx->flags &= ~IEEE80211_RX_RA_MATCH; + status->rx_flags &= ~IEEE80211_RX_RA_MATCH; } break; case NL80211_IFTYPE_ADHOC: @@ -2383,15 +2574,15 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata, return 1; } else if (!ieee80211_bssid_match(bssid, sdata->u.ibss.bssid)) { - if (!(rx->flags & IEEE80211_RX_IN_SCAN)) + if (!(status->rx_flags & IEEE80211_RX_IN_SCAN)) return 0; - rx->flags &= ~IEEE80211_RX_RA_MATCH; + status->rx_flags &= ~IEEE80211_RX_RA_MATCH; } else if (!multicast && compare_ether_addr(sdata->vif.addr, hdr->addr1) != 0) { if (!(sdata->dev->flags & IFF_PROMISC)) return 0; - rx->flags &= ~IEEE80211_RX_RA_MATCH; + status->rx_flags &= ~IEEE80211_RX_RA_MATCH; } else if (!rx->sta) { int rate_idx; if (status->flag & RX_FLAG_HT) @@ -2409,7 +2600,7 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata, if (!(sdata->dev->flags & IFF_PROMISC)) return 0; - rx->flags &= ~IEEE80211_RX_RA_MATCH; + status->rx_flags &= ~IEEE80211_RX_RA_MATCH; } break; case NL80211_IFTYPE_AP_VLAN: @@ -2420,9 +2611,9 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata, return 0; } else if (!ieee80211_bssid_match(bssid, sdata->vif.addr)) { - if (!(rx->flags & IEEE80211_RX_IN_SCAN)) + if (!(status->rx_flags & IEEE80211_RX_IN_SCAN)) return 0; - rx->flags &= ~IEEE80211_RX_RA_MATCH; + status->rx_flags &= ~IEEE80211_RX_RA_MATCH; } break; case NL80211_IFTYPE_WDS: @@ -2431,9 +2622,7 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata, if (compare_ether_addr(sdata->u.wds.remote_addr, hdr->addr2)) return 0; break; - case NL80211_IFTYPE_MONITOR: - case NL80211_IFTYPE_UNSPECIFIED: - case __NL80211_IFTYPE_AFTER_LAST: + default: /* should never get here */ WARN_ON(1); break; @@ -2443,12 +2632,56 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata, } /* + * This function returns whether or not the SKB + * was destined for RX processing or not, which, + * if consume is true, is equivalent to whether + * or not the skb was consumed. + */ +static bool ieee80211_prepare_and_rx_handle(struct ieee80211_rx_data *rx, + struct sk_buff *skb, bool consume) +{ + struct ieee80211_local *local = rx->local; + struct ieee80211_sub_if_data *sdata = rx->sdata; + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb); + struct ieee80211_hdr *hdr = (void *)skb->data; + int prepares; + + rx->skb = skb; + status->rx_flags |= IEEE80211_RX_RA_MATCH; + prepares = prepare_for_handlers(rx, hdr); + + if (!prepares) + return false; + + if (status->flag & RX_FLAG_MMIC_ERROR) { + if (status->rx_flags & IEEE80211_RX_RA_MATCH) + ieee80211_rx_michael_mic_report(hdr, rx); + return false; + } + + if (!consume) { + skb = skb_copy(skb, GFP_ATOMIC); + if (!skb) { + if (net_ratelimit()) + wiphy_debug(local->hw.wiphy, + "failed to copy multicast frame for %s\n", + sdata->name); + return true; + } + + rx->skb = skb; + } + + ieee80211_invoke_rx_handlers(rx); + return true; +} + +/* * This is the actual Rx frames handler. as it blongs to Rx path it must * be called with rcu_read_lock protection. */ static void __ieee80211_rx_handle_packet(struct ieee80211_hw *hw, - struct sk_buff *skb, - struct ieee80211_rate *rate) + struct sk_buff *skb) { struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb); struct ieee80211_local *local = hw_to_local(hw); @@ -2456,11 +2689,8 @@ static void __ieee80211_rx_handle_packet(struct ieee80211_hw *hw, struct ieee80211_hdr *hdr; __le16 fc; struct ieee80211_rx_data rx; - int prepares; - struct ieee80211_sub_if_data *prev = NULL; - struct sk_buff *skb_new; - struct sta_info *sta, *tmp; - bool found_sta = false; + struct ieee80211_sub_if_data *prev; + struct sta_info *sta, *tmp, *prev_sta; int err = 0; fc = ((struct ieee80211_hdr *)skb->data)->frame_control; @@ -2473,7 +2703,7 @@ static void __ieee80211_rx_handle_packet(struct ieee80211_hw *hw, if (unlikely(test_bit(SCAN_HW_SCANNING, &local->scanning) || test_bit(SCAN_OFF_CHANNEL, &local->scanning))) - rx.flags |= IEEE80211_RX_IN_SCAN; + status->rx_flags |= IEEE80211_RX_IN_SCAN; if (ieee80211_is_mgmt(fc)) err = skb_linearize(skb); @@ -2490,91 +2720,67 @@ static void __ieee80211_rx_handle_packet(struct ieee80211_hw *hw, ieee80211_verify_alignment(&rx); if (ieee80211_is_data(fc)) { + prev_sta = NULL; + for_each_sta_info(local, hdr->addr2, sta, tmp) { - rx.sta = sta; - found_sta = true; - rx.sdata = sta->sdata; - - rx.flags |= IEEE80211_RX_RA_MATCH; - prepares = prepare_for_handlers(rx.sdata, &rx, hdr); - if (prepares) { - if (status->flag & RX_FLAG_MMIC_ERROR) { - if (rx.flags & IEEE80211_RX_RA_MATCH) - ieee80211_rx_michael_mic_report(hdr, &rx); - } else - prev = rx.sdata; - } - } - } - if (!found_sta) { - list_for_each_entry_rcu(sdata, &local->interfaces, list) { - if (!ieee80211_sdata_running(sdata)) + if (!prev_sta) { + prev_sta = sta; continue; + } - if (sdata->vif.type == NL80211_IFTYPE_MONITOR || - sdata->vif.type == NL80211_IFTYPE_AP_VLAN) - continue; + rx.sta = prev_sta; + rx.sdata = prev_sta->sdata; + ieee80211_prepare_and_rx_handle(&rx, skb, false); - /* - * frame is destined for this interface, but if it's - * not also for the previous one we handle that after - * the loop to avoid copying the SKB once too much - */ + prev_sta = sta; + } - if (!prev) { - prev = sdata; - continue; - } + if (prev_sta) { + rx.sta = prev_sta; + rx.sdata = prev_sta->sdata; - rx.sta = sta_info_get_bss(prev, hdr->addr2); + if (ieee80211_prepare_and_rx_handle(&rx, skb, true)) + return; + } + } - rx.flags |= IEEE80211_RX_RA_MATCH; - prepares = prepare_for_handlers(prev, &rx, hdr); + prev = NULL; - if (!prepares) - goto next; + list_for_each_entry_rcu(sdata, &local->interfaces, list) { + if (!ieee80211_sdata_running(sdata)) + continue; - if (status->flag & RX_FLAG_MMIC_ERROR) { - rx.sdata = prev; - if (rx.flags & IEEE80211_RX_RA_MATCH) - ieee80211_rx_michael_mic_report(hdr, - &rx); - goto next; - } + if (sdata->vif.type == NL80211_IFTYPE_MONITOR || + sdata->vif.type == NL80211_IFTYPE_AP_VLAN) + continue; - /* - * frame was destined for the previous interface - * so invoke RX handlers for it - */ + /* + * frame is destined for this interface, but if it's + * not also for the previous one we handle that after + * the loop to avoid copying the SKB once too much + */ - skb_new = skb_copy(skb, GFP_ATOMIC); - if (!skb_new) { - if (net_ratelimit()) - printk(KERN_DEBUG "%s: failed to copy " - "multicast frame for %s\n", - wiphy_name(local->hw.wiphy), - prev->name); - goto next; - } - ieee80211_invoke_rx_handlers(prev, &rx, skb_new, rate); -next: + if (!prev) { prev = sdata; + continue; } - if (prev) { - rx.sta = sta_info_get_bss(prev, hdr->addr2); + rx.sta = sta_info_get_bss(prev, hdr->addr2); + rx.sdata = prev; + ieee80211_prepare_and_rx_handle(&rx, skb, false); - rx.flags |= IEEE80211_RX_RA_MATCH; - prepares = prepare_for_handlers(prev, &rx, hdr); + prev = sdata; + } - if (!prepares) - prev = NULL; - } + if (prev) { + rx.sta = sta_info_get_bss(prev, hdr->addr2); + rx.sdata = prev; + + if (ieee80211_prepare_and_rx_handle(&rx, skb, true)) + return; } - if (prev) - ieee80211_invoke_rx_handlers(prev, &rx, skb, rate); - else - dev_kfree_skb(skb); + + dev_kfree_skb(skb); } /* @@ -2615,30 +2821,41 @@ void ieee80211_rx(struct ieee80211_hw *hw, struct sk_buff *skb) if (WARN_ON(!local->started)) goto drop; - if (status->flag & RX_FLAG_HT) { + if (likely(!(status->flag & RX_FLAG_FAILED_PLCP_CRC))) { /* - * rate_idx is MCS index, which can be [0-76] as documented on: - * - * http://wireless.kernel.org/en/developers/Documentation/ieee80211/802.11n - * - * Anything else would be some sort of driver or hardware error. - * The driver should catch hardware errors. + * Validate the rate, unless a PLCP error means that + * we probably can't have a valid rate here anyway. */ - if (WARN((status->rate_idx < 0 || - status->rate_idx > 76), - "Rate marked as an HT rate but passed " - "status->rate_idx is not " - "an MCS index [0-76]: %d (0x%02x)\n", - status->rate_idx, - status->rate_idx)) - goto drop; - } else { - if (WARN_ON(status->rate_idx < 0 || - status->rate_idx >= sband->n_bitrates)) - goto drop; - rate = &sband->bitrates[status->rate_idx]; + + if (status->flag & RX_FLAG_HT) { + /* + * rate_idx is MCS index, which can be [0-76] + * as documented on: + * + * http://wireless.kernel.org/en/developers/Documentation/ieee80211/802.11n + * + * Anything else would be some sort of driver or + * hardware error. The driver should catch hardware + * errors. + */ + if (WARN((status->rate_idx < 0 || + status->rate_idx > 76), + "Rate marked as an HT rate but passed " + "status->rate_idx is not " + "an MCS index [0-76]: %d (0x%02x)\n", + status->rate_idx, + status->rate_idx)) + goto drop; + } else { + if (WARN_ON(status->rate_idx < 0 || + status->rate_idx >= sband->n_bitrates)) + goto drop; + rate = &sband->bitrates[status->rate_idx]; + } } + status->rx_flags = 0; + /* * key references and virtual interfaces are protected using RCU * and this requires that we are in a read-side RCU section during @@ -2658,7 +2875,7 @@ void ieee80211_rx(struct ieee80211_hw *hw, struct sk_buff *skb) return; } - __ieee80211_rx_handle_packet(hw, skb, rate); + __ieee80211_rx_handle_packet(hw, skb); rcu_read_unlock(); diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index 872d7b6ef6b..fb274db77e3 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -242,20 +242,19 @@ static bool ieee80211_prep_hw_scan(struct ieee80211_local *local) local->hw_scan_req->n_channels = n_chans; ielen = ieee80211_build_preq_ies(local, (u8 *)local->hw_scan_req->ie, - req->ie, req->ie_len, band); + req->ie, req->ie_len, band, (u32) -1, + 0); local->hw_scan_req->ie_len = ielen; return true; } -void ieee80211_scan_completed(struct ieee80211_hw *hw, bool aborted) +static bool __ieee80211_scan_completed(struct ieee80211_hw *hw, bool aborted, + bool was_hw_scan) { struct ieee80211_local *local = hw_to_local(hw); - bool was_hw_scan; - - trace_api_scan_completed(local, aborted); - mutex_lock(&local->scan_mtx); + lockdep_assert_held(&local->mtx); /* * It's ok to abort a not-yet-running scan (that @@ -266,17 +265,13 @@ void ieee80211_scan_completed(struct ieee80211_hw *hw, bool aborted) if (WARN_ON(!local->scanning && !aborted)) aborted = true; - if (WARN_ON(!local->scan_req)) { - mutex_unlock(&local->scan_mtx); - return; - } + if (WARN_ON(!local->scan_req)) + return false; - was_hw_scan = test_bit(SCAN_HW_SCANNING, &local->scanning); if (was_hw_scan && !aborted && ieee80211_prep_hw_scan(local)) { - ieee80211_queue_delayed_work(&local->hw, - &local->scan_work, 0); - mutex_unlock(&local->scan_mtx); - return; + int rc = drv_hw_scan(local, local->scan_sdata, local->hw_scan_req); + if (rc == 0) + return false; } kfree(local->hw_scan_req); @@ -290,26 +285,42 @@ void ieee80211_scan_completed(struct ieee80211_hw *hw, bool aborted) local->scanning = 0; local->scan_channel = NULL; - /* we only have to protect scan_req and hw/sw scan */ - mutex_unlock(&local->scan_mtx); - - ieee80211_hw_config(local, IEEE80211_CONF_CHANGE_CHANNEL); - if (was_hw_scan) - goto done; - - ieee80211_configure_filter(local); + return true; +} - drv_sw_scan_complete(local); +static void __ieee80211_scan_completed_finish(struct ieee80211_hw *hw, + bool was_hw_scan) +{ + struct ieee80211_local *local = hw_to_local(hw); - ieee80211_offchannel_return(local, true); + ieee80211_hw_config(local, IEEE80211_CONF_CHANGE_CHANNEL); + if (!was_hw_scan) { + ieee80211_configure_filter(local); + drv_sw_scan_complete(local); + ieee80211_offchannel_return(local, true); + } - done: + mutex_lock(&local->mtx); ieee80211_recalc_idle(local); + mutex_unlock(&local->mtx); + ieee80211_mlme_notify_scan_completed(local); ieee80211_ibss_notify_scan_completed(local); ieee80211_mesh_notify_scan_completed(local); ieee80211_queue_work(&local->hw, &local->work_work); } + +void ieee80211_scan_completed(struct ieee80211_hw *hw, bool aborted) +{ + struct ieee80211_local *local = hw_to_local(hw); + + trace_api_scan_completed(local, aborted); + + set_bit(SCAN_COMPLETED, &local->scanning); + if (aborted) + set_bit(SCAN_ABORTED, &local->scanning); + ieee80211_queue_delayed_work(&local->hw, &local->scan_work, 0); +} EXPORT_SYMBOL(ieee80211_scan_completed); static int ieee80211_start_sw_scan(struct ieee80211_local *local) @@ -353,6 +364,8 @@ static int __ieee80211_start_scan(struct ieee80211_sub_if_data *sdata, struct ieee80211_local *local = sdata->local; int rc; + lockdep_assert_held(&local->mtx); + if (local->scan_req) return -EBUSY; @@ -434,8 +447,8 @@ ieee80211_scan_get_channel_time(struct ieee80211_channel *chan) return IEEE80211_PROBE_DELAY + IEEE80211_CHANNEL_TIME; } -static int ieee80211_scan_state_decision(struct ieee80211_local *local, - unsigned long *next_delay) +static void ieee80211_scan_state_decision(struct ieee80211_local *local, + unsigned long *next_delay) { bool associated = false; bool tx_empty = true; @@ -445,12 +458,6 @@ static int ieee80211_scan_state_decision(struct ieee80211_local *local, struct ieee80211_sub_if_data *sdata; struct ieee80211_channel *next_chan; - /* if no more bands/channels left, complete scan and advance to the idle state */ - if (local->scan_channel_idx >= local->scan_req->n_channels) { - ieee80211_scan_completed(&local->hw, false); - return 1; - } - /* * check if at least one STA interface is associated, * check if at least one STA interface has pending tx frames @@ -522,7 +529,6 @@ static int ieee80211_scan_state_decision(struct ieee80211_local *local, } *next_delay = 0; - return 0; } static void ieee80211_scan_state_leave_oper_channel(struct ieee80211_local *local, @@ -638,21 +644,18 @@ void ieee80211_scan_work(struct work_struct *work) container_of(work, struct ieee80211_local, scan_work.work); struct ieee80211_sub_if_data *sdata = local->scan_sdata; unsigned long next_delay = 0; + bool aborted, hw_scan, finish; - mutex_lock(&local->scan_mtx); - if (!sdata || !local->scan_req) { - mutex_unlock(&local->scan_mtx); - return; - } + mutex_lock(&local->mtx); - if (local->hw_scan_req) { - int rc = drv_hw_scan(local, sdata, local->hw_scan_req); - mutex_unlock(&local->scan_mtx); - if (rc) - ieee80211_scan_completed(&local->hw, true); - return; + if (test_and_clear_bit(SCAN_COMPLETED, &local->scanning)) { + aborted = test_and_clear_bit(SCAN_ABORTED, &local->scanning); + goto out_complete; } + if (!sdata || !local->scan_req) + goto out; + if (local->scan_req && !local->scanning) { struct cfg80211_scan_request *req = local->scan_req; int rc; @@ -661,21 +664,21 @@ void ieee80211_scan_work(struct work_struct *work) local->scan_sdata = NULL; rc = __ieee80211_start_scan(sdata, req); - mutex_unlock(&local->scan_mtx); - - if (rc) - ieee80211_scan_completed(&local->hw, true); - return; + if (rc) { + /* need to complete scan in cfg80211 */ + local->scan_req = req; + aborted = true; + goto out_complete; + } else + goto out; } - mutex_unlock(&local->scan_mtx); - /* * Avoid re-scheduling when the sdata is going away. */ if (!ieee80211_sdata_running(sdata)) { - ieee80211_scan_completed(&local->hw, true); - return; + aborted = true; + goto out_complete; } /* @@ -685,8 +688,12 @@ void ieee80211_scan_work(struct work_struct *work) do { switch (local->next_scan_state) { case SCAN_DECISION: - if (ieee80211_scan_state_decision(local, &next_delay)) - return; + /* if no more bands/channels left, complete scan */ + if (local->scan_channel_idx >= local->scan_req->n_channels) { + aborted = false; + goto out_complete; + } + ieee80211_scan_state_decision(local, &next_delay); break; case SCAN_SET_CHANNEL: ieee80211_scan_state_set_channel(local, &next_delay); @@ -704,6 +711,19 @@ void ieee80211_scan_work(struct work_struct *work) } while (next_delay == 0); ieee80211_queue_delayed_work(&local->hw, &local->scan_work, next_delay); + mutex_unlock(&local->mtx); + return; + +out_complete: + hw_scan = test_bit(SCAN_HW_SCANNING, &local->scanning); + finish = __ieee80211_scan_completed(&local->hw, aborted, hw_scan); + mutex_unlock(&local->mtx); + if (finish) + __ieee80211_scan_completed_finish(&local->hw, hw_scan); + return; + +out: + mutex_unlock(&local->mtx); } int ieee80211_request_scan(struct ieee80211_sub_if_data *sdata, @@ -711,9 +731,9 @@ int ieee80211_request_scan(struct ieee80211_sub_if_data *sdata, { int res; - mutex_lock(&sdata->local->scan_mtx); + mutex_lock(&sdata->local->mtx); res = __ieee80211_start_scan(sdata, req); - mutex_unlock(&sdata->local->scan_mtx); + mutex_unlock(&sdata->local->mtx); return res; } @@ -726,7 +746,7 @@ int ieee80211_request_internal_scan(struct ieee80211_sub_if_data *sdata, int ret = -EBUSY; enum ieee80211_band band; - mutex_lock(&local->scan_mtx); + mutex_lock(&local->mtx); /* busy scanning */ if (local->scan_req) @@ -761,25 +781,44 @@ int ieee80211_request_internal_scan(struct ieee80211_sub_if_data *sdata, ret = __ieee80211_start_scan(sdata, sdata->local->int_scan_req); unlock: - mutex_unlock(&local->scan_mtx); + mutex_unlock(&local->mtx); return ret; } +/* + * Only call this function when a scan can't be queued -- under RTNL. + */ void ieee80211_scan_cancel(struct ieee80211_local *local) { bool abortscan; - - cancel_delayed_work_sync(&local->scan_work); + bool finish = false; /* - * Only call this function when a scan can't be - * queued -- mostly at suspend under RTNL. + * We are only canceling software scan, or deferred scan that was not + * yet really started (see __ieee80211_start_scan ). + * + * Regarding hardware scan: + * - we can not call __ieee80211_scan_completed() as when + * SCAN_HW_SCANNING bit is set this function change + * local->hw_scan_req to operate on 5G band, what race with + * driver which can use local->hw_scan_req + * + * - we can not cancel scan_work since driver can schedule it + * by ieee80211_scan_completed(..., true) to finish scan + * + * Hence low lever driver is responsible for canceling HW scan. */ - mutex_lock(&local->scan_mtx); - abortscan = test_bit(SCAN_SW_SCANNING, &local->scanning) || - (!local->scanning && local->scan_req); - mutex_unlock(&local->scan_mtx); + mutex_lock(&local->mtx); + abortscan = local->scan_req && !test_bit(SCAN_HW_SCANNING, &local->scanning); if (abortscan) - ieee80211_scan_completed(&local->hw, true); + finish = __ieee80211_scan_completed(&local->hw, true, false); + mutex_unlock(&local->mtx); + + if (abortscan) { + /* The scan is canceled, but stop work from being pending */ + cancel_delayed_work_sync(&local->scan_work); + } + if (finish) + __ieee80211_scan_completed_finish(&local->hw, false); } diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 6d86f0c1ad0..6d8f897d876 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -125,7 +125,7 @@ struct sta_info *sta_info_get_bss(struct ieee80211_sub_if_data *sdata, lockdep_is_held(&local->sta_mtx)); while (sta) { if ((sta->sdata == sdata || - sta->sdata->bss == sdata->bss) && + (sta->sdata->bss && sta->sdata->bss == sdata->bss)) && memcmp(sta->sta.addr, addr, ETH_ALEN) == 0) break; sta = rcu_dereference_check(sta->hnext, @@ -174,8 +174,7 @@ static void __sta_info_free(struct ieee80211_local *local, } #ifdef CONFIG_MAC80211_VERBOSE_DEBUG - printk(KERN_DEBUG "%s: Destroyed STA %pM\n", - wiphy_name(local->hw.wiphy), sta->sta.addr); + wiphy_debug(local->hw.wiphy, "Destroyed STA %pM\n", sta->sta.addr); #endif /* CONFIG_MAC80211_VERBOSE_DEBUG */ kfree(sta); @@ -262,8 +261,7 @@ struct sta_info *sta_info_alloc(struct ieee80211_sub_if_data *sdata, sta->last_seq_ctrl[i] = cpu_to_le16(USHRT_MAX); #ifdef CONFIG_MAC80211_VERBOSE_DEBUG - printk(KERN_DEBUG "%s: Allocated STA %pM\n", - wiphy_name(local->hw.wiphy), sta->sta.addr); + wiphy_debug(local->hw.wiphy, "Allocated STA %pM\n", sta->sta.addr); #endif /* CONFIG_MAC80211_VERBOSE_DEBUG */ #ifdef CONFIG_MAC80211_MESH @@ -282,7 +280,7 @@ static int sta_info_finish_insert(struct sta_info *sta, bool async) unsigned long flags; int err = 0; - WARN_ON(!mutex_is_locked(&local->sta_mtx)); + lockdep_assert_held(&local->sta_mtx); /* notify driver */ if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN) @@ -300,8 +298,9 @@ static int sta_info_finish_insert(struct sta_info *sta, bool async) sta->uploaded = true; #ifdef CONFIG_MAC80211_VERBOSE_DEBUG if (async) - printk(KERN_DEBUG "%s: Finished adding IBSS STA %pM\n", - wiphy_name(local->hw.wiphy), sta->sta.addr); + wiphy_debug(local->hw.wiphy, + "Finished adding IBSS STA %pM\n", + sta->sta.addr); #endif } @@ -411,8 +410,8 @@ int sta_info_insert_rcu(struct sta_info *sta) __acquires(RCU) spin_unlock_irqrestore(&local->sta_lock, flags); #ifdef CONFIG_MAC80211_VERBOSE_DEBUG - printk(KERN_DEBUG "%s: Added IBSS STA %pM\n", - wiphy_name(local->hw.wiphy), sta->sta.addr); + wiphy_debug(local->hw.wiphy, "Added IBSS STA %pM\n", + sta->sta.addr); #endif /* CONFIG_MAC80211_VERBOSE_DEBUG */ ieee80211_queue_work(&local->hw, &local->sta_finish_work); @@ -459,8 +458,7 @@ int sta_info_insert_rcu(struct sta_info *sta) __acquires(RCU) } #ifdef CONFIG_MAC80211_VERBOSE_DEBUG - printk(KERN_DEBUG "%s: Inserted STA %pM\n", - wiphy_name(local->hw.wiphy), sta->sta.addr); + wiphy_debug(local->hw.wiphy, "Inserted STA %pM\n", sta->sta.addr); #endif /* CONFIG_MAC80211_VERBOSE_DEBUG */ /* move reference to rcu-protected */ @@ -618,7 +616,7 @@ static int __must_check __sta_info_destroy(struct sta_info *sta) struct ieee80211_sub_if_data *sdata; struct sk_buff *skb; unsigned long flags; - int ret; + int ret, i; might_sleep(); @@ -635,7 +633,7 @@ static int __must_check __sta_info_destroy(struct sta_info *sta) * will be sufficient. */ set_sta_flags(sta, WLAN_STA_BLOCK_BA); - ieee80211_sta_tear_down_BA_sessions(sta); + ieee80211_sta_tear_down_BA_sessions(sta, true); spin_lock_irqsave(&local->sta_lock, flags); ret = sta_info_hash_del(local, sta); @@ -646,10 +644,10 @@ static int __must_check __sta_info_destroy(struct sta_info *sta) if (ret) return ret; - if (sta->key) { - ieee80211_key_free(local, sta->key); - WARN_ON(sta->key); - } + for (i = 0; i < NUM_DEFAULT_KEYS; i++) + ieee80211_key_free(local, sta->gtk[i]); + if (sta->ptk) + ieee80211_key_free(local, sta->ptk); sta->dead = true; @@ -690,8 +688,7 @@ static int __must_check __sta_info_destroy(struct sta_info *sta) #endif #ifdef CONFIG_MAC80211_VERBOSE_DEBUG - printk(KERN_DEBUG "%s: Removed STA %pM\n", - wiphy_name(local->hw.wiphy), sta->sta.addr); + wiphy_debug(local->hw.wiphy, "Removed STA %pM\n", sta->sta.addr); #endif /* CONFIG_MAC80211_VERBOSE_DEBUG */ cancel_work_sync(&sta->drv_unblock_wk); @@ -841,13 +838,20 @@ void ieee80211_sta_expire(struct ieee80211_sub_if_data *sdata, mutex_unlock(&local->sta_mtx); } -struct ieee80211_sta *ieee80211_find_sta_by_hw(struct ieee80211_hw *hw, - const u8 *addr) +struct ieee80211_sta *ieee80211_find_sta_by_ifaddr(struct ieee80211_hw *hw, + const u8 *addr, + const u8 *localaddr) { struct sta_info *sta, *nxt; - /* Just return a random station ... first in list ... */ + /* + * Just return a random station if localaddr is NULL + * ... first in list. + */ for_each_sta_info(hw_to_local(hw), addr, sta, nxt) { + if (localaddr && + compare_ether_addr(sta->sdata->vif.addr, localaddr) != 0) + continue; if (!sta->uploaded) return NULL; return &sta->sta; @@ -855,7 +859,7 @@ struct ieee80211_sta *ieee80211_find_sta_by_hw(struct ieee80211_hw *hw, return NULL; } -EXPORT_SYMBOL_GPL(ieee80211_find_sta_by_hw); +EXPORT_SYMBOL_GPL(ieee80211_find_sta_by_ifaddr); struct ieee80211_sta *ieee80211_find_sta(struct ieee80211_vif *vif, const u8 *addr) diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index 54262e72376..9265acadef3 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -79,6 +79,7 @@ enum ieee80211_sta_info_flags { * @dialog_token: dialog token for aggregation session * @state: session state (see above) * @stop_initiator: initiator of a session stop + * @tx_stop: TX DelBA frame when stopping * * This structure is protected by RCU and the per-station * spinlock. Assignments to the array holding it must hold @@ -95,6 +96,7 @@ struct tid_ampdu_tx { unsigned long state; u8 dialog_token; u8 stop_initiator; + bool tx_stop; }; /** @@ -103,6 +105,7 @@ struct tid_ampdu_tx { * @reorder_buf: buffer to reorder incoming aggregated MPDUs * @reorder_time: jiffies when skb was added * @session_timer: check if peer keeps Tx-ing on the TID (by timeout value) + * @reorder_timer: releases expired frames from the reorder buffer. * @head_seq_num: head sequence number in reordering buffer. * @stored_mpdu_num: number of MPDUs in reordering buffer * @ssn: Starting Sequence Number expected to be aggregated. @@ -110,20 +113,25 @@ struct tid_ampdu_tx { * @timeout: reset timer value (in TUs). * @dialog_token: dialog token for aggregation session * @rcu_head: RCU head used for freeing this struct + * @reorder_lock: serializes access to reorder buffer, see below. * * This structure is protected by RCU and the per-station * spinlock. Assignments to the array holding it must hold - * the spinlock, only the RX path can access it under RCU - * lock-free. The RX path, since it is single-threaded, - * can even modify the structure without locking since the - * only other modifications to it are done when the struct - * can not yet or no longer be found by the RX path. + * the spinlock. + * + * The @reorder_lock is used to protect the variables and + * arrays such as @reorder_buf, @reorder_time, @head_seq_num, + * @stored_mpdu_num and @reorder_time from being corrupted by + * concurrent access of the RX path and the expired frame + * release timer. */ struct tid_ampdu_rx { struct rcu_head rcu_head; + spinlock_t reorder_lock; struct sk_buff **reorder_buf; unsigned long *reorder_time; struct timer_list session_timer; + struct timer_list reorder_timer; u16 head_seq_num; u16 stored_mpdu_num; u16 ssn; @@ -191,7 +199,8 @@ enum plink_state { * @hnext: hash table linked list pointer * @local: pointer to the global information * @sdata: virtual interface this station belongs to - * @key: peer key negotiated with this station, if any + * @ptk: peer key negotiated with this station, if any + * @gtk: group keys negotiated with this station, if any * @rate_ctrl: rate control algorithm reference * @rate_ctrl_priv: rate control private per-STA pointer * @last_tx_rate: rate used for last transmit, to report to userspace as @@ -246,7 +255,8 @@ struct sta_info { struct sta_info *hnext; struct ieee80211_local *local; struct ieee80211_sub_if_data *sdata; - struct ieee80211_key *key; + struct ieee80211_key *gtk[NUM_DEFAULT_KEYS + NUM_DEFAULT_MGMT_KEYS]; + struct ieee80211_key *ptk; struct rate_control_ref *rate_ctrl; void *rate_ctrl_priv; spinlock_t lock; diff --git a/net/mac80211/status.c b/net/mac80211/status.c index 10caec5ea8f..3153c19893b 100644 --- a/net/mac80211/status.c +++ b/net/mac80211/status.c @@ -58,6 +58,7 @@ static void ieee80211_handle_filtered_frame(struct ieee80211_local *local, info->control.vif = &sta->sdata->vif; info->flags |= IEEE80211_TX_INTFL_NEED_TXPROCESSING | IEEE80211_TX_INTFL_RETRANSMISSION; + info->flags &= ~IEEE80211_TX_TEMPORARY_FLAGS; sta->tx_filtered_count++; @@ -114,11 +115,10 @@ static void ieee80211_handle_filtered_frame(struct ieee80211_local *local, #ifdef CONFIG_MAC80211_VERBOSE_DEBUG if (net_ratelimit()) - printk(KERN_DEBUG "%s: dropped TX filtered frame, " - "queue_len=%d PS=%d @%lu\n", - wiphy_name(local->hw.wiphy), - skb_queue_len(&sta->tx_filtered), - !!test_sta_flags(sta, WLAN_STA_PS_STA), jiffies); + wiphy_debug(local->hw.wiphy, + "dropped TX filtered frame, queue_len=%d PS=%d @%lu\n", + skb_queue_len(&sta->tx_filtered), + !!test_sta_flags(sta, WLAN_STA_PS_STA), jiffies); #endif dev_kfree_skb(skb); } @@ -176,7 +176,7 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb) for (i = 0; i < IEEE80211_TX_MAX_RATES; i++) { /* the HW cannot have attempted that rate */ - if (i >= hw->max_rates) { + if (i >= hw->max_report_rates) { info->status.rates[i].idx = -1; info->status.rates[i].count = 0; } else if (info->status.rates[i].idx >= 0) { @@ -296,7 +296,7 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb) } if (info->flags & IEEE80211_TX_INTFL_NL80211_FRAME_TX) - cfg80211_action_tx_status( + cfg80211_mgmt_tx_status( skb->dev, (unsigned long) skb, skb->data, skb->len, !!(info->flags & IEEE80211_TX_STAT_ACK), GFP_ATOMIC); @@ -377,7 +377,7 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb) skb2 = skb_clone(skb, GFP_ATOMIC); if (skb2) { skb2->dev = prev_dev; - netif_receive_skb(skb2); + netif_rx(skb2); } } @@ -386,7 +386,7 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb) } if (prev_dev) { skb->dev = prev_dev; - netif_receive_skb(skb); + netif_rx(skb); skb = NULL; } rcu_read_unlock(); diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index c54db966926..96c59430950 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -273,6 +273,9 @@ ieee80211_tx_h_check_assoc(struct ieee80211_tx_data *tx) */ return TX_DROP; + if (tx->sdata->vif.type == NL80211_IFTYPE_WDS) + return TX_CONTINUE; + if (tx->sdata->vif.type == NL80211_IFTYPE_MESH_POINT) return TX_CONTINUE; @@ -351,8 +354,8 @@ static void purge_old_ps_buffers(struct ieee80211_local *local) local->total_ps_buffered = total; #ifdef CONFIG_MAC80211_VERBOSE_PS_DEBUG - printk(KERN_DEBUG "%s: PS buffers full - purged %d frames\n", - wiphy_name(local->hw.wiphy), purged); + wiphy_debug(local->hw.wiphy, "PS buffers full - purged %d frames\n", + purged); #endif } @@ -509,6 +512,18 @@ ieee80211_tx_h_ps_buf(struct ieee80211_tx_data *tx) } static ieee80211_tx_result debug_noinline +ieee80211_tx_h_check_control_port_protocol(struct ieee80211_tx_data *tx) +{ + struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx->skb); + + if (unlikely(tx->sdata->control_port_protocol == tx->skb->protocol && + tx->sdata->control_port_no_encrypt)) + info->flags |= IEEE80211_TX_INTFL_DONT_ENCRYPT; + + return TX_CONTINUE; +} + +static ieee80211_tx_result debug_noinline ieee80211_tx_h_select_key(struct ieee80211_tx_data *tx) { struct ieee80211_key *key = NULL; @@ -517,7 +532,7 @@ ieee80211_tx_h_select_key(struct ieee80211_tx_data *tx) if (unlikely(info->flags & IEEE80211_TX_INTFL_DONT_ENCRYPT)) tx->key = NULL; - else if (tx->sta && (key = rcu_dereference(tx->sta->key))) + else if (tx->sta && (key = rcu_dereference(tx->sta->ptk))) tx->key = key; else if (ieee80211_is_mgmt(hdr->frame_control) && is_multicast_ether_addr(hdr->addr1) && @@ -527,7 +542,7 @@ ieee80211_tx_h_select_key(struct ieee80211_tx_data *tx) else if ((key = rcu_dereference(tx->sdata->default_key))) tx->key = key; else if (tx->sdata->drop_unencrypted && - (tx->skb->protocol != cpu_to_be16(ETH_P_PAE)) && + (tx->skb->protocol != tx->sdata->control_port_protocol) && !(info->flags & IEEE80211_TX_CTL_INJECTED) && (!ieee80211_is_robust_mgmt_frame(hdr) || (ieee80211_is_action(hdr->frame_control) && @@ -543,15 +558,16 @@ ieee80211_tx_h_select_key(struct ieee80211_tx_data *tx) tx->key->tx_rx_count++; /* TODO: add threshold stuff again */ - switch (tx->key->conf.alg) { - case ALG_WEP: + switch (tx->key->conf.cipher) { + case WLAN_CIPHER_SUITE_WEP40: + case WLAN_CIPHER_SUITE_WEP104: if (ieee80211_is_auth(hdr->frame_control)) break; - case ALG_TKIP: + case WLAN_CIPHER_SUITE_TKIP: if (!ieee80211_is_data_present(hdr->frame_control)) tx->key = NULL; break; - case ALG_CCMP: + case WLAN_CIPHER_SUITE_CCMP: if (!ieee80211_is_data_present(hdr->frame_control) && !ieee80211_use_mfp(hdr->frame_control, tx->sta, tx->skb)) @@ -561,7 +577,7 @@ ieee80211_tx_h_select_key(struct ieee80211_tx_data *tx) IEEE80211_KEY_FLAG_SW_MGMT) && ieee80211_is_mgmt(hdr->frame_control); break; - case ALG_AES_CMAC: + case WLAN_CIPHER_SUITE_AES_CMAC: if (!ieee80211_is_mgmt(hdr->frame_control)) tx->key = NULL; break; @@ -946,22 +962,31 @@ ieee80211_tx_h_stats(struct ieee80211_tx_data *tx) static ieee80211_tx_result debug_noinline ieee80211_tx_h_encrypt(struct ieee80211_tx_data *tx) { + struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx->skb); + if (!tx->key) return TX_CONTINUE; - switch (tx->key->conf.alg) { - case ALG_WEP: + switch (tx->key->conf.cipher) { + case WLAN_CIPHER_SUITE_WEP40: + case WLAN_CIPHER_SUITE_WEP104: return ieee80211_crypto_wep_encrypt(tx); - case ALG_TKIP: + case WLAN_CIPHER_SUITE_TKIP: return ieee80211_crypto_tkip_encrypt(tx); - case ALG_CCMP: + case WLAN_CIPHER_SUITE_CCMP: return ieee80211_crypto_ccmp_encrypt(tx); - case ALG_AES_CMAC: + case WLAN_CIPHER_SUITE_AES_CMAC: return ieee80211_crypto_aes_cmac_encrypt(tx); + default: + /* handle hw-only algorithm */ + if (info->control.hw_key) { + ieee80211_tx_set_protected(tx); + return TX_CONTINUE; + } + break; + } - /* not reached */ - WARN_ON(1); return TX_DROP; } @@ -1339,6 +1364,7 @@ static int invoke_tx_handlers(struct ieee80211_tx_data *tx) CALL_TXH(ieee80211_tx_h_dynamic_ps); CALL_TXH(ieee80211_tx_h_check_assoc); CALL_TXH(ieee80211_tx_h_ps_buf); + CALL_TXH(ieee80211_tx_h_check_control_port_protocol); CALL_TXH(ieee80211_tx_h_select_key); if (!(tx->local->hw.flags & IEEE80211_HW_HAS_RATE_CONTROL)) CALL_TXH(ieee80211_tx_h_rate_ctrl); @@ -1511,8 +1537,8 @@ static int ieee80211_skb_resize(struct ieee80211_local *local, I802_DEBUG_INC(local->tx_expand_skb_head); if (pskb_expand_head(skb, head_need, tail_need, GFP_ATOMIC)) { - printk(KERN_DEBUG "%s: failed to reallocate TX buffer\n", - wiphy_name(local->hw.wiphy)); + wiphy_debug(local->hw.wiphy, + "failed to reallocate TX buffer\n"); return -ENOMEM; } @@ -1586,6 +1612,7 @@ static void ieee80211_xmit(struct ieee80211_sub_if_data *sdata, return; } + hdr = (struct ieee80211_hdr *) skb->data; info->control.vif = &sdata->vif; if (ieee80211_vif_is_mesh(&sdata->vif) && @@ -1699,7 +1726,7 @@ netdev_tx_t ieee80211_subif_start_xmit(struct sk_buff *skb, u16 ethertype, hdrlen, meshhdrlen = 0; __le16 fc; struct ieee80211_hdr hdr; - struct ieee80211s_hdr mesh_hdr; + struct ieee80211s_hdr mesh_hdr __maybe_unused; const u8 *encaps_data; int encaps_len, skip_header_bytes; int nh_pos, h_pos; @@ -1816,7 +1843,8 @@ netdev_tx_t ieee80211_subif_start_xmit(struct sk_buff *skb, #endif case NL80211_IFTYPE_STATION: memcpy(hdr.addr1, sdata->u.mgd.bssid, ETH_ALEN); - if (sdata->u.mgd.use_4addr && ethertype != ETH_P_PAE) { + if (sdata->u.mgd.use_4addr && + cpu_to_be16(ethertype) != sdata->control_port_protocol) { fc |= cpu_to_le16(IEEE80211_FCTL_FROMDS | IEEE80211_FCTL_TODS); /* RA TA DA SA */ memcpy(hdr.addr2, sdata->vif.addr, ETH_ALEN); @@ -1869,7 +1897,7 @@ netdev_tx_t ieee80211_subif_start_xmit(struct sk_buff *skb, if (!ieee80211_vif_is_mesh(&sdata->vif) && unlikely(!is_multicast_ether_addr(hdr.addr1) && !(sta_flags & WLAN_STA_AUTHORIZED) && - !(ethertype == ETH_P_PAE && + !(cpu_to_be16(ethertype) == sdata->control_port_protocol && compare_ether_addr(sdata->vif.addr, skb->data + ETH_ALEN) == 0))) { #ifdef CONFIG_MAC80211_VERBOSE_DEBUG @@ -2068,8 +2096,7 @@ void ieee80211_tx_pending(unsigned long data) if (skb_queue_empty(&local->pending[i])) list_for_each_entry_rcu(sdata, &local->interfaces, list) - netif_tx_wake_queue( - netdev_get_tx_queue(sdata->dev, i)); + netif_wake_subqueue(sdata->dev, i); } spin_unlock_irqrestore(&local->queue_stop_reason_lock, flags); diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 748387d45bc..0b6fc92bc0d 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -283,8 +283,11 @@ static void __ieee80211_wake_queue(struct ieee80211_hw *hw, int queue, if (skb_queue_empty(&local->pending[queue])) { rcu_read_lock(); - list_for_each_entry_rcu(sdata, &local->interfaces, list) - netif_tx_wake_queue(netdev_get_tx_queue(sdata->dev, queue)); + list_for_each_entry_rcu(sdata, &local->interfaces, list) { + if (test_bit(SDATA_STATE_OFFCHANNEL, &sdata->state)) + continue; + netif_wake_subqueue(sdata->dev, queue); + } rcu_read_unlock(); } else tasklet_schedule(&local->tx_pending_tasklet); @@ -323,7 +326,7 @@ static void __ieee80211_stop_queue(struct ieee80211_hw *hw, int queue, rcu_read_lock(); list_for_each_entry_rcu(sdata, &local->interfaces, list) - netif_tx_stop_queue(netdev_get_tx_queue(sdata->dev, queue)); + netif_stop_subqueue(sdata->dev, queue); rcu_read_unlock(); } @@ -471,16 +474,10 @@ void ieee80211_iterate_active_interfaces( list_for_each_entry(sdata, &local->interfaces, list) { switch (sdata->vif.type) { - case __NL80211_IFTYPE_AFTER_LAST: - case NL80211_IFTYPE_UNSPECIFIED: case NL80211_IFTYPE_MONITOR: case NL80211_IFTYPE_AP_VLAN: continue; - case NL80211_IFTYPE_AP: - case NL80211_IFTYPE_STATION: - case NL80211_IFTYPE_ADHOC: - case NL80211_IFTYPE_WDS: - case NL80211_IFTYPE_MESH_POINT: + default: break; } if (ieee80211_sdata_running(sdata)) @@ -505,16 +502,10 @@ void ieee80211_iterate_active_interfaces_atomic( list_for_each_entry_rcu(sdata, &local->interfaces, list) { switch (sdata->vif.type) { - case __NL80211_IFTYPE_AFTER_LAST: - case NL80211_IFTYPE_UNSPECIFIED: case NL80211_IFTYPE_MONITOR: case NL80211_IFTYPE_AP_VLAN: continue; - case NL80211_IFTYPE_AP: - case NL80211_IFTYPE_STATION: - case NL80211_IFTYPE_ADHOC: - case NL80211_IFTYPE_WDS: - case NL80211_IFTYPE_MESH_POINT: + default: break; } if (ieee80211_sdata_running(sdata)) @@ -904,26 +895,34 @@ void ieee80211_send_auth(struct ieee80211_sub_if_data *sdata, int ieee80211_build_preq_ies(struct ieee80211_local *local, u8 *buffer, const u8 *ie, size_t ie_len, - enum ieee80211_band band) + enum ieee80211_band band, u32 rate_mask, + u8 channel) { struct ieee80211_supported_band *sband; u8 *pos; size_t offset = 0, noffset; int supp_rates_len, i; + u8 rates[32]; + int num_rates; + int ext_rates_len; sband = local->hw.wiphy->bands[band]; pos = buffer; - supp_rates_len = min_t(int, sband->n_bitrates, 8); + num_rates = 0; + for (i = 0; i < sband->n_bitrates; i++) { + if ((BIT(i) & rate_mask) == 0) + continue; /* skip rate */ + rates[num_rates++] = (u8) (sband->bitrates[i].bitrate / 5); + } + + supp_rates_len = min_t(int, num_rates, 8); *pos++ = WLAN_EID_SUPP_RATES; *pos++ = supp_rates_len; - - for (i = 0; i < supp_rates_len; i++) { - int rate = sband->bitrates[i].bitrate; - *pos++ = (u8) (rate / 5); - } + memcpy(pos, rates, supp_rates_len); + pos += supp_rates_len; /* insert "request information" if in custom IEs */ if (ie && ie_len) { @@ -941,14 +940,18 @@ int ieee80211_build_preq_ies(struct ieee80211_local *local, u8 *buffer, offset = noffset; } - if (sband->n_bitrates > i) { + ext_rates_len = num_rates - supp_rates_len; + if (ext_rates_len > 0) { *pos++ = WLAN_EID_EXT_SUPP_RATES; - *pos++ = sband->n_bitrates - i; + *pos++ = ext_rates_len; + memcpy(pos, rates + supp_rates_len, ext_rates_len); + pos += ext_rates_len; + } - for (; i < sband->n_bitrates; i++) { - int rate = sband->bitrates[i].bitrate; - *pos++ = (u8) (rate / 5); - } + if (channel && sband->band == IEEE80211_BAND_2GHZ) { + *pos++ = WLAN_EID_DS_PARAMS; + *pos++ = 1; + *pos++ = channel; } /* insert custom IEs that go before HT */ @@ -1017,6 +1020,7 @@ void ieee80211_send_probe_req(struct ieee80211_sub_if_data *sdata, u8 *dst, struct ieee80211_mgmt *mgmt; size_t buf_len; u8 *buf; + u8 chan; /* FIXME: come up with a proper value */ buf = kmalloc(200 + ie_len, GFP_KERNEL); @@ -1026,8 +1030,14 @@ void ieee80211_send_probe_req(struct ieee80211_sub_if_data *sdata, u8 *dst, return; } + chan = ieee80211_frequency_to_channel( + local->hw.conf.channel->center_freq); + buf_len = ieee80211_build_preq_ies(local, buf, ie, ie_len, - local->hw.conf.channel->band); + local->hw.conf.channel->band, + sdata->rc_rateidx_mask + [local->hw.conf.channel->band], + chan); skb = ieee80211_probereq_get(&local->hw, &sdata->vif, ssid, ssid_len, @@ -1189,7 +1199,9 @@ int ieee80211_reconfig(struct ieee80211_local *local) /* ignore virtual */ break; case NL80211_IFTYPE_UNSPECIFIED: - case __NL80211_IFTYPE_AFTER_LAST: + case NUM_NL80211_IFTYPES: + case NL80211_IFTYPE_P2P_CLIENT: + case NL80211_IFTYPE_P2P_GO: WARN_ON(1); break; } @@ -1209,7 +1221,7 @@ int ieee80211_reconfig(struct ieee80211_local *local) mutex_lock(&local->sta_mtx); list_for_each_entry(sta, &local->sta_list, list) { - ieee80211_sta_tear_down_BA_sessions(sta); + ieee80211_sta_tear_down_BA_sessions(sta, true); clear_sta_flags(sta, WLAN_STA_BLOCK_BA); } @@ -1285,17 +1297,13 @@ static int check_mgd_smps(struct ieee80211_if_managed *ifmgd, } /* must hold iflist_mtx */ -void ieee80211_recalc_smps(struct ieee80211_local *local, - struct ieee80211_sub_if_data *forsdata) +void ieee80211_recalc_smps(struct ieee80211_local *local) { struct ieee80211_sub_if_data *sdata; enum ieee80211_smps_mode smps_mode = IEEE80211_SMPS_OFF; int count = 0; - if (forsdata) - WARN_ON(!mutex_is_locked(&forsdata->u.mgd.mtx)); - - WARN_ON(!mutex_is_locked(&local->iflist_mtx)); + lockdep_assert_held(&local->iflist_mtx); /* * This function could be improved to handle multiple @@ -1308,22 +1316,12 @@ void ieee80211_recalc_smps(struct ieee80211_local *local, */ list_for_each_entry(sdata, &local->interfaces, list) { - if (!netif_running(sdata->dev)) + if (!ieee80211_sdata_running(sdata)) continue; if (sdata->vif.type != NL80211_IFTYPE_STATION) goto set; - if (sdata != forsdata) { - /* - * This nested is ok -- we are holding the iflist_mtx - * so can't get here twice or so. But it's required - * since normally we acquire it first and then the - * iflist_mtx. - */ - mutex_lock_nested(&sdata->u.mgd.mtx, SINGLE_DEPTH_NESTING); - count += check_mgd_smps(&sdata->u.mgd, &smps_mode); - mutex_unlock(&sdata->u.mgd.mtx); - } else - count += check_mgd_smps(&sdata->u.mgd, &smps_mode); + + count += check_mgd_smps(&sdata->u.mgd, &smps_mode); if (count > 1) { smps_mode = IEEE80211_SMPS_OFF; diff --git a/net/mac80211/wep.c b/net/mac80211/wep.c index 9ebc8d8a1f5..2ff6d1e3ed2 100644 --- a/net/mac80211/wep.c +++ b/net/mac80211/wep.c @@ -222,7 +222,7 @@ static int ieee80211_wep_decrypt(struct ieee80211_local *local, struct ieee80211_key *key) { u32 klen; - u8 *rc4key; + u8 rc4key[3 + WLAN_KEY_LEN_WEP104]; u8 keyidx; struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data; unsigned int hdrlen; @@ -240,15 +240,11 @@ static int ieee80211_wep_decrypt(struct ieee80211_local *local, keyidx = skb->data[hdrlen + 3] >> 6; - if (!key || keyidx != key->conf.keyidx || key->conf.alg != ALG_WEP) + if (!key || keyidx != key->conf.keyidx) return -1; klen = 3 + key->conf.keylen; - rc4key = kmalloc(klen, GFP_ATOMIC); - if (!rc4key) - return -1; - /* Prepend 24-bit IV to RC4 key */ memcpy(rc4key, skb->data + hdrlen, 3); @@ -260,8 +256,6 @@ static int ieee80211_wep_decrypt(struct ieee80211_local *local, len)) ret = -1; - kfree(rc4key); - /* Trim ICV */ skb_trim(skb, skb->len - WEP_ICV_LEN); diff --git a/net/mac80211/work.c b/net/mac80211/work.c index 81d4ad64184..ae344d1ba05 100644 --- a/net/mac80211/work.c +++ b/net/mac80211/work.c @@ -43,7 +43,7 @@ enum work_action { /* utils */ static inline void ASSERT_WORK_MTX(struct ieee80211_local *local) { - WARN_ON(!mutex_is_locked(&local->work_mtx)); + lockdep_assert_held(&local->mtx); } /* @@ -757,7 +757,7 @@ static void ieee80211_work_rx_queued_mgmt(struct ieee80211_local *local, mgmt = (struct ieee80211_mgmt *) skb->data; fc = le16_to_cpu(mgmt->frame_control); - mutex_lock(&local->work_mtx); + mutex_lock(&local->mtx); list_for_each_entry(wk, &local->work_list, list) { const u8 *bssid = NULL; @@ -833,7 +833,7 @@ static void ieee80211_work_rx_queued_mgmt(struct ieee80211_local *local, WARN(1, "unexpected: %d", rma); } - mutex_unlock(&local->work_mtx); + mutex_unlock(&local->mtx); if (rma != WORK_ACT_DONE) goto out; @@ -845,9 +845,9 @@ static void ieee80211_work_rx_queued_mgmt(struct ieee80211_local *local, case WORK_DONE_REQUEUE: synchronize_rcu(); wk->started = false; /* restart */ - mutex_lock(&local->work_mtx); + mutex_lock(&local->mtx); list_add_tail(&wk->list, &local->work_list); - mutex_unlock(&local->work_mtx); + mutex_unlock(&local->mtx); } out: @@ -888,9 +888,9 @@ static void ieee80211_work_work(struct work_struct *work) while ((skb = skb_dequeue(&local->work_skb_queue))) ieee80211_work_rx_queued_mgmt(local, skb); - ieee80211_recalc_idle(local); + mutex_lock(&local->mtx); - mutex_lock(&local->work_mtx); + ieee80211_recalc_idle(local); list_for_each_entry_safe(wk, tmp, &local->work_list, list) { bool started = wk->started; @@ -995,20 +995,16 @@ static void ieee80211_work_work(struct work_struct *work) run_again(local, jiffies + HZ/2); } - mutex_lock(&local->scan_mtx); - if (list_empty(&local->work_list) && local->scan_req && !local->scanning) ieee80211_queue_delayed_work(&local->hw, &local->scan_work, round_jiffies_relative(0)); - mutex_unlock(&local->scan_mtx); - - mutex_unlock(&local->work_mtx); - ieee80211_recalc_idle(local); + mutex_unlock(&local->mtx); + list_for_each_entry_safe(wk, tmp, &free_work, list) { wk->done(wk, NULL); list_del(&wk->list); @@ -1035,16 +1031,15 @@ void ieee80211_add_work(struct ieee80211_work *wk) wk->started = false; local = wk->sdata->local; - mutex_lock(&local->work_mtx); + mutex_lock(&local->mtx); list_add_tail(&wk->list, &local->work_list); - mutex_unlock(&local->work_mtx); + mutex_unlock(&local->mtx); ieee80211_queue_work(&local->hw, &local->work_work); } void ieee80211_work_init(struct ieee80211_local *local) { - mutex_init(&local->work_mtx); INIT_LIST_HEAD(&local->work_list); setup_timer(&local->work_timer, ieee80211_work_timer, (unsigned long)local); @@ -1057,7 +1052,7 @@ void ieee80211_work_purge(struct ieee80211_sub_if_data *sdata) struct ieee80211_local *local = sdata->local; struct ieee80211_work *wk; - mutex_lock(&local->work_mtx); + mutex_lock(&local->mtx); list_for_each_entry(wk, &local->work_list, list) { if (wk->sdata != sdata) continue; @@ -1065,19 +1060,19 @@ void ieee80211_work_purge(struct ieee80211_sub_if_data *sdata) wk->started = true; wk->timeout = jiffies; } - mutex_unlock(&local->work_mtx); + mutex_unlock(&local->mtx); /* run cleanups etc. */ ieee80211_work_work(&local->work_work); - mutex_lock(&local->work_mtx); + mutex_lock(&local->mtx); list_for_each_entry(wk, &local->work_list, list) { if (wk->sdata != sdata) continue; WARN_ON(1); break; } - mutex_unlock(&local->work_mtx); + mutex_unlock(&local->mtx); } ieee80211_rx_result ieee80211_work_rx_mgmt(struct ieee80211_sub_if_data *sdata, @@ -1163,7 +1158,7 @@ int ieee80211_wk_cancel_remain_on_channel(struct ieee80211_sub_if_data *sdata, struct ieee80211_work *wk, *tmp; bool found = false; - mutex_lock(&local->work_mtx); + mutex_lock(&local->mtx); list_for_each_entry_safe(wk, tmp, &local->work_list, list) { if ((unsigned long) wk == cookie) { wk->timeout = jiffies; @@ -1171,7 +1166,7 @@ int ieee80211_wk_cancel_remain_on_channel(struct ieee80211_sub_if_data *sdata, break; } } - mutex_unlock(&local->work_mtx); + mutex_unlock(&local->mtx); if (!found) return -ENOENT; diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c index 8d59d27d887..bee230d8fd1 100644 --- a/net/mac80211/wpa.c +++ b/net/mac80211/wpa.c @@ -36,8 +36,8 @@ ieee80211_tx_h_michael_mic_add(struct ieee80211_tx_data *tx) int tail; hdr = (struct ieee80211_hdr *)skb->data; - if (!tx->key || tx->key->conf.alg != ALG_TKIP || skb->len < 24 || - !ieee80211_is_data_present(hdr->frame_control)) + if (!tx->key || tx->key->conf.cipher != WLAN_CIPHER_SUITE_TKIP || + skb->len < 24 || !ieee80211_is_data_present(hdr->frame_control)) return TX_CONTINUE; hdrlen = ieee80211_hdrlen(hdr->frame_control); @@ -94,7 +94,7 @@ ieee80211_rx_h_michael_mic_verify(struct ieee80211_rx_data *rx) if (status->flag & RX_FLAG_MMIC_STRIPPED) return RX_CONTINUE; - if (!rx->key || rx->key->conf.alg != ALG_TKIP || + if (!rx->key || rx->key->conf.cipher != WLAN_CIPHER_SUITE_TKIP || !ieee80211_has_protected(hdr->frame_control) || !ieee80211_is_data_present(hdr->frame_control)) return RX_CONTINUE; @@ -117,7 +117,7 @@ ieee80211_rx_h_michael_mic_verify(struct ieee80211_rx_data *rx) key = &rx->key->conf.key[key_offset]; michael_mic(key, hdr, data, data_len, mic); if (memcmp(mic, data + data_len, MICHAEL_MIC_LEN) != 0 || wpa_test) { - if (!(rx->flags & IEEE80211_RX_RA_MATCH)) + if (!(status->rx_flags & IEEE80211_RX_RA_MATCH)) return RX_DROP_UNUSABLE; mac80211_ev_michael_mic_failure(rx->sdata, rx->key->conf.keyidx, @@ -221,19 +221,13 @@ ieee80211_crypto_tkip_decrypt(struct ieee80211_rx_data *rx) if (!rx->sta || skb->len - hdrlen < 12) return RX_DROP_UNUSABLE; - if (status->flag & RX_FLAG_DECRYPTED) { - if (status->flag & RX_FLAG_IV_STRIPPED) { - /* - * Hardware took care of all processing, including - * replay protection, and stripped the ICV/IV so - * we cannot do any checks here. - */ - return RX_CONTINUE; - } - - /* let TKIP code verify IV, but skip decryption */ + /* + * Let TKIP code verify IV, but skip decryption. + * In the case where hardware checks the IV as well, + * we don't even get here, see ieee80211_rx_h_decrypt() + */ + if (status->flag & RX_FLAG_DECRYPTED) hwaccel = 1; - } res = ieee80211_tkip_decrypt_data(rx->local->wep_rx_tfm, key, skb->data + hdrlen, @@ -447,10 +441,6 @@ ieee80211_crypto_ccmp_decrypt(struct ieee80211_rx_data *rx) if (!rx->sta || data_len < 0) return RX_DROP_UNUSABLE; - if ((status->flag & RX_FLAG_DECRYPTED) && - (status->flag & RX_FLAG_IV_STRIPPED)) - return RX_CONTINUE; - ccmp_hdr2pn(pn, skb->data + hdrlen); queue = ieee80211_is_mgmt(hdr->frame_control) ? @@ -564,10 +554,6 @@ ieee80211_crypto_aes_cmac_decrypt(struct ieee80211_rx_data *rx) if (!ieee80211_is_mgmt(hdr->frame_control)) return RX_CONTINUE; - if ((status->flag & RX_FLAG_DECRYPTED) && - (status->flag & RX_FLAG_IV_STRIPPED)) - return RX_CONTINUE; - if (skb->len < 24 + sizeof(*mmie)) return RX_DROP_UNUSABLE; diff --git a/net/netfilter/core.c b/net/netfilter/core.c index 78b505d33bf..85dabb86be6 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -27,7 +27,7 @@ static DEFINE_MUTEX(afinfo_mutex); -const struct nf_afinfo *nf_afinfo[NFPROTO_NUMPROTO] __read_mostly; +const struct nf_afinfo __rcu *nf_afinfo[NFPROTO_NUMPROTO] __read_mostly; EXPORT_SYMBOL(nf_afinfo); int nf_register_afinfo(const struct nf_afinfo *afinfo) @@ -105,10 +105,8 @@ EXPORT_SYMBOL(nf_register_hooks); void nf_unregister_hooks(struct nf_hook_ops *reg, unsigned int n) { - unsigned int i; - - for (i = 0; i < n; i++) - nf_unregister_hook(®[i]); + while (n-- > 0) + nf_unregister_hook(®[n]); } EXPORT_SYMBOL(nf_unregister_hooks); diff --git a/net/netfilter/ipvs/Kconfig b/net/netfilter/ipvs/Kconfig index 46a77d5c388..a22dac22705 100644 --- a/net/netfilter/ipvs/Kconfig +++ b/net/netfilter/ipvs/Kconfig @@ -3,7 +3,7 @@ # menuconfig IP_VS tristate "IP virtual server support" - depends on NET && INET && NETFILTER && NF_CONNTRACK + depends on NET && INET && NETFILTER ---help--- IP Virtual Server support will let you build a high-performance virtual server based on cluster of two or more real servers. This @@ -235,7 +235,8 @@ comment 'IPVS application helper' config IP_VS_FTP tristate "FTP protocol helper" - depends on IP_VS_PROTO_TCP && NF_NAT + depends on IP_VS_PROTO_TCP && NF_CONNTRACK && NF_NAT + select IP_VS_NFCT ---help--- FTP is a protocol that transfers IP address and/or port number in the payload. In the virtual server via Network Address Translation, @@ -247,4 +248,19 @@ config IP_VS_FTP If you want to compile it in kernel, say Y. To compile it as a module, choose M here. If unsure, say N. +config IP_VS_NFCT + bool "Netfilter connection tracking" + depends on NF_CONNTRACK + ---help--- + The Netfilter connection tracking support allows the IPVS + connection state to be exported to the Netfilter framework + for filtering purposes. + +config IP_VS_PE_SIP + tristate "SIP persistence engine" + depends on IP_VS_PROTO_UDP + depends on NF_CONNTRACK_SIP + ---help--- + Allow persistence based on the SIP Call-ID + endif # IP_VS diff --git a/net/netfilter/ipvs/Makefile b/net/netfilter/ipvs/Makefile index e3baefd7066..34ee602ddb6 100644 --- a/net/netfilter/ipvs/Makefile +++ b/net/netfilter/ipvs/Makefile @@ -9,10 +9,13 @@ ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_UDP) += ip_vs_proto_udp.o ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_AH_ESP) += ip_vs_proto_ah_esp.o ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_SCTP) += ip_vs_proto_sctp.o +ip_vs-extra_objs-y := +ip_vs-extra_objs-$(CONFIG_IP_VS_NFCT) += ip_vs_nfct.o + ip_vs-objs := ip_vs_conn.o ip_vs_core.o ip_vs_ctl.o ip_vs_sched.o \ ip_vs_xmit.o ip_vs_app.o ip_vs_sync.o \ - ip_vs_est.o ip_vs_proto.o \ - $(ip_vs_proto-objs-y) + ip_vs_est.o ip_vs_proto.o ip_vs_pe.o \ + $(ip_vs_proto-objs-y) $(ip_vs-extra_objs-y) # IPVS core @@ -32,3 +35,6 @@ obj-$(CONFIG_IP_VS_NQ) += ip_vs_nq.o # IPVS application helpers obj-$(CONFIG_IP_VS_FTP) += ip_vs_ftp.o + +# IPVS connection template retrievers +obj-$(CONFIG_IP_VS_PE_SIP) += ip_vs_pe_sip.o diff --git a/net/netfilter/ipvs/ip_vs_app.c b/net/netfilter/ipvs/ip_vs_app.c index e76f87f4aca..a475edee091 100644 --- a/net/netfilter/ipvs/ip_vs_app.c +++ b/net/netfilter/ipvs/ip_vs_app.c @@ -103,8 +103,8 @@ ip_vs_app_inc_new(struct ip_vs_app *app, __u16 proto, __u16 port) goto out; list_add(&inc->a_list, &app->incs_list); - IP_VS_DBG(9, "%s application %s:%u registered\n", - pp->name, inc->name, inc->port); + IP_VS_DBG(9, "%s App %s:%u registered\n", + pp->name, inc->name, ntohs(inc->port)); return 0; @@ -130,7 +130,7 @@ ip_vs_app_inc_release(struct ip_vs_app *inc) pp->unregister_app(inc); IP_VS_DBG(9, "%s App %s:%u unregistered\n", - pp->name, inc->name, inc->port); + pp->name, inc->name, ntohs(inc->port)); list_del(&inc->a_list); diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index b71c69a2db1..e9adecdc8ca 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -148,6 +148,42 @@ static unsigned int ip_vs_conn_hashkey(int af, unsigned proto, & ip_vs_conn_tab_mask; } +static unsigned int ip_vs_conn_hashkey_param(const struct ip_vs_conn_param *p, + bool inverse) +{ + const union nf_inet_addr *addr; + __be16 port; + + if (p->pe_data && p->pe->hashkey_raw) + return p->pe->hashkey_raw(p, ip_vs_conn_rnd, inverse) & + ip_vs_conn_tab_mask; + + if (likely(!inverse)) { + addr = p->caddr; + port = p->cport; + } else { + addr = p->vaddr; + port = p->vport; + } + + return ip_vs_conn_hashkey(p->af, p->protocol, addr, port); +} + +static unsigned int ip_vs_conn_hashkey_conn(const struct ip_vs_conn *cp) +{ + struct ip_vs_conn_param p; + + ip_vs_conn_fill_param(cp->af, cp->protocol, &cp->caddr, cp->cport, + NULL, 0, &p); + + if (cp->dest && cp->dest->svc->pe) { + p.pe = cp->dest->svc->pe; + p.pe_data = cp->pe_data; + p.pe_data_len = cp->pe_data_len; + } + + return ip_vs_conn_hashkey_param(&p, false); +} /* * Hashes ip_vs_conn in ip_vs_conn_tab by proto,addr,port. @@ -162,7 +198,7 @@ static inline int ip_vs_conn_hash(struct ip_vs_conn *cp) return 0; /* Hash by protocol, client address and port */ - hash = ip_vs_conn_hashkey(cp->af, cp->protocol, &cp->caddr, cp->cport); + hash = ip_vs_conn_hashkey_conn(cp); ct_write_lock(hash); spin_lock(&cp->lock); @@ -195,7 +231,7 @@ static inline int ip_vs_conn_unhash(struct ip_vs_conn *cp) int ret; /* unhash it and decrease its reference counter */ - hash = ip_vs_conn_hashkey(cp->af, cp->protocol, &cp->caddr, cp->cport); + hash = ip_vs_conn_hashkey_conn(cp); ct_write_lock(hash); spin_lock(&cp->lock); @@ -218,27 +254,26 @@ static inline int ip_vs_conn_unhash(struct ip_vs_conn *cp) /* * Gets ip_vs_conn associated with supplied parameters in the ip_vs_conn_tab. * Called for pkts coming from OUTside-to-INside. - * s_addr, s_port: pkt source address (foreign host) - * d_addr, d_port: pkt dest address (load balancer) + * p->caddr, p->cport: pkt source address (foreign host) + * p->vaddr, p->vport: pkt dest address (load balancer) */ -static inline struct ip_vs_conn *__ip_vs_conn_in_get -(int af, int protocol, const union nf_inet_addr *s_addr, __be16 s_port, - const union nf_inet_addr *d_addr, __be16 d_port) +static inline struct ip_vs_conn * +__ip_vs_conn_in_get(const struct ip_vs_conn_param *p) { unsigned hash; struct ip_vs_conn *cp; - hash = ip_vs_conn_hashkey(af, protocol, s_addr, s_port); + hash = ip_vs_conn_hashkey_param(p, false); ct_read_lock(hash); list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { - if (cp->af == af && - ip_vs_addr_equal(af, s_addr, &cp->caddr) && - ip_vs_addr_equal(af, d_addr, &cp->vaddr) && - s_port == cp->cport && d_port == cp->vport && - ((!s_port) ^ (!(cp->flags & IP_VS_CONN_F_NO_CPORT))) && - protocol == cp->protocol) { + if (cp->af == p->af && + ip_vs_addr_equal(p->af, p->caddr, &cp->caddr) && + ip_vs_addr_equal(p->af, p->vaddr, &cp->vaddr) && + p->cport == cp->cport && p->vport == cp->vport && + ((!p->cport) ^ (!(cp->flags & IP_VS_CONN_F_NO_CPORT))) && + p->protocol == cp->protocol) { /* HIT */ atomic_inc(&cp->refcnt); ct_read_unlock(hash); @@ -251,99 +286,111 @@ static inline struct ip_vs_conn *__ip_vs_conn_in_get return NULL; } -struct ip_vs_conn *ip_vs_conn_in_get -(int af, int protocol, const union nf_inet_addr *s_addr, __be16 s_port, - const union nf_inet_addr *d_addr, __be16 d_port) +struct ip_vs_conn *ip_vs_conn_in_get(const struct ip_vs_conn_param *p) { struct ip_vs_conn *cp; - cp = __ip_vs_conn_in_get(af, protocol, s_addr, s_port, d_addr, d_port); - if (!cp && atomic_read(&ip_vs_conn_no_cport_cnt)) - cp = __ip_vs_conn_in_get(af, protocol, s_addr, 0, d_addr, - d_port); + cp = __ip_vs_conn_in_get(p); + if (!cp && atomic_read(&ip_vs_conn_no_cport_cnt)) { + struct ip_vs_conn_param cport_zero_p = *p; + cport_zero_p.cport = 0; + cp = __ip_vs_conn_in_get(&cport_zero_p); + } IP_VS_DBG_BUF(9, "lookup/in %s %s:%d->%s:%d %s\n", - ip_vs_proto_name(protocol), - IP_VS_DBG_ADDR(af, s_addr), ntohs(s_port), - IP_VS_DBG_ADDR(af, d_addr), ntohs(d_port), + ip_vs_proto_name(p->protocol), + IP_VS_DBG_ADDR(p->af, p->caddr), ntohs(p->cport), + IP_VS_DBG_ADDR(p->af, p->vaddr), ntohs(p->vport), cp ? "hit" : "not hit"); return cp; } +static int +ip_vs_conn_fill_param_proto(int af, const struct sk_buff *skb, + const struct ip_vs_iphdr *iph, + unsigned int proto_off, int inverse, + struct ip_vs_conn_param *p) +{ + __be16 _ports[2], *pptr; + + pptr = skb_header_pointer(skb, proto_off, sizeof(_ports), _ports); + if (pptr == NULL) + return 1; + + if (likely(!inverse)) + ip_vs_conn_fill_param(af, iph->protocol, &iph->saddr, pptr[0], + &iph->daddr, pptr[1], p); + else + ip_vs_conn_fill_param(af, iph->protocol, &iph->daddr, pptr[1], + &iph->saddr, pptr[0], p); + return 0; +} + struct ip_vs_conn * ip_vs_conn_in_get_proto(int af, const struct sk_buff *skb, struct ip_vs_protocol *pp, const struct ip_vs_iphdr *iph, unsigned int proto_off, int inverse) { - __be16 _ports[2], *pptr; + struct ip_vs_conn_param p; - pptr = skb_header_pointer(skb, proto_off, sizeof(_ports), _ports); - if (pptr == NULL) + if (ip_vs_conn_fill_param_proto(af, skb, iph, proto_off, inverse, &p)) return NULL; - if (likely(!inverse)) - return ip_vs_conn_in_get(af, iph->protocol, - &iph->saddr, pptr[0], - &iph->daddr, pptr[1]); - else - return ip_vs_conn_in_get(af, iph->protocol, - &iph->daddr, pptr[1], - &iph->saddr, pptr[0]); + return ip_vs_conn_in_get(&p); } EXPORT_SYMBOL_GPL(ip_vs_conn_in_get_proto); /* Get reference to connection template */ -struct ip_vs_conn *ip_vs_ct_in_get -(int af, int protocol, const union nf_inet_addr *s_addr, __be16 s_port, - const union nf_inet_addr *d_addr, __be16 d_port) +struct ip_vs_conn *ip_vs_ct_in_get(const struct ip_vs_conn_param *p) { unsigned hash; struct ip_vs_conn *cp; - hash = ip_vs_conn_hashkey(af, protocol, s_addr, s_port); + hash = ip_vs_conn_hashkey_param(p, false); ct_read_lock(hash); list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { - if (cp->af == af && - ip_vs_addr_equal(af, s_addr, &cp->caddr) && + if (p->pe_data && p->pe->ct_match) { + if (p->pe->ct_match(p, cp)) + goto out; + continue; + } + + if (cp->af == p->af && + ip_vs_addr_equal(p->af, p->caddr, &cp->caddr) && /* protocol should only be IPPROTO_IP if - * d_addr is a fwmark */ - ip_vs_addr_equal(protocol == IPPROTO_IP ? AF_UNSPEC : af, - d_addr, &cp->vaddr) && - s_port == cp->cport && d_port == cp->vport && + * p->vaddr is a fwmark */ + ip_vs_addr_equal(p->protocol == IPPROTO_IP ? AF_UNSPEC : + p->af, p->vaddr, &cp->vaddr) && + p->cport == cp->cport && p->vport == cp->vport && cp->flags & IP_VS_CONN_F_TEMPLATE && - protocol == cp->protocol) { - /* HIT */ - atomic_inc(&cp->refcnt); + p->protocol == cp->protocol) goto out; - } } cp = NULL; out: + if (cp) + atomic_inc(&cp->refcnt); ct_read_unlock(hash); IP_VS_DBG_BUF(9, "template lookup/in %s %s:%d->%s:%d %s\n", - ip_vs_proto_name(protocol), - IP_VS_DBG_ADDR(af, s_addr), ntohs(s_port), - IP_VS_DBG_ADDR(af, d_addr), ntohs(d_port), + ip_vs_proto_name(p->protocol), + IP_VS_DBG_ADDR(p->af, p->caddr), ntohs(p->cport), + IP_VS_DBG_ADDR(p->af, p->vaddr), ntohs(p->vport), cp ? "hit" : "not hit"); return cp; } -/* - * Gets ip_vs_conn associated with supplied parameters in the ip_vs_conn_tab. - * Called for pkts coming from inside-to-OUTside. - * s_addr, s_port: pkt source address (inside host) - * d_addr, d_port: pkt dest address (foreign host) - */ -struct ip_vs_conn *ip_vs_conn_out_get -(int af, int protocol, const union nf_inet_addr *s_addr, __be16 s_port, - const union nf_inet_addr *d_addr, __be16 d_port) +/* Gets ip_vs_conn associated with supplied parameters in the ip_vs_conn_tab. + * Called for pkts coming from inside-to-OUTside. + * p->caddr, p->cport: pkt source address (inside host) + * p->vaddr, p->vport: pkt dest address (foreign host) */ +struct ip_vs_conn *ip_vs_conn_out_get(const struct ip_vs_conn_param *p) { unsigned hash; struct ip_vs_conn *cp, *ret=NULL; @@ -351,16 +398,16 @@ struct ip_vs_conn *ip_vs_conn_out_get /* * Check for "full" addressed entries */ - hash = ip_vs_conn_hashkey(af, protocol, d_addr, d_port); + hash = ip_vs_conn_hashkey_param(p, true); ct_read_lock(hash); list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { - if (cp->af == af && - ip_vs_addr_equal(af, d_addr, &cp->caddr) && - ip_vs_addr_equal(af, s_addr, &cp->daddr) && - d_port == cp->cport && s_port == cp->dport && - protocol == cp->protocol) { + if (cp->af == p->af && + ip_vs_addr_equal(p->af, p->vaddr, &cp->caddr) && + ip_vs_addr_equal(p->af, p->caddr, &cp->daddr) && + p->vport == cp->cport && p->cport == cp->dport && + p->protocol == cp->protocol) { /* HIT */ atomic_inc(&cp->refcnt); ret = cp; @@ -371,9 +418,9 @@ struct ip_vs_conn *ip_vs_conn_out_get ct_read_unlock(hash); IP_VS_DBG_BUF(9, "lookup/out %s %s:%d->%s:%d %s\n", - ip_vs_proto_name(protocol), - IP_VS_DBG_ADDR(af, s_addr), ntohs(s_port), - IP_VS_DBG_ADDR(af, d_addr), ntohs(d_port), + ip_vs_proto_name(p->protocol), + IP_VS_DBG_ADDR(p->af, p->caddr), ntohs(p->cport), + IP_VS_DBG_ADDR(p->af, p->vaddr), ntohs(p->vport), ret ? "hit" : "not hit"); return ret; @@ -385,20 +432,12 @@ ip_vs_conn_out_get_proto(int af, const struct sk_buff *skb, const struct ip_vs_iphdr *iph, unsigned int proto_off, int inverse) { - __be16 _ports[2], *pptr; + struct ip_vs_conn_param p; - pptr = skb_header_pointer(skb, proto_off, sizeof(_ports), _ports); - if (pptr == NULL) + if (ip_vs_conn_fill_param_proto(af, skb, iph, proto_off, inverse, &p)) return NULL; - if (likely(!inverse)) - return ip_vs_conn_out_get(af, iph->protocol, - &iph->saddr, pptr[0], - &iph->daddr, pptr[1]); - else - return ip_vs_conn_out_get(af, iph->protocol, - &iph->daddr, pptr[1], - &iph->saddr, pptr[0]); + return ip_vs_conn_out_get(&p); } EXPORT_SYMBOL_GPL(ip_vs_conn_out_get_proto); @@ -505,6 +544,8 @@ static inline int ip_vs_dest_totalconns(struct ip_vs_dest *dest) static inline void ip_vs_bind_dest(struct ip_vs_conn *cp, struct ip_vs_dest *dest) { + unsigned int conn_flags; + /* if dest is NULL, then return directly */ if (!dest) return; @@ -512,16 +553,20 @@ ip_vs_bind_dest(struct ip_vs_conn *cp, struct ip_vs_dest *dest) /* Increase the refcnt counter of the dest */ atomic_inc(&dest->refcnt); + conn_flags = atomic_read(&dest->conn_flags); + if (cp->protocol != IPPROTO_UDP) + conn_flags &= ~IP_VS_CONN_F_ONE_PACKET; /* Bind with the destination and its corresponding transmitter */ - if ((cp->flags & IP_VS_CONN_F_SYNC) && - (!(cp->flags & IP_VS_CONN_F_TEMPLATE))) + if (cp->flags & IP_VS_CONN_F_SYNC) { /* if the connection is not template and is created * by sync, preserve the activity flag. */ - cp->flags |= atomic_read(&dest->conn_flags) & - (~IP_VS_CONN_F_INACTIVE); - else - cp->flags |= atomic_read(&dest->conn_flags); + if (!(cp->flags & IP_VS_CONN_F_TEMPLATE)) + conn_flags &= ~IP_VS_CONN_F_INACTIVE; + /* connections inherit forwarding method from dest */ + cp->flags &= ~IP_VS_CONN_F_FWD_MASK; + } + cp->flags |= conn_flags; cp->dest = dest; IP_VS_DBG_BUF(7, "Bind-dest %s c:%s:%d v:%s:%d " @@ -717,6 +762,10 @@ static void ip_vs_conn_expire(unsigned long data) if (cp->control) ip_vs_control_del(cp); + if (cp->flags & IP_VS_CONN_F_NFCT) + ip_vs_conn_drop_conntrack(cp); + + kfree(cp->pe_data); if (unlikely(cp->app != NULL)) ip_vs_unbind_app(cp); ip_vs_unbind_dest(cp); @@ -751,13 +800,12 @@ void ip_vs_conn_expire_now(struct ip_vs_conn *cp) * Create a new connection entry and hash it into the ip_vs_conn_tab */ struct ip_vs_conn * -ip_vs_conn_new(int af, int proto, const union nf_inet_addr *caddr, __be16 cport, - const union nf_inet_addr *vaddr, __be16 vport, +ip_vs_conn_new(const struct ip_vs_conn_param *p, const union nf_inet_addr *daddr, __be16 dport, unsigned flags, struct ip_vs_dest *dest) { struct ip_vs_conn *cp; - struct ip_vs_protocol *pp = ip_vs_proto_get(proto); + struct ip_vs_protocol *pp = ip_vs_proto_get(p->protocol); cp = kmem_cache_zalloc(ip_vs_conn_cachep, GFP_ATOMIC); if (cp == NULL) { @@ -767,17 +815,21 @@ ip_vs_conn_new(int af, int proto, const union nf_inet_addr *caddr, __be16 cport, INIT_LIST_HEAD(&cp->c_list); setup_timer(&cp->timer, ip_vs_conn_expire, (unsigned long)cp); - cp->af = af; - cp->protocol = proto; - ip_vs_addr_copy(af, &cp->caddr, caddr); - cp->cport = cport; - ip_vs_addr_copy(af, &cp->vaddr, vaddr); - cp->vport = vport; + cp->af = p->af; + cp->protocol = p->protocol; + ip_vs_addr_copy(p->af, &cp->caddr, p->caddr); + cp->cport = p->cport; + ip_vs_addr_copy(p->af, &cp->vaddr, p->vaddr); + cp->vport = p->vport; /* proto should only be IPPROTO_IP if d_addr is a fwmark */ - ip_vs_addr_copy(proto == IPPROTO_IP ? AF_UNSPEC : af, + ip_vs_addr_copy(p->protocol == IPPROTO_IP ? AF_UNSPEC : p->af, &cp->daddr, daddr); cp->dport = dport; cp->flags = flags; + if (flags & IP_VS_CONN_F_TEMPLATE && p->pe_data) { + cp->pe_data = p->pe_data; + cp->pe_data_len = p->pe_data_len; + } spin_lock_init(&cp->lock); /* @@ -803,7 +855,7 @@ ip_vs_conn_new(int af, int proto, const union nf_inet_addr *caddr, __be16 cport, /* Bind its packet transmitter */ #ifdef CONFIG_IP_VS_IPV6 - if (af == AF_INET6) + if (p->af == AF_INET6) ip_vs_bind_xmit_v6(cp); else #endif @@ -812,13 +864,22 @@ ip_vs_conn_new(int af, int proto, const union nf_inet_addr *caddr, __be16 cport, if (unlikely(pp && atomic_read(&pp->appcnt))) ip_vs_bind_app(cp, pp); + /* + * Allow conntrack to be preserved. By default, conntrack + * is created and destroyed for every packet. + * Sometimes keeping conntrack can be useful for + * IP_VS_CONN_F_ONE_PACKET too. + */ + + if (ip_vs_conntrack_enabled()) + cp->flags |= IP_VS_CONN_F_NFCT; + /* Hash it in the ip_vs_conn_tab finally */ ip_vs_conn_hash(cp); return cp; } - /* * /proc/net/ip_vs_conn entries */ @@ -834,7 +895,7 @@ static void *ip_vs_conn_array(struct seq_file *seq, loff_t pos) list_for_each_entry(cp, &ip_vs_conn_tab[idx], c_list) { if (pos-- == 0) { seq->private = &ip_vs_conn_tab[idx]; - return cp; + return cp; } } ct_read_unlock_bh(idx); @@ -891,30 +952,45 @@ static int ip_vs_conn_seq_show(struct seq_file *seq, void *v) if (v == SEQ_START_TOKEN) seq_puts(seq, - "Pro FromIP FPrt ToIP TPrt DestIP DPrt State Expires\n"); + "Pro FromIP FPrt ToIP TPrt DestIP DPrt State Expires PEName PEData\n"); else { const struct ip_vs_conn *cp = v; + char pe_data[IP_VS_PENAME_MAXLEN + IP_VS_PEDATA_MAXLEN + 3]; + size_t len = 0; + + if (cp->dest && cp->pe_data && + cp->dest->svc->pe->show_pe_data) { + pe_data[0] = ' '; + len = strlen(cp->dest->svc->pe->name); + memcpy(pe_data + 1, cp->dest->svc->pe->name, len); + pe_data[len + 1] = ' '; + len += 2; + len += cp->dest->svc->pe->show_pe_data(cp, + pe_data + len); + } + pe_data[len] = '\0'; #ifdef CONFIG_IP_VS_IPV6 if (cp->af == AF_INET6) - seq_printf(seq, "%-3s %pI6 %04X %pI6 %04X %pI6 %04X %-11s %7lu\n", + seq_printf(seq, "%-3s %pI6 %04X %pI6 %04X " + "%pI6 %04X %-11s %7lu%s\n", ip_vs_proto_name(cp->protocol), &cp->caddr.in6, ntohs(cp->cport), &cp->vaddr.in6, ntohs(cp->vport), &cp->daddr.in6, ntohs(cp->dport), ip_vs_state_name(cp->protocol, cp->state), - (cp->timer.expires-jiffies)/HZ); + (cp->timer.expires-jiffies)/HZ, pe_data); else #endif seq_printf(seq, "%-3s %08X %04X %08X %04X" - " %08X %04X %-11s %7lu\n", + " %08X %04X %-11s %7lu%s\n", ip_vs_proto_name(cp->protocol), ntohl(cp->caddr.ip), ntohs(cp->cport), ntohl(cp->vaddr.ip), ntohs(cp->vport), ntohl(cp->daddr.ip), ntohs(cp->dport), ip_vs_state_name(cp->protocol, cp->state), - (cp->timer.expires-jiffies)/HZ); + (cp->timer.expires-jiffies)/HZ, pe_data); } return 0; } diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 4f8ddba4801..b4e51e9c5a0 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -40,6 +40,7 @@ #include <net/udp.h> #include <net/icmp.h> /* for icmp_send */ #include <net/route.h> +#include <net/ip6_checksum.h> #include <linux/netfilter.h> #include <linux/netfilter_ipv4.h> @@ -47,6 +48,7 @@ #ifdef CONFIG_IP_VS_IPV6 #include <net/ipv6.h> #include <linux/netfilter_ipv6.h> +#include <net/ip6_route.h> #endif #include <net/ip_vs.h> @@ -175,6 +177,18 @@ ip_vs_set_state(struct ip_vs_conn *cp, int direction, return pp->state_transition(cp, direction, skb, pp); } +static inline void +ip_vs_conn_fill_param_persist(const struct ip_vs_service *svc, + struct sk_buff *skb, int protocol, + const union nf_inet_addr *caddr, __be16 cport, + const union nf_inet_addr *vaddr, __be16 vport, + struct ip_vs_conn_param *p) +{ + ip_vs_conn_fill_param(svc->af, protocol, caddr, cport, vaddr, vport, p); + p->pe = svc->pe; + if (p->pe && p->pe->fill_param) + p->pe->fill_param(p, skb); +} /* * IPVS persistent scheduling function @@ -185,15 +199,16 @@ ip_vs_set_state(struct ip_vs_conn *cp, int direction, */ static struct ip_vs_conn * ip_vs_sched_persist(struct ip_vs_service *svc, - const struct sk_buff *skb, + struct sk_buff *skb, __be16 ports[2]) { struct ip_vs_conn *cp = NULL; struct ip_vs_iphdr iph; struct ip_vs_dest *dest; struct ip_vs_conn *ct; - __be16 dport; /* destination port to forward */ - __be16 flags; + __be16 dport = 0; /* destination port to forward */ + unsigned int flags; + struct ip_vs_conn_param param; union nf_inet_addr snet; /* source network of the client, after masking */ @@ -226,120 +241,75 @@ ip_vs_sched_persist(struct ip_vs_service *svc, * service, and a template like <caddr, 0, vaddr, vport, daddr, dport> * is created for other persistent services. */ - if (ports[1] == svc->port) { - /* Check if a template already exists */ - if (svc->port != FTPPORT) - ct = ip_vs_ct_in_get(svc->af, iph.protocol, &snet, 0, - &iph.daddr, ports[1]); - else - ct = ip_vs_ct_in_get(svc->af, iph.protocol, &snet, 0, - &iph.daddr, 0); - - if (!ct || !ip_vs_check_template(ct)) { - /* - * No template found or the dest of the connection - * template is not available. - */ - dest = svc->scheduler->schedule(svc, skb); - if (dest == NULL) { - IP_VS_DBG(1, "p-schedule: no dest found.\n"); - return NULL; - } - - /* - * Create a template like <protocol,caddr,0, - * vaddr,vport,daddr,dport> for non-ftp service, - * and <protocol,caddr,0,vaddr,0,daddr,0> - * for ftp service. + { + int protocol = iph.protocol; + const union nf_inet_addr *vaddr = &iph.daddr; + const union nf_inet_addr fwmark = { .ip = htonl(svc->fwmark) }; + __be16 vport = 0; + + if (ports[1] == svc->port) { + /* non-FTP template: + * <protocol, caddr, 0, vaddr, vport, daddr, dport> + * FTP template: + * <protocol, caddr, 0, vaddr, 0, daddr, 0> */ if (svc->port != FTPPORT) - ct = ip_vs_conn_new(svc->af, iph.protocol, - &snet, 0, - &iph.daddr, - ports[1], - &dest->addr, dest->port, - IP_VS_CONN_F_TEMPLATE, - dest); - else - ct = ip_vs_conn_new(svc->af, iph.protocol, - &snet, 0, - &iph.daddr, 0, - &dest->addr, 0, - IP_VS_CONN_F_TEMPLATE, - dest); - if (ct == NULL) - return NULL; - - ct->timeout = svc->timeout; + vport = ports[1]; } else { - /* set destination with the found template */ - dest = ct->dest; - } - dport = dest->port; - } else { - /* - * Note: persistent fwmark-based services and persistent - * port zero service are handled here. - * fwmark template: <IPPROTO_IP,caddr,0,fwmark,0,daddr,0> - * port zero template: <protocol,caddr,0,vaddr,0,daddr,0> - */ - if (svc->fwmark) { - union nf_inet_addr fwmark = { - .ip = htonl(svc->fwmark) - }; - - ct = ip_vs_ct_in_get(svc->af, IPPROTO_IP, &snet, 0, - &fwmark, 0); - } else - ct = ip_vs_ct_in_get(svc->af, iph.protocol, &snet, 0, - &iph.daddr, 0); - - if (!ct || !ip_vs_check_template(ct)) { - /* - * If it is not persistent port zero, return NULL, - * otherwise create a connection template. + /* Note: persistent fwmark-based services and + * persistent port zero service are handled here. + * fwmark template: + * <IPPROTO_IP,caddr,0,fwmark,0,daddr,0> + * port zero template: + * <protocol,caddr,0,vaddr,0,daddr,0> */ - if (svc->port) - return NULL; - - dest = svc->scheduler->schedule(svc, skb); - if (dest == NULL) { - IP_VS_DBG(1, "p-schedule: no dest found.\n"); - return NULL; + if (svc->fwmark) { + protocol = IPPROTO_IP; + vaddr = &fwmark; } + } + ip_vs_conn_fill_param_persist(svc, skb, protocol, &snet, 0, + vaddr, vport, ¶m); + } - /* - * Create a template according to the service - */ - if (svc->fwmark) { - union nf_inet_addr fwmark = { - .ip = htonl(svc->fwmark) - }; - - ct = ip_vs_conn_new(svc->af, IPPROTO_IP, - &snet, 0, - &fwmark, 0, - &dest->addr, 0, - IP_VS_CONN_F_TEMPLATE, - dest); - } else - ct = ip_vs_conn_new(svc->af, iph.protocol, - &snet, 0, - &iph.daddr, 0, - &dest->addr, 0, - IP_VS_CONN_F_TEMPLATE, - dest); - if (ct == NULL) - return NULL; - - ct->timeout = svc->timeout; - } else { - /* set destination with the found template */ - dest = ct->dest; + /* Check if a template already exists */ + ct = ip_vs_ct_in_get(¶m); + if (!ct || !ip_vs_check_template(ct)) { + /* No template found or the dest of the connection + * template is not available. + */ + dest = svc->scheduler->schedule(svc, skb); + if (!dest) { + IP_VS_DBG(1, "p-schedule: no dest found.\n"); + kfree(param.pe_data); + return NULL; } - dport = ports[1]; + + if (ports[1] == svc->port && svc->port != FTPPORT) + dport = dest->port; + + /* Create a template + * This adds param.pe_data to the template, + * and thus param.pe_data will be destroyed + * when the template expires */ + ct = ip_vs_conn_new(¶m, &dest->addr, dport, + IP_VS_CONN_F_TEMPLATE, dest); + if (ct == NULL) { + kfree(param.pe_data); + return NULL; + } + + ct->timeout = svc->timeout; + } else { + /* set destination with the found template */ + dest = ct->dest; + kfree(param.pe_data); } + dport = ports[1]; + if (dport == svc->port && dest->port) + dport = dest->port; + flags = (svc->flags & IP_VS_SVC_F_ONEPACKET && iph.protocol == IPPROTO_UDP)? IP_VS_CONN_F_ONE_PACKET : 0; @@ -347,12 +317,9 @@ ip_vs_sched_persist(struct ip_vs_service *svc, /* * Create a new connection according to the template */ - cp = ip_vs_conn_new(svc->af, iph.protocol, - &iph.saddr, ports[0], - &iph.daddr, ports[1], - &dest->addr, dport, - flags, - dest); + ip_vs_conn_fill_param(svc->af, iph.protocol, &iph.saddr, ports[0], + &iph.daddr, ports[1], ¶m); + cp = ip_vs_conn_new(¶m, &dest->addr, dport, flags, dest); if (cp == NULL) { ip_vs_conn_put(ct); return NULL; @@ -376,23 +343,53 @@ ip_vs_sched_persist(struct ip_vs_service *svc, * Protocols supported: TCP, UDP */ struct ip_vs_conn * -ip_vs_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) +ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, + struct ip_vs_protocol *pp, int *ignored) { struct ip_vs_conn *cp = NULL; struct ip_vs_iphdr iph; struct ip_vs_dest *dest; - __be16 _ports[2], *pptr, flags; + __be16 _ports[2], *pptr; + unsigned int flags; + *ignored = 1; ip_vs_fill_iphdr(svc->af, skb_network_header(skb), &iph); pptr = skb_header_pointer(skb, iph.len, sizeof(_ports), _ports); if (pptr == NULL) return NULL; /* + * FTPDATA needs this check when using local real server. + * Never schedule Active FTPDATA connections from real server. + * For LVS-NAT they must be already created. For other methods + * with persistence the connection is created on SYN+ACK. + */ + if (pptr[0] == FTPDATA) { + IP_VS_DBG_PKT(12, svc->af, pp, skb, 0, + "Not scheduling FTPDATA"); + return NULL; + } + + /* + * Do not schedule replies from local real server. It is risky + * for fwmark services but mostly for persistent services. + */ + if ((!skb->dev || skb->dev->flags & IFF_LOOPBACK) && + (svc->flags & IP_VS_SVC_F_PERSISTENT || svc->fwmark) && + (cp = pp->conn_in_get(svc->af, skb, pp, &iph, iph.len, 1))) { + IP_VS_DBG_PKT(12, svc->af, pp, skb, 0, + "Not scheduling reply for existing connection"); + __ip_vs_conn_put(cp); + return NULL; + } + + /* * Persistent service */ - if (svc->flags & IP_VS_SVC_F_PERSISTENT) + if (svc->flags & IP_VS_SVC_F_PERSISTENT) { + *ignored = 0; return ip_vs_sched_persist(svc, skb, pptr); + } /* * Non-persistent service @@ -405,6 +402,8 @@ ip_vs_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) return NULL; } + *ignored = 0; + dest = svc->scheduler->schedule(svc, skb); if (dest == NULL) { IP_VS_DBG(1, "Schedule: no dest found.\n"); @@ -418,14 +417,16 @@ ip_vs_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) /* * Create a connection entry. */ - cp = ip_vs_conn_new(svc->af, iph.protocol, - &iph.saddr, pptr[0], - &iph.daddr, pptr[1], - &dest->addr, dest->port ? dest->port : pptr[1], - flags, - dest); - if (cp == NULL) - return NULL; + { + struct ip_vs_conn_param p; + ip_vs_conn_fill_param(svc->af, iph.protocol, &iph.saddr, + pptr[0], &iph.daddr, pptr[1], &p); + cp = ip_vs_conn_new(&p, &dest->addr, + dest->port ? dest->port : pptr[1], + flags, dest); + if (!cp) + return NULL; + } IP_VS_DBG_BUF(6, "Schedule fwd:%c c:%s:%u v:%s:%u " "d:%s:%u conn->flags:%X conn->refcnt:%d\n", @@ -472,23 +473,26 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, if (sysctl_ip_vs_cache_bypass && svc->fwmark && unicast) { int ret, cs; struct ip_vs_conn *cp; - __u16 flags = (svc->flags & IP_VS_SVC_F_ONEPACKET && - iph.protocol == IPPROTO_UDP)? - IP_VS_CONN_F_ONE_PACKET : 0; + unsigned int flags = (svc->flags & IP_VS_SVC_F_ONEPACKET && + iph.protocol == IPPROTO_UDP)? + IP_VS_CONN_F_ONE_PACKET : 0; union nf_inet_addr daddr = { .all = { 0, 0, 0, 0 } }; ip_vs_service_put(svc); /* create a new connection entry */ IP_VS_DBG(6, "%s(): create a cache_bypass entry\n", __func__); - cp = ip_vs_conn_new(svc->af, iph.protocol, - &iph.saddr, pptr[0], - &iph.daddr, pptr[1], - &daddr, 0, - IP_VS_CONN_F_BYPASS | flags, - NULL); - if (cp == NULL) - return NF_DROP; + { + struct ip_vs_conn_param p; + ip_vs_conn_fill_param(svc->af, iph.protocol, + &iph.saddr, pptr[0], + &iph.daddr, pptr[1], &p); + cp = ip_vs_conn_new(&p, &daddr, 0, + IP_VS_CONN_F_BYPASS | flags, + NULL); + if (!cp) + return NF_DROP; + } /* statistics */ ip_vs_in_stats(cp, skb); @@ -526,9 +530,14 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, * ICMP_PORT_UNREACH is sent here no matter it is TCP/UDP. --WZ */ #ifdef CONFIG_IP_VS_IPV6 - if (svc->af == AF_INET6) + if (svc->af == AF_INET6) { + if (!skb->dev) { + struct net *net = dev_net(skb_dst(skb)->dev); + + skb->dev = net->loopback_dev; + } icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0); - else + } else #endif icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); @@ -540,6 +549,15 @@ __sum16 ip_vs_checksum_complete(struct sk_buff *skb, int offset) return csum_fold(skb_checksum(skb, offset, skb->len - offset, 0)); } +static inline enum ip_defrag_users ip_vs_defrag_user(unsigned int hooknum) +{ + if (NF_INET_LOCAL_IN == hooknum) + return IP_DEFRAG_VS_IN; + if (NF_INET_FORWARD == hooknum) + return IP_DEFRAG_VS_FWD; + return IP_DEFRAG_VS_OUT; +} + static inline int ip_vs_gather_frags(struct sk_buff *skb, u_int32_t user) { int err = ip_defrag(skb, user); @@ -600,10 +618,10 @@ void ip_vs_nat_icmp(struct sk_buff *skb, struct ip_vs_protocol *pp, skb->ip_summed = CHECKSUM_UNNECESSARY; if (inout) - IP_VS_DBG_PKT(11, pp, skb, (void *)ciph - (void *)iph, + IP_VS_DBG_PKT(11, AF_INET, pp, skb, (void *)ciph - (void *)iph, "Forwarding altered outgoing ICMP"); else - IP_VS_DBG_PKT(11, pp, skb, (void *)ciph - (void *)iph, + IP_VS_DBG_PKT(11, AF_INET, pp, skb, (void *)ciph - (void *)iph, "Forwarding altered incoming ICMP"); } @@ -637,17 +655,21 @@ void ip_vs_nat_icmp_v6(struct sk_buff *skb, struct ip_vs_protocol *pp, } /* And finally the ICMP checksum */ - icmph->icmp6_cksum = 0; - /* TODO IPv6: is this correct for ICMPv6? */ - ip_vs_checksum_complete(skb, icmp_offset); - skb->ip_summed = CHECKSUM_UNNECESSARY; + icmph->icmp6_cksum = ~csum_ipv6_magic(&iph->saddr, &iph->daddr, + skb->len - icmp_offset, + IPPROTO_ICMPV6, 0); + skb->csum_start = skb_network_header(skb) - skb->head + icmp_offset; + skb->csum_offset = offsetof(struct icmp6hdr, icmp6_cksum); + skb->ip_summed = CHECKSUM_PARTIAL; if (inout) - IP_VS_DBG_PKT(11, pp, skb, (void *)ciph - (void *)iph, - "Forwarding altered outgoing ICMPv6"); + IP_VS_DBG_PKT(11, AF_INET6, pp, skb, + (void *)ciph - (void *)iph, + "Forwarding altered outgoing ICMPv6"); else - IP_VS_DBG_PKT(11, pp, skb, (void *)ciph - (void *)iph, - "Forwarding altered incoming ICMPv6"); + IP_VS_DBG_PKT(11, AF_INET6, pp, skb, + (void *)ciph - (void *)iph, + "Forwarding altered incoming ICMPv6"); } #endif @@ -688,10 +710,25 @@ static int handle_response_icmp(int af, struct sk_buff *skb, #endif ip_vs_nat_icmp(skb, pp, cp, 1); +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) { + if (sysctl_ip_vs_snat_reroute && ip6_route_me_harder(skb) != 0) + goto out; + } else +#endif + if ((sysctl_ip_vs_snat_reroute || + skb_rtable(skb)->rt_flags & RTCF_LOCAL) && + ip_route_me_harder(skb, RTN_LOCAL) != 0) + goto out; + /* do the statistics and put it back */ ip_vs_out_stats(cp, skb); skb->ipvs_property = 1; + if (!(cp->flags & IP_VS_CONN_F_NFCT)) + ip_vs_notrack(skb); + else + ip_vs_update_conntrack(skb, cp, 0); verdict = NF_ACCEPT; out: @@ -705,7 +742,8 @@ out: * Find any that might be relevant, check against existing connections. * Currently handles error types - unreachable, quench, ttl exceeded. */ -static int ip_vs_out_icmp(struct sk_buff *skb, int *related) +static int ip_vs_out_icmp(struct sk_buff *skb, int *related, + unsigned int hooknum) { struct iphdr *iph; struct icmphdr _icmph, *ic; @@ -720,7 +758,7 @@ static int ip_vs_out_icmp(struct sk_buff *skb, int *related) /* reassemble IP fragments */ if (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) { - if (ip_vs_gather_frags(skb, IP_DEFRAG_VS_OUT)) + if (ip_vs_gather_frags(skb, ip_vs_defrag_user(hooknum))) return NF_STOLEN; } @@ -763,7 +801,8 @@ static int ip_vs_out_icmp(struct sk_buff *skb, int *related) pp->dont_defrag)) return NF_ACCEPT; - IP_VS_DBG_PKT(11, pp, skb, offset, "Checking outgoing ICMP for"); + IP_VS_DBG_PKT(11, AF_INET, pp, skb, offset, + "Checking outgoing ICMP for"); offset += cih->ihl * 4; @@ -779,7 +818,8 @@ static int ip_vs_out_icmp(struct sk_buff *skb, int *related) } #ifdef CONFIG_IP_VS_IPV6 -static int ip_vs_out_icmp_v6(struct sk_buff *skb, int *related) +static int ip_vs_out_icmp_v6(struct sk_buff *skb, int *related, + unsigned int hooknum) { struct ipv6hdr *iph; struct icmp6hdr _icmph, *ic; @@ -795,7 +835,7 @@ static int ip_vs_out_icmp_v6(struct sk_buff *skb, int *related) /* reassemble IP fragments */ if (ipv6_hdr(skb)->nexthdr == IPPROTO_FRAGMENT) { - if (ip_vs_gather_frags_v6(skb, IP_DEFRAG_VS_OUT)) + if (ip_vs_gather_frags_v6(skb, ip_vs_defrag_user(hooknum))) return NF_STOLEN; } @@ -838,7 +878,8 @@ static int ip_vs_out_icmp_v6(struct sk_buff *skb, int *related) if (unlikely(cih->nexthdr == IPPROTO_FRAGMENT && pp->dont_defrag)) return NF_ACCEPT; - IP_VS_DBG_PKT(11, pp, skb, offset, "Checking outgoing ICMPv6 for"); + IP_VS_DBG_PKT(11, AF_INET6, pp, skb, offset, + "Checking outgoing ICMPv6 for"); offset += sizeof(struct ipv6hdr); @@ -886,7 +927,7 @@ static unsigned int handle_response(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, struct ip_vs_conn *cp, int ihl) { - IP_VS_DBG_PKT(11, pp, skb, 0, "Outgoing packet"); + IP_VS_DBG_PKT(11, af, pp, skb, 0, "Outgoing packet"); if (!skb_make_writable(skb, ihl)) goto drop; @@ -905,6 +946,15 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, ip_send_check(ip_hdr(skb)); } + /* + * nf_iterate does not expect change in the skb->dst->dev. + * It looks like it is not fatal to enable this code for hooks + * where our handlers are at the end of the chain list and + * when all next handlers use skb->dst->dev and not outdev. + * It will definitely route properly the inout NAT traffic + * when multiple paths are used. + */ + /* For policy routing, packets originating from this * machine itself may be routed differently to packets * passing through. We want this packet to be routed as @@ -913,20 +963,25 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, */ #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) { - if (ip6_route_me_harder(skb) != 0) + if (sysctl_ip_vs_snat_reroute && ip6_route_me_harder(skb) != 0) goto drop; } else #endif - if (ip_route_me_harder(skb, RTN_LOCAL) != 0) + if ((sysctl_ip_vs_snat_reroute || + skb_rtable(skb)->rt_flags & RTCF_LOCAL) && + ip_route_me_harder(skb, RTN_LOCAL) != 0) goto drop; - IP_VS_DBG_PKT(10, pp, skb, 0, "After SNAT"); + IP_VS_DBG_PKT(10, af, pp, skb, 0, "After SNAT"); ip_vs_out_stats(cp, skb); ip_vs_set_state(cp, IP_VS_DIR_OUTPUT, skb, pp); - ip_vs_conn_put(cp); - skb->ipvs_property = 1; + if (!(cp->flags & IP_VS_CONN_F_NFCT)) + ip_vs_notrack(skb); + else + ip_vs_update_conntrack(skb, cp, 0); + ip_vs_conn_put(cp); LeaveFunction(11); return NF_ACCEPT; @@ -934,35 +989,46 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, drop: ip_vs_conn_put(cp); kfree_skb(skb); + LeaveFunction(11); return NF_STOLEN; } /* - * It is hooked at the NF_INET_FORWARD chain, used only for VS/NAT. * Check if outgoing packet belongs to the established ip_vs_conn. */ static unsigned int -ip_vs_out(unsigned int hooknum, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) +ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af) { struct ip_vs_iphdr iph; struct ip_vs_protocol *pp; struct ip_vs_conn *cp; - int af; EnterFunction(11); - af = (skb->protocol == htons(ETH_P_IP)) ? AF_INET : AF_INET6; - + /* Already marked as IPVS request or reply? */ if (skb->ipvs_property) return NF_ACCEPT; + /* Bad... Do not break raw sockets */ + if (unlikely(skb->sk != NULL && hooknum == NF_INET_LOCAL_OUT && + af == AF_INET)) { + struct sock *sk = skb->sk; + struct inet_sock *inet = inet_sk(skb->sk); + + if (inet && sk->sk_family == PF_INET && inet->nodefrag) + return NF_ACCEPT; + } + + if (unlikely(!skb_dst(skb))) + return NF_ACCEPT; + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) { if (unlikely(iph.protocol == IPPROTO_ICMPV6)) { - int related, verdict = ip_vs_out_icmp_v6(skb, &related); + int related; + int verdict = ip_vs_out_icmp_v6(skb, &related, + hooknum); if (related) return verdict; @@ -971,7 +1037,8 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, } else #endif if (unlikely(iph.protocol == IPPROTO_ICMP)) { - int related, verdict = ip_vs_out_icmp(skb, &related); + int related; + int verdict = ip_vs_out_icmp(skb, &related, hooknum); if (related) return verdict; @@ -985,19 +1052,19 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, /* reassemble IP fragments */ #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) { - if (unlikely(iph.protocol == IPPROTO_ICMPV6)) { - int related, verdict = ip_vs_out_icmp_v6(skb, &related); - - if (related) - return verdict; - - ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); + if (ipv6_hdr(skb)->nexthdr == IPPROTO_FRAGMENT) { + if (ip_vs_gather_frags_v6(skb, + ip_vs_defrag_user(hooknum))) + return NF_STOLEN; } + + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); } else #endif if (unlikely(ip_hdr(skb)->frag_off & htons(IP_MF|IP_OFFSET) && !pp->dont_defrag)) { - if (ip_vs_gather_frags(skb, IP_DEFRAG_VS_OUT)) + if (ip_vs_gather_frags(skb, + ip_vs_defrag_user(hooknum))) return NF_STOLEN; ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); @@ -1008,55 +1075,123 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, */ cp = pp->conn_out_get(af, skb, pp, &iph, iph.len, 0); - if (unlikely(!cp)) { - if (sysctl_ip_vs_nat_icmp_send && - (pp->protocol == IPPROTO_TCP || - pp->protocol == IPPROTO_UDP || - pp->protocol == IPPROTO_SCTP)) { - __be16 _ports[2], *pptr; - - pptr = skb_header_pointer(skb, iph.len, - sizeof(_ports), _ports); - if (pptr == NULL) - return NF_ACCEPT; /* Not for me */ - if (ip_vs_lookup_real_service(af, iph.protocol, - &iph.saddr, - pptr[0])) { - /* - * Notify the real server: there is no - * existing entry if it is not RST - * packet or not TCP packet. - */ - if ((iph.protocol != IPPROTO_TCP && - iph.protocol != IPPROTO_SCTP) - || ((iph.protocol == IPPROTO_TCP - && !is_tcp_reset(skb, iph.len)) - || (iph.protocol == IPPROTO_SCTP - && !is_sctp_abort(skb, - iph.len)))) { + if (likely(cp)) + return handle_response(af, skb, pp, cp, iph.len); + if (sysctl_ip_vs_nat_icmp_send && + (pp->protocol == IPPROTO_TCP || + pp->protocol == IPPROTO_UDP || + pp->protocol == IPPROTO_SCTP)) { + __be16 _ports[2], *pptr; + + pptr = skb_header_pointer(skb, iph.len, + sizeof(_ports), _ports); + if (pptr == NULL) + return NF_ACCEPT; /* Not for me */ + if (ip_vs_lookup_real_service(af, iph.protocol, + &iph.saddr, + pptr[0])) { + /* + * Notify the real server: there is no + * existing entry if it is not RST + * packet or not TCP packet. + */ + if ((iph.protocol != IPPROTO_TCP && + iph.protocol != IPPROTO_SCTP) + || ((iph.protocol == IPPROTO_TCP + && !is_tcp_reset(skb, iph.len)) + || (iph.protocol == IPPROTO_SCTP + && !is_sctp_abort(skb, + iph.len)))) { #ifdef CONFIG_IP_VS_IPV6 - if (af == AF_INET6) - icmpv6_send(skb, - ICMPV6_DEST_UNREACH, - ICMPV6_PORT_UNREACH, - 0); - else + if (af == AF_INET6) { + struct net *net = + dev_net(skb_dst(skb)->dev); + + if (!skb->dev) + skb->dev = net->loopback_dev; + icmpv6_send(skb, + ICMPV6_DEST_UNREACH, + ICMPV6_PORT_UNREACH, + 0); + } else #endif - icmp_send(skb, - ICMP_DEST_UNREACH, - ICMP_PORT_UNREACH, 0); - return NF_DROP; - } + icmp_send(skb, + ICMP_DEST_UNREACH, + ICMP_PORT_UNREACH, 0); + return NF_DROP; } } - IP_VS_DBG_PKT(12, pp, skb, 0, - "packet continues traversal as normal"); - return NF_ACCEPT; } + IP_VS_DBG_PKT(12, af, pp, skb, 0, + "ip_vs_out: packet continues traversal as normal"); + return NF_ACCEPT; +} + +/* + * It is hooked at the NF_INET_FORWARD and NF_INET_LOCAL_IN chain, + * used only for VS/NAT. + * Check if packet is reply for established ip_vs_conn. + */ +static unsigned int +ip_vs_reply4(unsigned int hooknum, struct sk_buff *skb, + const struct net_device *in, const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + return ip_vs_out(hooknum, skb, AF_INET); +} + +/* + * It is hooked at the NF_INET_LOCAL_OUT chain, used only for VS/NAT. + * Check if packet is reply for established ip_vs_conn. + */ +static unsigned int +ip_vs_local_reply4(unsigned int hooknum, struct sk_buff *skb, + const struct net_device *in, const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + unsigned int verdict; + + /* Disable BH in LOCAL_OUT until all places are fixed */ + local_bh_disable(); + verdict = ip_vs_out(hooknum, skb, AF_INET); + local_bh_enable(); + return verdict; +} + +#ifdef CONFIG_IP_VS_IPV6 + +/* + * It is hooked at the NF_INET_FORWARD and NF_INET_LOCAL_IN chain, + * used only for VS/NAT. + * Check if packet is reply for established ip_vs_conn. + */ +static unsigned int +ip_vs_reply6(unsigned int hooknum, struct sk_buff *skb, + const struct net_device *in, const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + return ip_vs_out(hooknum, skb, AF_INET6); +} - return handle_response(af, skb, pp, cp, iph.len); +/* + * It is hooked at the NF_INET_LOCAL_OUT chain, used only for VS/NAT. + * Check if packet is reply for established ip_vs_conn. + */ +static unsigned int +ip_vs_local_reply6(unsigned int hooknum, struct sk_buff *skb, + const struct net_device *in, const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + unsigned int verdict; + + /* Disable BH in LOCAL_OUT until all places are fixed */ + local_bh_disable(); + verdict = ip_vs_out(hooknum, skb, AF_INET6); + local_bh_enable(); + return verdict; } +#endif /* * Handle ICMP messages in the outside-to-inside direction (incoming). @@ -1080,8 +1215,7 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum) /* reassemble IP fragments */ if (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) { - if (ip_vs_gather_frags(skb, hooknum == NF_INET_LOCAL_IN ? - IP_DEFRAG_VS_IN : IP_DEFRAG_VS_FWD)) + if (ip_vs_gather_frags(skb, ip_vs_defrag_user(hooknum))) return NF_STOLEN; } @@ -1124,7 +1258,8 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum) pp->dont_defrag)) return NF_ACCEPT; - IP_VS_DBG_PKT(11, pp, skb, offset, "Checking incoming ICMP for"); + IP_VS_DBG_PKT(11, AF_INET, pp, skb, offset, + "Checking incoming ICMP for"); offset += cih->ihl * 4; @@ -1158,7 +1293,14 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum) if (IPPROTO_TCP == cih->protocol || IPPROTO_UDP == cih->protocol) offset += 2 * sizeof(__u16); verdict = ip_vs_icmp_xmit(skb, cp, pp, offset); - /* do not touch skb anymore */ + /* LOCALNODE from FORWARD hook is not supported */ + if (verdict == NF_ACCEPT && hooknum == NF_INET_FORWARD && + skb_rtable(skb)->rt_flags & RTCF_LOCAL) { + IP_VS_DBG(1, "%s(): " + "local delivery to %pI4 but in FORWARD\n", + __func__, &skb_rtable(skb)->rt_dst); + verdict = NF_DROP; + } out: __ip_vs_conn_put(cp); @@ -1179,14 +1321,13 @@ ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) struct ip_vs_protocol *pp; unsigned int offset, verdict; union nf_inet_addr snet; + struct rt6_info *rt; *related = 1; /* reassemble IP fragments */ if (ipv6_hdr(skb)->nexthdr == IPPROTO_FRAGMENT) { - if (ip_vs_gather_frags_v6(skb, hooknum == NF_INET_LOCAL_IN ? - IP_DEFRAG_VS_IN : - IP_DEFRAG_VS_FWD)) + if (ip_vs_gather_frags_v6(skb, ip_vs_defrag_user(hooknum))) return NF_STOLEN; } @@ -1229,7 +1370,8 @@ ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) if (unlikely(cih->nexthdr == IPPROTO_FRAGMENT && pp->dont_defrag)) return NF_ACCEPT; - IP_VS_DBG_PKT(11, pp, skb, offset, "Checking incoming ICMPv6 for"); + IP_VS_DBG_PKT(11, AF_INET6, pp, skb, offset, + "Checking incoming ICMPv6 for"); offset += sizeof(struct ipv6hdr); @@ -1257,7 +1399,15 @@ ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) IPPROTO_SCTP == cih->nexthdr) offset += 2 * sizeof(__u16); verdict = ip_vs_icmp_xmit_v6(skb, cp, pp, offset); - /* do not touch skb anymore */ + /* LOCALNODE from FORWARD hook is not supported */ + if (verdict == NF_ACCEPT && hooknum == NF_INET_FORWARD && + (rt = (struct rt6_info *) skb_dst(skb)) && + rt->rt6i_dev && rt->rt6i_dev->flags & IFF_LOOPBACK) { + IP_VS_DBG(1, "%s(): " + "local delivery to %pI6 but in FORWARD\n", + __func__, &rt->rt6i_dst); + verdict = NF_DROP; + } __ip_vs_conn_put(cp); @@ -1271,35 +1421,49 @@ ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) * and send it on its way... */ static unsigned int -ip_vs_in(unsigned int hooknum, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) +ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) { struct ip_vs_iphdr iph; struct ip_vs_protocol *pp; struct ip_vs_conn *cp; - int ret, restart, af, pkts; + int ret, restart, pkts; - af = (skb->protocol == htons(ETH_P_IP)) ? AF_INET : AF_INET6; - - ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); + /* Already marked as IPVS request or reply? */ + if (skb->ipvs_property) + return NF_ACCEPT; /* - * Big tappo: only PACKET_HOST, including loopback for local client - * Don't handle local packets on IPv6 for now + * Big tappo: + * - remote client: only PACKET_HOST + * - route: used for struct net when skb->dev is unset */ - if (unlikely(skb->pkt_type != PACKET_HOST)) { - IP_VS_DBG_BUF(12, "packet type=%d proto=%d daddr=%s ignored\n", - skb->pkt_type, - iph.protocol, - IP_VS_DBG_ADDR(af, &iph.daddr)); + if (unlikely((skb->pkt_type != PACKET_HOST && + hooknum != NF_INET_LOCAL_OUT) || + !skb_dst(skb))) { + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); + IP_VS_DBG_BUF(12, "packet type=%d proto=%d daddr=%s" + " ignored in hook %u\n", + skb->pkt_type, iph.protocol, + IP_VS_DBG_ADDR(af, &iph.daddr), hooknum); return NF_ACCEPT; } + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); + + /* Bad... Do not break raw sockets */ + if (unlikely(skb->sk != NULL && hooknum == NF_INET_LOCAL_OUT && + af == AF_INET)) { + struct sock *sk = skb->sk; + struct inet_sock *inet = inet_sk(skb->sk); + + if (inet && sk->sk_family == PF_INET && inet->nodefrag) + return NF_ACCEPT; + } #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) { if (unlikely(iph.protocol == IPPROTO_ICMPV6)) { - int related, verdict = ip_vs_in_icmp_v6(skb, &related, hooknum); + int related; + int verdict = ip_vs_in_icmp_v6(skb, &related, hooknum); if (related) return verdict; @@ -1308,7 +1472,8 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, } else #endif if (unlikely(iph.protocol == IPPROTO_ICMP)) { - int related, verdict = ip_vs_in_icmp(skb, &related, hooknum); + int related; + int verdict = ip_vs_in_icmp(skb, &related, hooknum); if (related) return verdict; @@ -1328,23 +1493,18 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, if (unlikely(!cp)) { int v; - /* For local client packets, it could be a response */ - cp = pp->conn_out_get(af, skb, pp, &iph, iph.len, 0); - if (cp) - return handle_response(af, skb, pp, cp, iph.len); - if (!pp->conn_schedule(af, skb, pp, &v, &cp)) return v; } if (unlikely(!cp)) { /* sorry, all this trouble for a no-hit :) */ - IP_VS_DBG_PKT(12, pp, skb, 0, - "packet continues traversal as normal"); + IP_VS_DBG_PKT(12, af, pp, skb, 0, + "ip_vs_in: packet continues traversal as normal"); return NF_ACCEPT; } - IP_VS_DBG_PKT(11, pp, skb, 0, "Incoming packet"); + IP_VS_DBG_PKT(11, af, pp, skb, 0, "Incoming packet"); /* Check the server status */ if (cp->dest && !(cp->dest->flags & IP_VS_DEST_F_AVAILABLE)) { @@ -1380,8 +1540,7 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, if (af == AF_INET && (ip_vs_sync_state & IP_VS_STATE_MASTER) && cp->protocol == IPPROTO_SCTP) { if ((cp->state == IP_VS_SCTP_S_ESTABLISHED && - (atomic_read(&cp->in_pkts) % - sysctl_ip_vs_sync_threshold[1] + (pkts % sysctl_ip_vs_sync_threshold[1] == sysctl_ip_vs_sync_threshold[0])) || (cp->old_state != cp->state && ((cp->state == IP_VS_SCTP_S_CLOSED) || @@ -1392,7 +1551,8 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, } } - if (af == AF_INET && + /* Keep this block last: TCP and others with pp->num_states <= 1 */ + else if (af == AF_INET && (ip_vs_sync_state & IP_VS_STATE_MASTER) && (((cp->protocol != IPPROTO_TCP || cp->state == IP_VS_TCP_S_ESTABLISHED) && @@ -1411,6 +1571,72 @@ out: return ret; } +/* + * AF_INET handler in NF_INET_LOCAL_IN chain + * Schedule and forward packets from remote clients + */ +static unsigned int +ip_vs_remote_request4(unsigned int hooknum, struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + return ip_vs_in(hooknum, skb, AF_INET); +} + +/* + * AF_INET handler in NF_INET_LOCAL_OUT chain + * Schedule and forward packets from local clients + */ +static unsigned int +ip_vs_local_request4(unsigned int hooknum, struct sk_buff *skb, + const struct net_device *in, const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + unsigned int verdict; + + /* Disable BH in LOCAL_OUT until all places are fixed */ + local_bh_disable(); + verdict = ip_vs_in(hooknum, skb, AF_INET); + local_bh_enable(); + return verdict; +} + +#ifdef CONFIG_IP_VS_IPV6 + +/* + * AF_INET6 handler in NF_INET_LOCAL_IN chain + * Schedule and forward packets from remote clients + */ +static unsigned int +ip_vs_remote_request6(unsigned int hooknum, struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + return ip_vs_in(hooknum, skb, AF_INET6); +} + +/* + * AF_INET6 handler in NF_INET_LOCAL_OUT chain + * Schedule and forward packets from local clients + */ +static unsigned int +ip_vs_local_request6(unsigned int hooknum, struct sk_buff *skb, + const struct net_device *in, const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + unsigned int verdict; + + /* Disable BH in LOCAL_OUT until all places are fixed */ + local_bh_disable(); + verdict = ip_vs_in(hooknum, skb, AF_INET6); + local_bh_enable(); + return verdict; +} + +#endif + /* * It is hooked at the NF_INET_FORWARD chain, in order to catch ICMP @@ -1451,23 +1677,39 @@ ip_vs_forward_icmp_v6(unsigned int hooknum, struct sk_buff *skb, static struct nf_hook_ops ip_vs_ops[] __read_mostly = { + /* After packet filtering, change source only for VS/NAT */ + { + .hook = ip_vs_reply4, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_INET_LOCAL_IN, + .priority = 99, + }, /* After packet filtering, forward packet through VS/DR, VS/TUN, * or VS/NAT(change destination), so that filtering rules can be * applied to IPVS. */ { - .hook = ip_vs_in, + .hook = ip_vs_remote_request4, .owner = THIS_MODULE, .pf = PF_INET, - .hooknum = NF_INET_LOCAL_IN, - .priority = 100, + .hooknum = NF_INET_LOCAL_IN, + .priority = 101, }, - /* After packet filtering, change source only for VS/NAT */ + /* Before ip_vs_in, change source only for VS/NAT */ { - .hook = ip_vs_out, + .hook = ip_vs_local_reply4, .owner = THIS_MODULE, .pf = PF_INET, - .hooknum = NF_INET_FORWARD, - .priority = 100, + .hooknum = NF_INET_LOCAL_OUT, + .priority = -99, + }, + /* After mangle, schedule and forward local requests */ + { + .hook = ip_vs_local_request4, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_INET_LOCAL_OUT, + .priority = -98, }, /* After packet filtering (but before ip_vs_out_icmp), catch icmp * destined for 0.0.0.0/0, which is for incoming IPVS connections */ @@ -1475,27 +1717,51 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = { .hook = ip_vs_forward_icmp, .owner = THIS_MODULE, .pf = PF_INET, - .hooknum = NF_INET_FORWARD, - .priority = 99, + .hooknum = NF_INET_FORWARD, + .priority = 99, + }, + /* After packet filtering, change source only for VS/NAT */ + { + .hook = ip_vs_reply4, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_INET_FORWARD, + .priority = 100, }, #ifdef CONFIG_IP_VS_IPV6 + /* After packet filtering, change source only for VS/NAT */ + { + .hook = ip_vs_reply6, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_INET_LOCAL_IN, + .priority = 99, + }, /* After packet filtering, forward packet through VS/DR, VS/TUN, * or VS/NAT(change destination), so that filtering rules can be * applied to IPVS. */ { - .hook = ip_vs_in, + .hook = ip_vs_remote_request6, .owner = THIS_MODULE, .pf = PF_INET6, - .hooknum = NF_INET_LOCAL_IN, - .priority = 100, + .hooknum = NF_INET_LOCAL_IN, + .priority = 101, }, - /* After packet filtering, change source only for VS/NAT */ + /* Before ip_vs_in, change source only for VS/NAT */ + { + .hook = ip_vs_local_reply6, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_INET_LOCAL_OUT, + .priority = -99, + }, + /* After mangle, schedule and forward local requests */ { - .hook = ip_vs_out, + .hook = ip_vs_local_request6, .owner = THIS_MODULE, .pf = PF_INET6, - .hooknum = NF_INET_FORWARD, - .priority = 100, + .hooknum = NF_INET_LOCAL_OUT, + .priority = -98, }, /* After packet filtering (but before ip_vs_out_icmp), catch icmp * destined for 0.0.0.0/0, which is for incoming IPVS connections */ @@ -1503,8 +1769,16 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = { .hook = ip_vs_forward_icmp_v6, .owner = THIS_MODULE, .pf = PF_INET6, - .hooknum = NF_INET_FORWARD, - .priority = 99, + .hooknum = NF_INET_FORWARD, + .priority = 99, + }, + /* After packet filtering, change source only for VS/NAT */ + { + .hook = ip_vs_reply6, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_INET_FORWARD, + .priority = 100, }, #endif }; diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 0f0c079c422..5f5daa30b0a 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -61,7 +61,7 @@ static DEFINE_RWLOCK(__ip_vs_svc_lock); static DEFINE_RWLOCK(__ip_vs_rs_lock); /* lock for state and timeout tables */ -static DEFINE_RWLOCK(__ip_vs_securetcp_lock); +static DEFINE_SPINLOCK(ip_vs_securetcp_lock); /* lock for drop entry handling */ static DEFINE_SPINLOCK(__ip_vs_dropentry_lock); @@ -88,6 +88,10 @@ int sysctl_ip_vs_expire_nodest_conn = 0; int sysctl_ip_vs_expire_quiescent_template = 0; int sysctl_ip_vs_sync_threshold[2] = { 3, 50 }; int sysctl_ip_vs_nat_icmp_send = 0; +#ifdef CONFIG_IP_VS_NFCT +int sysctl_ip_vs_conntrack; +#endif +int sysctl_ip_vs_snat_reroute = 1; #ifdef CONFIG_IP_VS_DEBUG @@ -204,7 +208,7 @@ static void update_defense_level(void) spin_unlock(&__ip_vs_droppacket_lock); /* secure_tcp */ - write_lock(&__ip_vs_securetcp_lock); + spin_lock(&ip_vs_securetcp_lock); switch (sysctl_ip_vs_secure_tcp) { case 0: if (old_secure_tcp >= 2) @@ -238,7 +242,7 @@ static void update_defense_level(void) old_secure_tcp = sysctl_ip_vs_secure_tcp; if (to_change >= 0) ip_vs_protocol_timeout_change(sysctl_ip_vs_secure_tcp>1); - write_unlock(&__ip_vs_securetcp_lock); + spin_unlock(&ip_vs_securetcp_lock); local_bh_enable(); } @@ -401,7 +405,7 @@ static int ip_vs_svc_unhash(struct ip_vs_service *svc) * Get service by {proto,addr,port} in the service table. */ static inline struct ip_vs_service * -__ip_vs_service_get(int af, __u16 protocol, const union nf_inet_addr *vaddr, +__ip_vs_service_find(int af, __u16 protocol, const union nf_inet_addr *vaddr, __be16 vport) { unsigned hash; @@ -416,7 +420,6 @@ __ip_vs_service_get(int af, __u16 protocol, const union nf_inet_addr *vaddr, && (svc->port == vport) && (svc->protocol == protocol)) { /* HIT */ - atomic_inc(&svc->usecnt); return svc; } } @@ -429,7 +432,7 @@ __ip_vs_service_get(int af, __u16 protocol, const union nf_inet_addr *vaddr, * Get service by {fwmark} in the service table. */ static inline struct ip_vs_service * -__ip_vs_svc_fwm_get(int af, __u32 fwmark) +__ip_vs_svc_fwm_find(int af, __u32 fwmark) { unsigned hash; struct ip_vs_service *svc; @@ -440,7 +443,6 @@ __ip_vs_svc_fwm_get(int af, __u32 fwmark) list_for_each_entry(svc, &ip_vs_svc_fwm_table[hash], f_list) { if (svc->fwmark == fwmark && svc->af == af) { /* HIT */ - atomic_inc(&svc->usecnt); return svc; } } @@ -459,14 +461,14 @@ ip_vs_service_get(int af, __u32 fwmark, __u16 protocol, /* * Check the table hashed by fwmark first */ - if (fwmark && (svc = __ip_vs_svc_fwm_get(af, fwmark))) + if (fwmark && (svc = __ip_vs_svc_fwm_find(af, fwmark))) goto out; /* * Check the table hashed by <protocol,addr,port> * for "full" addressed entries */ - svc = __ip_vs_service_get(af, protocol, vaddr, vport); + svc = __ip_vs_service_find(af, protocol, vaddr, vport); if (svc == NULL && protocol == IPPROTO_TCP @@ -476,7 +478,7 @@ ip_vs_service_get(int af, __u32 fwmark, __u16 protocol, * Check if ftp service entry exists, the packet * might belong to FTP data connections. */ - svc = __ip_vs_service_get(af, protocol, vaddr, FTPPORT); + svc = __ip_vs_service_find(af, protocol, vaddr, FTPPORT); } if (svc == NULL @@ -484,10 +486,12 @@ ip_vs_service_get(int af, __u32 fwmark, __u16 protocol, /* * Check if the catch-all port (port zero) exists */ - svc = __ip_vs_service_get(af, protocol, vaddr, 0); + svc = __ip_vs_service_find(af, protocol, vaddr, 0); } out: + if (svc) + atomic_inc(&svc->usecnt); read_unlock(&__ip_vs_svc_lock); IP_VS_DBG_BUF(9, "lookup service: fwm %u %s %s:%u %s\n", @@ -506,14 +510,19 @@ __ip_vs_bind_svc(struct ip_vs_dest *dest, struct ip_vs_service *svc) dest->svc = svc; } -static inline void +static void __ip_vs_unbind_svc(struct ip_vs_dest *dest) { struct ip_vs_service *svc = dest->svc; dest->svc = NULL; - if (atomic_dec_and_test(&svc->refcnt)) + if (atomic_dec_and_test(&svc->refcnt)) { + IP_VS_DBG_BUF(3, "Removing service %u/%s:%u usecnt=%d\n", + svc->fwmark, + IP_VS_DBG_ADDR(svc->af, &svc->addr), + ntohs(svc->port), atomic_read(&svc->usecnt)); kfree(svc); + } } @@ -758,31 +767,18 @@ ip_vs_zero_stats(struct ip_vs_stats *stats) * Update a destination in the given service */ static void -__ip_vs_update_dest(struct ip_vs_service *svc, - struct ip_vs_dest *dest, struct ip_vs_dest_user_kern *udest) +__ip_vs_update_dest(struct ip_vs_service *svc, struct ip_vs_dest *dest, + struct ip_vs_dest_user_kern *udest, int add) { int conn_flags; /* set the weight and the flags */ atomic_set(&dest->weight, udest->weight); - conn_flags = udest->conn_flags | IP_VS_CONN_F_INACTIVE; - - /* check if local node and update the flags */ -#ifdef CONFIG_IP_VS_IPV6 - if (svc->af == AF_INET6) { - if (__ip_vs_addr_is_local_v6(&udest->addr.in6)) { - conn_flags = (conn_flags & ~IP_VS_CONN_F_FWD_MASK) - | IP_VS_CONN_F_LOCALNODE; - } - } else -#endif - if (inet_addr_type(&init_net, udest->addr.ip) == RTN_LOCAL) { - conn_flags = (conn_flags & ~IP_VS_CONN_F_FWD_MASK) - | IP_VS_CONN_F_LOCALNODE; - } + conn_flags = udest->conn_flags & IP_VS_CONN_F_DEST_MASK; + conn_flags |= IP_VS_CONN_F_INACTIVE; /* set the IP_VS_CONN_F_NOOUTPUT flag if not masquerading/NAT */ - if ((conn_flags & IP_VS_CONN_F_FWD_MASK) != 0) { + if ((conn_flags & IP_VS_CONN_F_FWD_MASK) != IP_VS_CONN_F_MASQ) { conn_flags |= IP_VS_CONN_F_NOOUTPUT; } else { /* @@ -813,6 +809,29 @@ __ip_vs_update_dest(struct ip_vs_service *svc, dest->flags &= ~IP_VS_DEST_F_OVERLOAD; dest->u_threshold = udest->u_threshold; dest->l_threshold = udest->l_threshold; + + spin_lock(&dest->dst_lock); + ip_vs_dst_reset(dest); + spin_unlock(&dest->dst_lock); + + if (add) + ip_vs_new_estimator(&dest->stats); + + write_lock_bh(&__ip_vs_svc_lock); + + /* Wait until all other svc users go away */ + IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 0); + + if (add) { + list_add(&dest->n_list, &svc->destinations); + svc->num_dests++; + } + + /* call the update_service, because server weight may be changed */ + if (svc->scheduler->update_service) + svc->scheduler->update_service(svc); + + write_unlock_bh(&__ip_vs_svc_lock); } @@ -843,7 +862,7 @@ ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest, return -EINVAL; } - dest = kzalloc(sizeof(struct ip_vs_dest), GFP_ATOMIC); + dest = kzalloc(sizeof(struct ip_vs_dest), GFP_KERNEL); if (dest == NULL) { pr_err("%s(): no memory.\n", __func__); return -ENOMEM; @@ -860,13 +879,12 @@ ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest, atomic_set(&dest->activeconns, 0); atomic_set(&dest->inactconns, 0); atomic_set(&dest->persistconns, 0); - atomic_set(&dest->refcnt, 0); + atomic_set(&dest->refcnt, 1); INIT_LIST_HEAD(&dest->d_list); spin_lock_init(&dest->dst_lock); spin_lock_init(&dest->stats.lock); - __ip_vs_update_dest(svc, dest, udest); - ip_vs_new_estimator(&dest->stats); + __ip_vs_update_dest(svc, dest, udest, 1); *dest_p = dest; @@ -926,65 +944,22 @@ ip_vs_add_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) IP_VS_DBG_ADDR(svc->af, &dest->vaddr), ntohs(dest->vport)); - __ip_vs_update_dest(svc, dest, udest); - /* * Get the destination from the trash */ list_del(&dest->n_list); - ip_vs_new_estimator(&dest->stats); - - write_lock_bh(&__ip_vs_svc_lock); - + __ip_vs_update_dest(svc, dest, udest, 1); + ret = 0; + } else { /* - * Wait until all other svc users go away. + * Allocate and initialize the dest structure */ - IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1); - - list_add(&dest->n_list, &svc->destinations); - svc->num_dests++; - - /* call the update_service function of its scheduler */ - if (svc->scheduler->update_service) - svc->scheduler->update_service(svc); - - write_unlock_bh(&__ip_vs_svc_lock); - return 0; - } - - /* - * Allocate and initialize the dest structure - */ - ret = ip_vs_new_dest(svc, udest, &dest); - if (ret) { - return ret; + ret = ip_vs_new_dest(svc, udest, &dest); } - - /* - * Add the dest entry into the list - */ - atomic_inc(&dest->refcnt); - - write_lock_bh(&__ip_vs_svc_lock); - - /* - * Wait until all other svc users go away. - */ - IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1); - - list_add(&dest->n_list, &svc->destinations); - svc->num_dests++; - - /* call the update_service function of its scheduler */ - if (svc->scheduler->update_service) - svc->scheduler->update_service(svc); - - write_unlock_bh(&__ip_vs_svc_lock); - LeaveFunction(2); - return 0; + return ret; } @@ -1023,19 +998,7 @@ ip_vs_edit_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) return -ENOENT; } - __ip_vs_update_dest(svc, dest, udest); - - write_lock_bh(&__ip_vs_svc_lock); - - /* Wait until all other svc users go away */ - IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1); - - /* call the update_service, because server weight may be changed */ - if (svc->scheduler->update_service) - svc->scheduler->update_service(svc); - - write_unlock_bh(&__ip_vs_svc_lock); - + __ip_vs_update_dest(svc, dest, udest, 0); LeaveFunction(2); return 0; @@ -1062,6 +1025,10 @@ static void __ip_vs_del_dest(struct ip_vs_dest *dest) * the destination into the trash. */ if (atomic_dec_and_test(&dest->refcnt)) { + IP_VS_DBG_BUF(3, "Removing destination %u/%s:%u\n", + dest->vfwmark, + IP_VS_DBG_ADDR(dest->af, &dest->addr), + ntohs(dest->port)); ip_vs_dst_reset(dest); /* simply decrease svc->refcnt here, let the caller check and release the service if nobody refers to it. @@ -1128,7 +1095,7 @@ ip_vs_del_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) /* * Wait until all other svc users go away. */ - IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1); + IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 0); /* * Unlink dest from the service @@ -1157,6 +1124,7 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u, { int ret = 0; struct ip_vs_scheduler *sched = NULL; + struct ip_vs_pe *pe = NULL; struct ip_vs_service *svc = NULL; /* increase the module use count */ @@ -1167,7 +1135,17 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u, if (sched == NULL) { pr_info("Scheduler module ip_vs_%s not found\n", u->sched_name); ret = -ENOENT; - goto out_mod_dec; + goto out_err; + } + + if (u->pe_name && *u->pe_name) { + pe = ip_vs_pe_get(u->pe_name); + if (pe == NULL) { + pr_info("persistence engine module ip_vs_pe_%s " + "not found\n", u->pe_name); + ret = -ENOENT; + goto out_err; + } } #ifdef CONFIG_IP_VS_IPV6 @@ -1177,7 +1155,7 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u, } #endif - svc = kzalloc(sizeof(struct ip_vs_service), GFP_ATOMIC); + svc = kzalloc(sizeof(struct ip_vs_service), GFP_KERNEL); if (svc == NULL) { IP_VS_DBG(1, "%s(): no memory\n", __func__); ret = -ENOMEM; @@ -1185,7 +1163,7 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u, } /* I'm the first user of the service */ - atomic_set(&svc->usecnt, 1); + atomic_set(&svc->usecnt, 0); atomic_set(&svc->refcnt, 0); svc->af = u->af; @@ -1207,6 +1185,10 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u, goto out_err; sched = NULL; + /* Bind the ct retriever */ + ip_vs_bind_pe(svc, pe); + pe = NULL; + /* Update the virtual service counters */ if (svc->port == FTPPORT) atomic_inc(&ip_vs_ftpsvc_counter); @@ -1227,10 +1209,9 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u, *svc_p = svc; return 0; - out_err: + out_err: if (svc != NULL) { - if (svc->scheduler) - ip_vs_unbind_scheduler(svc); + ip_vs_unbind_scheduler(svc); if (svc->inc) { local_bh_disable(); ip_vs_app_inc_put(svc->inc); @@ -1239,8 +1220,8 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u, kfree(svc); } ip_vs_scheduler_put(sched); + ip_vs_pe_put(pe); - out_mod_dec: /* decrease the module use count */ ip_vs_use_count_dec(); @@ -1255,6 +1236,7 @@ static int ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u) { struct ip_vs_scheduler *sched, *old_sched; + struct ip_vs_pe *pe = NULL, *old_pe = NULL; int ret = 0; /* @@ -1267,6 +1249,17 @@ ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u) } old_sched = sched; + if (u->pe_name && *u->pe_name) { + pe = ip_vs_pe_get(u->pe_name); + if (pe == NULL) { + pr_info("persistence engine module ip_vs_pe_%s " + "not found\n", u->pe_name); + ret = -ENOENT; + goto out; + } + old_pe = pe; + } + #ifdef CONFIG_IP_VS_IPV6 if (u->af == AF_INET6 && (u->netmask < 1 || u->netmask > 128)) { ret = -EINVAL; @@ -1279,7 +1272,7 @@ ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u) /* * Wait until all other svc users go away. */ - IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1); + IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 0); /* * Set the flags and timeout value @@ -1318,15 +1311,17 @@ ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u) } } + old_pe = svc->pe; + if (pe != old_pe) { + ip_vs_unbind_pe(svc); + ip_vs_bind_pe(svc, pe); + } + out_unlock: write_unlock_bh(&__ip_vs_svc_lock); -#ifdef CONFIG_IP_VS_IPV6 out: -#endif - - if (old_sched) - ip_vs_scheduler_put(old_sched); - + ip_vs_scheduler_put(old_sched); + ip_vs_pe_put(old_pe); return ret; } @@ -1340,6 +1335,9 @@ static void __ip_vs_del_service(struct ip_vs_service *svc) { struct ip_vs_dest *dest, *nxt; struct ip_vs_scheduler *old_sched; + struct ip_vs_pe *old_pe; + + pr_info("%s: enter\n", __func__); /* Count only IPv4 services for old get/setsockopt interface */ if (svc->af == AF_INET) @@ -1350,8 +1348,12 @@ static void __ip_vs_del_service(struct ip_vs_service *svc) /* Unbind scheduler */ old_sched = svc->scheduler; ip_vs_unbind_scheduler(svc); - if (old_sched) - ip_vs_scheduler_put(old_sched); + ip_vs_scheduler_put(old_sched); + + /* Unbind persistence engine */ + old_pe = svc->pe; + ip_vs_unbind_pe(svc); + ip_vs_pe_put(old_pe); /* Unbind app inc */ if (svc->inc) { @@ -1378,21 +1380,23 @@ static void __ip_vs_del_service(struct ip_vs_service *svc) /* * Free the service if nobody refers to it */ - if (atomic_read(&svc->refcnt) == 0) + if (atomic_read(&svc->refcnt) == 0) { + IP_VS_DBG_BUF(3, "Removing service %u/%s:%u usecnt=%d\n", + svc->fwmark, + IP_VS_DBG_ADDR(svc->af, &svc->addr), + ntohs(svc->port), atomic_read(&svc->usecnt)); kfree(svc); + } /* decrease the module use count */ ip_vs_use_count_dec(); } /* - * Delete a service from the service list + * Unlink a service from list and try to delete it if its refcnt reached 0 */ -static int ip_vs_del_service(struct ip_vs_service *svc) +static void ip_vs_unlink_service(struct ip_vs_service *svc) { - if (svc == NULL) - return -EEXIST; - /* * Unhash it from the service table */ @@ -1403,11 +1407,21 @@ static int ip_vs_del_service(struct ip_vs_service *svc) /* * Wait until all the svc users go away. */ - IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1); + IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 0); __ip_vs_del_service(svc); write_unlock_bh(&__ip_vs_svc_lock); +} + +/* + * Delete a service from the service list + */ +static int ip_vs_del_service(struct ip_vs_service *svc) +{ + if (svc == NULL) + return -EEXIST; + ip_vs_unlink_service(svc); return 0; } @@ -1426,14 +1440,7 @@ static int ip_vs_flush(void) */ for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { list_for_each_entry_safe(svc, nxt, &ip_vs_svc_table[idx], s_list) { - write_lock_bh(&__ip_vs_svc_lock); - ip_vs_svc_unhash(svc); - /* - * Wait until all the svc users go away. - */ - IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 0); - __ip_vs_del_service(svc); - write_unlock_bh(&__ip_vs_svc_lock); + ip_vs_unlink_service(svc); } } @@ -1443,14 +1450,7 @@ static int ip_vs_flush(void) for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { list_for_each_entry_safe(svc, nxt, &ip_vs_svc_fwm_table[idx], f_list) { - write_lock_bh(&__ip_vs_svc_lock); - ip_vs_svc_unhash(svc); - /* - * Wait until all the svc users go away. - */ - IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 0); - __ip_vs_del_service(svc); - write_unlock_bh(&__ip_vs_svc_lock); + ip_vs_unlink_service(svc); } } @@ -1579,6 +1579,15 @@ static struct ctl_table vs_vars[] = { .mode = 0644, .proc_handler = proc_do_defense_mode, }, +#ifdef CONFIG_IP_VS_NFCT + { + .procname = "conntrack", + .data = &sysctl_ip_vs_conntrack, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, +#endif { .procname = "secure_tcp", .data = &sysctl_ip_vs_secure_tcp, @@ -1586,6 +1595,13 @@ static struct ctl_table vs_vars[] = { .mode = 0644, .proc_handler = proc_do_defense_mode, }, + { + .procname = "snat_reroute", + .data = &sysctl_ip_vs_snat_reroute, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, #if 0 { .procname = "timeout_established", @@ -2041,6 +2057,8 @@ static const unsigned char set_arglen[SET_CMDID(IP_VS_SO_SET_MAX)+1] = { static void ip_vs_copy_usvc_compat(struct ip_vs_service_user_kern *usvc, struct ip_vs_service_user *usvc_compat) { + memset(usvc, 0, sizeof(*usvc)); + usvc->af = AF_INET; usvc->protocol = usvc_compat->protocol; usvc->addr.ip = usvc_compat->addr; @@ -2058,6 +2076,8 @@ static void ip_vs_copy_usvc_compat(struct ip_vs_service_user_kern *usvc, static void ip_vs_copy_udest_compat(struct ip_vs_dest_user_kern *udest, struct ip_vs_dest_user *udest_compat) { + memset(udest, 0, sizeof(*udest)); + udest->addr.ip = udest_compat->addr; udest->port = udest_compat->port; udest->conn_flags = udest_compat->conn_flags; @@ -2147,10 +2167,10 @@ do_ip_vs_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) /* Lookup the exact service by <protocol, addr, port> or fwmark */ if (usvc.fwmark == 0) - svc = __ip_vs_service_get(usvc.af, usvc.protocol, - &usvc.addr, usvc.port); + svc = __ip_vs_service_find(usvc.af, usvc.protocol, + &usvc.addr, usvc.port); else - svc = __ip_vs_svc_fwm_get(usvc.af, usvc.fwmark); + svc = __ip_vs_svc_fwm_find(usvc.af, usvc.fwmark); if (cmd != IP_VS_SO_SET_ADD && (svc == NULL || svc->protocol != usvc.protocol)) { @@ -2189,9 +2209,6 @@ do_ip_vs_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) ret = -EINVAL; } - if (svc) - ip_vs_service_put(svc); - out_unlock: mutex_unlock(&__ip_vs_mutex); out_dec: @@ -2284,10 +2301,10 @@ __ip_vs_get_dest_entries(const struct ip_vs_get_dests *get, int ret = 0; if (get->fwmark) - svc = __ip_vs_svc_fwm_get(AF_INET, get->fwmark); + svc = __ip_vs_svc_fwm_find(AF_INET, get->fwmark); else - svc = __ip_vs_service_get(AF_INET, get->protocol, &addr, - get->port); + svc = __ip_vs_service_find(AF_INET, get->protocol, &addr, + get->port); if (svc) { int count = 0; @@ -2315,7 +2332,6 @@ __ip_vs_get_dest_entries(const struct ip_vs_get_dests *get, } count++; } - ip_vs_service_put(svc); } else ret = -ESRCH; return ret; @@ -2436,15 +2452,14 @@ do_ip_vs_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) entry = (struct ip_vs_service_entry *)arg; addr.ip = entry->addr; if (entry->fwmark) - svc = __ip_vs_svc_fwm_get(AF_INET, entry->fwmark); + svc = __ip_vs_svc_fwm_find(AF_INET, entry->fwmark); else - svc = __ip_vs_service_get(AF_INET, entry->protocol, - &addr, entry->port); + svc = __ip_vs_service_find(AF_INET, entry->protocol, + &addr, entry->port); if (svc) { ip_vs_copy_service(entry, svc); if (copy_to_user(user, entry, sizeof(*entry)) != 0) ret = -EFAULT; - ip_vs_service_put(svc); } else ret = -ESRCH; } @@ -2559,6 +2574,8 @@ static const struct nla_policy ip_vs_svc_policy[IPVS_SVC_ATTR_MAX + 1] = { [IPVS_SVC_ATTR_FWMARK] = { .type = NLA_U32 }, [IPVS_SVC_ATTR_SCHED_NAME] = { .type = NLA_NUL_STRING, .len = IP_VS_SCHEDNAME_MAXLEN }, + [IPVS_SVC_ATTR_PE_NAME] = { .type = NLA_NUL_STRING, + .len = IP_VS_PENAME_MAXLEN }, [IPVS_SVC_ATTR_FLAGS] = { .type = NLA_BINARY, .len = sizeof(struct ip_vs_flags) }, [IPVS_SVC_ATTR_TIMEOUT] = { .type = NLA_U32 }, @@ -2635,6 +2652,8 @@ static int ip_vs_genl_fill_service(struct sk_buff *skb, } NLA_PUT_STRING(skb, IPVS_SVC_ATTR_SCHED_NAME, svc->scheduler->name); + if (svc->pe) + NLA_PUT_STRING(skb, IPVS_SVC_ATTR_PE_NAME, svc->pe->name); NLA_PUT(skb, IPVS_SVC_ATTR_FLAGS, sizeof(flags), &flags); NLA_PUT_U32(skb, IPVS_SVC_ATTR_TIMEOUT, svc->timeout / HZ); NLA_PUT_U32(skb, IPVS_SVC_ATTR_NETMASK, svc->netmask); @@ -2711,10 +2730,12 @@ nla_put_failure: } static int ip_vs_genl_parse_service(struct ip_vs_service_user_kern *usvc, - struct nlattr *nla, int full_entry) + struct nlattr *nla, int full_entry, + struct ip_vs_service **ret_svc) { struct nlattr *attrs[IPVS_SVC_ATTR_MAX + 1]; struct nlattr *nla_af, *nla_port, *nla_fwmark, *nla_protocol, *nla_addr; + struct ip_vs_service *svc; /* Parse mandatory identifying service fields first */ if (nla == NULL || @@ -2750,14 +2771,21 @@ static int ip_vs_genl_parse_service(struct ip_vs_service_user_kern *usvc, usvc->fwmark = 0; } + if (usvc->fwmark) + svc = __ip_vs_svc_fwm_find(usvc->af, usvc->fwmark); + else + svc = __ip_vs_service_find(usvc->af, usvc->protocol, + &usvc->addr, usvc->port); + *ret_svc = svc; + /* If a full entry was requested, check for the additional fields */ if (full_entry) { - struct nlattr *nla_sched, *nla_flags, *nla_timeout, + struct nlattr *nla_sched, *nla_flags, *nla_pe, *nla_timeout, *nla_netmask; struct ip_vs_flags flags; - struct ip_vs_service *svc; nla_sched = attrs[IPVS_SVC_ATTR_SCHED_NAME]; + nla_pe = attrs[IPVS_SVC_ATTR_PE_NAME]; nla_flags = attrs[IPVS_SVC_ATTR_FLAGS]; nla_timeout = attrs[IPVS_SVC_ATTR_TIMEOUT]; nla_netmask = attrs[IPVS_SVC_ATTR_NETMASK]; @@ -2768,21 +2796,14 @@ static int ip_vs_genl_parse_service(struct ip_vs_service_user_kern *usvc, nla_memcpy(&flags, nla_flags, sizeof(flags)); /* prefill flags from service if it already exists */ - if (usvc->fwmark) - svc = __ip_vs_svc_fwm_get(usvc->af, usvc->fwmark); - else - svc = __ip_vs_service_get(usvc->af, usvc->protocol, - &usvc->addr, usvc->port); - if (svc) { + if (svc) usvc->flags = svc->flags; - ip_vs_service_put(svc); - } else - usvc->flags = 0; /* set new flags from userland */ usvc->flags = (usvc->flags & ~flags.mask) | (flags.flags & flags.mask); usvc->sched_name = nla_data(nla_sched); + usvc->pe_name = nla_pe ? nla_data(nla_pe) : NULL; usvc->timeout = nla_get_u32(nla_timeout); usvc->netmask = nla_get_u32(nla_netmask); } @@ -2793,17 +2814,11 @@ static int ip_vs_genl_parse_service(struct ip_vs_service_user_kern *usvc, static struct ip_vs_service *ip_vs_genl_find_service(struct nlattr *nla) { struct ip_vs_service_user_kern usvc; + struct ip_vs_service *svc; int ret; - ret = ip_vs_genl_parse_service(&usvc, nla, 0); - if (ret) - return ERR_PTR(ret); - - if (usvc.fwmark) - return __ip_vs_svc_fwm_get(usvc.af, usvc.fwmark); - else - return __ip_vs_service_get(usvc.af, usvc.protocol, - &usvc.addr, usvc.port); + ret = ip_vs_genl_parse_service(&usvc, nla, 0, &svc); + return ret ? ERR_PTR(ret) : svc; } static int ip_vs_genl_fill_dest(struct sk_buff *skb, struct ip_vs_dest *dest) @@ -2894,7 +2909,6 @@ static int ip_vs_genl_dump_dests(struct sk_buff *skb, nla_put_failure: cb->args[0] = idx; - ip_vs_service_put(svc); out_err: mutex_unlock(&__ip_vs_mutex); @@ -3107,17 +3121,10 @@ static int ip_vs_genl_set_cmd(struct sk_buff *skb, struct genl_info *info) ret = ip_vs_genl_parse_service(&usvc, info->attrs[IPVS_CMD_ATTR_SERVICE], - need_full_svc); + need_full_svc, &svc); if (ret) goto out; - /* Lookup the exact service by <protocol, addr, port> or fwmark */ - if (usvc.fwmark == 0) - svc = __ip_vs_service_get(usvc.af, usvc.protocol, - &usvc.addr, usvc.port); - else - svc = __ip_vs_svc_fwm_get(usvc.af, usvc.fwmark); - /* Unless we're adding a new service, the service must already exist */ if ((cmd != IPVS_CMD_NEW_SERVICE) && (svc == NULL)) { ret = -ESRCH; @@ -3151,6 +3158,7 @@ static int ip_vs_genl_set_cmd(struct sk_buff *skb, struct genl_info *info) break; case IPVS_CMD_DEL_SERVICE: ret = ip_vs_del_service(svc); + /* do not use svc, it can be freed */ break; case IPVS_CMD_NEW_DEST: ret = ip_vs_add_dest(svc, &udest); @@ -3169,8 +3177,6 @@ static int ip_vs_genl_set_cmd(struct sk_buff *skb, struct genl_info *info) } out: - if (svc) - ip_vs_service_put(svc); mutex_unlock(&__ip_vs_mutex); return ret; @@ -3216,7 +3222,6 @@ static int ip_vs_genl_get_cmd(struct sk_buff *skb, struct genl_info *info) goto out_err; } else if (svc) { ret = ip_vs_genl_fill_service(msg, svc); - ip_vs_service_put(svc); if (ret) goto nla_put_failure; } else { @@ -3385,6 +3390,16 @@ int __init ip_vs_control_init(void) EnterFunction(2); + /* Initialize ip_vs_svc_table, ip_vs_svc_fwm_table, ip_vs_rtable */ + for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { + INIT_LIST_HEAD(&ip_vs_svc_table[idx]); + INIT_LIST_HEAD(&ip_vs_svc_fwm_table[idx]); + } + for(idx = 0; idx < IP_VS_RTAB_SIZE; idx++) { + INIT_LIST_HEAD(&ip_vs_rtable[idx]); + } + smp_wmb(); + ret = nf_register_sockopt(&ip_vs_sockopts); if (ret) { pr_err("cannot register sockopt.\n"); @@ -3403,15 +3418,6 @@ int __init ip_vs_control_init(void) sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars); - /* Initialize ip_vs_svc_table, ip_vs_svc_fwm_table, ip_vs_rtable */ - for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { - INIT_LIST_HEAD(&ip_vs_svc_table[idx]); - INIT_LIST_HEAD(&ip_vs_svc_fwm_table[idx]); - } - for(idx = 0; idx < IP_VS_RTAB_SIZE; idx++) { - INIT_LIST_HEAD(&ip_vs_rtable[idx]); - } - ip_vs_new_estimator(&ip_vs_stats); /* Hook the defense timer */ diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c index f228a17ec64..75455000ad1 100644 --- a/net/netfilter/ipvs/ip_vs_ftp.c +++ b/net/netfilter/ipvs/ip_vs_ftp.c @@ -20,17 +20,6 @@ * * Author: Wouter Gadeyne * - * - * Code for ip_vs_expect_related and ip_vs_expect_callback is taken from - * http://www.ssi.bg/~ja/nfct/: - * - * ip_vs_nfct.c: Netfilter connection tracking support for IPVS - * - * Portions Copyright (C) 2001-2002 - * Antefacto Ltd, 181 Parnell St, Dublin 1, Ireland. - * - * Portions Copyright (C) 2003-2008 - * Julian Anastasov */ #define KMSG_COMPONENT "IPVS" @@ -45,6 +34,7 @@ #include <linux/netfilter.h> #include <net/netfilter/nf_conntrack.h> #include <net/netfilter/nf_conntrack_expect.h> +#include <net/netfilter/nf_nat.h> #include <net/netfilter/nf_nat_helper.h> #include <linux/gfp.h> #include <net/protocol.h> @@ -57,16 +47,6 @@ #define SERVER_STRING "227 Entering Passive Mode (" #define CLIENT_STRING "PORT " -#define FMT_TUPLE "%pI4:%u->%pI4:%u/%u" -#define ARG_TUPLE(T) &(T)->src.u3.ip, ntohs((T)->src.u.all), \ - &(T)->dst.u3.ip, ntohs((T)->dst.u.all), \ - (T)->dst.protonum - -#define FMT_CONN "%pI4:%u->%pI4:%u->%pI4:%u/%u:%u" -#define ARG_CONN(C) &((C)->caddr.ip), ntohs((C)->cport), \ - &((C)->vaddr.ip), ntohs((C)->vport), \ - &((C)->daddr.ip), ntohs((C)->dport), \ - (C)->protocol, (C)->state /* * List of ports (up to IP_VS_APP_MAX_PORTS) to be handled by helper @@ -84,6 +64,8 @@ static int ip_vs_ftp_pasv; static int ip_vs_ftp_init_conn(struct ip_vs_app *app, struct ip_vs_conn *cp) { + /* We use connection tracking for the command connection */ + cp->flags |= IP_VS_CONN_F_NFCT; return 0; } @@ -148,120 +130,6 @@ static int ip_vs_ftp_get_addrport(char *data, char *data_limit, } /* - * Called from init_conntrack() as expectfn handler. - */ -static void -ip_vs_expect_callback(struct nf_conn *ct, - struct nf_conntrack_expect *exp) -{ - struct nf_conntrack_tuple *orig, new_reply; - struct ip_vs_conn *cp; - - if (exp->tuple.src.l3num != PF_INET) - return; - - /* - * We assume that no NF locks are held before this callback. - * ip_vs_conn_out_get and ip_vs_conn_in_get should match their - * expectations even if they use wildcard values, now we provide the - * actual values from the newly created original conntrack direction. - * The conntrack is confirmed when packet reaches IPVS hooks. - */ - - /* RS->CLIENT */ - orig = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; - cp = ip_vs_conn_out_get(exp->tuple.src.l3num, orig->dst.protonum, - &orig->src.u3, orig->src.u.tcp.port, - &orig->dst.u3, orig->dst.u.tcp.port); - if (cp) { - /* Change reply CLIENT->RS to CLIENT->VS */ - new_reply = ct->tuplehash[IP_CT_DIR_REPLY].tuple; - IP_VS_DBG(7, "%s(): ct=%p, status=0x%lX, tuples=" FMT_TUPLE ", " - FMT_TUPLE ", found inout cp=" FMT_CONN "\n", - __func__, ct, ct->status, - ARG_TUPLE(orig), ARG_TUPLE(&new_reply), - ARG_CONN(cp)); - new_reply.dst.u3 = cp->vaddr; - new_reply.dst.u.tcp.port = cp->vport; - IP_VS_DBG(7, "%s(): ct=%p, new tuples=" FMT_TUPLE ", " FMT_TUPLE - ", inout cp=" FMT_CONN "\n", - __func__, ct, - ARG_TUPLE(orig), ARG_TUPLE(&new_reply), - ARG_CONN(cp)); - goto alter; - } - - /* CLIENT->VS */ - cp = ip_vs_conn_in_get(exp->tuple.src.l3num, orig->dst.protonum, - &orig->src.u3, orig->src.u.tcp.port, - &orig->dst.u3, orig->dst.u.tcp.port); - if (cp) { - /* Change reply VS->CLIENT to RS->CLIENT */ - new_reply = ct->tuplehash[IP_CT_DIR_REPLY].tuple; - IP_VS_DBG(7, "%s(): ct=%p, status=0x%lX, tuples=" FMT_TUPLE ", " - FMT_TUPLE ", found outin cp=" FMT_CONN "\n", - __func__, ct, ct->status, - ARG_TUPLE(orig), ARG_TUPLE(&new_reply), - ARG_CONN(cp)); - new_reply.src.u3 = cp->daddr; - new_reply.src.u.tcp.port = cp->dport; - IP_VS_DBG(7, "%s(): ct=%p, new tuples=" FMT_TUPLE ", " - FMT_TUPLE ", outin cp=" FMT_CONN "\n", - __func__, ct, - ARG_TUPLE(orig), ARG_TUPLE(&new_reply), - ARG_CONN(cp)); - goto alter; - } - - IP_VS_DBG(7, "%s(): ct=%p, status=0x%lX, tuple=" FMT_TUPLE - " - unknown expect\n", - __func__, ct, ct->status, ARG_TUPLE(orig)); - return; - -alter: - /* Never alter conntrack for non-NAT conns */ - if (IP_VS_FWD_METHOD(cp) == IP_VS_CONN_F_MASQ) - nf_conntrack_alter_reply(ct, &new_reply); - ip_vs_conn_put(cp); - return; -} - -/* - * Create NF conntrack expectation with wildcard (optional) source port. - * Then the default callback function will alter the reply and will confirm - * the conntrack entry when the first packet comes. - */ -static void -ip_vs_expect_related(struct sk_buff *skb, struct nf_conn *ct, - struct ip_vs_conn *cp, u_int8_t proto, - const __be16 *port, int from_rs) -{ - struct nf_conntrack_expect *exp; - - BUG_ON(!ct || ct == &nf_conntrack_untracked); - - exp = nf_ct_expect_alloc(ct); - if (!exp) - return; - - if (from_rs) - nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, - nf_ct_l3num(ct), &cp->daddr, &cp->caddr, - proto, port, &cp->cport); - else - nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, - nf_ct_l3num(ct), &cp->caddr, &cp->vaddr, - proto, port, &cp->vport); - - exp->expectfn = ip_vs_expect_callback; - - IP_VS_DBG(7, "%s(): ct=%p, expect tuple=" FMT_TUPLE "\n", - __func__, ct, ARG_TUPLE(&exp->tuple)); - nf_ct_expect_related(exp); - nf_ct_expect_put(exp); -} - -/* * Look at outgoing ftp packets to catch the response to a PASV command * from the server (inside-to-outside). * When we see one, we build a connection entry with the client address, @@ -327,14 +195,19 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp, /* * Now update or create an connection entry for it */ - n_cp = ip_vs_conn_out_get(AF_INET, iph->protocol, &from, port, - &cp->caddr, 0); + { + struct ip_vs_conn_param p; + ip_vs_conn_fill_param(AF_INET, iph->protocol, + &from, port, &cp->caddr, 0, &p); + n_cp = ip_vs_conn_out_get(&p); + } if (!n_cp) { - n_cp = ip_vs_conn_new(AF_INET, IPPROTO_TCP, - &cp->caddr, 0, - &cp->vaddr, port, - &from, port, - IP_VS_CONN_F_NO_CPORT, + struct ip_vs_conn_param p; + ip_vs_conn_fill_param(AF_INET, IPPROTO_TCP, &cp->caddr, + 0, &cp->vaddr, port, &p); + n_cp = ip_vs_conn_new(&p, &from, port, + IP_VS_CONN_F_NO_CPORT | + IP_VS_CONN_F_NFCT, cp->dest); if (!n_cp) return 0; @@ -359,7 +232,7 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp, buf_len = strlen(buf); ct = nf_ct_get(skb, &ctinfo); - if (ct && !nf_ct_is_untracked(ct)) { + if (ct && !nf_ct_is_untracked(ct) && nfct_nat(ct)) { /* If mangling fails this function will return 0 * which will cause the packet to be dropped. * Mangling can only fail under memory pressure, @@ -369,9 +242,14 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp, ret = nf_nat_mangle_tcp_packet(skb, ct, ctinfo, start-data, end-start, buf, buf_len); - if (ret) - ip_vs_expect_related(skb, ct, n_cp, - IPPROTO_TCP, NULL, 0); + if (ret) { + ip_vs_nfct_expect_related(skb, ct, n_cp, + IPPROTO_TCP, 0, 0); + if (skb->ip_summed == CHECKSUM_COMPLETE) + skb->ip_summed = CHECKSUM_UNNECESSARY; + /* csum is updated */ + ret = 1; + } } /* @@ -409,7 +287,6 @@ static int ip_vs_ftp_in(struct ip_vs_app *app, struct ip_vs_conn *cp, union nf_inet_addr to; __be16 port; struct ip_vs_conn *n_cp; - struct nf_conn *ct; #ifdef CONFIG_IP_VS_IPV6 /* This application helper doesn't work with IPv6 yet, @@ -479,28 +356,24 @@ static int ip_vs_ftp_in(struct ip_vs_app *app, struct ip_vs_conn *cp, ip_vs_proto_name(iph->protocol), &to.ip, ntohs(port), &cp->vaddr.ip, 0); - n_cp = ip_vs_conn_in_get(AF_INET, iph->protocol, - &to, port, - &cp->vaddr, htons(ntohs(cp->vport)-1)); - if (!n_cp) { - n_cp = ip_vs_conn_new(AF_INET, IPPROTO_TCP, - &to, port, + { + struct ip_vs_conn_param p; + ip_vs_conn_fill_param(AF_INET, iph->protocol, &to, port, &cp->vaddr, htons(ntohs(cp->vport)-1), - &cp->daddr, htons(ntohs(cp->dport)-1), - 0, - cp->dest); - if (!n_cp) - return 0; + &p); + n_cp = ip_vs_conn_in_get(&p); + if (!n_cp) { + n_cp = ip_vs_conn_new(&p, &cp->daddr, + htons(ntohs(cp->dport)-1), + IP_VS_CONN_F_NFCT, cp->dest); + if (!n_cp) + return 0; - /* add its controller */ - ip_vs_control_add(n_cp, cp); + /* add its controller */ + ip_vs_control_add(n_cp, cp); + } } - ct = (struct nf_conn *)skb->nfct; - if (ct && ct != &nf_conntrack_untracked) - ip_vs_expect_related(skb, ct, n_cp, - IPPROTO_TCP, &n_cp->dport, 1); - /* * Move tunnel to listen state */ diff --git a/net/netfilter/ipvs/ip_vs_nfct.c b/net/netfilter/ipvs/ip_vs_nfct.c new file mode 100644 index 00000000000..4680647cd45 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_nfct.c @@ -0,0 +1,292 @@ +/* + * ip_vs_nfct.c: Netfilter connection tracking support for IPVS + * + * Portions Copyright (C) 2001-2002 + * Antefacto Ltd, 181 Parnell St, Dublin 1, Ireland. + * + * Portions Copyright (C) 2003-2010 + * Julian Anastasov + * + * + * This code is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * + * Authors: + * Ben North <ben@redfrontdoor.org> + * Julian Anastasov <ja@ssi.bg> Reorganize and sync with latest kernels + * Hannes Eder <heder@google.com> Extend NFCT support for FTP, ipvs match + * + * + * Current status: + * + * - provide conntrack confirmation for new and related connections, by + * this way we can see their proper conntrack state in all hooks + * - support for all forwarding methods, not only NAT + * - FTP support (NAT), ability to support other NAT apps with expectations + * - to correctly create expectations for related NAT connections the proper + * NF conntrack support must be already installed, eg. ip_vs_ftp requires + * nf_conntrack_ftp ... iptables_nat for the same ports (but no iptables + * NAT rules are needed) + * - alter reply for NAT when forwarding packet in original direction: + * conntrack from client in NEW or RELATED (Passive FTP DATA) state or + * when RELATED conntrack is created from real server (Active FTP DATA) + * - if iptables_nat is not loaded the Passive FTP will not work (the + * PASV response can not be NAT-ed) but Active FTP should work + * + */ + +#define KMSG_COMPONENT "IPVS" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/compiler.h> +#include <linux/vmalloc.h> +#include <linux/skbuff.h> +#include <net/ip.h> +#include <linux/netfilter.h> +#include <linux/netfilter_ipv4.h> +#include <net/ip_vs.h> +#include <net/netfilter/nf_conntrack_core.h> +#include <net/netfilter/nf_conntrack_expect.h> +#include <net/netfilter/nf_conntrack_helper.h> +#include <net/netfilter/nf_conntrack_zones.h> + + +#define FMT_TUPLE "%pI4:%u->%pI4:%u/%u" +#define ARG_TUPLE(T) &(T)->src.u3.ip, ntohs((T)->src.u.all), \ + &(T)->dst.u3.ip, ntohs((T)->dst.u.all), \ + (T)->dst.protonum + +#define FMT_CONN "%pI4:%u->%pI4:%u->%pI4:%u/%u:%u" +#define ARG_CONN(C) &((C)->caddr.ip), ntohs((C)->cport), \ + &((C)->vaddr.ip), ntohs((C)->vport), \ + &((C)->daddr.ip), ntohs((C)->dport), \ + (C)->protocol, (C)->state + +void +ip_vs_update_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp, int outin) +{ + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = ct = nf_ct_get(skb, &ctinfo); + struct nf_conntrack_tuple new_tuple; + + if (ct == NULL || nf_ct_is_confirmed(ct) || nf_ct_is_untracked(ct) || + nf_ct_is_dying(ct)) + return; + + /* Never alter conntrack for non-NAT conns */ + if (IP_VS_FWD_METHOD(cp) != IP_VS_CONN_F_MASQ) + return; + + /* Alter reply only in original direction */ + if (CTINFO2DIR(ctinfo) != IP_CT_DIR_ORIGINAL) + return; + + /* + * The connection is not yet in the hashtable, so we update it. + * CIP->VIP will remain the same, so leave the tuple in + * IP_CT_DIR_ORIGINAL untouched. When the reply comes back from the + * real-server we will see RIP->DIP. + */ + new_tuple = ct->tuplehash[IP_CT_DIR_REPLY].tuple; + /* + * This will also take care of UDP and other protocols. + */ + if (outin) { + new_tuple.src.u3 = cp->daddr; + if (new_tuple.dst.protonum != IPPROTO_ICMP && + new_tuple.dst.protonum != IPPROTO_ICMPV6) + new_tuple.src.u.tcp.port = cp->dport; + } else { + new_tuple.dst.u3 = cp->vaddr; + if (new_tuple.dst.protonum != IPPROTO_ICMP && + new_tuple.dst.protonum != IPPROTO_ICMPV6) + new_tuple.dst.u.tcp.port = cp->vport; + } + IP_VS_DBG(7, "%s: Updating conntrack ct=%p, status=0x%lX, " + "ctinfo=%d, old reply=" FMT_TUPLE + ", new reply=" FMT_TUPLE ", cp=" FMT_CONN "\n", + __func__, ct, ct->status, ctinfo, + ARG_TUPLE(&ct->tuplehash[IP_CT_DIR_REPLY].tuple), + ARG_TUPLE(&new_tuple), ARG_CONN(cp)); + nf_conntrack_alter_reply(ct, &new_tuple); +} + +int ip_vs_confirm_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp) +{ + return nf_conntrack_confirm(skb); +} + +/* + * Called from init_conntrack() as expectfn handler. + */ +static void ip_vs_nfct_expect_callback(struct nf_conn *ct, + struct nf_conntrack_expect *exp) +{ + struct nf_conntrack_tuple *orig, new_reply; + struct ip_vs_conn *cp; + struct ip_vs_conn_param p; + + if (exp->tuple.src.l3num != PF_INET) + return; + + /* + * We assume that no NF locks are held before this callback. + * ip_vs_conn_out_get and ip_vs_conn_in_get should match their + * expectations even if they use wildcard values, now we provide the + * actual values from the newly created original conntrack direction. + * The conntrack is confirmed when packet reaches IPVS hooks. + */ + + /* RS->CLIENT */ + orig = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; + ip_vs_conn_fill_param(exp->tuple.src.l3num, orig->dst.protonum, + &orig->src.u3, orig->src.u.tcp.port, + &orig->dst.u3, orig->dst.u.tcp.port, &p); + cp = ip_vs_conn_out_get(&p); + if (cp) { + /* Change reply CLIENT->RS to CLIENT->VS */ + new_reply = ct->tuplehash[IP_CT_DIR_REPLY].tuple; + IP_VS_DBG(7, "%s: ct=%p, status=0x%lX, tuples=" FMT_TUPLE ", " + FMT_TUPLE ", found inout cp=" FMT_CONN "\n", + __func__, ct, ct->status, + ARG_TUPLE(orig), ARG_TUPLE(&new_reply), + ARG_CONN(cp)); + new_reply.dst.u3 = cp->vaddr; + new_reply.dst.u.tcp.port = cp->vport; + IP_VS_DBG(7, "%s: ct=%p, new tuples=" FMT_TUPLE ", " FMT_TUPLE + ", inout cp=" FMT_CONN "\n", + __func__, ct, + ARG_TUPLE(orig), ARG_TUPLE(&new_reply), + ARG_CONN(cp)); + goto alter; + } + + /* CLIENT->VS */ + cp = ip_vs_conn_in_get(&p); + if (cp) { + /* Change reply VS->CLIENT to RS->CLIENT */ + new_reply = ct->tuplehash[IP_CT_DIR_REPLY].tuple; + IP_VS_DBG(7, "%s: ct=%p, status=0x%lX, tuples=" FMT_TUPLE ", " + FMT_TUPLE ", found outin cp=" FMT_CONN "\n", + __func__, ct, ct->status, + ARG_TUPLE(orig), ARG_TUPLE(&new_reply), + ARG_CONN(cp)); + new_reply.src.u3 = cp->daddr; + new_reply.src.u.tcp.port = cp->dport; + IP_VS_DBG(7, "%s: ct=%p, new tuples=" FMT_TUPLE ", " + FMT_TUPLE ", outin cp=" FMT_CONN "\n", + __func__, ct, + ARG_TUPLE(orig), ARG_TUPLE(&new_reply), + ARG_CONN(cp)); + goto alter; + } + + IP_VS_DBG(7, "%s: ct=%p, status=0x%lX, tuple=" FMT_TUPLE + " - unknown expect\n", + __func__, ct, ct->status, ARG_TUPLE(orig)); + return; + +alter: + /* Never alter conntrack for non-NAT conns */ + if (IP_VS_FWD_METHOD(cp) == IP_VS_CONN_F_MASQ) + nf_conntrack_alter_reply(ct, &new_reply); + ip_vs_conn_put(cp); + return; +} + +/* + * Create NF conntrack expectation with wildcard (optional) source port. + * Then the default callback function will alter the reply and will confirm + * the conntrack entry when the first packet comes. + * Use port 0 to expect connection from any port. + */ +void ip_vs_nfct_expect_related(struct sk_buff *skb, struct nf_conn *ct, + struct ip_vs_conn *cp, u_int8_t proto, + const __be16 port, int from_rs) +{ + struct nf_conntrack_expect *exp; + + if (ct == NULL || nf_ct_is_untracked(ct)) + return; + + exp = nf_ct_expect_alloc(ct); + if (!exp) + return; + + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), + from_rs ? &cp->daddr : &cp->caddr, + from_rs ? &cp->caddr : &cp->vaddr, + proto, port ? &port : NULL, + from_rs ? &cp->cport : &cp->vport); + + exp->expectfn = ip_vs_nfct_expect_callback; + + IP_VS_DBG(7, "%s: ct=%p, expect tuple=" FMT_TUPLE "\n", + __func__, ct, ARG_TUPLE(&exp->tuple)); + nf_ct_expect_related(exp); + nf_ct_expect_put(exp); +} +EXPORT_SYMBOL(ip_vs_nfct_expect_related); + +/* + * Our connection was terminated, try to drop the conntrack immediately + */ +void ip_vs_conn_drop_conntrack(struct ip_vs_conn *cp) +{ + struct nf_conntrack_tuple_hash *h; + struct nf_conn *ct; + struct nf_conntrack_tuple tuple; + + if (!cp->cport) + return; + + tuple = (struct nf_conntrack_tuple) { + .dst = { .protonum = cp->protocol, .dir = IP_CT_DIR_ORIGINAL } }; + tuple.src.u3 = cp->caddr; + tuple.src.u.all = cp->cport; + tuple.src.l3num = cp->af; + tuple.dst.u3 = cp->vaddr; + tuple.dst.u.all = cp->vport; + + IP_VS_DBG(7, "%s: dropping conntrack with tuple=" FMT_TUPLE + " for conn " FMT_CONN "\n", + __func__, ARG_TUPLE(&tuple), ARG_CONN(cp)); + + h = nf_conntrack_find_get(&init_net, NF_CT_DEFAULT_ZONE, &tuple); + if (h) { + ct = nf_ct_tuplehash_to_ctrack(h); + /* Show what happens instead of calling nf_ct_kill() */ + if (del_timer(&ct->timeout)) { + IP_VS_DBG(7, "%s: ct=%p, deleted conntrack timer for tuple=" + FMT_TUPLE "\n", + __func__, ct, ARG_TUPLE(&tuple)); + if (ct->timeout.function) + ct->timeout.function(ct->timeout.data); + } else { + IP_VS_DBG(7, "%s: ct=%p, no conntrack timer for tuple=" + FMT_TUPLE "\n", + __func__, ct, ARG_TUPLE(&tuple)); + } + nf_ct_put(ct); + } else { + IP_VS_DBG(7, "%s: no conntrack for tuple=" FMT_TUPLE "\n", + __func__, ARG_TUPLE(&tuple)); + } +} + diff --git a/net/netfilter/ipvs/ip_vs_pe.c b/net/netfilter/ipvs/ip_vs_pe.c new file mode 100644 index 00000000000..3414af70ee1 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_pe.c @@ -0,0 +1,147 @@ +#define KMSG_COMPONENT "IPVS" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/module.h> +#include <linux/spinlock.h> +#include <linux/interrupt.h> +#include <asm/string.h> +#include <linux/kmod.h> +#include <linux/sysctl.h> + +#include <net/ip_vs.h> + +/* IPVS pe list */ +static LIST_HEAD(ip_vs_pe); + +/* lock for service table */ +static DEFINE_SPINLOCK(ip_vs_pe_lock); + +/* Bind a service with a pe */ +void ip_vs_bind_pe(struct ip_vs_service *svc, struct ip_vs_pe *pe) +{ + svc->pe = pe; +} + +/* Unbind a service from its pe */ +void ip_vs_unbind_pe(struct ip_vs_service *svc) +{ + svc->pe = NULL; +} + +/* Get pe in the pe list by name */ +static struct ip_vs_pe * +ip_vs_pe_getbyname(const char *pe_name) +{ + struct ip_vs_pe *pe; + + IP_VS_DBG(2, "%s(): pe_name \"%s\"\n", __func__, + pe_name); + + spin_lock_bh(&ip_vs_pe_lock); + + list_for_each_entry(pe, &ip_vs_pe, n_list) { + /* Test and get the modules atomically */ + if (pe->module && + !try_module_get(pe->module)) { + /* This pe is just deleted */ + continue; + } + if (strcmp(pe_name, pe->name)==0) { + /* HIT */ + spin_unlock_bh(&ip_vs_pe_lock); + return pe; + } + if (pe->module) + module_put(pe->module); + } + + spin_unlock_bh(&ip_vs_pe_lock); + return NULL; +} + +/* Lookup pe and try to load it if it doesn't exist */ +struct ip_vs_pe *ip_vs_pe_get(const char *name) +{ + struct ip_vs_pe *pe; + + /* Search for the pe by name */ + pe = ip_vs_pe_getbyname(name); + + /* If pe not found, load the module and search again */ + if (!pe) { + request_module("ip_vs_pe_%s", name); + pe = ip_vs_pe_getbyname(name); + } + + return pe; +} + +void ip_vs_pe_put(struct ip_vs_pe *pe) +{ + if (pe && pe->module) + module_put(pe->module); +} + +/* Register a pe in the pe list */ +int register_ip_vs_pe(struct ip_vs_pe *pe) +{ + struct ip_vs_pe *tmp; + + /* increase the module use count */ + ip_vs_use_count_inc(); + + spin_lock_bh(&ip_vs_pe_lock); + + if (!list_empty(&pe->n_list)) { + spin_unlock_bh(&ip_vs_pe_lock); + ip_vs_use_count_dec(); + pr_err("%s(): [%s] pe already linked\n", + __func__, pe->name); + return -EINVAL; + } + + /* Make sure that the pe with this name doesn't exist + * in the pe list. + */ + list_for_each_entry(tmp, &ip_vs_pe, n_list) { + if (strcmp(tmp->name, pe->name) == 0) { + spin_unlock_bh(&ip_vs_pe_lock); + ip_vs_use_count_dec(); + pr_err("%s(): [%s] pe already existed " + "in the system\n", __func__, pe->name); + return -EINVAL; + } + } + /* Add it into the d-linked pe list */ + list_add(&pe->n_list, &ip_vs_pe); + spin_unlock_bh(&ip_vs_pe_lock); + + pr_info("[%s] pe registered.\n", pe->name); + + return 0; +} +EXPORT_SYMBOL_GPL(register_ip_vs_pe); + +/* Unregister a pe from the pe list */ +int unregister_ip_vs_pe(struct ip_vs_pe *pe) +{ + spin_lock_bh(&ip_vs_pe_lock); + if (list_empty(&pe->n_list)) { + spin_unlock_bh(&ip_vs_pe_lock); + pr_err("%s(): [%s] pe is not in the list. failed\n", + __func__, pe->name); + return -EINVAL; + } + + /* Remove it from the d-linked pe list */ + list_del(&pe->n_list); + spin_unlock_bh(&ip_vs_pe_lock); + + /* decrease the module use count */ + ip_vs_use_count_dec(); + + pr_info("[%s] pe unregistered.\n", pe->name); + + return 0; +} +EXPORT_SYMBOL_GPL(unregister_ip_vs_pe); diff --git a/net/netfilter/ipvs/ip_vs_pe_sip.c b/net/netfilter/ipvs/ip_vs_pe_sip.c new file mode 100644 index 00000000000..b8b4e9620f3 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_pe_sip.c @@ -0,0 +1,169 @@ +#define KMSG_COMPONENT "IPVS" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/module.h> +#include <linux/kernel.h> + +#include <net/ip_vs.h> +#include <net/netfilter/nf_conntrack.h> +#include <linux/netfilter/nf_conntrack_sip.h> + +#ifdef CONFIG_IP_VS_DEBUG +static const char *ip_vs_dbg_callid(char *buf, size_t buf_len, + const char *callid, size_t callid_len, + int *idx) +{ + size_t len = min(min(callid_len, (size_t)64), buf_len - *idx - 1); + memcpy(buf + *idx, callid, len); + buf[*idx+len] = '\0'; + *idx += len + 1; + return buf + *idx - len; +} + +#define IP_VS_DEBUG_CALLID(callid, len) \ + ip_vs_dbg_callid(ip_vs_dbg_buf, sizeof(ip_vs_dbg_buf), \ + callid, len, &ip_vs_dbg_idx) +#endif + +static int get_callid(const char *dptr, unsigned int dataoff, + unsigned int datalen, + unsigned int *matchoff, unsigned int *matchlen) +{ + /* Find callid */ + while (1) { + int ret = ct_sip_get_header(NULL, dptr, dataoff, datalen, + SIP_HDR_CALL_ID, matchoff, + matchlen); + if (ret > 0) + break; + if (!ret) + return 0; + dataoff += *matchoff; + } + + /* Empty callid is useless */ + if (!*matchlen) + return -EINVAL; + + /* Too large is useless */ + if (*matchlen > IP_VS_PEDATA_MAXLEN) + return -EINVAL; + + /* SIP headers are always followed by a line terminator */ + if (*matchoff + *matchlen == datalen) + return -EINVAL; + + /* RFC 2543 allows lines to be terminated with CR, LF or CRLF, + * RFC 3261 allows only CRLF, we support both. */ + if (*(dptr + *matchoff + *matchlen) != '\r' && + *(dptr + *matchoff + *matchlen) != '\n') + return -EINVAL; + + IP_VS_DBG_BUF(9, "SIP callid %s (%d bytes)\n", + IP_VS_DEBUG_CALLID(dptr + *matchoff, *matchlen), + *matchlen); + return 0; +} + +static int +ip_vs_sip_fill_param(struct ip_vs_conn_param *p, struct sk_buff *skb) +{ + struct ip_vs_iphdr iph; + unsigned int dataoff, datalen, matchoff, matchlen; + const char *dptr; + + ip_vs_fill_iphdr(p->af, skb_network_header(skb), &iph); + + /* Only useful with UDP */ + if (iph.protocol != IPPROTO_UDP) + return -EINVAL; + + /* No Data ? */ + dataoff = iph.len + sizeof(struct udphdr); + if (dataoff >= skb->len) + return -EINVAL; + + dptr = skb->data + dataoff; + datalen = skb->len - dataoff; + + if (get_callid(dptr, dataoff, datalen, &matchoff, &matchlen)) + return -EINVAL; + + p->pe_data = kmalloc(matchlen, GFP_ATOMIC); + if (!p->pe_data) + return -ENOMEM; + + /* N.B: pe_data is only set on success, + * this allows fallback to the default persistence logic on failure + */ + memcpy(p->pe_data, dptr + matchoff, matchlen); + p->pe_data_len = matchlen; + + return 0; +} + +static bool ip_vs_sip_ct_match(const struct ip_vs_conn_param *p, + struct ip_vs_conn *ct) + +{ + bool ret = 0; + + if (ct->af == p->af && + ip_vs_addr_equal(p->af, p->caddr, &ct->caddr) && + /* protocol should only be IPPROTO_IP if + * d_addr is a fwmark */ + ip_vs_addr_equal(p->protocol == IPPROTO_IP ? AF_UNSPEC : p->af, + p->vaddr, &ct->vaddr) && + ct->vport == p->vport && + ct->flags & IP_VS_CONN_F_TEMPLATE && + ct->protocol == p->protocol && + ct->pe_data && ct->pe_data_len == p->pe_data_len && + !memcmp(ct->pe_data, p->pe_data, p->pe_data_len)) + ret = 1; + + IP_VS_DBG_BUF(9, "SIP template match %s %s->%s:%d %s\n", + ip_vs_proto_name(p->protocol), + IP_VS_DEBUG_CALLID(p->pe_data, p->pe_data_len), + IP_VS_DBG_ADDR(p->af, p->vaddr), ntohs(p->vport), + ret ? "hit" : "not hit"); + + return ret; +} + +static u32 ip_vs_sip_hashkey_raw(const struct ip_vs_conn_param *p, + u32 initval, bool inverse) +{ + return jhash(p->pe_data, p->pe_data_len, initval); +} + +static int ip_vs_sip_show_pe_data(const struct ip_vs_conn *cp, char *buf) +{ + memcpy(buf, cp->pe_data, cp->pe_data_len); + return cp->pe_data_len; +} + +static struct ip_vs_pe ip_vs_sip_pe = +{ + .name = "sip", + .refcnt = ATOMIC_INIT(0), + .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_sip_pe.n_list), + .fill_param = ip_vs_sip_fill_param, + .ct_match = ip_vs_sip_ct_match, + .hashkey_raw = ip_vs_sip_hashkey_raw, + .show_pe_data = ip_vs_sip_show_pe_data, +}; + +static int __init ip_vs_sip_init(void) +{ + return register_ip_vs_pe(&ip_vs_sip_pe); +} + +static void __exit ip_vs_sip_cleanup(void) +{ + unregister_ip_vs_pe(&ip_vs_sip_pe); +} + +module_init(ip_vs_sip_init); +module_exit(ip_vs_sip_cleanup); +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/ipvs/ip_vs_proto.c b/net/netfilter/ipvs/ip_vs_proto.c index 027f654799f..c5399839087 100644 --- a/net/netfilter/ipvs/ip_vs_proto.c +++ b/net/netfilter/ipvs/ip_vs_proto.c @@ -172,8 +172,8 @@ ip_vs_tcpudp_debug_packet_v4(struct ip_vs_protocol *pp, else if (ih->frag_off & htons(IP_OFFSET)) sprintf(buf, "%pI4->%pI4 frag", &ih->saddr, &ih->daddr); else { - __be16 _ports[2], *pptr -; + __be16 _ports[2], *pptr; + pptr = skb_header_pointer(skb, offset + ih->ihl*4, sizeof(_ports), _ports); if (pptr == NULL) @@ -223,13 +223,13 @@ ip_vs_tcpudp_debug_packet_v6(struct ip_vs_protocol *pp, void -ip_vs_tcpudp_debug_packet(struct ip_vs_protocol *pp, +ip_vs_tcpudp_debug_packet(int af, struct ip_vs_protocol *pp, const struct sk_buff *skb, int offset, const char *msg) { #ifdef CONFIG_IP_VS_IPV6 - if (skb->protocol == htons(ETH_P_IPV6)) + if (af == AF_INET6) ip_vs_tcpudp_debug_packet_v6(pp, skb, offset, msg); else #endif diff --git a/net/netfilter/ipvs/ip_vs_proto_ah_esp.c b/net/netfilter/ipvs/ip_vs_proto_ah_esp.c index 1892dfc12fd..3a0461117d3 100644 --- a/net/netfilter/ipvs/ip_vs_proto_ah_esp.c +++ b/net/netfilter/ipvs/ip_vs_proto_ah_esp.c @@ -40,6 +40,19 @@ struct isakmp_hdr { #define PORT_ISAKMP 500 +static void +ah_esp_conn_fill_param_proto(int af, const struct ip_vs_iphdr *iph, + int inverse, struct ip_vs_conn_param *p) +{ + if (likely(!inverse)) + ip_vs_conn_fill_param(af, IPPROTO_UDP, + &iph->saddr, htons(PORT_ISAKMP), + &iph->daddr, htons(PORT_ISAKMP), p); + else + ip_vs_conn_fill_param(af, IPPROTO_UDP, + &iph->daddr, htons(PORT_ISAKMP), + &iph->saddr, htons(PORT_ISAKMP), p); +} static struct ip_vs_conn * ah_esp_conn_in_get(int af, const struct sk_buff *skb, struct ip_vs_protocol *pp, @@ -47,21 +60,10 @@ ah_esp_conn_in_get(int af, const struct sk_buff *skb, struct ip_vs_protocol *pp, int inverse) { struct ip_vs_conn *cp; + struct ip_vs_conn_param p; - if (likely(!inverse)) { - cp = ip_vs_conn_in_get(af, IPPROTO_UDP, - &iph->saddr, - htons(PORT_ISAKMP), - &iph->daddr, - htons(PORT_ISAKMP)); - } else { - cp = ip_vs_conn_in_get(af, IPPROTO_UDP, - &iph->daddr, - htons(PORT_ISAKMP), - &iph->saddr, - htons(PORT_ISAKMP)); - } - + ah_esp_conn_fill_param_proto(af, iph, inverse, &p); + cp = ip_vs_conn_in_get(&p); if (!cp) { /* * We are not sure if the packet is from our @@ -87,21 +89,10 @@ ah_esp_conn_out_get(int af, const struct sk_buff *skb, int inverse) { struct ip_vs_conn *cp; + struct ip_vs_conn_param p; - if (likely(!inverse)) { - cp = ip_vs_conn_out_get(af, IPPROTO_UDP, - &iph->saddr, - htons(PORT_ISAKMP), - &iph->daddr, - htons(PORT_ISAKMP)); - } else { - cp = ip_vs_conn_out_get(af, IPPROTO_UDP, - &iph->daddr, - htons(PORT_ISAKMP), - &iph->saddr, - htons(PORT_ISAKMP)); - } - + ah_esp_conn_fill_param_proto(af, iph, inverse, &p); + cp = ip_vs_conn_out_get(&p); if (!cp) { IP_VS_DBG_BUF(12, "Unknown ISAKMP entry for inout packet " "%s%s %s->%s\n", @@ -126,54 +117,6 @@ ah_esp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, return 0; } - -static void -ah_esp_debug_packet_v4(struct ip_vs_protocol *pp, const struct sk_buff *skb, - int offset, const char *msg) -{ - char buf[256]; - struct iphdr _iph, *ih; - - ih = skb_header_pointer(skb, offset, sizeof(_iph), &_iph); - if (ih == NULL) - sprintf(buf, "TRUNCATED"); - else - sprintf(buf, "%pI4->%pI4", &ih->saddr, &ih->daddr); - - pr_debug("%s: %s %s\n", msg, pp->name, buf); -} - -#ifdef CONFIG_IP_VS_IPV6 -static void -ah_esp_debug_packet_v6(struct ip_vs_protocol *pp, const struct sk_buff *skb, - int offset, const char *msg) -{ - char buf[256]; - struct ipv6hdr _iph, *ih; - - ih = skb_header_pointer(skb, offset, sizeof(_iph), &_iph); - if (ih == NULL) - sprintf(buf, "TRUNCATED"); - else - sprintf(buf, "%pI6->%pI6", &ih->saddr, &ih->daddr); - - pr_debug("%s: %s %s\n", msg, pp->name, buf); -} -#endif - -static void -ah_esp_debug_packet(struct ip_vs_protocol *pp, const struct sk_buff *skb, - int offset, const char *msg) -{ -#ifdef CONFIG_IP_VS_IPV6 - if (skb->protocol == htons(ETH_P_IPV6)) - ah_esp_debug_packet_v6(pp, skb, offset, msg); - else -#endif - ah_esp_debug_packet_v4(pp, skb, offset, msg); -} - - static void ah_esp_init(struct ip_vs_protocol *pp) { /* nothing to do now */ @@ -204,7 +147,7 @@ struct ip_vs_protocol ip_vs_protocol_ah = { .register_app = NULL, .unregister_app = NULL, .app_conn_bind = NULL, - .debug_packet = ah_esp_debug_packet, + .debug_packet = ip_vs_tcpudp_debug_packet, .timeout_change = NULL, /* ISAKMP */ .set_state_timeout = NULL, }; @@ -228,7 +171,7 @@ struct ip_vs_protocol ip_vs_protocol_esp = { .register_app = NULL, .unregister_app = NULL, .app_conn_bind = NULL, - .debug_packet = ah_esp_debug_packet, + .debug_packet = ip_vs_tcpudp_debug_packet, .timeout_change = NULL, /* ISAKMP */ }; #endif diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c index 4c0855cb006..1ea96bcd342 100644 --- a/net/netfilter/ipvs/ip_vs_proto_sctp.c +++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c @@ -31,6 +31,8 @@ sctp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, if ((sch->type == SCTP_CID_INIT) && (svc = ip_vs_service_get(af, skb->mark, iph.protocol, &iph.daddr, sh->dest))) { + int ignored; + if (ip_vs_todrop()) { /* * It seems that we are very loaded. @@ -44,8 +46,8 @@ sctp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, * Let the virtual server select a real server for the * incoming connection, and create a connection entry. */ - *cpp = ip_vs_schedule(svc, skb); - if (!*cpp) { + *cpp = ip_vs_schedule(svc, skb, pp, &ignored); + if (!*cpp && !ignored) { *verdict = ip_vs_leave(svc, skb, pp); return 0; } @@ -61,6 +63,7 @@ sctp_snat_handler(struct sk_buff *skb, { sctp_sctphdr_t *sctph; unsigned int sctphoff; + struct sk_buff *iter; __be32 crc32; #ifdef CONFIG_IP_VS_IPV6 @@ -89,8 +92,8 @@ sctp_snat_handler(struct sk_buff *skb, /* Calculate the checksum */ crc32 = sctp_start_cksum((u8 *) sctph, skb_headlen(skb) - sctphoff); - for (skb = skb_shinfo(skb)->frag_list; skb; skb = skb->next) - crc32 = sctp_update_cksum((u8 *) skb->data, skb_headlen(skb), + skb_walk_frags(skb, iter) + crc32 = sctp_update_cksum((u8 *) iter->data, skb_headlen(iter), crc32); crc32 = sctp_end_cksum(crc32); sctph->checksum = crc32; @@ -102,9 +105,9 @@ static int sctp_dnat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp, struct ip_vs_conn *cp) { - sctp_sctphdr_t *sctph; unsigned int sctphoff; + struct sk_buff *iter; __be32 crc32; #ifdef CONFIG_IP_VS_IPV6 @@ -133,8 +136,8 @@ sctp_dnat_handler(struct sk_buff *skb, /* Calculate the checksum */ crc32 = sctp_start_cksum((u8 *) sctph, skb_headlen(skb) - sctphoff); - for (skb = skb_shinfo(skb)->frag_list; skb; skb = skb->next) - crc32 = sctp_update_cksum((u8 *) skb->data, skb_headlen(skb), + skb_walk_frags(skb, iter) + crc32 = sctp_update_cksum((u8 *) iter->data, skb_headlen(iter), crc32); crc32 = sctp_end_cksum(crc32); sctph->checksum = crc32; @@ -145,9 +148,9 @@ sctp_dnat_handler(struct sk_buff *skb, static int sctp_csum_check(int af, struct sk_buff *skb, struct ip_vs_protocol *pp) { - struct sk_buff *list = skb_shinfo(skb)->frag_list; unsigned int sctphoff; struct sctphdr *sh, _sctph; + struct sk_buff *iter; __le32 cmp; __le32 val; __u32 tmp; @@ -166,15 +169,15 @@ sctp_csum_check(int af, struct sk_buff *skb, struct ip_vs_protocol *pp) cmp = sh->checksum; tmp = sctp_start_cksum((__u8 *) sh, skb_headlen(skb)); - for (; list; list = list->next) - tmp = sctp_update_cksum((__u8 *) list->data, - skb_headlen(list), tmp); + skb_walk_frags(skb, iter) + tmp = sctp_update_cksum((__u8 *) iter->data, + skb_headlen(iter), tmp); val = sctp_end_cksum(tmp); if (val != cmp) { /* CRC failure, dump it. */ - IP_VS_DBG_RL_PKT(0, pp, skb, 0, + IP_VS_DBG_RL_PKT(0, af, pp, skb, 0, "Failed checksum for"); return 0; } diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c index 282d24de859..f6c5200e214 100644 --- a/net/netfilter/ipvs/ip_vs_proto_tcp.c +++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c @@ -43,9 +43,12 @@ tcp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, return 0; } + /* No !th->ack check to allow scheduling on SYN+ACK for Active FTP */ if (th->syn && (svc = ip_vs_service_get(af, skb->mark, iph.protocol, &iph.daddr, th->dest))) { + int ignored; + if (ip_vs_todrop()) { /* * It seems that we are very loaded. @@ -60,8 +63,8 @@ tcp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, * Let the virtual server select a real server for the * incoming connection, and create a connection entry. */ - *cpp = ip_vs_schedule(svc, skb); - if (!*cpp) { + *cpp = ip_vs_schedule(svc, skb, pp, &ignored); + if (!*cpp && !ignored) { *verdict = ip_vs_leave(svc, skb, pp); return 0; } @@ -101,15 +104,15 @@ tcp_partial_csum_update(int af, struct tcphdr *tcph, #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) tcph->check = - csum_fold(ip_vs_check_diff16(oldip->ip6, newip->ip6, + ~csum_fold(ip_vs_check_diff16(oldip->ip6, newip->ip6, ip_vs_check_diff2(oldlen, newlen, - ~csum_unfold(tcph->check)))); + csum_unfold(tcph->check)))); else #endif tcph->check = - csum_fold(ip_vs_check_diff4(oldip->ip, newip->ip, + ~csum_fold(ip_vs_check_diff4(oldip->ip, newip->ip, ip_vs_check_diff2(oldlen, newlen, - ~csum_unfold(tcph->check)))); + csum_unfold(tcph->check)))); } @@ -120,6 +123,7 @@ tcp_snat_handler(struct sk_buff *skb, struct tcphdr *tcph; unsigned int tcphoff; int oldlen; + int payload_csum = 0; #ifdef CONFIG_IP_VS_IPV6 if (cp->af == AF_INET6) @@ -134,13 +138,20 @@ tcp_snat_handler(struct sk_buff *skb, return 0; if (unlikely(cp->app != NULL)) { + int ret; + /* Some checks before mangling */ if (pp->csum_check && !pp->csum_check(cp->af, skb, pp)) return 0; /* Call application helper if needed */ - if (!ip_vs_app_pkt_out(cp, skb)) + if (!(ret = ip_vs_app_pkt_out(cp, skb))) return 0; + /* ret=2: csum update is needed after payload mangling */ + if (ret == 1) + oldlen = skb->len - tcphoff; + else + payload_csum = 1; } tcph = (void *)skb_network_header(skb) + tcphoff; @@ -151,12 +162,13 @@ tcp_snat_handler(struct sk_buff *skb, tcp_partial_csum_update(cp->af, tcph, &cp->daddr, &cp->vaddr, htons(oldlen), htons(skb->len - tcphoff)); - } else if (!cp->app) { + } else if (!payload_csum) { /* Only port and addr are changed, do fast csum update */ tcp_fast_csum_update(cp->af, tcph, &cp->daddr, &cp->vaddr, cp->dport, cp->vport); if (skb->ip_summed == CHECKSUM_COMPLETE) - skb->ip_summed = CHECKSUM_NONE; + skb->ip_summed = (cp->app && pp->csum_check) ? + CHECKSUM_UNNECESSARY : CHECKSUM_NONE; } else { /* full checksum calculation */ tcph->check = 0; @@ -174,6 +186,7 @@ tcp_snat_handler(struct sk_buff *skb, skb->len - tcphoff, cp->protocol, skb->csum); + skb->ip_summed = CHECKSUM_UNNECESSARY; IP_VS_DBG(11, "O-pkt: %s O-csum=%d (+%zd)\n", pp->name, tcph->check, @@ -190,6 +203,7 @@ tcp_dnat_handler(struct sk_buff *skb, struct tcphdr *tcph; unsigned int tcphoff; int oldlen; + int payload_csum = 0; #ifdef CONFIG_IP_VS_IPV6 if (cp->af == AF_INET6) @@ -204,6 +218,8 @@ tcp_dnat_handler(struct sk_buff *skb, return 0; if (unlikely(cp->app != NULL)) { + int ret; + /* Some checks before mangling */ if (pp->csum_check && !pp->csum_check(cp->af, skb, pp)) return 0; @@ -212,8 +228,13 @@ tcp_dnat_handler(struct sk_buff *skb, * Attempt ip_vs_app call. * It will fix ip_vs_conn and iph ack_seq stuff */ - if (!ip_vs_app_pkt_in(cp, skb)) + if (!(ret = ip_vs_app_pkt_in(cp, skb))) return 0; + /* ret=2: csum update is needed after payload mangling */ + if (ret == 1) + oldlen = skb->len - tcphoff; + else + payload_csum = 1; } tcph = (void *)skb_network_header(skb) + tcphoff; @@ -223,15 +244,16 @@ tcp_dnat_handler(struct sk_buff *skb, * Adjust TCP checksums */ if (skb->ip_summed == CHECKSUM_PARTIAL) { - tcp_partial_csum_update(cp->af, tcph, &cp->daddr, &cp->vaddr, + tcp_partial_csum_update(cp->af, tcph, &cp->vaddr, &cp->daddr, htons(oldlen), htons(skb->len - tcphoff)); - } else if (!cp->app) { + } else if (!payload_csum) { /* Only port and addr are changed, do fast csum update */ tcp_fast_csum_update(cp->af, tcph, &cp->vaddr, &cp->daddr, cp->vport, cp->dport); if (skb->ip_summed == CHECKSUM_COMPLETE) - skb->ip_summed = CHECKSUM_NONE; + skb->ip_summed = (cp->app && pp->csum_check) ? + CHECKSUM_UNNECESSARY : CHECKSUM_NONE; } else { /* full checksum calculation */ tcph->check = 0; @@ -278,7 +300,7 @@ tcp_csum_check(int af, struct sk_buff *skb, struct ip_vs_protocol *pp) skb->len - tcphoff, ipv6_hdr(skb)->nexthdr, skb->csum)) { - IP_VS_DBG_RL_PKT(0, pp, skb, 0, + IP_VS_DBG_RL_PKT(0, af, pp, skb, 0, "Failed checksum for"); return 0; } @@ -289,7 +311,7 @@ tcp_csum_check(int af, struct sk_buff *skb, struct ip_vs_protocol *pp) skb->len - tcphoff, ip_hdr(skb)->protocol, skb->csum)) { - IP_VS_DBG_RL_PKT(0, pp, skb, 0, + IP_VS_DBG_RL_PKT(0, af, pp, skb, 0, "Failed checksum for"); return 0; } diff --git a/net/netfilter/ipvs/ip_vs_proto_udp.c b/net/netfilter/ipvs/ip_vs_proto_udp.c index 8553231b5d4..9d106a06bb0 100644 --- a/net/netfilter/ipvs/ip_vs_proto_udp.c +++ b/net/netfilter/ipvs/ip_vs_proto_udp.c @@ -46,6 +46,8 @@ udp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, svc = ip_vs_service_get(af, skb->mark, iph.protocol, &iph.daddr, uh->dest); if (svc) { + int ignored; + if (ip_vs_todrop()) { /* * It seems that we are very loaded. @@ -60,8 +62,8 @@ udp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, * Let the virtual server select a real server for the * incoming connection, and create a connection entry. */ - *cpp = ip_vs_schedule(svc, skb); - if (!*cpp) { + *cpp = ip_vs_schedule(svc, skb, pp, &ignored); + if (!*cpp && !ignored) { *verdict = ip_vs_leave(svc, skb, pp); return 0; } @@ -102,15 +104,15 @@ udp_partial_csum_update(int af, struct udphdr *uhdr, #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) uhdr->check = - csum_fold(ip_vs_check_diff16(oldip->ip6, newip->ip6, + ~csum_fold(ip_vs_check_diff16(oldip->ip6, newip->ip6, ip_vs_check_diff2(oldlen, newlen, - ~csum_unfold(uhdr->check)))); + csum_unfold(uhdr->check)))); else #endif uhdr->check = - csum_fold(ip_vs_check_diff4(oldip->ip, newip->ip, + ~csum_fold(ip_vs_check_diff4(oldip->ip, newip->ip, ip_vs_check_diff2(oldlen, newlen, - ~csum_unfold(uhdr->check)))); + csum_unfold(uhdr->check)))); } @@ -121,6 +123,7 @@ udp_snat_handler(struct sk_buff *skb, struct udphdr *udph; unsigned int udphoff; int oldlen; + int payload_csum = 0; #ifdef CONFIG_IP_VS_IPV6 if (cp->af == AF_INET6) @@ -135,6 +138,8 @@ udp_snat_handler(struct sk_buff *skb, return 0; if (unlikely(cp->app != NULL)) { + int ret; + /* Some checks before mangling */ if (pp->csum_check && !pp->csum_check(cp->af, skb, pp)) return 0; @@ -142,8 +147,13 @@ udp_snat_handler(struct sk_buff *skb, /* * Call application helper if needed */ - if (!ip_vs_app_pkt_out(cp, skb)) + if (!(ret = ip_vs_app_pkt_out(cp, skb))) return 0; + /* ret=2: csum update is needed after payload mangling */ + if (ret == 1) + oldlen = skb->len - udphoff; + else + payload_csum = 1; } udph = (void *)skb_network_header(skb) + udphoff; @@ -156,12 +166,13 @@ udp_snat_handler(struct sk_buff *skb, udp_partial_csum_update(cp->af, udph, &cp->daddr, &cp->vaddr, htons(oldlen), htons(skb->len - udphoff)); - } else if (!cp->app && (udph->check != 0)) { + } else if (!payload_csum && (udph->check != 0)) { /* Only port and addr are changed, do fast csum update */ udp_fast_csum_update(cp->af, udph, &cp->daddr, &cp->vaddr, cp->dport, cp->vport); if (skb->ip_summed == CHECKSUM_COMPLETE) - skb->ip_summed = CHECKSUM_NONE; + skb->ip_summed = (cp->app && pp->csum_check) ? + CHECKSUM_UNNECESSARY : CHECKSUM_NONE; } else { /* full checksum calculation */ udph->check = 0; @@ -181,6 +192,7 @@ udp_snat_handler(struct sk_buff *skb, skb->csum); if (udph->check == 0) udph->check = CSUM_MANGLED_0; + skb->ip_summed = CHECKSUM_UNNECESSARY; IP_VS_DBG(11, "O-pkt: %s O-csum=%d (+%zd)\n", pp->name, udph->check, (char*)&(udph->check) - (char*)udph); @@ -196,6 +208,7 @@ udp_dnat_handler(struct sk_buff *skb, struct udphdr *udph; unsigned int udphoff; int oldlen; + int payload_csum = 0; #ifdef CONFIG_IP_VS_IPV6 if (cp->af == AF_INET6) @@ -210,6 +223,8 @@ udp_dnat_handler(struct sk_buff *skb, return 0; if (unlikely(cp->app != NULL)) { + int ret; + /* Some checks before mangling */ if (pp->csum_check && !pp->csum_check(cp->af, skb, pp)) return 0; @@ -218,8 +233,13 @@ udp_dnat_handler(struct sk_buff *skb, * Attempt ip_vs_app call. * It will fix ip_vs_conn */ - if (!ip_vs_app_pkt_in(cp, skb)) + if (!(ret = ip_vs_app_pkt_in(cp, skb))) return 0; + /* ret=2: csum update is needed after payload mangling */ + if (ret == 1) + oldlen = skb->len - udphoff; + else + payload_csum = 1; } udph = (void *)skb_network_header(skb) + udphoff; @@ -229,15 +249,16 @@ udp_dnat_handler(struct sk_buff *skb, * Adjust UDP checksums */ if (skb->ip_summed == CHECKSUM_PARTIAL) { - udp_partial_csum_update(cp->af, udph, &cp->daddr, &cp->vaddr, + udp_partial_csum_update(cp->af, udph, &cp->vaddr, &cp->daddr, htons(oldlen), htons(skb->len - udphoff)); - } else if (!cp->app && (udph->check != 0)) { + } else if (!payload_csum && (udph->check != 0)) { /* Only port and addr are changed, do fast csum update */ udp_fast_csum_update(cp->af, udph, &cp->vaddr, &cp->daddr, cp->vport, cp->dport); if (skb->ip_summed == CHECKSUM_COMPLETE) - skb->ip_summed = CHECKSUM_NONE; + skb->ip_summed = (cp->app && pp->csum_check) ? + CHECKSUM_UNNECESSARY : CHECKSUM_NONE; } else { /* full checksum calculation */ udph->check = 0; @@ -293,7 +314,7 @@ udp_csum_check(int af, struct sk_buff *skb, struct ip_vs_protocol *pp) skb->len - udphoff, ipv6_hdr(skb)->nexthdr, skb->csum)) { - IP_VS_DBG_RL_PKT(0, pp, skb, 0, + IP_VS_DBG_RL_PKT(0, af, pp, skb, 0, "Failed checksum for"); return 0; } @@ -304,7 +325,7 @@ udp_csum_check(int af, struct sk_buff *skb, struct ip_vs_protocol *pp) skb->len - udphoff, ip_hdr(skb)->protocol, skb->csum)) { - IP_VS_DBG_RL_PKT(0, pp, skb, 0, + IP_VS_DBG_RL_PKT(0, af, pp, skb, 0, "Failed checksum for"); return 0; } diff --git a/net/netfilter/ipvs/ip_vs_sched.c b/net/netfilter/ipvs/ip_vs_sched.c index bbc1ac79595..076ebe00435 100644 --- a/net/netfilter/ipvs/ip_vs_sched.c +++ b/net/netfilter/ipvs/ip_vs_sched.c @@ -35,7 +35,7 @@ static LIST_HEAD(ip_vs_schedulers); /* lock for service table */ -static DEFINE_RWLOCK(__ip_vs_sched_lock); +static DEFINE_SPINLOCK(ip_vs_sched_lock); /* @@ -46,15 +46,6 @@ int ip_vs_bind_scheduler(struct ip_vs_service *svc, { int ret; - if (svc == NULL) { - pr_err("%s(): svc arg NULL\n", __func__); - return -EINVAL; - } - if (scheduler == NULL) { - pr_err("%s(): scheduler arg NULL\n", __func__); - return -EINVAL; - } - svc->scheduler = scheduler; if (scheduler->init_service) { @@ -74,18 +65,10 @@ int ip_vs_bind_scheduler(struct ip_vs_service *svc, */ int ip_vs_unbind_scheduler(struct ip_vs_service *svc) { - struct ip_vs_scheduler *sched; + struct ip_vs_scheduler *sched = svc->scheduler; - if (svc == NULL) { - pr_err("%s(): svc arg NULL\n", __func__); - return -EINVAL; - } - - sched = svc->scheduler; - if (sched == NULL) { - pr_err("%s(): svc isn't bound\n", __func__); - return -EINVAL; - } + if (!sched) + return 0; if (sched->done_service) { if (sched->done_service(svc) != 0) { @@ -108,7 +91,7 @@ static struct ip_vs_scheduler *ip_vs_sched_getbyname(const char *sched_name) IP_VS_DBG(2, "%s(): sched_name \"%s\"\n", __func__, sched_name); - read_lock_bh(&__ip_vs_sched_lock); + spin_lock_bh(&ip_vs_sched_lock); list_for_each_entry(sched, &ip_vs_schedulers, n_list) { /* @@ -122,14 +105,14 @@ static struct ip_vs_scheduler *ip_vs_sched_getbyname(const char *sched_name) } if (strcmp(sched_name, sched->name)==0) { /* HIT */ - read_unlock_bh(&__ip_vs_sched_lock); + spin_unlock_bh(&ip_vs_sched_lock); return sched; } if (sched->module) module_put(sched->module); } - read_unlock_bh(&__ip_vs_sched_lock); + spin_unlock_bh(&ip_vs_sched_lock); return NULL; } @@ -159,7 +142,7 @@ struct ip_vs_scheduler *ip_vs_scheduler_get(const char *sched_name) void ip_vs_scheduler_put(struct ip_vs_scheduler *scheduler) { - if (scheduler->module) + if (scheduler && scheduler->module) module_put(scheduler->module); } @@ -184,10 +167,10 @@ int register_ip_vs_scheduler(struct ip_vs_scheduler *scheduler) /* increase the module use count */ ip_vs_use_count_inc(); - write_lock_bh(&__ip_vs_sched_lock); + spin_lock_bh(&ip_vs_sched_lock); if (!list_empty(&scheduler->n_list)) { - write_unlock_bh(&__ip_vs_sched_lock); + spin_unlock_bh(&ip_vs_sched_lock); ip_vs_use_count_dec(); pr_err("%s(): [%s] scheduler already linked\n", __func__, scheduler->name); @@ -200,7 +183,7 @@ int register_ip_vs_scheduler(struct ip_vs_scheduler *scheduler) */ list_for_each_entry(sched, &ip_vs_schedulers, n_list) { if (strcmp(scheduler->name, sched->name) == 0) { - write_unlock_bh(&__ip_vs_sched_lock); + spin_unlock_bh(&ip_vs_sched_lock); ip_vs_use_count_dec(); pr_err("%s(): [%s] scheduler already existed " "in the system\n", __func__, scheduler->name); @@ -211,7 +194,7 @@ int register_ip_vs_scheduler(struct ip_vs_scheduler *scheduler) * Add it into the d-linked scheduler list */ list_add(&scheduler->n_list, &ip_vs_schedulers); - write_unlock_bh(&__ip_vs_sched_lock); + spin_unlock_bh(&ip_vs_sched_lock); pr_info("[%s] scheduler registered.\n", scheduler->name); @@ -229,9 +212,9 @@ int unregister_ip_vs_scheduler(struct ip_vs_scheduler *scheduler) return -EINVAL; } - write_lock_bh(&__ip_vs_sched_lock); + spin_lock_bh(&ip_vs_sched_lock); if (list_empty(&scheduler->n_list)) { - write_unlock_bh(&__ip_vs_sched_lock); + spin_unlock_bh(&ip_vs_sched_lock); pr_err("%s(): [%s] scheduler is not in the list. failed\n", __func__, scheduler->name); return -EINVAL; @@ -241,7 +224,7 @@ int unregister_ip_vs_scheduler(struct ip_vs_scheduler *scheduler) * Remove it from the d-linked scheduler list */ list_del(&scheduler->n_list); - write_unlock_bh(&__ip_vs_sched_lock); + spin_unlock_bh(&ip_vs_sched_lock); /* decrease the module use count */ ip_vs_use_count_dec(); diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index 7ba06939829..ab85aedea17 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -288,6 +288,16 @@ void ip_vs_sync_conn(struct ip_vs_conn *cp) ip_vs_sync_conn(cp->control); } +static inline int +ip_vs_conn_fill_param_sync(int af, int protocol, + const union nf_inet_addr *caddr, __be16 cport, + const union nf_inet_addr *vaddr, __be16 vport, + struct ip_vs_conn_param *p) +{ + /* XXX: Need to take into account persistence engine */ + ip_vs_conn_fill_param(af, protocol, caddr, cport, vaddr, vport, p); + return 0; +} /* * Process received multicast message and create the corresponding @@ -301,6 +311,7 @@ static void ip_vs_process_message(const char *buffer, const size_t buflen) struct ip_vs_conn *cp; struct ip_vs_protocol *pp; struct ip_vs_dest *dest; + struct ip_vs_conn_param param; char *p; int i; @@ -370,18 +381,20 @@ static void ip_vs_process_message(const char *buffer, const size_t buflen) } } - if (!(flags & IP_VS_CONN_F_TEMPLATE)) - cp = ip_vs_conn_in_get(AF_INET, s->protocol, - (union nf_inet_addr *)&s->caddr, - s->cport, - (union nf_inet_addr *)&s->vaddr, - s->vport); - else - cp = ip_vs_ct_in_get(AF_INET, s->protocol, - (union nf_inet_addr *)&s->caddr, - s->cport, - (union nf_inet_addr *)&s->vaddr, - s->vport); + { + if (ip_vs_conn_fill_param_sync(AF_INET, s->protocol, + (union nf_inet_addr *)&s->caddr, + s->cport, + (union nf_inet_addr *)&s->vaddr, + s->vport, ¶m)) { + pr_err("ip_vs_conn_fill_param_sync failed"); + return; + } + if (!(flags & IP_VS_CONN_F_TEMPLATE)) + cp = ip_vs_conn_in_get(¶m); + else + cp = ip_vs_ct_in_get(¶m); + } if (!cp) { /* * Find the appropriate destination for the connection. @@ -406,14 +419,9 @@ static void ip_vs_process_message(const char *buffer, const size_t buflen) else flags &= ~IP_VS_CONN_F_INACTIVE; } - cp = ip_vs_conn_new(AF_INET, s->protocol, - (union nf_inet_addr *)&s->caddr, - s->cport, - (union nf_inet_addr *)&s->vaddr, - s->vport, + cp = ip_vs_conn_new(¶m, (union nf_inet_addr *)&s->daddr, - s->dport, - flags, dest); + s->dport, flags, dest); if (dest) atomic_dec(&dest->refcnt); if (!cp) { diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 21e1a5e9b9d..de04ea39cde 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -11,6 +11,16 @@ * * Changes: * + * Description of forwarding methods: + * - all transmitters are called from LOCAL_IN (remote clients) and + * LOCAL_OUT (local clients) but for ICMP can be called from FORWARD + * - not all connections have destination server, for example, + * connections in backup server when fwmark is used + * - bypass connections use daddr from packet + * LOCAL_OUT rules: + * - skb->dev is NULL, skb->protocol is not set (both are set in POST_ROUTING) + * - skb->pkt_type is not set yet + * - the only place where we can see skb->sk != NULL */ #define KMSG_COMPONENT "IPVS" @@ -26,9 +36,9 @@ #include <net/route.h> /* for ip_route_output */ #include <net/ipv6.h> #include <net/ip6_route.h> +#include <net/addrconf.h> #include <linux/icmpv6.h> #include <linux/netfilter.h> -#include <net/netfilter/nf_conntrack.h> #include <linux/netfilter_ipv4.h> #include <net/ip_vs.h> @@ -38,26 +48,27 @@ * Destination cache to speed up outgoing route lookup */ static inline void -__ip_vs_dst_set(struct ip_vs_dest *dest, u32 rtos, struct dst_entry *dst) +__ip_vs_dst_set(struct ip_vs_dest *dest, u32 rtos, struct dst_entry *dst, + u32 dst_cookie) { struct dst_entry *old_dst; old_dst = dest->dst_cache; dest->dst_cache = dst; dest->dst_rtos = rtos; + dest->dst_cookie = dst_cookie; dst_release(old_dst); } static inline struct dst_entry * -__ip_vs_dst_check(struct ip_vs_dest *dest, u32 rtos, u32 cookie) +__ip_vs_dst_check(struct ip_vs_dest *dest, u32 rtos) { struct dst_entry *dst = dest->dst_cache; if (!dst) return NULL; - if ((dst->obsolete - || (dest->af == AF_INET && rtos != dest->dst_rtos)) && - dst->ops->check(dst, cookie) == NULL) { + if ((dst->obsolete || rtos != dest->dst_rtos) && + dst->ops->check(dst, dest->dst_cookie) == NULL) { dest->dst_cache = NULL; dst_release(dst); return NULL; @@ -66,16 +77,24 @@ __ip_vs_dst_check(struct ip_vs_dest *dest, u32 rtos, u32 cookie) return dst; } +/* + * Get route to destination or remote server + * rt_mode: flags, &1=Allow local dest, &2=Allow non-local dest, + * &4=Allow redirect from remote daddr to local + */ static struct rtable * -__ip_vs_get_out_rt(struct ip_vs_conn *cp, u32 rtos) +__ip_vs_get_out_rt(struct sk_buff *skb, struct ip_vs_dest *dest, + __be32 daddr, u32 rtos, int rt_mode) { + struct net *net = dev_net(skb_dst(skb)->dev); struct rtable *rt; /* Route to the other host */ - struct ip_vs_dest *dest = cp->dest; + struct rtable *ort; /* Original route */ + int local; if (dest) { spin_lock(&dest->dst_lock); if (!(rt = (struct rtable *) - __ip_vs_dst_check(dest, rtos, 0))) { + __ip_vs_dst_check(dest, rtos))) { struct flowi fl = { .oif = 0, .nl_u = { @@ -85,13 +104,13 @@ __ip_vs_get_out_rt(struct ip_vs_conn *cp, u32 rtos) .tos = rtos, } }, }; - if (ip_route_output_key(&init_net, &rt, &fl)) { + if (ip_route_output_key(net, &rt, &fl)) { spin_unlock(&dest->dst_lock); IP_VS_DBG_RL("ip_route_output error, dest: %pI4\n", &dest->addr.ip); return NULL; } - __ip_vs_dst_set(dest, rtos, dst_clone(&rt->dst)); + __ip_vs_dst_set(dest, rtos, dst_clone(&rt->dst), 0); IP_VS_DBG(10, "new dst %pI4, refcnt=%d, rtos=%X\n", &dest->addr.ip, atomic_read(&rt->dst.__refcnt), rtos); @@ -102,78 +121,199 @@ __ip_vs_get_out_rt(struct ip_vs_conn *cp, u32 rtos) .oif = 0, .nl_u = { .ip4_u = { - .daddr = cp->daddr.ip, + .daddr = daddr, .saddr = 0, .tos = rtos, } }, }; - if (ip_route_output_key(&init_net, &rt, &fl)) { + if (ip_route_output_key(net, &rt, &fl)) { IP_VS_DBG_RL("ip_route_output error, dest: %pI4\n", - &cp->daddr.ip); + &daddr); return NULL; } } + local = rt->rt_flags & RTCF_LOCAL; + if (!((local ? 1 : 2) & rt_mode)) { + IP_VS_DBG_RL("Stopping traffic to %s address, dest: %pI4\n", + (rt->rt_flags & RTCF_LOCAL) ? + "local":"non-local", &rt->rt_dst); + ip_rt_put(rt); + return NULL; + } + if (local && !(rt_mode & 4) && !((ort = skb_rtable(skb)) && + ort->rt_flags & RTCF_LOCAL)) { + IP_VS_DBG_RL("Redirect from non-local address %pI4 to local " + "requires NAT method, dest: %pI4\n", + &ip_hdr(skb)->daddr, &rt->rt_dst); + ip_rt_put(rt); + return NULL; + } + if (unlikely(!local && ipv4_is_loopback(ip_hdr(skb)->saddr))) { + IP_VS_DBG_RL("Stopping traffic from loopback address %pI4 " + "to non-local address, dest: %pI4\n", + &ip_hdr(skb)->saddr, &rt->rt_dst); + ip_rt_put(rt); + return NULL; + } + return rt; } +/* Reroute packet to local IPv4 stack after DNAT */ +static int +__ip_vs_reroute_locally(struct sk_buff *skb) +{ + struct rtable *rt = skb_rtable(skb); + struct net_device *dev = rt->dst.dev; + struct net *net = dev_net(dev); + struct iphdr *iph = ip_hdr(skb); + + if (rt->fl.iif) { + unsigned long orefdst = skb->_skb_refdst; + + if (ip_route_input(skb, iph->daddr, iph->saddr, + iph->tos, skb->dev)) + return 0; + refdst_drop(orefdst); + } else { + struct flowi fl = { + .oif = 0, + .nl_u = { + .ip4_u = { + .daddr = iph->daddr, + .saddr = iph->saddr, + .tos = RT_TOS(iph->tos), + } + }, + .mark = skb->mark, + }; + struct rtable *rt; + + if (ip_route_output_key(net, &rt, &fl)) + return 0; + if (!(rt->rt_flags & RTCF_LOCAL)) { + ip_rt_put(rt); + return 0; + } + /* Drop old route. */ + skb_dst_drop(skb); + skb_dst_set(skb, &rt->dst); + } + return 1; +} + #ifdef CONFIG_IP_VS_IPV6 + +static inline int __ip_vs_is_local_route6(struct rt6_info *rt) +{ + return rt->rt6i_dev && rt->rt6i_dev->flags & IFF_LOOPBACK; +} + +static struct dst_entry * +__ip_vs_route_output_v6(struct net *net, struct in6_addr *daddr, + struct in6_addr *ret_saddr, int do_xfrm) +{ + struct dst_entry *dst; + struct flowi fl = { + .oif = 0, + .nl_u = { + .ip6_u = { + .daddr = *daddr, + }, + }, + }; + + dst = ip6_route_output(net, NULL, &fl); + if (dst->error) + goto out_err; + if (!ret_saddr) + return dst; + if (ipv6_addr_any(&fl.fl6_src) && + ipv6_dev_get_saddr(net, ip6_dst_idev(dst)->dev, + &fl.fl6_dst, 0, &fl.fl6_src) < 0) + goto out_err; + if (do_xfrm && xfrm_lookup(net, &dst, &fl, NULL, 0) < 0) + goto out_err; + ipv6_addr_copy(ret_saddr, &fl.fl6_src); + return dst; + +out_err: + dst_release(dst); + IP_VS_DBG_RL("ip6_route_output error, dest: %pI6\n", daddr); + return NULL; +} + +/* + * Get route to destination or remote server + * rt_mode: flags, &1=Allow local dest, &2=Allow non-local dest, + * &4=Allow redirect from remote daddr to local + */ static struct rt6_info * -__ip_vs_get_out_rt_v6(struct ip_vs_conn *cp) +__ip_vs_get_out_rt_v6(struct sk_buff *skb, struct ip_vs_dest *dest, + struct in6_addr *daddr, struct in6_addr *ret_saddr, + int do_xfrm, int rt_mode) { + struct net *net = dev_net(skb_dst(skb)->dev); struct rt6_info *rt; /* Route to the other host */ - struct ip_vs_dest *dest = cp->dest; + struct rt6_info *ort; /* Original route */ + struct dst_entry *dst; + int local; if (dest) { spin_lock(&dest->dst_lock); - rt = (struct rt6_info *)__ip_vs_dst_check(dest, 0, 0); + rt = (struct rt6_info *)__ip_vs_dst_check(dest, 0); if (!rt) { - struct flowi fl = { - .oif = 0, - .nl_u = { - .ip6_u = { - .daddr = dest->addr.in6, - .saddr = { - .s6_addr32 = - { 0, 0, 0, 0 }, - }, - }, - }, - }; + u32 cookie; - rt = (struct rt6_info *)ip6_route_output(&init_net, - NULL, &fl); - if (!rt) { + dst = __ip_vs_route_output_v6(net, &dest->addr.in6, + &dest->dst_saddr, + do_xfrm); + if (!dst) { spin_unlock(&dest->dst_lock); - IP_VS_DBG_RL("ip6_route_output error, dest: %pI6\n", - &dest->addr.in6); return NULL; } - __ip_vs_dst_set(dest, 0, dst_clone(&rt->dst)); - IP_VS_DBG(10, "new dst %pI6, refcnt=%d\n", - &dest->addr.in6, + rt = (struct rt6_info *) dst; + cookie = rt->rt6i_node ? rt->rt6i_node->fn_sernum : 0; + __ip_vs_dst_set(dest, 0, dst_clone(&rt->dst), cookie); + IP_VS_DBG(10, "new dst %pI6, src %pI6, refcnt=%d\n", + &dest->addr.in6, &dest->dst_saddr, atomic_read(&rt->dst.__refcnt)); } + if (ret_saddr) + ipv6_addr_copy(ret_saddr, &dest->dst_saddr); spin_unlock(&dest->dst_lock); } else { - struct flowi fl = { - .oif = 0, - .nl_u = { - .ip6_u = { - .daddr = cp->daddr.in6, - .saddr = { - .s6_addr32 = { 0, 0, 0, 0 }, - }, - }, - }, - }; - - rt = (struct rt6_info *)ip6_route_output(&init_net, NULL, &fl); - if (!rt) { - IP_VS_DBG_RL("ip6_route_output error, dest: %pI6\n", - &cp->daddr.in6); + dst = __ip_vs_route_output_v6(net, daddr, ret_saddr, do_xfrm); + if (!dst) return NULL; - } + rt = (struct rt6_info *) dst; + } + + local = __ip_vs_is_local_route6(rt); + if (!((local ? 1 : 2) & rt_mode)) { + IP_VS_DBG_RL("Stopping traffic to %s address, dest: %pI6\n", + local ? "local":"non-local", daddr); + dst_release(&rt->dst); + return NULL; + } + if (local && !(rt_mode & 4) && + !((ort = (struct rt6_info *) skb_dst(skb)) && + __ip_vs_is_local_route6(ort))) { + IP_VS_DBG_RL("Redirect from non-local address %pI6 to local " + "requires NAT method, dest: %pI6\n", + &ipv6_hdr(skb)->daddr, daddr); + dst_release(&rt->dst); + return NULL; + } + if (unlikely(!local && (!skb->dev || skb->dev->flags & IFF_LOOPBACK) && + ipv6_addr_type(&ipv6_hdr(skb)->saddr) & + IPV6_ADDR_LOOPBACK)) { + IP_VS_DBG_RL("Stopping traffic from loopback address %pI6 " + "to non-local address, dest: %pI6\n", + &ipv6_hdr(skb)->saddr, daddr); + dst_release(&rt->dst); + return NULL; } return rt; @@ -194,12 +334,44 @@ ip_vs_dst_reset(struct ip_vs_dest *dest) dst_release(old_dst); } -#define IP_VS_XMIT(pf, skb, rt) \ +#define IP_VS_XMIT_TUNNEL(skb, cp) \ +({ \ + int __ret = NF_ACCEPT; \ + \ + (skb)->ipvs_property = 1; \ + if (unlikely((cp)->flags & IP_VS_CONN_F_NFCT)) \ + __ret = ip_vs_confirm_conntrack(skb, cp); \ + if (__ret == NF_ACCEPT) { \ + nf_reset(skb); \ + skb_forward_csum(skb); \ + } \ + __ret; \ +}) + +#define IP_VS_XMIT_NAT(pf, skb, cp, local) \ +do { \ + (skb)->ipvs_property = 1; \ + if (likely(!((cp)->flags & IP_VS_CONN_F_NFCT))) \ + ip_vs_notrack(skb); \ + else \ + ip_vs_update_conntrack(skb, cp, 1); \ + if (local) \ + return NF_ACCEPT; \ + skb_forward_csum(skb); \ + NF_HOOK(pf, NF_INET_LOCAL_OUT, (skb), NULL, \ + skb_dst(skb)->dev, dst_output); \ +} while (0) + +#define IP_VS_XMIT(pf, skb, cp, local) \ do { \ (skb)->ipvs_property = 1; \ + if (likely(!((cp)->flags & IP_VS_CONN_F_NFCT))) \ + ip_vs_notrack(skb); \ + if (local) \ + return NF_ACCEPT; \ skb_forward_csum(skb); \ NF_HOOK(pf, NF_INET_LOCAL_OUT, (skb), NULL, \ - (rt)->dst.dev, dst_output); \ + skb_dst(skb)->dev, dst_output); \ } while (0) @@ -211,7 +383,7 @@ ip_vs_null_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, struct ip_vs_protocol *pp) { /* we do not touch skb and do not need pskb ptr */ - return NF_ACCEPT; + IP_VS_XMIT(NFPROTO_IPV4, skb, cp, 1); } @@ -226,24 +398,13 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, { struct rtable *rt; /* Route to the other host */ struct iphdr *iph = ip_hdr(skb); - u8 tos = iph->tos; int mtu; - struct flowi fl = { - .oif = 0, - .nl_u = { - .ip4_u = { - .daddr = iph->daddr, - .saddr = 0, - .tos = RT_TOS(tos), } }, - }; EnterFunction(10); - if (ip_route_output_key(&init_net, &rt, &fl)) { - IP_VS_DBG_RL("%s(): ip_route_output error, dest: %pI4\n", - __func__, &iph->daddr); + if (!(rt = __ip_vs_get_out_rt(skb, NULL, iph->daddr, + RT_TOS(iph->tos), 2))) goto tx_error_icmp; - } /* MTU checking */ mtu = dst_mtu(&rt->dst); @@ -271,7 +432,7 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; - IP_VS_XMIT(NFPROTO_IPV4, skb, rt); + IP_VS_XMIT(NFPROTO_IPV4, skb, cp, 0); LeaveFunction(10); return NF_STOLEN; @@ -292,28 +453,22 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, struct rt6_info *rt; /* Route to the other host */ struct ipv6hdr *iph = ipv6_hdr(skb); int mtu; - struct flowi fl = { - .oif = 0, - .nl_u = { - .ip6_u = { - .daddr = iph->daddr, - .saddr = { .s6_addr32 = {0, 0, 0, 0} }, } }, - }; EnterFunction(10); - rt = (struct rt6_info *)ip6_route_output(&init_net, NULL, &fl); - if (!rt) { - IP_VS_DBG_RL("%s(): ip6_route_output error, dest: %pI6\n", - __func__, &iph->daddr); + if (!(rt = __ip_vs_get_out_rt_v6(skb, NULL, &iph->daddr, NULL, 0, 2))) goto tx_error_icmp; - } /* MTU checking */ mtu = dst_mtu(&rt->dst); if (skb->len > mtu) { - dst_release(&rt->dst); + if (!skb->dev) { + struct net *net = dev_net(skb_dst(skb)->dev); + + skb->dev = net->loopback_dev; + } icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); + dst_release(&rt->dst); IP_VS_DBG_RL("%s(): frag needed\n", __func__); goto tx_error; } @@ -335,7 +490,7 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; - IP_VS_XMIT(NFPROTO_IPV6, skb, rt); + IP_VS_XMIT(NFPROTO_IPV6, skb, cp, 0); LeaveFunction(10); return NF_STOLEN; @@ -349,30 +504,6 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, } #endif -static void -ip_vs_update_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp) -{ - struct nf_conn *ct = (struct nf_conn *)skb->nfct; - struct nf_conntrack_tuple new_tuple; - - if (ct == NULL || nf_ct_is_untracked(ct) || nf_ct_is_confirmed(ct)) - return; - - /* - * The connection is not yet in the hashtable, so we update it. - * CIP->VIP will remain the same, so leave the tuple in - * IP_CT_DIR_ORIGINAL untouched. When the reply comes back from the - * real-server we will see RIP->DIP. - */ - new_tuple = ct->tuplehash[IP_CT_DIR_REPLY].tuple; - new_tuple.src.u3 = cp->daddr; - /* - * This will also take care of UDP and other protocols. - */ - new_tuple.src.u.tcp.port = cp->dport; - nf_conntrack_alter_reply(ct, &new_tuple); -} - /* * NAT transmitter (only for outside-to-inside nat forwarding) * Not used for related ICMP @@ -384,6 +515,7 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, struct rtable *rt; /* Route to the other host */ int mtu; struct iphdr *iph = ip_hdr(skb); + int local; EnterFunction(10); @@ -397,16 +529,42 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, IP_VS_DBG(10, "filled cport=%d\n", ntohs(*p)); } - if (!(rt = __ip_vs_get_out_rt(cp, RT_TOS(iph->tos)))) + if (!(rt = __ip_vs_get_out_rt(skb, cp->dest, cp->daddr.ip, + RT_TOS(iph->tos), 1|2|4))) goto tx_error_icmp; + local = rt->rt_flags & RTCF_LOCAL; + /* + * Avoid duplicate tuple in reply direction for NAT traffic + * to local address when connection is sync-ed + */ +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + if (cp->flags & IP_VS_CONN_F_SYNC && local) { + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = ct = nf_ct_get(skb, &ctinfo); + + if (ct && !nf_ct_is_untracked(ct)) { + IP_VS_DBG_RL_PKT(10, AF_INET, pp, skb, 0, + "ip_vs_nat_xmit(): " + "stopping DNAT to local address"); + goto tx_error_put; + } + } +#endif + + /* From world but DNAT to loopback address? */ + if (local && ipv4_is_loopback(rt->rt_dst) && skb_rtable(skb)->fl.iif) { + IP_VS_DBG_RL_PKT(1, AF_INET, pp, skb, 0, "ip_vs_nat_xmit(): " + "stopping DNAT to loopback address"); + goto tx_error_put; + } /* MTU checking */ mtu = dst_mtu(&rt->dst); if ((skb->len > mtu) && (iph->frag_off & htons(IP_DF))) { - ip_rt_put(rt); icmp_send(skb, ICMP_DEST_UNREACH,ICMP_FRAG_NEEDED, htonl(mtu)); - IP_VS_DBG_RL_PKT(0, pp, skb, 0, "ip_vs_nat_xmit(): frag needed for"); - goto tx_error; + IP_VS_DBG_RL_PKT(0, AF_INET, pp, skb, 0, + "ip_vs_nat_xmit(): frag needed for"); + goto tx_error_put; } /* copy-on-write the packet before mangling it */ @@ -416,19 +574,28 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, if (skb_cow(skb, rt->dst.dev->hard_header_len)) goto tx_error_put; - /* drop old route */ - skb_dst_drop(skb); - skb_dst_set(skb, &rt->dst); - /* mangle the packet */ if (pp->dnat_handler && !pp->dnat_handler(skb, pp, cp)) - goto tx_error; + goto tx_error_put; ip_hdr(skb)->daddr = cp->daddr.ip; ip_send_check(ip_hdr(skb)); - IP_VS_DBG_PKT(10, pp, skb, 0, "After DNAT"); + if (!local) { + /* drop old route */ + skb_dst_drop(skb); + skb_dst_set(skb, &rt->dst); + } else { + ip_rt_put(rt); + /* + * Some IPv4 replies get local address from routes, + * not from iph, so while we DNAT after routing + * we need this second input/output route. + */ + if (!__ip_vs_reroute_locally(skb)) + goto tx_error; + } - ip_vs_update_conntrack(skb, cp); + IP_VS_DBG_PKT(10, AF_INET, pp, skb, 0, "After DNAT"); /* FIXME: when application helper enlarges the packet and the length is larger than the MTU of outgoing device, there will be still @@ -437,7 +604,7 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; - IP_VS_XMIT(NFPROTO_IPV4, skb, rt); + IP_VS_XMIT_NAT(NFPROTO_IPV4, skb, cp, local); LeaveFunction(10); return NF_STOLEN; @@ -445,8 +612,8 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, tx_error_icmp: dst_link_failure(skb); tx_error: - LeaveFunction(10); kfree_skb(skb); + LeaveFunction(10); return NF_STOLEN; tx_error_put: ip_rt_put(rt); @@ -460,6 +627,7 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, { struct rt6_info *rt; /* Route to the other host */ int mtu; + int local; EnterFunction(10); @@ -474,18 +642,49 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, IP_VS_DBG(10, "filled cport=%d\n", ntohs(*p)); } - rt = __ip_vs_get_out_rt_v6(cp); - if (!rt) + if (!(rt = __ip_vs_get_out_rt_v6(skb, cp->dest, &cp->daddr.in6, NULL, + 0, 1|2|4))) goto tx_error_icmp; + local = __ip_vs_is_local_route6(rt); + /* + * Avoid duplicate tuple in reply direction for NAT traffic + * to local address when connection is sync-ed + */ +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + if (cp->flags & IP_VS_CONN_F_SYNC && local) { + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = ct = nf_ct_get(skb, &ctinfo); + + if (ct && !nf_ct_is_untracked(ct)) { + IP_VS_DBG_RL_PKT(10, AF_INET6, pp, skb, 0, + "ip_vs_nat_xmit_v6(): " + "stopping DNAT to local address"); + goto tx_error_put; + } + } +#endif + + /* From world but DNAT to loopback address? */ + if (local && skb->dev && !(skb->dev->flags & IFF_LOOPBACK) && + ipv6_addr_type(&rt->rt6i_dst.addr) & IPV6_ADDR_LOOPBACK) { + IP_VS_DBG_RL_PKT(1, AF_INET6, pp, skb, 0, + "ip_vs_nat_xmit_v6(): " + "stopping DNAT to loopback address"); + goto tx_error_put; + } /* MTU checking */ mtu = dst_mtu(&rt->dst); if (skb->len > mtu) { - dst_release(&rt->dst); + if (!skb->dev) { + struct net *net = dev_net(skb_dst(skb)->dev); + + skb->dev = net->loopback_dev; + } icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); - IP_VS_DBG_RL_PKT(0, pp, skb, 0, + IP_VS_DBG_RL_PKT(0, AF_INET6, pp, skb, 0, "ip_vs_nat_xmit_v6(): frag needed for"); - goto tx_error; + goto tx_error_put; } /* copy-on-write the packet before mangling it */ @@ -495,18 +694,21 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, if (skb_cow(skb, rt->dst.dev->hard_header_len)) goto tx_error_put; - /* drop old route */ - skb_dst_drop(skb); - skb_dst_set(skb, &rt->dst); - /* mangle the packet */ if (pp->dnat_handler && !pp->dnat_handler(skb, pp, cp)) goto tx_error; - ipv6_hdr(skb)->daddr = cp->daddr.in6; + ipv6_addr_copy(&ipv6_hdr(skb)->daddr, &cp->daddr.in6); - IP_VS_DBG_PKT(10, pp, skb, 0, "After DNAT"); + if (!local || !skb->dev) { + /* drop the old route when skb is not shared */ + skb_dst_drop(skb); + skb_dst_set(skb, &rt->dst); + } else { + /* destined to loopback, do we need to change route? */ + dst_release(&rt->dst); + } - ip_vs_update_conntrack(skb, cp); + IP_VS_DBG_PKT(10, AF_INET6, pp, skb, 0, "After DNAT"); /* FIXME: when application helper enlarges the packet and the length is larger than the MTU of outgoing device, there will be still @@ -515,7 +717,7 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; - IP_VS_XMIT(NFPROTO_IPV6, skb, rt); + IP_VS_XMIT_NAT(NFPROTO_IPV6, skb, cp, local); LeaveFunction(10); return NF_STOLEN; @@ -561,30 +763,27 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, struct iphdr *old_iph = ip_hdr(skb); u8 tos = old_iph->tos; __be16 df = old_iph->frag_off; - sk_buff_data_t old_transport_header = skb->transport_header; struct iphdr *iph; /* Our new IP header */ unsigned int max_headroom; /* The extra header space needed */ int mtu; + int ret; EnterFunction(10); - if (skb->protocol != htons(ETH_P_IP)) { - IP_VS_DBG_RL("%s(): protocol error, " - "ETH_P_IP: %d, skb protocol: %d\n", - __func__, htons(ETH_P_IP), skb->protocol); - goto tx_error; - } - - if (!(rt = __ip_vs_get_out_rt(cp, RT_TOS(tos)))) + if (!(rt = __ip_vs_get_out_rt(skb, cp->dest, cp->daddr.ip, + RT_TOS(tos), 1|2))) goto tx_error_icmp; + if (rt->rt_flags & RTCF_LOCAL) { + ip_rt_put(rt); + IP_VS_XMIT(NFPROTO_IPV4, skb, cp, 1); + } tdev = rt->dst.dev; mtu = dst_mtu(&rt->dst) - sizeof(struct iphdr); if (mtu < 68) { - ip_rt_put(rt); IP_VS_DBG_RL("%s(): mtu less than 68\n", __func__); - goto tx_error; + goto tx_error_put; } if (skb_dst(skb)) skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); @@ -594,9 +793,8 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, if ((old_iph->frag_off & htons(IP_DF)) && mtu < ntohs(old_iph->tot_len)) { icmp_send(skb, ICMP_DEST_UNREACH,ICMP_FRAG_NEEDED, htonl(mtu)); - ip_rt_put(rt); IP_VS_DBG_RL("%s(): frag needed\n", __func__); - goto tx_error; + goto tx_error_put; } /* @@ -619,7 +817,7 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, old_iph = ip_hdr(skb); } - skb->transport_header = old_transport_header; + skb->transport_header = skb->network_header; /* fix old IP header checksum */ ip_send_check(old_iph); @@ -649,7 +847,11 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; - ip_local_out(skb); + ret = IP_VS_XMIT_TUNNEL(skb, cp); + if (ret == NF_ACCEPT) + ip_local_out(skb); + else if (ret == NF_DROP) + kfree_skb(skb); LeaveFunction(10); @@ -661,6 +863,9 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, kfree_skb(skb); LeaveFunction(10); return NF_STOLEN; +tx_error_put: + ip_rt_put(rt); + goto tx_error; } #ifdef CONFIG_IP_VS_IPV6 @@ -669,43 +874,44 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, struct ip_vs_protocol *pp) { struct rt6_info *rt; /* Route to the other host */ + struct in6_addr saddr; /* Source for tunnel */ struct net_device *tdev; /* Device to other host */ struct ipv6hdr *old_iph = ipv6_hdr(skb); - sk_buff_data_t old_transport_header = skb->transport_header; struct ipv6hdr *iph; /* Our new IP header */ unsigned int max_headroom; /* The extra header space needed */ int mtu; + int ret; EnterFunction(10); - if (skb->protocol != htons(ETH_P_IPV6)) { - IP_VS_DBG_RL("%s(): protocol error, " - "ETH_P_IPV6: %d, skb protocol: %d\n", - __func__, htons(ETH_P_IPV6), skb->protocol); - goto tx_error; - } - - rt = __ip_vs_get_out_rt_v6(cp); - if (!rt) + if (!(rt = __ip_vs_get_out_rt_v6(skb, cp->dest, &cp->daddr.in6, + &saddr, 1, 1|2))) goto tx_error_icmp; + if (__ip_vs_is_local_route6(rt)) { + dst_release(&rt->dst); + IP_VS_XMIT(NFPROTO_IPV6, skb, cp, 1); + } tdev = rt->dst.dev; mtu = dst_mtu(&rt->dst) - sizeof(struct ipv6hdr); - /* TODO IPv6: do we need this check in IPv6? */ - if (mtu < 1280) { - dst_release(&rt->dst); - IP_VS_DBG_RL("%s(): mtu less than 1280\n", __func__); - goto tx_error; + if (mtu < IPV6_MIN_MTU) { + IP_VS_DBG_RL("%s(): mtu less than %d\n", __func__, + IPV6_MIN_MTU); + goto tx_error_put; } if (skb_dst(skb)) skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); if (mtu < ntohs(old_iph->payload_len) + sizeof(struct ipv6hdr)) { + if (!skb->dev) { + struct net *net = dev_net(skb_dst(skb)->dev); + + skb->dev = net->loopback_dev; + } icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); - dst_release(&rt->dst); IP_VS_DBG_RL("%s(): frag needed\n", __func__); - goto tx_error; + goto tx_error_put; } /* @@ -728,7 +934,7 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, old_iph = ipv6_hdr(skb); } - skb->transport_header = old_transport_header; + skb->transport_header = skb->network_header; skb_push(skb, sizeof(struct ipv6hdr)); skb_reset_network_header(skb); @@ -748,14 +954,18 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, be16_add_cpu(&iph->payload_len, sizeof(*old_iph)); iph->priority = old_iph->priority; memset(&iph->flow_lbl, 0, sizeof(iph->flow_lbl)); - iph->daddr = rt->rt6i_dst.addr; - iph->saddr = cp->vaddr.in6; /* rt->rt6i_src.addr; */ + ipv6_addr_copy(&iph->daddr, &cp->daddr.in6); + ipv6_addr_copy(&iph->saddr, &saddr); iph->hop_limit = old_iph->hop_limit; /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; - ip6_local_out(skb); + ret = IP_VS_XMIT_TUNNEL(skb, cp); + if (ret == NF_ACCEPT) + ip6_local_out(skb); + else if (ret == NF_DROP) + kfree_skb(skb); LeaveFunction(10); @@ -767,6 +977,9 @@ tx_error: kfree_skb(skb); LeaveFunction(10); return NF_STOLEN; +tx_error_put: + dst_release(&rt->dst); + goto tx_error; } #endif @@ -785,8 +998,13 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, EnterFunction(10); - if (!(rt = __ip_vs_get_out_rt(cp, RT_TOS(iph->tos)))) + if (!(rt = __ip_vs_get_out_rt(skb, cp->dest, cp->daddr.ip, + RT_TOS(iph->tos), 1|2))) goto tx_error_icmp; + if (rt->rt_flags & RTCF_LOCAL) { + ip_rt_put(rt); + IP_VS_XMIT(NFPROTO_IPV4, skb, cp, 1); + } /* MTU checking */ mtu = dst_mtu(&rt->dst); @@ -814,7 +1032,7 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; - IP_VS_XMIT(NFPROTO_IPV4, skb, rt); + IP_VS_XMIT(NFPROTO_IPV4, skb, cp, 0); LeaveFunction(10); return NF_STOLEN; @@ -837,13 +1055,22 @@ ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, EnterFunction(10); - rt = __ip_vs_get_out_rt_v6(cp); - if (!rt) + if (!(rt = __ip_vs_get_out_rt_v6(skb, cp->dest, &cp->daddr.in6, NULL, + 0, 1|2))) goto tx_error_icmp; + if (__ip_vs_is_local_route6(rt)) { + dst_release(&rt->dst); + IP_VS_XMIT(NFPROTO_IPV6, skb, cp, 1); + } /* MTU checking */ mtu = dst_mtu(&rt->dst); if (skb->len > mtu) { + if (!skb->dev) { + struct net *net = dev_net(skb_dst(skb)->dev); + + skb->dev = net->loopback_dev; + } icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); dst_release(&rt->dst); IP_VS_DBG_RL("%s(): frag needed\n", __func__); @@ -867,7 +1094,7 @@ ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; - IP_VS_XMIT(NFPROTO_IPV6, skb, rt); + IP_VS_XMIT(NFPROTO_IPV6, skb, cp, 0); LeaveFunction(10); return NF_STOLEN; @@ -893,6 +1120,7 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, struct rtable *rt; /* Route to the other host */ int mtu; int rc; + int local; EnterFunction(10); @@ -913,16 +1141,43 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, * mangle and send the packet here (only for VS/NAT) */ - if (!(rt = __ip_vs_get_out_rt(cp, RT_TOS(ip_hdr(skb)->tos)))) + if (!(rt = __ip_vs_get_out_rt(skb, cp->dest, cp->daddr.ip, + RT_TOS(ip_hdr(skb)->tos), 1|2|4))) goto tx_error_icmp; + local = rt->rt_flags & RTCF_LOCAL; + + /* + * Avoid duplicate tuple in reply direction for NAT traffic + * to local address when connection is sync-ed + */ +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + if (cp->flags & IP_VS_CONN_F_SYNC && local) { + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = ct = nf_ct_get(skb, &ctinfo); + + if (ct && !nf_ct_is_untracked(ct)) { + IP_VS_DBG(10, "%s(): " + "stopping DNAT to local address %pI4\n", + __func__, &cp->daddr.ip); + goto tx_error_put; + } + } +#endif + + /* From world but DNAT to loopback address? */ + if (local && ipv4_is_loopback(rt->rt_dst) && skb_rtable(skb)->fl.iif) { + IP_VS_DBG(1, "%s(): " + "stopping DNAT to loopback %pI4\n", + __func__, &cp->daddr.ip); + goto tx_error_put; + } /* MTU checking */ mtu = dst_mtu(&rt->dst); if ((skb->len > mtu) && (ip_hdr(skb)->frag_off & htons(IP_DF))) { - ip_rt_put(rt); icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); IP_VS_DBG_RL("%s(): frag needed\n", __func__); - goto tx_error; + goto tx_error_put; } /* copy-on-write the packet before mangling it */ @@ -932,16 +1187,27 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, if (skb_cow(skb, rt->dst.dev->hard_header_len)) goto tx_error_put; - /* drop the old route when skb is not shared */ - skb_dst_drop(skb); - skb_dst_set(skb, &rt->dst); - ip_vs_nat_icmp(skb, pp, cp, 0); + if (!local) { + /* drop the old route when skb is not shared */ + skb_dst_drop(skb); + skb_dst_set(skb, &rt->dst); + } else { + ip_rt_put(rt); + /* + * Some IPv4 replies get local address from routes, + * not from iph, so while we DNAT after routing + * we need this second input/output route. + */ + if (!__ip_vs_reroute_locally(skb)) + goto tx_error; + } + /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; - IP_VS_XMIT(NFPROTO_IPV4, skb, rt); + IP_VS_XMIT_NAT(NFPROTO_IPV4, skb, cp, local); rc = NF_STOLEN; goto out; @@ -967,6 +1233,7 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, struct rt6_info *rt; /* Route to the other host */ int mtu; int rc; + int local; EnterFunction(10); @@ -987,17 +1254,49 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, * mangle and send the packet here (only for VS/NAT) */ - rt = __ip_vs_get_out_rt_v6(cp); - if (!rt) + if (!(rt = __ip_vs_get_out_rt_v6(skb, cp->dest, &cp->daddr.in6, NULL, + 0, 1|2|4))) goto tx_error_icmp; + local = __ip_vs_is_local_route6(rt); + /* + * Avoid duplicate tuple in reply direction for NAT traffic + * to local address when connection is sync-ed + */ +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + if (cp->flags & IP_VS_CONN_F_SYNC && local) { + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = ct = nf_ct_get(skb, &ctinfo); + + if (ct && !nf_ct_is_untracked(ct)) { + IP_VS_DBG(10, "%s(): " + "stopping DNAT to local address %pI6\n", + __func__, &cp->daddr.in6); + goto tx_error_put; + } + } +#endif + + /* From world but DNAT to loopback address? */ + if (local && skb->dev && !(skb->dev->flags & IFF_LOOPBACK) && + ipv6_addr_type(&rt->rt6i_dst.addr) & IPV6_ADDR_LOOPBACK) { + IP_VS_DBG(1, "%s(): " + "stopping DNAT to loopback %pI6\n", + __func__, &cp->daddr.in6); + goto tx_error_put; + } + /* MTU checking */ mtu = dst_mtu(&rt->dst); if (skb->len > mtu) { - dst_release(&rt->dst); + if (!skb->dev) { + struct net *net = dev_net(skb_dst(skb)->dev); + + skb->dev = net->loopback_dev; + } icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); IP_VS_DBG_RL("%s(): frag needed\n", __func__); - goto tx_error; + goto tx_error_put; } /* copy-on-write the packet before mangling it */ @@ -1007,16 +1306,21 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, if (skb_cow(skb, rt->dst.dev->hard_header_len)) goto tx_error_put; - /* drop the old route when skb is not shared */ - skb_dst_drop(skb); - skb_dst_set(skb, &rt->dst); - ip_vs_nat_icmp_v6(skb, pp, cp, 0); + if (!local || !skb->dev) { + /* drop the old route when skb is not shared */ + skb_dst_drop(skb); + skb_dst_set(skb, &rt->dst); + } else { + /* destined to loopback, do we need to change route? */ + dst_release(&rt->dst); + } + /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; - IP_VS_XMIT(NFPROTO_IPV6, skb, rt); + IP_VS_XMIT_NAT(NFPROTO_IPV6, skb, cp, local); rc = NF_STOLEN; goto out; diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index df3eedb142f..1eacf8d9966 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -65,32 +65,42 @@ EXPORT_SYMBOL_GPL(nf_conntrack_max); DEFINE_PER_CPU(struct nf_conn, nf_conntrack_untracked); EXPORT_PER_CPU_SYMBOL(nf_conntrack_untracked); -static int nf_conntrack_hash_rnd_initted; -static unsigned int nf_conntrack_hash_rnd; +static unsigned int nf_conntrack_hash_rnd __read_mostly; -static u_int32_t __hash_conntrack(const struct nf_conntrack_tuple *tuple, - u16 zone, unsigned int size, unsigned int rnd) +static u32 hash_conntrack_raw(const struct nf_conntrack_tuple *tuple, u16 zone) { unsigned int n; - u_int32_t h; /* The direction must be ignored, so we hash everything up to the * destination ports (which is a multiple of 4) and treat the last * three bytes manually. */ n = (sizeof(tuple->src) + sizeof(tuple->dst.u3)) / sizeof(u32); - h = jhash2((u32 *)tuple, n, - zone ^ rnd ^ (((__force __u16)tuple->dst.u.all << 16) | - tuple->dst.protonum)); + return jhash2((u32 *)tuple, n, zone ^ nf_conntrack_hash_rnd ^ + (((__force __u16)tuple->dst.u.all << 16) | + tuple->dst.protonum)); +} + +static u32 __hash_bucket(u32 hash, unsigned int size) +{ + return ((u64)hash * size) >> 32; +} + +static u32 hash_bucket(u32 hash, const struct net *net) +{ + return __hash_bucket(hash, net->ct.htable_size); +} - return ((u64)h * size) >> 32; +static u_int32_t __hash_conntrack(const struct nf_conntrack_tuple *tuple, + u16 zone, unsigned int size) +{ + return __hash_bucket(hash_conntrack_raw(tuple, zone), size); } static inline u_int32_t hash_conntrack(const struct net *net, u16 zone, const struct nf_conntrack_tuple *tuple) { - return __hash_conntrack(tuple, zone, net->ct.htable_size, - nf_conntrack_hash_rnd); + return __hash_conntrack(tuple, zone, net->ct.htable_size); } bool @@ -292,20 +302,20 @@ static void death_by_timeout(unsigned long ul_conntrack) * OR * - Caller must lock nf_conntrack_lock before calling this function */ -struct nf_conntrack_tuple_hash * -__nf_conntrack_find(struct net *net, u16 zone, - const struct nf_conntrack_tuple *tuple) +static struct nf_conntrack_tuple_hash * +____nf_conntrack_find(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple, u32 hash) { struct nf_conntrack_tuple_hash *h; struct hlist_nulls_node *n; - unsigned int hash = hash_conntrack(net, zone, tuple); + unsigned int bucket = hash_bucket(hash, net); /* Disable BHs the entire time since we normally need to disable them * at least once for the stats anyway. */ local_bh_disable(); begin: - hlist_nulls_for_each_entry_rcu(h, n, &net->ct.hash[hash], hnnode) { + hlist_nulls_for_each_entry_rcu(h, n, &net->ct.hash[bucket], hnnode) { if (nf_ct_tuple_equal(tuple, &h->tuple) && nf_ct_zone(nf_ct_tuplehash_to_ctrack(h)) == zone) { NF_CT_STAT_INC(net, found); @@ -319,7 +329,7 @@ begin: * not the expected one, we must restart lookup. * We probably met an item that was moved to another chain. */ - if (get_nulls_value(n) != hash) { + if (get_nulls_value(n) != bucket) { NF_CT_STAT_INC(net, search_restart); goto begin; } @@ -327,19 +337,27 @@ begin: return NULL; } + +struct nf_conntrack_tuple_hash * +__nf_conntrack_find(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple) +{ + return ____nf_conntrack_find(net, zone, tuple, + hash_conntrack_raw(tuple, zone)); +} EXPORT_SYMBOL_GPL(__nf_conntrack_find); /* Find a connection corresponding to a tuple. */ -struct nf_conntrack_tuple_hash * -nf_conntrack_find_get(struct net *net, u16 zone, - const struct nf_conntrack_tuple *tuple) +static struct nf_conntrack_tuple_hash * +__nf_conntrack_find_get(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple, u32 hash) { struct nf_conntrack_tuple_hash *h; struct nf_conn *ct; rcu_read_lock(); begin: - h = __nf_conntrack_find(net, zone, tuple); + h = ____nf_conntrack_find(net, zone, tuple, hash); if (h) { ct = nf_ct_tuplehash_to_ctrack(h); if (unlikely(nf_ct_is_dying(ct) || @@ -357,6 +375,14 @@ begin: return h; } + +struct nf_conntrack_tuple_hash * +nf_conntrack_find_get(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple) +{ + return __nf_conntrack_find_get(net, zone, tuple, + hash_conntrack_raw(tuple, zone)); +} EXPORT_SYMBOL_GPL(nf_conntrack_find_get); static void __nf_conntrack_hash_insert(struct nf_conn *ct, @@ -409,8 +435,11 @@ __nf_conntrack_confirm(struct sk_buff *skb) return NF_ACCEPT; zone = nf_ct_zone(ct); - hash = hash_conntrack(net, zone, &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); - repl_hash = hash_conntrack(net, zone, &ct->tuplehash[IP_CT_DIR_REPLY].tuple); + /* reuse the hash saved before */ + hash = *(unsigned long *)&ct->tuplehash[IP_CT_DIR_REPLY].hnnode.pprev; + hash = hash_bucket(hash, net); + repl_hash = hash_conntrack(net, zone, + &ct->tuplehash[IP_CT_DIR_REPLY].tuple); /* We're not in hash table, and we refuse to set up related connections for unconfirmed conns. But packet copies and @@ -567,17 +596,29 @@ static noinline int early_drop(struct net *net, unsigned int hash) return dropped; } -struct nf_conn *nf_conntrack_alloc(struct net *net, u16 zone, - const struct nf_conntrack_tuple *orig, - const struct nf_conntrack_tuple *repl, - gfp_t gfp) +static struct nf_conn * +__nf_conntrack_alloc(struct net *net, u16 zone, + const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_tuple *repl, + gfp_t gfp, u32 hash) { struct nf_conn *ct; - if (unlikely(!nf_conntrack_hash_rnd_initted)) { - get_random_bytes(&nf_conntrack_hash_rnd, - sizeof(nf_conntrack_hash_rnd)); - nf_conntrack_hash_rnd_initted = 1; + if (unlikely(!nf_conntrack_hash_rnd)) { + unsigned int rand; + + /* + * Why not initialize nf_conntrack_rnd in a "init()" function ? + * Because there isn't enough entropy when system initializing, + * and we initialize it as late as possible. + */ + do { + get_random_bytes(&rand, sizeof(rand)); + } while (!rand); + cmpxchg(&nf_conntrack_hash_rnd, 0, rand); + + /* recompute the hash as nf_conntrack_hash_rnd is initialized */ + hash = hash_conntrack_raw(orig, zone); } /* We don't want any race condition at early drop stage */ @@ -585,8 +626,7 @@ struct nf_conn *nf_conntrack_alloc(struct net *net, u16 zone, if (nf_conntrack_max && unlikely(atomic_read(&net->ct.count) > nf_conntrack_max)) { - unsigned int hash = hash_conntrack(net, zone, orig); - if (!early_drop(net, hash)) { + if (!early_drop(net, hash_bucket(hash, net))) { atomic_dec(&net->ct.count); if (net_ratelimit()) printk(KERN_WARNING @@ -616,7 +656,8 @@ struct nf_conn *nf_conntrack_alloc(struct net *net, u16 zone, ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple = *orig; ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode.pprev = NULL; ct->tuplehash[IP_CT_DIR_REPLY].tuple = *repl; - ct->tuplehash[IP_CT_DIR_REPLY].hnnode.pprev = NULL; + /* save hash for reusing when confirming */ + *(unsigned long *)(&ct->tuplehash[IP_CT_DIR_REPLY].hnnode.pprev) = hash; /* Don't set timer yet: wait for confirmation */ setup_timer(&ct->timeout, death_by_timeout, (unsigned long)ct); write_pnet(&ct->ct_net, net); @@ -643,6 +684,14 @@ out_free: return ERR_PTR(-ENOMEM); #endif } + +struct nf_conn *nf_conntrack_alloc(struct net *net, u16 zone, + const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_tuple *repl, + gfp_t gfp) +{ + return __nf_conntrack_alloc(net, zone, orig, repl, gfp, 0); +} EXPORT_SYMBOL_GPL(nf_conntrack_alloc); void nf_conntrack_free(struct nf_conn *ct) @@ -664,7 +713,7 @@ init_conntrack(struct net *net, struct nf_conn *tmpl, struct nf_conntrack_l3proto *l3proto, struct nf_conntrack_l4proto *l4proto, struct sk_buff *skb, - unsigned int dataoff) + unsigned int dataoff, u32 hash) { struct nf_conn *ct; struct nf_conn_help *help; @@ -678,7 +727,8 @@ init_conntrack(struct net *net, struct nf_conn *tmpl, return NULL; } - ct = nf_conntrack_alloc(net, zone, tuple, &repl_tuple, GFP_ATOMIC); + ct = __nf_conntrack_alloc(net, zone, tuple, &repl_tuple, GFP_ATOMIC, + hash); if (IS_ERR(ct)) { pr_debug("Can't allocate conntrack.\n"); return (struct nf_conntrack_tuple_hash *)ct; @@ -755,6 +805,7 @@ resolve_normal_ct(struct net *net, struct nf_conn *tmpl, struct nf_conntrack_tuple_hash *h; struct nf_conn *ct; u16 zone = tmpl ? nf_ct_zone(tmpl) : NF_CT_DEFAULT_ZONE; + u32 hash; if (!nf_ct_get_tuple(skb, skb_network_offset(skb), dataoff, l3num, protonum, &tuple, l3proto, @@ -764,10 +815,11 @@ resolve_normal_ct(struct net *net, struct nf_conn *tmpl, } /* look for tuple match */ - h = nf_conntrack_find_get(net, zone, &tuple); + hash = hash_conntrack_raw(&tuple, zone); + h = __nf_conntrack_find_get(net, zone, &tuple, hash); if (!h) { h = init_conntrack(net, tmpl, &tuple, l3proto, l4proto, - skb, dataoff); + skb, dataoff, hash); if (!h) return NULL; if (IS_ERR(h)) @@ -1307,8 +1359,7 @@ int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) ct = nf_ct_tuplehash_to_ctrack(h); hlist_nulls_del_rcu(&h->hnnode); bucket = __hash_conntrack(&h->tuple, nf_ct_zone(ct), - hashsize, - nf_conntrack_hash_rnd); + hashsize); hlist_nulls_add_head_rcu(&h->hnnode, &hash[bucket]); } } diff --git a/net/netfilter/nf_conntrack_ecache.c b/net/netfilter/nf_conntrack_ecache.c index cdcc7649476..5702de35e2b 100644 --- a/net/netfilter/nf_conntrack_ecache.c +++ b/net/netfilter/nf_conntrack_ecache.c @@ -26,10 +26,10 @@ static DEFINE_MUTEX(nf_ct_ecache_mutex); -struct nf_ct_event_notifier *nf_conntrack_event_cb __read_mostly; +struct nf_ct_event_notifier __rcu *nf_conntrack_event_cb __read_mostly; EXPORT_SYMBOL_GPL(nf_conntrack_event_cb); -struct nf_exp_event_notifier *nf_expect_event_cb __read_mostly; +struct nf_exp_event_notifier __rcu *nf_expect_event_cb __read_mostly; EXPORT_SYMBOL_GPL(nf_expect_event_cb); /* deliver cached events and clear cache entry - must be called with locally diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index acb29ccaa41..46e8966912b 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -38,25 +38,30 @@ static int nf_ct_expect_hash_rnd_initted __read_mostly; static struct kmem_cache *nf_ct_expect_cachep __read_mostly; +static HLIST_HEAD(nf_ct_userspace_expect_list); + /* nf_conntrack_expect helper functions */ -void nf_ct_unlink_expect(struct nf_conntrack_expect *exp) +void nf_ct_unlink_expect_report(struct nf_conntrack_expect *exp, + u32 pid, int report) { struct nf_conn_help *master_help = nfct_help(exp->master); struct net *net = nf_ct_exp_net(exp); - NF_CT_ASSERT(master_help); NF_CT_ASSERT(!timer_pending(&exp->timeout)); hlist_del_rcu(&exp->hnode); net->ct.expect_count--; hlist_del(&exp->lnode); - master_help->expecting[exp->class]--; + if (!(exp->flags & NF_CT_EXPECT_USERSPACE)) + master_help->expecting[exp->class]--; + + nf_ct_expect_event_report(IPEXP_DESTROY, exp, pid, report); nf_ct_expect_put(exp); NF_CT_STAT_INC(net, expect_delete); } -EXPORT_SYMBOL_GPL(nf_ct_unlink_expect); +EXPORT_SYMBOL_GPL(nf_ct_unlink_expect_report); static void nf_ct_expectation_timed_out(unsigned long ul_expect) { @@ -320,16 +325,21 @@ static void nf_ct_expect_insert(struct nf_conntrack_expect *exp) atomic_inc(&exp->use); - hlist_add_head(&exp->lnode, &master_help->expectations); - master_help->expecting[exp->class]++; + if (master_help) { + hlist_add_head(&exp->lnode, &master_help->expectations); + master_help->expecting[exp->class]++; + } else if (exp->flags & NF_CT_EXPECT_USERSPACE) + hlist_add_head(&exp->lnode, &nf_ct_userspace_expect_list); hlist_add_head_rcu(&exp->hnode, &net->ct.expect_hash[h]); net->ct.expect_count++; setup_timer(&exp->timeout, nf_ct_expectation_timed_out, (unsigned long)exp); - p = &master_help->helper->expect_policy[exp->class]; - exp->timeout.expires = jiffies + p->timeout * HZ; + if (master_help) { + p = &master_help->helper->expect_policy[exp->class]; + exp->timeout.expires = jiffies + p->timeout * HZ; + } add_timer(&exp->timeout); atomic_inc(&exp->use); @@ -380,7 +390,9 @@ static inline int __nf_ct_expect_check(struct nf_conntrack_expect *expect) unsigned int h; int ret = 1; - if (!master_help->helper) { + /* Don't allow expectations created from kernel-space with no helper */ + if (!(expect->flags & NF_CT_EXPECT_USERSPACE) && + (!master_help || (master_help && !master_help->helper))) { ret = -ESHUTDOWN; goto out; } @@ -398,13 +410,16 @@ static inline int __nf_ct_expect_check(struct nf_conntrack_expect *expect) } } /* Will be over limit? */ - p = &master_help->helper->expect_policy[expect->class]; - if (p->max_expected && - master_help->expecting[expect->class] >= p->max_expected) { - evict_oldest_expect(master, expect); - if (master_help->expecting[expect->class] >= p->max_expected) { - ret = -EMFILE; - goto out; + if (master_help) { + p = &master_help->helper->expect_policy[expect->class]; + if (p->max_expected && + master_help->expecting[expect->class] >= p->max_expected) { + evict_oldest_expect(master, expect); + if (master_help->expecting[expect->class] + >= p->max_expected) { + ret = -EMFILE; + goto out; + } } } @@ -439,6 +454,21 @@ out: } EXPORT_SYMBOL_GPL(nf_ct_expect_related_report); +void nf_ct_remove_userspace_expectations(void) +{ + struct nf_conntrack_expect *exp; + struct hlist_node *n, *next; + + hlist_for_each_entry_safe(exp, n, next, + &nf_ct_userspace_expect_list, lnode) { + if (del_timer(&exp->timeout)) { + nf_ct_unlink_expect(exp); + nf_ct_expect_put(exp); + } + } +} +EXPORT_SYMBOL_GPL(nf_ct_remove_userspace_expectations); + #ifdef CONFIG_PROC_FS struct ct_expect_iter_state { struct seq_net_private p; @@ -529,8 +559,12 @@ static int exp_seq_show(struct seq_file *s, void *v) seq_printf(s, "PERMANENT"); delim = ","; } - if (expect->flags & NF_CT_EXPECT_INACTIVE) + if (expect->flags & NF_CT_EXPECT_INACTIVE) { seq_printf(s, "%sINACTIVE", delim); + delim = ","; + } + if (expect->flags & NF_CT_EXPECT_USERSPACE) + seq_printf(s, "%sUSERSPACE", delim); helper = rcu_dereference(nfct_help(expect->master)->helper); if (helper) { diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c index 7dcf7a40419..bd82450c193 100644 --- a/net/netfilter/nf_conntrack_extend.c +++ b/net/netfilter/nf_conntrack_extend.c @@ -16,7 +16,7 @@ #include <linux/skbuff.h> #include <net/netfilter/nf_conntrack_extend.h> -static struct nf_ct_ext_type *nf_ct_ext_types[NF_CT_EXT_NUM]; +static struct nf_ct_ext_type __rcu *nf_ct_ext_types[NF_CT_EXT_NUM]; static DEFINE_MUTEX(nf_ct_ext_type_mutex); void __nf_ct_ext_destroy(struct nf_conn *ct) @@ -48,15 +48,17 @@ nf_ct_ext_create(struct nf_ct_ext **ext, enum nf_ct_ext_id id, gfp_t gfp) { unsigned int off, len; struct nf_ct_ext_type *t; + size_t alloc_size; rcu_read_lock(); t = rcu_dereference(nf_ct_ext_types[id]); BUG_ON(t == NULL); off = ALIGN(sizeof(struct nf_ct_ext), t->align); len = off + t->len; + alloc_size = t->alloc_size; rcu_read_unlock(); - *ext = kzalloc(t->alloc_size, gfp); + *ext = kzalloc(alloc_size, gfp); if (!*ext) return NULL; diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 5bae1cd15ee..b729ace1dcc 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -22,6 +22,7 @@ #include <linux/rculist_nulls.h> #include <linux/types.h> #include <linux/timer.h> +#include <linux/security.h> #include <linux/skbuff.h> #include <linux/errno.h> #include <linux/netlink.h> @@ -245,16 +246,31 @@ nla_put_failure: #ifdef CONFIG_NF_CONNTRACK_SECMARK static inline int -ctnetlink_dump_secmark(struct sk_buff *skb, const struct nf_conn *ct) +ctnetlink_dump_secctx(struct sk_buff *skb, const struct nf_conn *ct) { - NLA_PUT_BE32(skb, CTA_SECMARK, htonl(ct->secmark)); - return 0; + struct nlattr *nest_secctx; + int len, ret; + char *secctx; + + ret = security_secid_to_secctx(ct->secmark, &secctx, &len); + if (ret) + return ret; + + ret = -1; + nest_secctx = nla_nest_start(skb, CTA_SECCTX | NLA_F_NESTED); + if (!nest_secctx) + goto nla_put_failure; + NLA_PUT_STRING(skb, CTA_SECCTX_NAME, secctx); + nla_nest_end(skb, nest_secctx); + + ret = 0; nla_put_failure: - return -1; + security_release_secctx(secctx, len); + return ret; } #else -#define ctnetlink_dump_secmark(a, b) (0) +#define ctnetlink_dump_secctx(a, b) (0) #endif #define master_tuple(ct) &(ct->master->tuplehash[IP_CT_DIR_ORIGINAL].tuple) @@ -391,7 +407,7 @@ ctnetlink_fill_info(struct sk_buff *skb, u32 pid, u32 seq, ctnetlink_dump_protoinfo(skb, ct) < 0 || ctnetlink_dump_helpinfo(skb, ct) < 0 || ctnetlink_dump_mark(skb, ct) < 0 || - ctnetlink_dump_secmark(skb, ct) < 0 || + ctnetlink_dump_secctx(skb, ct) < 0 || ctnetlink_dump_id(skb, ct) < 0 || ctnetlink_dump_use(skb, ct) < 0 || ctnetlink_dump_master(skb, ct) < 0 || @@ -437,6 +453,17 @@ ctnetlink_counters_size(const struct nf_conn *ct) ; } +#ifdef CONFIG_NF_CONNTRACK_SECMARK +static int ctnetlink_nlmsg_secctx_size(const struct nf_conn *ct) +{ + int len; + + security_secid_to_secctx(ct->secmark, NULL, &len); + + return sizeof(char) * len; +} +#endif + static inline size_t ctnetlink_nlmsg_size(const struct nf_conn *ct) { @@ -453,7 +480,8 @@ ctnetlink_nlmsg_size(const struct nf_conn *ct) + nla_total_size(0) /* CTA_HELP */ + nla_total_size(NF_CT_HELPER_NAME_LEN) /* CTA_HELP_NAME */ #ifdef CONFIG_NF_CONNTRACK_SECMARK - + nla_total_size(sizeof(u_int32_t)) /* CTA_SECMARK */ + + nla_total_size(0) /* CTA_SECCTX */ + + nla_total_size(ctnetlink_nlmsg_secctx_size(ct)) /* CTA_SECCTX_NAME */ #endif #ifdef CONFIG_NF_NAT_NEEDED + 2 * nla_total_size(0) /* CTA_NAT_SEQ_ADJ_ORIG|REPL */ @@ -556,7 +584,7 @@ ctnetlink_conntrack_event(unsigned int events, struct nf_ct_event *item) #ifdef CONFIG_NF_CONNTRACK_SECMARK if ((events & (1 << IPCT_SECMARK) || ct->secmark) - && ctnetlink_dump_secmark(skb, ct) < 0) + && ctnetlink_dump_secctx(skb, ct) < 0) goto nla_put_failure; #endif @@ -1560,8 +1588,8 @@ ctnetlink_exp_dump_expect(struct sk_buff *skb, const struct nf_conntrack_expect *exp) { struct nf_conn *master = exp->master; - struct nf_conntrack_helper *helper; long timeout = (exp->timeout.expires - jiffies) / HZ; + struct nf_conn_help *help; if (timeout < 0) timeout = 0; @@ -1577,9 +1605,15 @@ ctnetlink_exp_dump_expect(struct sk_buff *skb, NLA_PUT_BE32(skb, CTA_EXPECT_TIMEOUT, htonl(timeout)); NLA_PUT_BE32(skb, CTA_EXPECT_ID, htonl((unsigned long)exp)); - helper = rcu_dereference(nfct_help(master)->helper); - if (helper) - NLA_PUT_STRING(skb, CTA_EXPECT_HELP_NAME, helper->name); + NLA_PUT_BE32(skb, CTA_EXPECT_FLAGS, htonl(exp->flags)); + help = nfct_help(master); + if (help) { + struct nf_conntrack_helper *helper; + + helper = rcu_dereference(help->helper); + if (helper) + NLA_PUT_STRING(skb, CTA_EXPECT_HELP_NAME, helper->name); + } return 0; @@ -1626,17 +1660,20 @@ ctnetlink_expect_event(unsigned int events, struct nf_exp_event *item) struct nlmsghdr *nlh; struct nfgenmsg *nfmsg; struct sk_buff *skb; - unsigned int type; + unsigned int type, group; int flags = 0; - if (events & (1 << IPEXP_NEW)) { + if (events & (1 << IPEXP_DESTROY)) { + type = IPCTNL_MSG_EXP_DELETE; + group = NFNLGRP_CONNTRACK_EXP_DESTROY; + } else if (events & (1 << IPEXP_NEW)) { type = IPCTNL_MSG_EXP_NEW; flags = NLM_F_CREATE|NLM_F_EXCL; + group = NFNLGRP_CONNTRACK_EXP_NEW; } else return 0; - if (!item->report && - !nfnetlink_has_listeners(net, NFNLGRP_CONNTRACK_EXP_NEW)) + if (!item->report && !nfnetlink_has_listeners(net, group)) return 0; skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC); @@ -1659,8 +1696,7 @@ ctnetlink_expect_event(unsigned int events, struct nf_exp_event *item) rcu_read_unlock(); nlmsg_end(skb, nlh); - nfnetlink_send(skb, net, item->pid, NFNLGRP_CONNTRACK_EXP_NEW, - item->report, GFP_ATOMIC); + nfnetlink_send(skb, net, item->pid, group, item->report, GFP_ATOMIC); return 0; nla_put_failure: @@ -1733,6 +1769,8 @@ static const struct nla_policy exp_nla_policy[CTA_EXPECT_MAX+1] = { [CTA_EXPECT_TIMEOUT] = { .type = NLA_U32 }, [CTA_EXPECT_ID] = { .type = NLA_U32 }, [CTA_EXPECT_HELP_NAME] = { .type = NLA_NUL_STRING }, + [CTA_EXPECT_ZONE] = { .type = NLA_U16 }, + [CTA_EXPECT_FLAGS] = { .type = NLA_U32 }, }; static int @@ -1841,7 +1879,13 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, } /* after list removal, usage count == 1 */ - nf_ct_unexpect_related(exp); + spin_lock_bh(&nf_conntrack_lock); + if (del_timer(&exp->timeout)) { + nf_ct_unlink_expect_report(exp, NETLINK_CB(skb).pid, + nlmsg_report(nlh)); + nf_ct_expect_put(exp); + } + spin_unlock_bh(&nf_conntrack_lock); /* have to put what we 'get' above. * after this line usage count == 0 */ nf_ct_expect_put(exp); @@ -1858,7 +1902,9 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, m_help = nfct_help(exp->master); if (!strcmp(m_help->helper->name, name) && del_timer(&exp->timeout)) { - nf_ct_unlink_expect(exp); + nf_ct_unlink_expect_report(exp, + NETLINK_CB(skb).pid, + nlmsg_report(nlh)); nf_ct_expect_put(exp); } } @@ -1872,7 +1918,9 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, &net->ct.expect_hash[i], hnode) { if (del_timer(&exp->timeout)) { - nf_ct_unlink_expect(exp); + nf_ct_unlink_expect_report(exp, + NETLINK_CB(skb).pid, + nlmsg_report(nlh)); nf_ct_expect_put(exp); } } @@ -1918,23 +1966,35 @@ ctnetlink_create_expect(struct net *net, u16 zone, if (!h) return -ENOENT; ct = nf_ct_tuplehash_to_ctrack(h); - help = nfct_help(ct); - - if (!help || !help->helper) { - /* such conntrack hasn't got any helper, abort */ - err = -EOPNOTSUPP; - goto out; - } - exp = nf_ct_expect_alloc(ct); if (!exp) { err = -ENOMEM; goto out; } + help = nfct_help(ct); + if (!help) { + if (!cda[CTA_EXPECT_TIMEOUT]) { + err = -EINVAL; + goto out; + } + exp->timeout.expires = + jiffies + ntohl(nla_get_be32(cda[CTA_EXPECT_TIMEOUT])) * HZ; + + exp->flags = NF_CT_EXPECT_USERSPACE; + if (cda[CTA_EXPECT_FLAGS]) { + exp->flags |= + ntohl(nla_get_be32(cda[CTA_EXPECT_FLAGS])); + } + } else { + if (cda[CTA_EXPECT_FLAGS]) { + exp->flags = ntohl(nla_get_be32(cda[CTA_EXPECT_FLAGS])); + exp->flags &= ~NF_CT_EXPECT_USERSPACE; + } else + exp->flags = 0; + } exp->class = 0; exp->expectfn = NULL; - exp->flags = 0; exp->master = ct; exp->helper = NULL; memcpy(&exp->tuple, &tuple, sizeof(struct nf_conntrack_tuple)); @@ -2102,6 +2162,7 @@ static void __exit ctnetlink_exit(void) { pr_info("ctnetlink: unregistering from nfnetlink.\n"); + nf_ct_remove_userspace_expectations(); #ifdef CONFIG_NF_CONNTRACK_EVENTS nf_ct_expect_unregister_notifier(&ctnl_notifier_exp); nf_conntrack_unregister_notifier(&ctnl_notifier); diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index 5886ba1d52a..ed6d9295802 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -28,8 +28,8 @@ #include <net/netfilter/nf_conntrack_l4proto.h> #include <net/netfilter/nf_conntrack_core.h> -static struct nf_conntrack_l4proto **nf_ct_protos[PF_MAX] __read_mostly; -struct nf_conntrack_l3proto *nf_ct_l3protos[AF_MAX] __read_mostly; +static struct nf_conntrack_l4proto __rcu **nf_ct_protos[PF_MAX] __read_mostly; +struct nf_conntrack_l3proto __rcu *nf_ct_l3protos[AF_MAX] __read_mostly; EXPORT_SYMBOL_GPL(nf_ct_l3protos); static DEFINE_MUTEX(nf_ct_proto_mutex); diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 53d892210a0..bcf47eb518e 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -130,6 +130,44 @@ static int digits_len(const struct nf_conn *ct, const char *dptr, return len; } +static int iswordc(const char c) +{ + if (isalnum(c) || c == '!' || c == '"' || c == '%' || + (c >= '(' && c <= '/') || c == ':' || c == '<' || c == '>' || + c == '?' || (c >= '[' && c <= ']') || c == '_' || c == '`' || + c == '{' || c == '}' || c == '~') + return 1; + return 0; +} + +static int word_len(const char *dptr, const char *limit) +{ + int len = 0; + while (dptr < limit && iswordc(*dptr)) { + dptr++; + len++; + } + return len; +} + +static int callid_len(const struct nf_conn *ct, const char *dptr, + const char *limit, int *shift) +{ + int len, domain_len; + + len = word_len(dptr, limit); + dptr += len; + if (!len || dptr == limit || *dptr != '@') + return len; + dptr++; + len++; + + domain_len = word_len(dptr, limit); + if (!domain_len) + return 0; + return len + domain_len; +} + /* get media type + port length */ static int media_len(const struct nf_conn *ct, const char *dptr, const char *limit, int *shift) @@ -152,6 +190,9 @@ static int parse_addr(const struct nf_conn *ct, const char *cp, const char *end; int ret = 0; + if (!ct) + return 0; + memset(addr, 0, sizeof(*addr)); switch (nf_ct_l3num(ct)) { case AF_INET: @@ -296,6 +337,7 @@ static const struct sip_header ct_sip_hdrs[] = { [SIP_HDR_VIA_TCP] = SIP_HDR("Via", "v", "TCP ", epaddr_len), [SIP_HDR_EXPIRES] = SIP_HDR("Expires", NULL, NULL, digits_len), [SIP_HDR_CONTENT_LENGTH] = SIP_HDR("Content-Length", "l", NULL, digits_len), + [SIP_HDR_CALL_ID] = SIP_HDR("Call-Id", "i", NULL, callid_len), }; static const char *sip_follow_continuation(const char *dptr, const char *limit) @@ -1376,7 +1418,7 @@ static int sip_help_tcp(struct sk_buff *skb, unsigned int protoff, unsigned int msglen, origlen; const char *dptr, *end; s16 diff, tdiff = 0; - int ret; + int ret = NF_ACCEPT; typeof(nf_nat_sip_seq_adjust_hook) nf_nat_sip_seq_adjust; if (ctinfo != IP_CT_ESTABLISHED && diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index eb973fcd67a..0fb65705b44 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -15,6 +15,7 @@ #include <linux/seq_file.h> #include <linux/percpu.h> #include <linux/netdevice.h> +#include <linux/security.h> #include <net/net_namespace.h> #ifdef CONFIG_SYSCTL #include <linux/sysctl.h> @@ -108,6 +109,29 @@ static void ct_seq_stop(struct seq_file *s, void *v) rcu_read_unlock(); } +#ifdef CONFIG_NF_CONNTRACK_SECMARK +static int ct_show_secctx(struct seq_file *s, const struct nf_conn *ct) +{ + int ret; + u32 len; + char *secctx; + + ret = security_secid_to_secctx(ct->secmark, &secctx, &len); + if (ret) + return ret; + + ret = seq_printf(s, "secctx=%s ", secctx); + + security_release_secctx(secctx, len); + return ret; +} +#else +static inline int ct_show_secctx(struct seq_file *s, const struct nf_conn *ct) +{ + return 0; +} +#endif + /* return 0 on success, 1 in case of error */ static int ct_seq_show(struct seq_file *s, void *v) { @@ -168,10 +192,8 @@ static int ct_seq_show(struct seq_file *s, void *v) goto release; #endif -#ifdef CONFIG_NF_CONNTRACK_SECMARK - if (seq_printf(s, "secmark=%u ", ct->secmark)) + if (ct_show_secctx(s, ct)) goto release; -#endif #ifdef CONFIG_NF_CONNTRACK_ZONES if (seq_printf(s, "zone=%u ", nf_ct_zone(ct))) diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c index 7df37fd786b..b07393eab88 100644 --- a/net/netfilter/nf_log.c +++ b/net/netfilter/nf_log.c @@ -16,7 +16,7 @@ #define NF_LOG_PREFIXLEN 128 #define NFLOGGER_NAME_LEN 64 -static const struct nf_logger *nf_loggers[NFPROTO_NUMPROTO] __read_mostly; +static const struct nf_logger __rcu *nf_loggers[NFPROTO_NUMPROTO] __read_mostly; static struct list_head nf_loggers_l[NFPROTO_NUMPROTO] __read_mostly; static DEFINE_MUTEX(nf_log_mutex); diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index 78b3cf9c519..74aebed5bd2 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -18,7 +18,7 @@ * long term mutex. The handler must provide an an outfn() to accept packets * for queueing and must reinject all packets it receives, no matter what. */ -static const struct nf_queue_handler *queue_handler[NFPROTO_NUMPROTO] __read_mostly; +static const struct nf_queue_handler __rcu *queue_handler[NFPROTO_NUMPROTO] __read_mostly; static DEFINE_MUTEX(queue_handler_mutex); diff --git a/net/netfilter/nf_tproxy_core.c b/net/netfilter/nf_tproxy_core.c index 5490fc37c92..4d87befb04c 100644 --- a/net/netfilter/nf_tproxy_core.c +++ b/net/netfilter/nf_tproxy_core.c @@ -18,41 +18,6 @@ #include <net/udp.h> #include <net/netfilter/nf_tproxy_core.h> -struct sock * -nf_tproxy_get_sock_v4(struct net *net, const u8 protocol, - const __be32 saddr, const __be32 daddr, - const __be16 sport, const __be16 dport, - const struct net_device *in, bool listening_only) -{ - struct sock *sk; - - /* look up socket */ - switch (protocol) { - case IPPROTO_TCP: - if (listening_only) - sk = __inet_lookup_listener(net, &tcp_hashinfo, - daddr, ntohs(dport), - in->ifindex); - else - sk = __inet_lookup(net, &tcp_hashinfo, - saddr, sport, daddr, dport, - in->ifindex); - break; - case IPPROTO_UDP: - sk = udp4_lib_lookup(net, saddr, sport, daddr, dport, - in->ifindex); - break; - default: - WARN_ON(1); - sk = NULL; - } - - pr_debug("tproxy socket lookup: proto %u %08x:%u -> %08x:%u, listener only: %d, sock %p\n", - protocol, ntohl(saddr), ntohs(sport), ntohl(daddr), ntohs(dport), listening_only, sk); - - return sk; -} -EXPORT_SYMBOL_GPL(nf_tproxy_get_sock_v4); static void nf_tproxy_destructor(struct sk_buff *skb) @@ -70,7 +35,11 @@ nf_tproxy_destructor(struct sk_buff *skb) int nf_tproxy_assign_sock(struct sk_buff *skb, struct sock *sk) { - if (inet_sk(sk)->transparent) { + bool transparent = (sk->sk_state == TCP_TIME_WAIT) ? + inet_twsk(sk)->tw_transparent : + inet_sk(sk)->transparent; + + if (transparent) { skb_orphan(skb); skb->sk = sk; skb->destructor = nf_tproxy_destructor; diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index e34622fa000..80463507420 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -116,10 +116,8 @@ EXPORT_SYMBOL(xt_register_targets); void xt_unregister_targets(struct xt_target *target, unsigned int n) { - unsigned int i; - - for (i = 0; i < n; i++) - xt_unregister_target(&target[i]); + while (n-- > 0) + xt_unregister_target(&target[n]); } EXPORT_SYMBOL(xt_unregister_targets); @@ -174,10 +172,8 @@ EXPORT_SYMBOL(xt_register_matches); void xt_unregister_matches(struct xt_match *match, unsigned int n) { - unsigned int i; - - for (i = 0; i < n; i++) - xt_unregister_match(&match[i]); + while (n-- > 0) + xt_unregister_match(&match[n]); } EXPORT_SYMBOL(xt_unregister_matches); diff --git a/net/netfilter/xt_CT.c b/net/netfilter/xt_CT.c index 0cb6053f02f..782e51986a6 100644 --- a/net/netfilter/xt_CT.c +++ b/net/netfilter/xt_CT.c @@ -9,7 +9,6 @@ #include <linux/module.h> #include <linux/gfp.h> #include <linux/skbuff.h> -#include <linux/selinux.h> #include <linux/netfilter_ipv4/ip_tables.h> #include <linux/netfilter_ipv6/ip6_tables.h> #include <linux/netfilter/x_tables.h> diff --git a/net/netfilter/xt_SECMARK.c b/net/netfilter/xt_SECMARK.c index 23b2d6c486b..9faf5e050b7 100644 --- a/net/netfilter/xt_SECMARK.c +++ b/net/netfilter/xt_SECMARK.c @@ -14,8 +14,8 @@ */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/module.h> +#include <linux/security.h> #include <linux/skbuff.h> -#include <linux/selinux.h> #include <linux/netfilter/x_tables.h> #include <linux/netfilter/xt_SECMARK.h> @@ -39,9 +39,8 @@ secmark_tg(struct sk_buff *skb, const struct xt_action_param *par) switch (mode) { case SECMARK_MODE_SEL: - secmark = info->u.sel.selsid; + secmark = info->secid; break; - default: BUG(); } @@ -50,33 +49,33 @@ secmark_tg(struct sk_buff *skb, const struct xt_action_param *par) return XT_CONTINUE; } -static int checkentry_selinux(struct xt_secmark_target_info *info) +static int checkentry_lsm(struct xt_secmark_target_info *info) { int err; - struct xt_secmark_target_selinux_info *sel = &info->u.sel; - sel->selctx[SECMARK_SELCTX_MAX - 1] = '\0'; + info->secctx[SECMARK_SECCTX_MAX - 1] = '\0'; + info->secid = 0; - err = selinux_string_to_sid(sel->selctx, &sel->selsid); + err = security_secctx_to_secid(info->secctx, strlen(info->secctx), + &info->secid); if (err) { if (err == -EINVAL) - pr_info("invalid SELinux context \'%s\'\n", - sel->selctx); + pr_info("invalid security context \'%s\'\n", info->secctx); return err; } - if (!sel->selsid) { - pr_info("unable to map SELinux context \'%s\'\n", sel->selctx); + if (!info->secid) { + pr_info("unable to map security context \'%s\'\n", info->secctx); return -ENOENT; } - err = selinux_secmark_relabel_packet_permission(sel->selsid); + err = security_secmark_relabel_packet(info->secid); if (err) { pr_info("unable to obtain relabeling permission\n"); return err; } - selinux_secmark_refcount_inc(); + security_secmark_refcount_inc(); return 0; } @@ -100,16 +99,16 @@ static int secmark_tg_check(const struct xt_tgchk_param *par) switch (info->mode) { case SECMARK_MODE_SEL: - err = checkentry_selinux(info); - if (err <= 0) - return err; break; - default: pr_info("invalid mode: %hu\n", info->mode); return -EINVAL; } + err = checkentry_lsm(info); + if (err) + return err; + if (!mode) mode = info->mode; return 0; @@ -119,7 +118,7 @@ static void secmark_tg_destroy(const struct xt_tgdtor_param *par) { switch (mode) { case SECMARK_MODE_SEL: - selinux_secmark_refcount_dec(); + security_secmark_refcount_dec(); } } diff --git a/net/netfilter/xt_TPROXY.c b/net/netfilter/xt_TPROXY.c index c61294d85fd..19c482caf30 100644 --- a/net/netfilter/xt_TPROXY.c +++ b/net/netfilter/xt_TPROXY.c @@ -1,7 +1,7 @@ /* * Transparent proxy support for Linux/iptables * - * Copyright (c) 2006-2007 BalaBit IT Ltd. + * Copyright (c) 2006-2010 BalaBit IT Ltd. * Author: Balazs Scheidler, Krisztian Kovacs * * This program is free software; you can redistribute it and/or modify @@ -16,19 +16,96 @@ #include <net/checksum.h> #include <net/udp.h> #include <net/inet_sock.h> - +#include <linux/inetdevice.h> #include <linux/netfilter/x_tables.h> #include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter/xt_TPROXY.h> #include <net/netfilter/ipv4/nf_defrag_ipv4.h> +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) +#include <net/if_inet6.h> +#include <net/addrconf.h> +#include <linux/netfilter_ipv6/ip6_tables.h> +#include <net/netfilter/ipv6/nf_defrag_ipv6.h> +#endif + #include <net/netfilter/nf_tproxy_core.h> +#include <linux/netfilter/xt_TPROXY.h> + +static inline __be32 +tproxy_laddr4(struct sk_buff *skb, __be32 user_laddr, __be32 daddr) +{ + struct in_device *indev; + __be32 laddr; + + if (user_laddr) + return user_laddr; + + laddr = 0; + rcu_read_lock(); + indev = __in_dev_get_rcu(skb->dev); + for_primary_ifa(indev) { + laddr = ifa->ifa_local; + break; + } endfor_ifa(indev); + rcu_read_unlock(); + + return laddr ? laddr : daddr; +} + +/** + * tproxy_handle_time_wait4() - handle IPv4 TCP TIME_WAIT reopen redirections + * @skb: The skb being processed. + * @laddr: IPv4 address to redirect to or zero. + * @lport: TCP port to redirect to or zero. + * @sk: The TIME_WAIT TCP socket found by the lookup. + * + * We have to handle SYN packets arriving to TIME_WAIT sockets + * differently: instead of reopening the connection we should rather + * redirect the new connection to the proxy if there's a listener + * socket present. + * + * tproxy_handle_time_wait4() consumes the socket reference passed in. + * + * Returns the listener socket if there's one, the TIME_WAIT socket if + * no such listener is found, or NULL if the TCP header is incomplete. + */ +static struct sock * +tproxy_handle_time_wait4(struct sk_buff *skb, __be32 laddr, __be16 lport, + struct sock *sk) +{ + const struct iphdr *iph = ip_hdr(skb); + struct tcphdr _hdr, *hp; + + hp = skb_header_pointer(skb, ip_hdrlen(skb), sizeof(_hdr), &_hdr); + if (hp == NULL) { + inet_twsk_put(inet_twsk(sk)); + return NULL; + } + + if (hp->syn && !hp->rst && !hp->ack && !hp->fin) { + /* SYN to a TIME_WAIT socket, we'd rather redirect it + * to a listener socket if there's one */ + struct sock *sk2; + + sk2 = nf_tproxy_get_sock_v4(dev_net(skb->dev), iph->protocol, + iph->saddr, laddr ? laddr : iph->daddr, + hp->source, lport ? lport : hp->dest, + skb->dev, NFT_LOOKUP_LISTENER); + if (sk2) { + inet_twsk_deschedule(inet_twsk(sk), &tcp_death_row); + inet_twsk_put(inet_twsk(sk)); + sk = sk2; + } + } + + return sk; +} static unsigned int -tproxy_tg(struct sk_buff *skb, const struct xt_action_param *par) +tproxy_tg4(struct sk_buff *skb, __be32 laddr, __be16 lport, + u_int32_t mark_mask, u_int32_t mark_value) { const struct iphdr *iph = ip_hdr(skb); - const struct xt_tproxy_target_info *tgi = par->targinfo; struct udphdr _hdr, *hp; struct sock *sk; @@ -36,12 +113,195 @@ tproxy_tg(struct sk_buff *skb, const struct xt_action_param *par) if (hp == NULL) return NF_DROP; + /* check if there's an ongoing connection on the packet + * addresses, this happens if the redirect already happened + * and the current packet belongs to an already established + * connection */ sk = nf_tproxy_get_sock_v4(dev_net(skb->dev), iph->protocol, - iph->saddr, - tgi->laddr ? tgi->laddr : iph->daddr, - hp->source, - tgi->lport ? tgi->lport : hp->dest, - par->in, true); + iph->saddr, iph->daddr, + hp->source, hp->dest, + skb->dev, NFT_LOOKUP_ESTABLISHED); + + laddr = tproxy_laddr4(skb, laddr, iph->daddr); + if (!lport) + lport = hp->dest; + + /* UDP has no TCP_TIME_WAIT state, so we never enter here */ + if (sk && sk->sk_state == TCP_TIME_WAIT) + /* reopening a TIME_WAIT connection needs special handling */ + sk = tproxy_handle_time_wait4(skb, laddr, lport, sk); + else if (!sk) + /* no, there's no established connection, check if + * there's a listener on the redirected addr/port */ + sk = nf_tproxy_get_sock_v4(dev_net(skb->dev), iph->protocol, + iph->saddr, laddr, + hp->source, lport, + skb->dev, NFT_LOOKUP_LISTENER); + + /* NOTE: assign_sock consumes our sk reference */ + if (sk && nf_tproxy_assign_sock(skb, sk)) { + /* This should be in a separate target, but we don't do multiple + targets on the same rule yet */ + skb->mark = (skb->mark & ~mark_mask) ^ mark_value; + + pr_debug("redirecting: proto %hhu %pI4:%hu -> %pI4:%hu, mark: %x\n", + iph->protocol, &iph->daddr, ntohs(hp->dest), + &laddr, ntohs(lport), skb->mark); + return NF_ACCEPT; + } + + pr_debug("no socket, dropping: proto %hhu %pI4:%hu -> %pI4:%hu, mark: %x\n", + iph->protocol, &iph->saddr, ntohs(hp->source), + &iph->daddr, ntohs(hp->dest), skb->mark); + return NF_DROP; +} + +static unsigned int +tproxy_tg4_v0(struct sk_buff *skb, const struct xt_action_param *par) +{ + const struct xt_tproxy_target_info *tgi = par->targinfo; + + return tproxy_tg4(skb, tgi->laddr, tgi->lport, tgi->mark_mask, tgi->mark_value); +} + +static unsigned int +tproxy_tg4_v1(struct sk_buff *skb, const struct xt_action_param *par) +{ + const struct xt_tproxy_target_info_v1 *tgi = par->targinfo; + + return tproxy_tg4(skb, tgi->laddr.ip, tgi->lport, tgi->mark_mask, tgi->mark_value); +} + +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + +static inline const struct in6_addr * +tproxy_laddr6(struct sk_buff *skb, const struct in6_addr *user_laddr, + const struct in6_addr *daddr) +{ + struct inet6_dev *indev; + struct inet6_ifaddr *ifa; + struct in6_addr *laddr; + + if (!ipv6_addr_any(user_laddr)) + return user_laddr; + laddr = NULL; + + rcu_read_lock(); + indev = __in6_dev_get(skb->dev); + if (indev) + list_for_each_entry(ifa, &indev->addr_list, if_list) { + if (ifa->flags & (IFA_F_TENTATIVE | IFA_F_DEPRECATED)) + continue; + + laddr = &ifa->addr; + break; + } + rcu_read_unlock(); + + return laddr ? laddr : daddr; +} + +/** + * tproxy_handle_time_wait6() - handle IPv6 TCP TIME_WAIT reopen redirections + * @skb: The skb being processed. + * @tproto: Transport protocol. + * @thoff: Transport protocol header offset. + * @par: Iptables target parameters. + * @sk: The TIME_WAIT TCP socket found by the lookup. + * + * We have to handle SYN packets arriving to TIME_WAIT sockets + * differently: instead of reopening the connection we should rather + * redirect the new connection to the proxy if there's a listener + * socket present. + * + * tproxy_handle_time_wait6() consumes the socket reference passed in. + * + * Returns the listener socket if there's one, the TIME_WAIT socket if + * no such listener is found, or NULL if the TCP header is incomplete. + */ +static struct sock * +tproxy_handle_time_wait6(struct sk_buff *skb, int tproto, int thoff, + const struct xt_action_param *par, + struct sock *sk) +{ + const struct ipv6hdr *iph = ipv6_hdr(skb); + struct tcphdr _hdr, *hp; + const struct xt_tproxy_target_info_v1 *tgi = par->targinfo; + + hp = skb_header_pointer(skb, thoff, sizeof(_hdr), &_hdr); + if (hp == NULL) { + inet_twsk_put(inet_twsk(sk)); + return NULL; + } + + if (hp->syn && !hp->rst && !hp->ack && !hp->fin) { + /* SYN to a TIME_WAIT socket, we'd rather redirect it + * to a listener socket if there's one */ + struct sock *sk2; + + sk2 = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto, + &iph->saddr, + tproxy_laddr6(skb, &tgi->laddr.in6, &iph->daddr), + hp->source, + tgi->lport ? tgi->lport : hp->dest, + skb->dev, NFT_LOOKUP_LISTENER); + if (sk2) { + inet_twsk_deschedule(inet_twsk(sk), &tcp_death_row); + inet_twsk_put(inet_twsk(sk)); + sk = sk2; + } + } + + return sk; +} + +static unsigned int +tproxy_tg6_v1(struct sk_buff *skb, const struct xt_action_param *par) +{ + const struct ipv6hdr *iph = ipv6_hdr(skb); + const struct xt_tproxy_target_info_v1 *tgi = par->targinfo; + struct udphdr _hdr, *hp; + struct sock *sk; + const struct in6_addr *laddr; + __be16 lport; + int thoff; + int tproto; + + tproto = ipv6_find_hdr(skb, &thoff, -1, NULL); + if (tproto < 0) { + pr_debug("unable to find transport header in IPv6 packet, dropping\n"); + return NF_DROP; + } + + hp = skb_header_pointer(skb, thoff, sizeof(_hdr), &_hdr); + if (hp == NULL) { + pr_debug("unable to grab transport header contents in IPv6 packet, dropping\n"); + return NF_DROP; + } + + /* check if there's an ongoing connection on the packet + * addresses, this happens if the redirect already happened + * and the current packet belongs to an already established + * connection */ + sk = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto, + &iph->saddr, &iph->daddr, + hp->source, hp->dest, + par->in, NFT_LOOKUP_ESTABLISHED); + + laddr = tproxy_laddr6(skb, &tgi->laddr.in6, &iph->daddr); + lport = tgi->lport ? tgi->lport : hp->dest; + + /* UDP has no TCP_TIME_WAIT state, so we never enter here */ + if (sk && sk->sk_state == TCP_TIME_WAIT) + /* reopening a TIME_WAIT connection needs special handling */ + sk = tproxy_handle_time_wait6(skb, tproto, thoff, par, sk); + else if (!sk) + /* no there's no established connection, check if + * there's a listener on the redirected addr/port */ + sk = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto, + &iph->saddr, laddr, + hp->source, lport, + par->in, NFT_LOOKUP_LISTENER); /* NOTE: assign_sock consumes our sk reference */ if (sk && nf_tproxy_assign_sock(skb, sk)) { @@ -49,19 +309,34 @@ tproxy_tg(struct sk_buff *skb, const struct xt_action_param *par) targets on the same rule yet */ skb->mark = (skb->mark & ~tgi->mark_mask) ^ tgi->mark_value; - pr_debug("redirecting: proto %u %08x:%u -> %08x:%u, mark: %x\n", - iph->protocol, ntohl(iph->daddr), ntohs(hp->dest), - ntohl(tgi->laddr), ntohs(tgi->lport), skb->mark); + pr_debug("redirecting: proto %hhu %pI6:%hu -> %pI6:%hu, mark: %x\n", + tproto, &iph->saddr, ntohs(hp->source), + laddr, ntohs(lport), skb->mark); return NF_ACCEPT; } - pr_debug("no socket, dropping: proto %u %08x:%u -> %08x:%u, mark: %x\n", - iph->protocol, ntohl(iph->daddr), ntohs(hp->dest), - ntohl(tgi->laddr), ntohs(tgi->lport), skb->mark); + pr_debug("no socket, dropping: proto %hhu %pI6:%hu -> %pI6:%hu, mark: %x\n", + tproto, &iph->saddr, ntohs(hp->source), + &iph->daddr, ntohs(hp->dest), skb->mark); + return NF_DROP; } -static int tproxy_tg_check(const struct xt_tgchk_param *par) +static int tproxy_tg6_check(const struct xt_tgchk_param *par) +{ + const struct ip6t_ip6 *i = par->entryinfo; + + if ((i->proto == IPPROTO_TCP || i->proto == IPPROTO_UDP) + && !(i->flags & IP6T_INV_PROTO)) + return 0; + + pr_info("Can be used only in combination with " + "either -p tcp or -p udp\n"); + return -EINVAL; +} +#endif + +static int tproxy_tg4_check(const struct xt_tgchk_param *par) { const struct ipt_ip *i = par->entryinfo; @@ -74,31 +349,64 @@ static int tproxy_tg_check(const struct xt_tgchk_param *par) return -EINVAL; } -static struct xt_target tproxy_tg_reg __read_mostly = { - .name = "TPROXY", - .family = AF_INET, - .table = "mangle", - .target = tproxy_tg, - .targetsize = sizeof(struct xt_tproxy_target_info), - .checkentry = tproxy_tg_check, - .hooks = 1 << NF_INET_PRE_ROUTING, - .me = THIS_MODULE, +static struct xt_target tproxy_tg_reg[] __read_mostly = { + { + .name = "TPROXY", + .family = NFPROTO_IPV4, + .table = "mangle", + .target = tproxy_tg4_v0, + .revision = 0, + .targetsize = sizeof(struct xt_tproxy_target_info), + .checkentry = tproxy_tg4_check, + .hooks = 1 << NF_INET_PRE_ROUTING, + .me = THIS_MODULE, + }, + { + .name = "TPROXY", + .family = NFPROTO_IPV4, + .table = "mangle", + .target = tproxy_tg4_v1, + .revision = 1, + .targetsize = sizeof(struct xt_tproxy_target_info_v1), + .checkentry = tproxy_tg4_check, + .hooks = 1 << NF_INET_PRE_ROUTING, + .me = THIS_MODULE, + }, +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + { + .name = "TPROXY", + .family = NFPROTO_IPV6, + .table = "mangle", + .target = tproxy_tg6_v1, + .revision = 1, + .targetsize = sizeof(struct xt_tproxy_target_info_v1), + .checkentry = tproxy_tg6_check, + .hooks = 1 << NF_INET_PRE_ROUTING, + .me = THIS_MODULE, + }, +#endif + }; static int __init tproxy_tg_init(void) { nf_defrag_ipv4_enable(); - return xt_register_target(&tproxy_tg_reg); +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + nf_defrag_ipv6_enable(); +#endif + + return xt_register_targets(tproxy_tg_reg, ARRAY_SIZE(tproxy_tg_reg)); } static void __exit tproxy_tg_exit(void) { - xt_unregister_target(&tproxy_tg_reg); + xt_unregister_targets(tproxy_tg_reg, ARRAY_SIZE(tproxy_tg_reg)); } module_init(tproxy_tg_init); module_exit(tproxy_tg_exit); MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Krisztian Kovacs"); +MODULE_AUTHOR("Balazs Scheidler, Krisztian Kovacs"); MODULE_DESCRIPTION("Netfilter transparent proxy (TPROXY) target module."); MODULE_ALIAS("ipt_TPROXY"); +MODULE_ALIAS("ip6t_TPROXY"); diff --git a/net/netfilter/xt_hashlimit.c b/net/netfilter/xt_hashlimit.c index b46a8390896..9228ee0dc11 100644 --- a/net/netfilter/xt_hashlimit.c +++ b/net/netfilter/xt_hashlimit.c @@ -448,6 +448,7 @@ hashlimit_init_dst(const struct xt_hashlimit_htable *hinfo, { __be16 _ports[2], *ports; u8 nexthdr; + int poff; memset(dst, 0, sizeof(*dst)); @@ -492,19 +493,13 @@ hashlimit_init_dst(const struct xt_hashlimit_htable *hinfo, return 0; } - switch (nexthdr) { - case IPPROTO_TCP: - case IPPROTO_UDP: - case IPPROTO_UDPLITE: - case IPPROTO_SCTP: - case IPPROTO_DCCP: - ports = skb_header_pointer(skb, protoff, sizeof(_ports), + poff = proto_ports_offset(nexthdr); + if (poff >= 0) { + ports = skb_header_pointer(skb, protoff + poff, sizeof(_ports), &_ports); - break; - default: + } else { _ports[0] = _ports[1] = 0; ports = _ports; - break; } if (!ports) return -1; diff --git a/net/netfilter/xt_ipvs.c b/net/netfilter/xt_ipvs.c index 7a4d66db95a..9127a3d8aa3 100644 --- a/net/netfilter/xt_ipvs.c +++ b/net/netfilter/xt_ipvs.c @@ -16,7 +16,6 @@ #include <linux/ip_vs.h> #include <linux/types.h> #include <linux/netfilter/x_tables.h> -#include <linux/netfilter/x_tables.h> #include <linux/netfilter/xt_ipvs.h> #include <net/netfilter/nf_conntrack.h> diff --git a/net/netfilter/xt_recent.c b/net/netfilter/xt_recent.c index 76aec6a4476..d2ff15a2412 100644 --- a/net/netfilter/xt_recent.c +++ b/net/netfilter/xt_recent.c @@ -567,6 +567,7 @@ static const struct file_operations recent_mt_fops = { .write = recent_mt_proc_write, .release = seq_release_private, .owner = THIS_MODULE, + .llseek = seq_lseek, }; static int __net_init recent_proc_net_init(struct net *net) diff --git a/net/netfilter/xt_socket.c b/net/netfilter/xt_socket.c index 1ca89908cba..2dbd4c85773 100644 --- a/net/netfilter/xt_socket.c +++ b/net/netfilter/xt_socket.c @@ -14,6 +14,7 @@ #include <linux/skbuff.h> #include <linux/netfilter/x_tables.h> #include <linux/netfilter_ipv4/ip_tables.h> +#include <linux/netfilter_ipv6/ip6_tables.h> #include <net/tcp.h> #include <net/udp.h> #include <net/icmp.h> @@ -21,6 +22,7 @@ #include <net/inet_sock.h> #include <net/netfilter/nf_tproxy_core.h> #include <net/netfilter/ipv4/nf_defrag_ipv4.h> +#include <net/netfilter/ipv6/nf_defrag_ipv6.h> #include <linux/netfilter/xt_socket.h> @@ -30,7 +32,7 @@ #endif static int -extract_icmp_fields(const struct sk_buff *skb, +extract_icmp4_fields(const struct sk_buff *skb, u8 *protocol, __be32 *raddr, __be32 *laddr, @@ -86,7 +88,6 @@ extract_icmp_fields(const struct sk_buff *skb, return 0; } - static bool socket_match(const struct sk_buff *skb, struct xt_action_param *par, const struct xt_socket_mtinfo1 *info) @@ -115,7 +116,7 @@ socket_match(const struct sk_buff *skb, struct xt_action_param *par, dport = hp->dest; } else if (iph->protocol == IPPROTO_ICMP) { - if (extract_icmp_fields(skb, &protocol, &saddr, &daddr, + if (extract_icmp4_fields(skb, &protocol, &saddr, &daddr, &sport, &dport)) return false; } else { @@ -142,7 +143,7 @@ socket_match(const struct sk_buff *skb, struct xt_action_param *par, #endif sk = nf_tproxy_get_sock_v4(dev_net(skb->dev), protocol, - saddr, daddr, sport, dport, par->in, false); + saddr, daddr, sport, dport, par->in, NFT_LOOKUP_ANY); if (sk != NULL) { bool wildcard; bool transparent = true; @@ -165,32 +166,157 @@ socket_match(const struct sk_buff *skb, struct xt_action_param *par, sk = NULL; } - pr_debug("proto %u %08x:%u -> %08x:%u (orig %08x:%u) sock %p\n", - protocol, ntohl(saddr), ntohs(sport), - ntohl(daddr), ntohs(dport), - ntohl(iph->daddr), hp ? ntohs(hp->dest) : 0, sk); + pr_debug("proto %hhu %pI4:%hu -> %pI4:%hu (orig %pI4:%hu) sock %p\n", + protocol, &saddr, ntohs(sport), + &daddr, ntohs(dport), + &iph->daddr, hp ? ntohs(hp->dest) : 0, sk); return (sk != NULL); } static bool -socket_mt_v0(const struct sk_buff *skb, struct xt_action_param *par) +socket_mt4_v0(const struct sk_buff *skb, struct xt_action_param *par) { return socket_match(skb, par, NULL); } static bool -socket_mt_v1(const struct sk_buff *skb, struct xt_action_param *par) +socket_mt4_v1(const struct sk_buff *skb, struct xt_action_param *par) { return socket_match(skb, par, par->matchinfo); } +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + +static int +extract_icmp6_fields(const struct sk_buff *skb, + unsigned int outside_hdrlen, + u8 *protocol, + struct in6_addr **raddr, + struct in6_addr **laddr, + __be16 *rport, + __be16 *lport) +{ + struct ipv6hdr *inside_iph, _inside_iph; + struct icmp6hdr *icmph, _icmph; + __be16 *ports, _ports[2]; + u8 inside_nexthdr; + int inside_hdrlen; + + icmph = skb_header_pointer(skb, outside_hdrlen, + sizeof(_icmph), &_icmph); + if (icmph == NULL) + return 1; + + if (icmph->icmp6_type & ICMPV6_INFOMSG_MASK) + return 1; + + inside_iph = skb_header_pointer(skb, outside_hdrlen + sizeof(_icmph), sizeof(_inside_iph), &_inside_iph); + if (inside_iph == NULL) + return 1; + inside_nexthdr = inside_iph->nexthdr; + + inside_hdrlen = ipv6_skip_exthdr(skb, outside_hdrlen + sizeof(_icmph) + sizeof(_inside_iph), &inside_nexthdr); + if (inside_hdrlen < 0) + return 1; /* hjm: Packet has no/incomplete transport layer headers. */ + + if (inside_nexthdr != IPPROTO_TCP && + inside_nexthdr != IPPROTO_UDP) + return 1; + + ports = skb_header_pointer(skb, inside_hdrlen, + sizeof(_ports), &_ports); + if (ports == NULL) + return 1; + + /* the inside IP packet is the one quoted from our side, thus + * its saddr is the local address */ + *protocol = inside_nexthdr; + *laddr = &inside_iph->saddr; + *lport = ports[0]; + *raddr = &inside_iph->daddr; + *rport = ports[1]; + + return 0; +} + +static bool +socket_mt6_v1(const struct sk_buff *skb, struct xt_action_param *par) +{ + struct ipv6hdr *iph = ipv6_hdr(skb); + struct udphdr _hdr, *hp = NULL; + struct sock *sk; + struct in6_addr *daddr, *saddr; + __be16 dport, sport; + int thoff; + u8 tproto; + const struct xt_socket_mtinfo1 *info = (struct xt_socket_mtinfo1 *) par->matchinfo; + + tproto = ipv6_find_hdr(skb, &thoff, -1, NULL); + if (tproto < 0) { + pr_debug("unable to find transport header in IPv6 packet, dropping\n"); + return NF_DROP; + } + + if (tproto == IPPROTO_UDP || tproto == IPPROTO_TCP) { + hp = skb_header_pointer(skb, thoff, + sizeof(_hdr), &_hdr); + if (hp == NULL) + return false; + + saddr = &iph->saddr; + sport = hp->source; + daddr = &iph->daddr; + dport = hp->dest; + + } else if (tproto == IPPROTO_ICMPV6) { + if (extract_icmp6_fields(skb, thoff, &tproto, &saddr, &daddr, + &sport, &dport)) + return false; + } else { + return false; + } + + sk = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto, + saddr, daddr, sport, dport, par->in, NFT_LOOKUP_ANY); + if (sk != NULL) { + bool wildcard; + bool transparent = true; + + /* Ignore sockets listening on INADDR_ANY */ + wildcard = (sk->sk_state != TCP_TIME_WAIT && + ipv6_addr_any(&inet6_sk(sk)->rcv_saddr)); + + /* Ignore non-transparent sockets, + if XT_SOCKET_TRANSPARENT is used */ + if (info && info->flags & XT_SOCKET_TRANSPARENT) + transparent = ((sk->sk_state != TCP_TIME_WAIT && + inet_sk(sk)->transparent) || + (sk->sk_state == TCP_TIME_WAIT && + inet_twsk(sk)->tw_transparent)); + + nf_tproxy_put_sock(sk); + + if (wildcard || !transparent) + sk = NULL; + } + + pr_debug("proto %hhu %pI6:%hu -> %pI6:%hu " + "(orig %pI6:%hu) sock %p\n", + tproto, saddr, ntohs(sport), + daddr, ntohs(dport), + &iph->daddr, hp ? ntohs(hp->dest) : 0, sk); + + return (sk != NULL); +} +#endif + static struct xt_match socket_mt_reg[] __read_mostly = { { .name = "socket", .revision = 0, .family = NFPROTO_IPV4, - .match = socket_mt_v0, + .match = socket_mt4_v0, .hooks = (1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_IN), .me = THIS_MODULE, @@ -199,17 +325,33 @@ static struct xt_match socket_mt_reg[] __read_mostly = { .name = "socket", .revision = 1, .family = NFPROTO_IPV4, - .match = socket_mt_v1, + .match = socket_mt4_v1, .matchsize = sizeof(struct xt_socket_mtinfo1), .hooks = (1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_IN), .me = THIS_MODULE, }, +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + { + .name = "socket", + .revision = 1, + .family = NFPROTO_IPV6, + .match = socket_mt6_v1, + .matchsize = sizeof(struct xt_socket_mtinfo1), + .hooks = (1 << NF_INET_PRE_ROUTING) | + (1 << NF_INET_LOCAL_IN), + .me = THIS_MODULE, + }, +#endif }; static int __init socket_mt_init(void) { nf_defrag_ipv4_enable(); +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + nf_defrag_ipv6_enable(); +#endif + return xt_register_matches(socket_mt_reg, ARRAY_SIZE(socket_mt_reg)); } @@ -225,3 +367,4 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Krisztian Kovacs, Balazs Scheidler"); MODULE_DESCRIPTION("x_tables socket match module"); MODULE_ALIAS("ipt_socket"); +MODULE_ALIAS("ip6t_socket"); diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 2cbf380377d..cd96ed3ccee 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1406,7 +1406,7 @@ static int netlink_recvmsg(struct kiocb *kiocb, struct socket *sock, struct netlink_sock *nlk = nlk_sk(sk); int noblock = flags&MSG_DONTWAIT; size_t copied; - struct sk_buff *skb; + struct sk_buff *skb, *data_skb; int err; if (flags&MSG_OOB) @@ -1418,59 +1418,35 @@ static int netlink_recvmsg(struct kiocb *kiocb, struct socket *sock, if (skb == NULL) goto out; + data_skb = skb; + #ifdef CONFIG_COMPAT_NETLINK_MESSAGES if (unlikely(skb_shinfo(skb)->frag_list)) { - bool need_compat = !!(flags & MSG_CMSG_COMPAT); - /* - * If this skb has a frag_list, then here that means that - * we will have to use the frag_list skb for compat tasks - * and the regular skb for non-compat tasks. + * If this skb has a frag_list, then here that means that we + * will have to use the frag_list skb's data for compat tasks + * and the regular skb's data for normal (non-compat) tasks. * - * The skb might (and likely will) be cloned, so we can't - * just reset frag_list and go on with things -- we need to - * keep that. For the compat case that's easy -- simply get - * a reference to the compat skb and free the regular one - * including the frag. For the non-compat case, we need to - * avoid sending the frag to the user -- so assign NULL but - * restore it below before freeing the skb. + * If we need to send the compat skb, assign it to the + * 'data_skb' variable so that it will be used below for data + * copying. We keep 'skb' for everything else, including + * freeing both later. */ - if (need_compat) { - struct sk_buff *compskb = skb_shinfo(skb)->frag_list; - skb_get(compskb); - kfree_skb(skb); - skb = compskb; - } else { - /* - * Before setting frag_list to NULL, we must get a - * private copy of skb if shared (because of MSG_PEEK) - */ - if (skb_shared(skb)) { - struct sk_buff *nskb; - - nskb = pskb_copy(skb, GFP_KERNEL); - kfree_skb(skb); - skb = nskb; - err = -ENOMEM; - if (!skb) - goto out; - } - kfree_skb(skb_shinfo(skb)->frag_list); - skb_shinfo(skb)->frag_list = NULL; - } + if (flags & MSG_CMSG_COMPAT) + data_skb = skb_shinfo(skb)->frag_list; } #endif msg->msg_namelen = 0; - copied = skb->len; + copied = data_skb->len; if (len < copied) { msg->msg_flags |= MSG_TRUNC; copied = len; } - skb_reset_transport_header(skb); - err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied); + skb_reset_transport_header(data_skb); + err = skb_copy_datagram_iovec(data_skb, 0, msg->msg_iov, copied); if (msg->msg_name) { struct sockaddr_nl *addr = (struct sockaddr_nl *)msg->msg_name; @@ -1490,7 +1466,7 @@ static int netlink_recvmsg(struct kiocb *kiocb, struct socket *sock, } siocb->scm->creds = *NETLINK_CREDS(skb); if (flags & MSG_TRUNC) - copied = skb->len; + copied = data_skb->len; skb_free_datagram(sk, skb); @@ -2126,6 +2102,26 @@ static void __net_exit netlink_net_exit(struct net *net) #endif } +static void __init netlink_add_usersock_entry(void) +{ + unsigned long *listeners; + int groups = 32; + + listeners = kzalloc(NLGRPSZ(groups) + sizeof(struct listeners_rcu_head), + GFP_KERNEL); + if (!listeners) + panic("netlink_add_usersock_entry: Cannot allocate listneres\n"); + + netlink_table_grab(); + + nl_table[NETLINK_USERSOCK].groups = groups; + nl_table[NETLINK_USERSOCK].listeners = listeners; + nl_table[NETLINK_USERSOCK].module = THIS_MODULE; + nl_table[NETLINK_USERSOCK].registered = 1; + + netlink_table_ungrab(); +} + static struct pernet_operations __net_initdata netlink_net_ops = { .init = netlink_net_init, .exit = netlink_net_exit, @@ -2174,6 +2170,8 @@ static int __init netlink_proto_init(void) hash->rehash_time = jiffies; } + netlink_add_usersock_entry(); + sock_register(&netlink_family_ops); register_pernet_subsys(&netlink_net_ops); /* The netlink device handler may be needed early. */ diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index 26ed3e8587c..1781d99145e 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -547,8 +547,20 @@ static int genl_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) info.userhdr = nlmsg_data(nlh) + GENL_HDRLEN; info.attrs = family->attrbuf; genl_info_net_set(&info, net); + memset(&info.user_ptr, 0, sizeof(info.user_ptr)); - return ops->doit(skb, &info); + if (family->pre_doit) { + err = family->pre_doit(ops, skb, &info); + if (err) + return err; + } + + err = ops->doit(skb, &info); + + if (family->post_doit) + family->post_doit(ops, skb, &info); + + return err; } static void genl_rcv(struct sk_buff *skb) diff --git a/net/nonet.c b/net/nonet.c index 92e76640c7c..b1a73fda9c1 100644 --- a/net/nonet.c +++ b/net/nonet.c @@ -22,4 +22,5 @@ static int sock_no_open(struct inode *irrelevant, struct file *dontcare) const struct file_operations bad_sock_fops = { .owner = THIS_MODULE, .open = sock_no_open, + .llseek = noop_llseek, }; diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 9a17f28b125..3616f27b9d4 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -488,7 +488,7 @@ retry: skb->dev = dev; skb->priority = sk->sk_priority; skb->mark = sk->sk_mark; - err = sock_tx_timestamp(msg, sk, skb_tx(skb)); + err = sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags); if (err < 0) goto out_unlock; @@ -1209,7 +1209,7 @@ static int packet_snd(struct socket *sock, err = skb_copy_datagram_from_iovec(skb, offset, msg->msg_iov, 0, len); if (err) goto out_free; - err = sock_tx_timestamp(msg, sk, skb_tx(skb)); + err = sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags); if (err < 0) goto out_free; diff --git a/net/phonet/Kconfig b/net/phonet/Kconfig index 6ec7d55b176..0d9b8a220a7 100644 --- a/net/phonet/Kconfig +++ b/net/phonet/Kconfig @@ -14,3 +14,15 @@ config PHONET To compile this driver as a module, choose M here: the module will be called phonet. If unsure, say N. + +config PHONET_PIPECTRLR + bool "Phonet Pipe Controller (EXPERIMENTAL)" + depends on PHONET && EXPERIMENTAL + default N + help + The Pipe Controller implementation in Phonet stack to support Pipe + data with Nokia Slim modems like WG2.5 used on ST-Ericsson U8500 + platform. + + This option is incompatible with older Nokia modems. + Say N here unless you really know what you are doing. diff --git a/net/phonet/af_phonet.c b/net/phonet/af_phonet.c index 73aee7f2fcd..fd95beb72f5 100644 --- a/net/phonet/af_phonet.c +++ b/net/phonet/af_phonet.c @@ -251,6 +251,16 @@ int pn_skb_send(struct sock *sk, struct sk_buff *skb, else if (phonet_address_lookup(net, daddr) == 0) { dev = phonet_device_get(net); skb->pkt_type = PACKET_LOOPBACK; + } else if (pn_sockaddr_get_object(target) == 0) { + /* Resource routing (small race until phonet_rcv()) */ + struct sock *sk = pn_find_sock_by_res(net, + target->spn_resource); + if (sk) { + sock_put(sk); + dev = phonet_device_get(net); + skb->pkt_type = PACKET_LOOPBACK; + } else + dev = phonet_route_output(net, daddr); } else dev = phonet_route_output(net, daddr); @@ -383,6 +393,13 @@ static int phonet_rcv(struct sk_buff *skb, struct net_device *dev, goto out; } + /* resource routing */ + if (pn_sockaddr_get_object(&sa) == 0) { + struct sock *sk = pn_find_sock_by_res(net, sa.spn_resource); + if (sk) + return sk_receive_skb(sk, skb, 0); + } + /* check if we are the destination */ if (phonet_address_lookup(net, pn_sockaddr_get_addr(&sa)) == 0) { /* Phonet packet input */ diff --git a/net/phonet/datagram.c b/net/phonet/datagram.c index 1bd38db4fe1..2f032381bd4 100644 --- a/net/phonet/datagram.c +++ b/net/phonet/datagram.c @@ -52,6 +52,19 @@ static int pn_ioctl(struct sock *sk, int cmd, unsigned long arg) answ = skb ? skb->len : 0; release_sock(sk); return put_user(answ, (int __user *)arg); + + case SIOCPNADDRESOURCE: + case SIOCPNDELRESOURCE: { + u32 res; + if (get_user(res, (u32 __user *)arg)) + return -EFAULT; + if (res >= 256) + return -EINVAL; + if (cmd == SIOCPNADDRESOURCE) + return pn_sock_bind_res(sk, res); + else + return pn_sock_unbind_res(sk, res); + } } return -ENOIOCTLCMD; diff --git a/net/phonet/pep.c b/net/phonet/pep.c index b2a3ae6cad7..3e60f2e4e6c 100644 --- a/net/phonet/pep.c +++ b/net/phonet/pep.c @@ -109,6 +109,210 @@ static int pep_reply(struct sock *sk, struct sk_buff *oskb, } #define PAD 0x00 + +#ifdef CONFIG_PHONET_PIPECTRLR +static u8 pipe_negotiate_fc(u8 *host_fc, u8 *remote_fc, int len) +{ + int i, j; + u8 base_fc, final_fc; + + for (i = 0; i < len; i++) { + base_fc = host_fc[i]; + for (j = 0; j < len; j++) { + if (remote_fc[j] == base_fc) { + final_fc = base_fc; + goto done; + } + } + } + return -EINVAL; + +done: + return final_fc; + +} + +static int pipe_get_flow_info(struct sock *sk, struct sk_buff *skb, + u8 *pref_rx_fc, u8 *req_tx_fc) +{ + struct pnpipehdr *hdr; + u8 n_sb; + + if (!pskb_may_pull(skb, sizeof(*hdr) + 4)) + return -EINVAL; + + hdr = pnp_hdr(skb); + n_sb = hdr->data[4]; + + __skb_pull(skb, sizeof(*hdr) + 4); + while (n_sb > 0) { + u8 type, buf[3], len = sizeof(buf); + u8 *data = pep_get_sb(skb, &type, &len, buf); + + if (data == NULL) + return -EINVAL; + + switch (type) { + case PN_PIPE_SB_REQUIRED_FC_TX: + if (len < 3 || (data[2] | data[3] | data[4]) > 3) + break; + req_tx_fc[0] = data[2]; + req_tx_fc[1] = data[3]; + req_tx_fc[2] = data[4]; + break; + + case PN_PIPE_SB_PREFERRED_FC_RX: + if (len < 3 || (data[2] | data[3] | data[4]) > 3) + break; + pref_rx_fc[0] = data[2]; + pref_rx_fc[1] = data[3]; + pref_rx_fc[2] = data[4]; + break; + + } + n_sb--; + } + return 0; +} + +static int pipe_handler_send_req(struct sock *sk, u8 utid, + u8 msg_id, gfp_t priority) +{ + int len; + struct pnpipehdr *ph; + struct sk_buff *skb; + struct pep_sock *pn = pep_sk(sk); + + static const u8 data[4] = { + PAD, PAD, PAD, PAD, + }; + + switch (msg_id) { + case PNS_PEP_CONNECT_REQ: + len = sizeof(data); + break; + + case PNS_PEP_DISCONNECT_REQ: + case PNS_PEP_ENABLE_REQ: + case PNS_PEP_DISABLE_REQ: + len = 0; + break; + + default: + return -EINVAL; + } + + skb = alloc_skb(MAX_PNPIPE_HEADER + len, priority); + if (!skb) + return -ENOMEM; + skb_set_owner_w(skb, sk); + + skb_reserve(skb, MAX_PNPIPE_HEADER); + if (len) { + __skb_put(skb, len); + skb_copy_to_linear_data(skb, data, len); + } + __skb_push(skb, sizeof(*ph)); + skb_reset_transport_header(skb); + ph = pnp_hdr(skb); + ph->utid = utid; + ph->message_id = msg_id; + ph->pipe_handle = pn->pipe_handle; + ph->error_code = PN_PIPE_NO_ERROR; + + return pn_skb_send(sk, skb, &pn->remote_pep); +} + +static int pipe_handler_send_created_ind(struct sock *sk, + u8 utid, u8 msg_id) +{ + int err_code; + struct pnpipehdr *ph; + struct sk_buff *skb; + + struct pep_sock *pn = pep_sk(sk); + static u8 data[4] = { + 0x03, 0x04, + }; + data[2] = pn->tx_fc; + data[3] = pn->rx_fc; + + /* + * actually, below is number of sub-blocks and not error code. + * Pipe_created_ind message format does not have any + * error code field. However, the Phonet stack will always send + * an error code as part of pnpipehdr. So, use that err_code to + * specify the number of sub-blocks. + */ + err_code = 0x01; + + skb = alloc_skb(MAX_PNPIPE_HEADER + sizeof(data), GFP_ATOMIC); + if (!skb) + return -ENOMEM; + skb_set_owner_w(skb, sk); + + skb_reserve(skb, MAX_PNPIPE_HEADER); + __skb_put(skb, sizeof(data)); + skb_copy_to_linear_data(skb, data, sizeof(data)); + __skb_push(skb, sizeof(*ph)); + skb_reset_transport_header(skb); + ph = pnp_hdr(skb); + ph->utid = utid; + ph->message_id = msg_id; + ph->pipe_handle = pn->pipe_handle; + ph->error_code = err_code; + + return pn_skb_send(sk, skb, &pn->remote_pep); +} + +static int pipe_handler_send_ind(struct sock *sk, u8 utid, u8 msg_id) +{ + int err_code; + struct pnpipehdr *ph; + struct sk_buff *skb; + struct pep_sock *pn = pep_sk(sk); + + /* + * actually, below is a filler. + * Pipe_enabled/disabled_ind message format does not have any + * error code field. However, the Phonet stack will always send + * an error code as part of pnpipehdr. So, use that err_code to + * specify the filler value. + */ + err_code = 0x0; + + skb = alloc_skb(MAX_PNPIPE_HEADER, GFP_ATOMIC); + if (!skb) + return -ENOMEM; + skb_set_owner_w(skb, sk); + + skb_reserve(skb, MAX_PNPIPE_HEADER); + __skb_push(skb, sizeof(*ph)); + skb_reset_transport_header(skb); + ph = pnp_hdr(skb); + ph->utid = utid; + ph->message_id = msg_id; + ph->pipe_handle = pn->pipe_handle; + ph->error_code = err_code; + + return pn_skb_send(sk, skb, &pn->remote_pep); +} + +static int pipe_handler_enable_pipe(struct sock *sk, int enable) +{ + int utid, req; + + if (enable) { + utid = PNS_PIPE_ENABLE_UTID; + req = PNS_PEP_ENABLE_REQ; + } else { + utid = PNS_PIPE_DISABLE_UTID; + req = PNS_PEP_DISABLE_REQ; + } + return pipe_handler_send_req(sk, utid, req, GFP_ATOMIC); +} +#endif + static int pep_accept_conn(struct sock *sk, struct sk_buff *skb) { static const u8 data[20] = { @@ -192,7 +396,11 @@ static int pipe_snd_status(struct sock *sk, u8 type, u8 status, gfp_t priority) ph->data[3] = PAD; ph->data[4] = status; +#ifdef CONFIG_PHONET_PIPECTRLR + return pn_skb_send(sk, skb, &pn->remote_pep); +#else return pn_skb_send(sk, skb, &pipe_srv); +#endif } /* Send our RX flow control information to the sender. @@ -225,12 +433,13 @@ static void pipe_grant_credits(struct sock *sk) static int pipe_rcv_status(struct sock *sk, struct sk_buff *skb) { struct pep_sock *pn = pep_sk(sk); - struct pnpipehdr *hdr = pnp_hdr(skb); + struct pnpipehdr *hdr; int wake = 0; if (!pskb_may_pull(skb, sizeof(*hdr) + 4)) return -EINVAL; + hdr = pnp_hdr(skb); if (hdr->data[0] != PN_PEP_TYPE_COMMON) { LIMIT_NETDEBUG(KERN_DEBUG"Phonet unknown PEP type: %u\n", (unsigned)hdr->data[0]); @@ -323,11 +532,35 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb) sk->sk_state_change(sk); break; +#ifdef CONFIG_PHONET_PIPECTRLR + case PNS_PEP_DISCONNECT_RESP: + pn->pipe_state = PIPE_IDLE; + sk->sk_state = TCP_CLOSE; + break; +#endif + case PNS_PEP_ENABLE_REQ: /* Wait for PNS_PIPE_(ENABLED|REDIRECTED)_IND */ pep_reply(sk, skb, PN_PIPE_NO_ERROR, NULL, 0, GFP_ATOMIC); break; +#ifdef CONFIG_PHONET_PIPECTRLR + case PNS_PEP_ENABLE_RESP: + pn->pipe_state = PIPE_ENABLED; + pipe_handler_send_ind(sk, PNS_PIPE_ENABLED_IND_UTID, + PNS_PIPE_ENABLED_IND); + + if (!pn_flow_safe(pn->tx_fc)) { + atomic_set(&pn->tx_credits, 1); + sk->sk_write_space(sk); + } + if (sk->sk_state == TCP_ESTABLISHED) + break; /* Nothing to do */ + sk->sk_state = TCP_ESTABLISHED; + pipe_grant_credits(sk); + break; +#endif + case PNS_PEP_RESET_REQ: switch (hdr->state_after_reset) { case PN_PIPE_DISABLE: @@ -346,6 +579,17 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb) pep_reply(sk, skb, PN_PIPE_NO_ERROR, NULL, 0, GFP_ATOMIC); break; +#ifdef CONFIG_PHONET_PIPECTRLR + case PNS_PEP_DISABLE_RESP: + pn->pipe_state = PIPE_DISABLED; + atomic_set(&pn->tx_credits, 0); + pipe_handler_send_ind(sk, PNS_PIPE_DISABLED_IND_UTID, + PNS_PIPE_DISABLED_IND); + sk->sk_state = TCP_SYN_RECV; + pn->rx_credits = 0; + break; +#endif + case PNS_PEP_CTRL_REQ: if (skb_queue_len(&pn->ctrlreq_queue) >= PNPIPE_CTRLREQ_MAX) { atomic_inc(&sk->sk_drops); @@ -437,6 +681,42 @@ static void pipe_destruct(struct sock *sk) skb_queue_purge(&pn->ctrlreq_queue); } +#ifdef CONFIG_PHONET_PIPECTRLR +static int pep_connresp_rcv(struct sock *sk, struct sk_buff *skb) +{ + struct pep_sock *pn = pep_sk(sk); + u8 host_pref_rx_fc[3] = {3, 2, 1}, host_req_tx_fc[3] = {3, 2, 1}; + u8 remote_pref_rx_fc[3], remote_req_tx_fc[3]; + u8 negotiated_rx_fc, negotiated_tx_fc; + int ret; + + pipe_get_flow_info(sk, skb, remote_pref_rx_fc, + remote_req_tx_fc); + negotiated_tx_fc = pipe_negotiate_fc(remote_req_tx_fc, + host_pref_rx_fc, + sizeof(host_pref_rx_fc)); + negotiated_rx_fc = pipe_negotiate_fc(host_req_tx_fc, + remote_pref_rx_fc, + sizeof(host_pref_rx_fc)); + + pn->pipe_state = PIPE_DISABLED; + sk->sk_state = TCP_SYN_RECV; + sk->sk_backlog_rcv = pipe_do_rcv; + sk->sk_destruct = pipe_destruct; + pn->rx_credits = 0; + pn->rx_fc = negotiated_rx_fc; + pn->tx_fc = negotiated_tx_fc; + sk->sk_state_change(sk); + + ret = pipe_handler_send_created_ind(sk, + PNS_PIPE_CREATED_IND_UTID, + PNS_PIPE_CREATED_IND + ); + + return ret; +} +#endif + static int pep_connreq_rcv(struct sock *sk, struct sk_buff *skb) { struct sock *newsk; @@ -600,6 +880,12 @@ static int pep_do_rcv(struct sock *sk, struct sk_buff *skb) err = pep_connreq_rcv(sk, skb); break; +#ifdef CONFIG_PHONET_PIPECTRLR + case PNS_PEP_CONNECT_RESP: + err = pep_connresp_rcv(sk, skb); + break; +#endif + case PNS_PEP_DISCONNECT_REQ: pep_reply(sk, skb, PN_PIPE_NO_ERROR, NULL, 0, GFP_ATOMIC); break; @@ -620,6 +906,28 @@ drop: return err; } +static int pipe_do_remove(struct sock *sk) +{ + struct pep_sock *pn = pep_sk(sk); + struct pnpipehdr *ph; + struct sk_buff *skb; + + skb = alloc_skb(MAX_PNPIPE_HEADER, GFP_KERNEL); + if (!skb) + return -ENOMEM; + + skb_reserve(skb, MAX_PNPIPE_HEADER); + __skb_push(skb, sizeof(*ph)); + skb_reset_transport_header(skb); + ph = pnp_hdr(skb); + ph->utid = 0; + ph->message_id = PNS_PIPE_REMOVE_REQ; + ph->pipe_handle = pn->pipe_handle; + ph->data[0] = PAD; + + return pn_skb_send(sk, skb, &pipe_srv); +} + /* associated socket ceases to exist */ static void pep_sock_close(struct sock *sk, long timeout) { @@ -638,7 +946,22 @@ static void pep_sock_close(struct sock *sk, long timeout) sk_for_each_safe(sknode, p, n, &pn->ackq) sk_del_node_init(sknode); sk->sk_state = TCP_CLOSE; + } else if ((1 << sk->sk_state) & (TCPF_SYN_RECV|TCPF_ESTABLISHED)) + /* Forcefully remove dangling Phonet pipe */ + pipe_do_remove(sk); + +#ifdef CONFIG_PHONET_PIPECTRLR + if (pn->pipe_state != PIPE_IDLE) { + /* send pep disconnect request */ + pipe_handler_send_req(sk, + PNS_PEP_DISCONNECT_UTID, PNS_PEP_DISCONNECT_REQ, + GFP_KERNEL); + + pn->pipe_state = PIPE_IDLE; + sk->sk_state = TCP_CLOSE; } +#endif + ifindex = pn->ifindex; pn->ifindex = 0; release_sock(sk); @@ -715,6 +1038,20 @@ out: return newsk; } +#ifdef CONFIG_PHONET_PIPECTRLR +static int pep_sock_connect(struct sock *sk, struct sockaddr *addr, int len) +{ + struct pep_sock *pn = pep_sk(sk); + struct sockaddr_pn *spn = (struct sockaddr_pn *)addr; + + memcpy(&pn->remote_pep, spn, sizeof(struct sockaddr_pn)); + + return pipe_handler_send_req(sk, + PNS_PEP_CONNECT_UTID, PNS_PEP_CONNECT_REQ, + GFP_ATOMIC); +} +#endif + static int pep_ioctl(struct sock *sk, int cmd, unsigned long arg) { struct pep_sock *pn = pep_sk(sk); @@ -766,6 +1103,18 @@ static int pep_setsockopt(struct sock *sk, int level, int optname, lock_sock(sk); switch (optname) { +#ifdef CONFIG_PHONET_PIPECTRLR + case PNPIPE_PIPE_HANDLE: + if (val) { + if (pn->pipe_state > PIPE_IDLE) { + err = -EFAULT; + break; + } + pn->pipe_handle = val; + break; + } +#endif + case PNPIPE_ENCAP: if (val && val != PNPIPE_ENCAP_IP) { err = -EINVAL; @@ -791,6 +1140,17 @@ static int pep_setsockopt(struct sock *sk, int level, int optname, err = 0; } goto out_norel; + +#ifdef CONFIG_PHONET_PIPECTRLR + case PNPIPE_ENABLE: + if (pn->pipe_state <= PIPE_IDLE) { + err = -ENOTCONN; + break; + } + err = pipe_handler_enable_pipe(sk, val); + break; +#endif + default: err = -ENOPROTOOPT; } @@ -815,9 +1175,19 @@ static int pep_getsockopt(struct sock *sk, int level, int optname, case PNPIPE_ENCAP: val = pn->ifindex ? PNPIPE_ENCAP_IP : PNPIPE_ENCAP_NONE; break; + case PNPIPE_IFINDEX: val = pn->ifindex; break; + +#ifdef CONFIG_PHONET_PIPECTRLR + case PNPIPE_ENABLE: + if (pn->pipe_state <= PIPE_IDLE) + return -ENOTCONN; + val = pn->pipe_state != PIPE_DISABLED; + break; +#endif + default: return -ENOPROTOOPT; } @@ -834,6 +1204,7 @@ static int pipe_skb_send(struct sock *sk, struct sk_buff *skb) { struct pep_sock *pn = pep_sk(sk); struct pnpipehdr *ph; + int err; if (pn_flow_safe(pn->tx_fc) && !atomic_add_unless(&pn->tx_credits, -1, 0)) { @@ -851,8 +1222,16 @@ static int pipe_skb_send(struct sock *sk, struct sk_buff *skb) } else ph->message_id = PNS_PIPE_DATA; ph->pipe_handle = pn->pipe_handle; +#ifdef CONFIG_PHONET_PIPECTRLR + err = pn_skb_send(sk, skb, &pn->remote_pep); +#else + err = pn_skb_send(sk, skb, &pipe_srv); +#endif + + if (err && pn_flow_safe(pn->tx_fc)) + atomic_inc(&pn->tx_credits); + return err; - return pn_skb_send(sk, skb, &pipe_srv); } static int pep_sendmsg(struct kiocb *iocb, struct sock *sk, @@ -872,7 +1251,7 @@ static int pep_sendmsg(struct kiocb *iocb, struct sock *sk, skb = sock_alloc_send_skb(sk, MAX_PNPIPE_HEADER + len, flags & MSG_DONTWAIT, &err); if (!skb) - return -ENOBUFS; + return err; skb_reserve(skb, MAX_PHONET_HEADER + 3); err = memcpy_fromiovec(skb_put(skb, len), msg->msg_iov, len); @@ -1044,6 +1423,8 @@ static void pep_sock_unhash(struct sock *sk) struct sock *skparent = NULL; lock_sock(sk); + +#ifndef CONFIG_PHONET_PIPECTRLR if ((1 << sk->sk_state) & ~(TCPF_CLOSE|TCPF_LISTEN)) { skparent = pn->listener; release_sock(sk); @@ -1053,6 +1434,7 @@ static void pep_sock_unhash(struct sock *sk) sk_del_node_init(sk); sk = skparent; } +#endif /* Unhash a listening sock only when it is closed * and all of its active connected pipes are closed. */ if (hlist_empty(&pn->hlist)) @@ -1066,6 +1448,9 @@ static void pep_sock_unhash(struct sock *sk) static struct proto pep_proto = { .close = pep_sock_close, .accept = pep_sock_accept, +#ifdef CONFIG_PHONET_PIPECTRLR + .connect = pep_sock_connect, +#endif .ioctl = pep_ioctl, .init = pep_init, .setsockopt = pep_setsockopt, diff --git a/net/phonet/pn_dev.c b/net/phonet/pn_dev.c index b18e48fae97..947038ddd04 100644 --- a/net/phonet/pn_dev.c +++ b/net/phonet/pn_dev.c @@ -292,8 +292,7 @@ static void phonet_route_autodel(struct net_device *dev) if (bitmap_empty(deleted, 64)) return; /* short-circuit RCU */ synchronize_rcu(); - for (i = find_first_bit(deleted, 64); i < 64; - i = find_next_bit(deleted, 64, i + 1)) { + for_each_set_bit(i, deleted, 64) { rtm_phonet_notify(RTM_DELROUTE, dev, i); dev_put(dev); } @@ -374,6 +373,7 @@ int __init phonet_device_init(void) if (err) return err; + proc_net_fops_create(&init_net, "pnresource", 0, &pn_res_seq_fops); register_netdevice_notifier(&phonet_device_notifier); err = phonet_netlink_register(); if (err) @@ -386,6 +386,7 @@ void phonet_device_exit(void) rtnl_unregister_all(PF_PHONET); unregister_netdevice_notifier(&phonet_device_notifier); unregister_pernet_device(&phonet_net_ops); + proc_net_remove(&init_net, "pnresource"); } int phonet_route_add(struct net_device *dev, u8 daddr) diff --git a/net/phonet/socket.c b/net/phonet/socket.c index 6e9848bf037..25f746d20c1 100644 --- a/net/phonet/socket.c +++ b/net/phonet/socket.c @@ -158,6 +158,7 @@ void pn_sock_unhash(struct sock *sk) spin_lock_bh(&pnsocks.lock); sk_del_node_init(sk); spin_unlock_bh(&pnsocks.lock); + pn_sock_unbind_all_res(sk); } EXPORT_SYMBOL(pn_sock_unhash); @@ -224,6 +225,101 @@ static int pn_socket_autobind(struct socket *sock) return 0; /* socket was already bound */ } +#ifdef CONFIG_PHONET_PIPECTRLR +static int pn_socket_connect(struct socket *sock, struct sockaddr *addr, + int len, int flags) +{ + struct sock *sk = sock->sk; + struct sockaddr_pn *spn = (struct sockaddr_pn *)addr; + long timeo; + int err; + + if (len < sizeof(struct sockaddr_pn)) + return -EINVAL; + if (spn->spn_family != AF_PHONET) + return -EAFNOSUPPORT; + + lock_sock(sk); + + switch (sock->state) { + case SS_UNCONNECTED: + sk->sk_state = TCP_CLOSE; + break; + case SS_CONNECTING: + switch (sk->sk_state) { + case TCP_SYN_RECV: + sock->state = SS_CONNECTED; + err = -EISCONN; + goto out; + case TCP_CLOSE: + err = -EALREADY; + if (flags & O_NONBLOCK) + goto out; + goto wait_connect; + } + break; + case SS_CONNECTED: + switch (sk->sk_state) { + case TCP_SYN_RECV: + err = -EISCONN; + goto out; + case TCP_CLOSE: + sock->state = SS_UNCONNECTED; + break; + } + break; + case SS_DISCONNECTING: + case SS_FREE: + break; + } + sk->sk_state = TCP_CLOSE; + sk_stream_kill_queues(sk); + + sock->state = SS_CONNECTING; + err = sk->sk_prot->connect(sk, addr, len); + if (err < 0) { + sock->state = SS_UNCONNECTED; + sk->sk_state = TCP_CLOSE; + goto out; + } + + err = -EINPROGRESS; +wait_connect: + if (sk->sk_state != TCP_SYN_RECV && (flags & O_NONBLOCK)) + goto out; + + timeo = sock_sndtimeo(sk, flags & O_NONBLOCK); + release_sock(sk); + + err = -ERESTARTSYS; + timeo = wait_event_interruptible_timeout(*sk_sleep(sk), + sk->sk_state != TCP_CLOSE, + timeo); + + lock_sock(sk); + if (timeo < 0) + goto out; /* -ERESTARTSYS */ + + err = -ETIMEDOUT; + if (timeo == 0 && sk->sk_state != TCP_SYN_RECV) + goto out; + + if (sk->sk_state != TCP_SYN_RECV) { + sock->state = SS_UNCONNECTED; + err = sock_error(sk); + if (!err) + err = -ECONNREFUSED; + goto out; + } + sock->state = SS_CONNECTED; + err = 0; + +out: + release_sock(sk); + return err; +} +#endif + static int pn_socket_accept(struct socket *sock, struct socket *newsock, int flags) { @@ -281,7 +377,9 @@ static unsigned int pn_socket_poll(struct file *file, struct socket *sock, if (!mask && sk->sk_state == TCP_CLOSE_WAIT) return POLLHUP; - if (sk->sk_state == TCP_ESTABLISHED && atomic_read(&pn->tx_credits)) + if (sk->sk_state == TCP_ESTABLISHED && + atomic_read(&sk->sk_wmem_alloc) < sk->sk_sndbuf && + atomic_read(&pn->tx_credits)) mask |= POLLOUT | POLLWRNORM | POLLWRBAND; return mask; @@ -390,7 +488,11 @@ const struct proto_ops phonet_stream_ops = { .owner = THIS_MODULE, .release = pn_socket_release, .bind = pn_socket_bind, +#ifdef CONFIG_PHONET_PIPECTRLR + .connect = pn_socket_connect, +#else .connect = sock_no_connect, +#endif .socketpair = sock_no_socketpair, .accept = pn_socket_accept, .getname = pn_socket_getname, @@ -563,3 +665,188 @@ const struct file_operations pn_sock_seq_fops = { .release = seq_release_net, }; #endif + +static struct { + struct sock *sk[256]; +} pnres; + +/* + * Find and hold socket based on resource. + */ +struct sock *pn_find_sock_by_res(struct net *net, u8 res) +{ + struct sock *sk; + + if (!net_eq(net, &init_net)) + return NULL; + + rcu_read_lock(); + sk = rcu_dereference(pnres.sk[res]); + if (sk) + sock_hold(sk); + rcu_read_unlock(); + return sk; +} + +static DEFINE_MUTEX(resource_mutex); + +int pn_sock_bind_res(struct sock *sk, u8 res) +{ + int ret = -EADDRINUSE; + + if (!net_eq(sock_net(sk), &init_net)) + return -ENOIOCTLCMD; + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + if (pn_socket_autobind(sk->sk_socket)) + return -EAGAIN; + + mutex_lock(&resource_mutex); + if (pnres.sk[res] == NULL) { + sock_hold(sk); + rcu_assign_pointer(pnres.sk[res], sk); + ret = 0; + } + mutex_unlock(&resource_mutex); + return ret; +} + +int pn_sock_unbind_res(struct sock *sk, u8 res) +{ + int ret = -ENOENT; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + mutex_lock(&resource_mutex); + if (pnres.sk[res] == sk) { + rcu_assign_pointer(pnres.sk[res], NULL); + ret = 0; + } + mutex_unlock(&resource_mutex); + + if (ret == 0) { + synchronize_rcu(); + sock_put(sk); + } + return ret; +} + +void pn_sock_unbind_all_res(struct sock *sk) +{ + unsigned res, match = 0; + + mutex_lock(&resource_mutex); + for (res = 0; res < 256; res++) { + if (pnres.sk[res] == sk) { + rcu_assign_pointer(pnres.sk[res], NULL); + match++; + } + } + mutex_unlock(&resource_mutex); + + if (match == 0) + return; + synchronize_rcu(); + while (match > 0) { + sock_put(sk); + match--; + } +} + +#ifdef CONFIG_PROC_FS +static struct sock **pn_res_get_idx(struct seq_file *seq, loff_t pos) +{ + struct net *net = seq_file_net(seq); + unsigned i; + + if (!net_eq(net, &init_net)) + return NULL; + + for (i = 0; i < 256; i++) { + if (pnres.sk[i] == NULL) + continue; + if (!pos) + return pnres.sk + i; + pos--; + } + return NULL; +} + +static struct sock **pn_res_get_next(struct seq_file *seq, struct sock **sk) +{ + struct net *net = seq_file_net(seq); + unsigned i; + + BUG_ON(!net_eq(net, &init_net)); + + for (i = (sk - pnres.sk) + 1; i < 256; i++) + if (pnres.sk[i]) + return pnres.sk + i; + return NULL; +} + +static void *pn_res_seq_start(struct seq_file *seq, loff_t *pos) + __acquires(resource_mutex) +{ + mutex_lock(&resource_mutex); + return *pos ? pn_res_get_idx(seq, *pos - 1) : SEQ_START_TOKEN; +} + +static void *pn_res_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct sock **sk; + + if (v == SEQ_START_TOKEN) + sk = pn_res_get_idx(seq, 0); + else + sk = pn_res_get_next(seq, v); + (*pos)++; + return sk; +} + +static void pn_res_seq_stop(struct seq_file *seq, void *v) + __releases(resource_mutex) +{ + mutex_unlock(&resource_mutex); +} + +static int pn_res_seq_show(struct seq_file *seq, void *v) +{ + int len; + + if (v == SEQ_START_TOKEN) + seq_printf(seq, "%s%n", "rs uid inode", &len); + else { + struct sock **psk = v; + struct sock *sk = *psk; + + seq_printf(seq, "%02X %5d %lu%n", + (int) (psk - pnres.sk), sock_i_uid(sk), + sock_i_ino(sk), &len); + } + seq_printf(seq, "%*s\n", 63 - len, ""); + return 0; +} + +static const struct seq_operations pn_res_seq_ops = { + .start = pn_res_seq_start, + .next = pn_res_seq_next, + .stop = pn_res_seq_stop, + .show = pn_res_seq_show, +}; + +static int pn_res_open(struct inode *inode, struct file *file) +{ + return seq_open_net(inode, file, &pn_res_seq_ops, + sizeof(struct seq_net_private)); +} + +const struct file_operations pn_res_seq_fops = { + .owner = THIS_MODULE, + .open = pn_res_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_net, +}; +#endif diff --git a/net/rds/af_rds.c b/net/rds/af_rds.c index aebfecbdb84..bb6ad81b671 100644 --- a/net/rds/af_rds.c +++ b/net/rds/af_rds.c @@ -39,7 +39,15 @@ #include <net/sock.h> #include "rds.h" -#include "rdma.h" + +char *rds_str_array(char **array, size_t elements, size_t index) +{ + if ((index < elements) && array[index]) + return array[index]; + else + return "unknown"; +} +EXPORT_SYMBOL(rds_str_array); /* this is just used for stats gathering :/ */ static DEFINE_SPINLOCK(rds_sock_lock); @@ -62,7 +70,7 @@ static int rds_release(struct socket *sock) struct rds_sock *rs; unsigned long flags; - if (sk == NULL) + if (!sk) goto out; rs = rds_sk_to_rs(sk); @@ -73,7 +81,15 @@ static int rds_release(struct socket *sock) * with the socket. */ rds_clear_recv_queue(rs); rds_cong_remove_socket(rs); + + /* + * the binding lookup hash uses rcu, we need to + * make sure we sychronize_rcu before we free our + * entry + */ rds_remove_bound(rs); + synchronize_rcu(); + rds_send_drop_to(rs, NULL); rds_rdma_drop_keys(rs); rds_notify_queue_get(rs, NULL); @@ -83,6 +99,8 @@ static int rds_release(struct socket *sock) rds_sock_count--; spin_unlock_irqrestore(&rds_sock_lock, flags); + rds_trans_put(rs->rs_transport); + sock->sk = NULL; sock_put(sk); out: @@ -514,7 +532,7 @@ out: spin_unlock_irqrestore(&rds_sock_lock, flags); } -static void __exit rds_exit(void) +static void rds_exit(void) { sock_unregister(rds_family_ops.family); proto_unregister(&rds_proto); @@ -529,7 +547,7 @@ static void __exit rds_exit(void) } module_exit(rds_exit); -static int __init rds_init(void) +static int rds_init(void) { int ret; diff --git a/net/rds/bind.c b/net/rds/bind.c index 5d95fc007f1..2f6b3fcc79f 100644 --- a/net/rds/bind.c +++ b/net/rds/bind.c @@ -34,45 +34,52 @@ #include <net/sock.h> #include <linux/in.h> #include <linux/if_arp.h> +#include <linux/jhash.h> #include "rds.h" -/* - * XXX this probably still needs more work.. no INADDR_ANY, and rbtrees aren't - * particularly zippy. - * - * This is now called for every incoming frame so we arguably care much more - * about it than we used to. - */ +#define BIND_HASH_SIZE 1024 +static struct hlist_head bind_hash_table[BIND_HASH_SIZE]; static DEFINE_SPINLOCK(rds_bind_lock); -static struct rb_root rds_bind_tree = RB_ROOT; -static struct rds_sock *rds_bind_tree_walk(__be32 addr, __be16 port, - struct rds_sock *insert) +static struct hlist_head *hash_to_bucket(__be32 addr, __be16 port) +{ + return bind_hash_table + (jhash_2words((u32)addr, (u32)port, 0) & + (BIND_HASH_SIZE - 1)); +} + +static struct rds_sock *rds_bind_lookup(__be32 addr, __be16 port, + struct rds_sock *insert) { - struct rb_node **p = &rds_bind_tree.rb_node; - struct rb_node *parent = NULL; struct rds_sock *rs; + struct hlist_node *node; + struct hlist_head *head = hash_to_bucket(addr, port); u64 cmp; u64 needle = ((u64)be32_to_cpu(addr) << 32) | be16_to_cpu(port); - while (*p) { - parent = *p; - rs = rb_entry(parent, struct rds_sock, rs_bound_node); - + rcu_read_lock(); + hlist_for_each_entry_rcu(rs, node, head, rs_bound_node) { cmp = ((u64)be32_to_cpu(rs->rs_bound_addr) << 32) | be16_to_cpu(rs->rs_bound_port); - if (needle < cmp) - p = &(*p)->rb_left; - else if (needle > cmp) - p = &(*p)->rb_right; - else + if (cmp == needle) { + rcu_read_unlock(); return rs; + } } + rcu_read_unlock(); if (insert) { - rb_link_node(&insert->rs_bound_node, parent, p); - rb_insert_color(&insert->rs_bound_node, &rds_bind_tree); + /* + * make sure our addr and port are set before + * we are added to the list, other people + * in rcu will find us as soon as the + * hlist_add_head_rcu is done + */ + insert->rs_bound_addr = addr; + insert->rs_bound_port = port; + rds_sock_addref(insert); + + hlist_add_head_rcu(&insert->rs_bound_node, head); } return NULL; } @@ -86,15 +93,13 @@ static struct rds_sock *rds_bind_tree_walk(__be32 addr, __be16 port, struct rds_sock *rds_find_bound(__be32 addr, __be16 port) { struct rds_sock *rs; - unsigned long flags; - spin_lock_irqsave(&rds_bind_lock, flags); - rs = rds_bind_tree_walk(addr, port, NULL); + rs = rds_bind_lookup(addr, port, NULL); + if (rs && !sock_flag(rds_rs_to_sk(rs), SOCK_DEAD)) rds_sock_addref(rs); else rs = NULL; - spin_unlock_irqrestore(&rds_bind_lock, flags); rdsdebug("returning rs %p for %pI4:%u\n", rs, &addr, ntohs(port)); @@ -121,22 +126,15 @@ static int rds_add_bound(struct rds_sock *rs, __be32 addr, __be16 *port) do { if (rover == 0) rover++; - if (rds_bind_tree_walk(addr, cpu_to_be16(rover), rs) == NULL) { - *port = cpu_to_be16(rover); + if (!rds_bind_lookup(addr, cpu_to_be16(rover), rs)) { + *port = rs->rs_bound_port; ret = 0; + rdsdebug("rs %p binding to %pI4:%d\n", + rs, &addr, (int)ntohs(*port)); break; } } while (rover++ != last); - if (ret == 0) { - rs->rs_bound_addr = addr; - rs->rs_bound_port = *port; - rds_sock_addref(rs); - - rdsdebug("rs %p binding to %pI4:%d\n", - rs, &addr, (int)ntohs(*port)); - } - spin_unlock_irqrestore(&rds_bind_lock, flags); return ret; @@ -153,7 +151,7 @@ void rds_remove_bound(struct rds_sock *rs) rs, &rs->rs_bound_addr, ntohs(rs->rs_bound_port)); - rb_erase(&rs->rs_bound_node, &rds_bind_tree); + hlist_del_init_rcu(&rs->rs_bound_node); rds_sock_put(rs); rs->rs_bound_addr = 0; } @@ -184,7 +182,7 @@ int rds_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) goto out; trans = rds_trans_get_preferred(sin->sin_addr.s_addr); - if (trans == NULL) { + if (!trans) { ret = -EADDRNOTAVAIL; rds_remove_bound(rs); if (printk_ratelimit()) @@ -198,5 +196,9 @@ int rds_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) out: release_sock(sk); + + /* we might have called rds_remove_bound on error */ + if (ret) + synchronize_rcu(); return ret; } diff --git a/net/rds/cong.c b/net/rds/cong.c index 0871a29f078..75ea686f27d 100644 --- a/net/rds/cong.c +++ b/net/rds/cong.c @@ -141,7 +141,7 @@ static struct rds_cong_map *rds_cong_from_addr(__be32 addr) unsigned long flags; map = kzalloc(sizeof(struct rds_cong_map), GFP_KERNEL); - if (map == NULL) + if (!map) return NULL; map->m_addr = addr; @@ -159,7 +159,7 @@ static struct rds_cong_map *rds_cong_from_addr(__be32 addr) ret = rds_cong_tree_walk(addr, map); spin_unlock_irqrestore(&rds_cong_lock, flags); - if (ret == NULL) { + if (!ret) { ret = map; map = NULL; } @@ -205,7 +205,7 @@ int rds_cong_get_maps(struct rds_connection *conn) conn->c_lcong = rds_cong_from_addr(conn->c_laddr); conn->c_fcong = rds_cong_from_addr(conn->c_faddr); - if (conn->c_lcong == NULL || conn->c_fcong == NULL) + if (!(conn->c_lcong && conn->c_fcong)) return -ENOMEM; return 0; @@ -221,7 +221,7 @@ void rds_cong_queue_updates(struct rds_cong_map *map) list_for_each_entry(conn, &map->m_conn_list, c_map_item) { if (!test_and_set_bit(0, &conn->c_map_queued)) { rds_stats_inc(s_cong_update_queued); - queue_delayed_work(rds_wq, &conn->c_send_w, 0); + rds_send_xmit(conn); } } diff --git a/net/rds/connection.c b/net/rds/connection.c index 7619b671ca2..9334d892366 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -37,7 +37,6 @@ #include "rds.h" #include "loop.h" -#include "rdma.h" #define RDS_CONNECTION_HASH_BITS 12 #define RDS_CONNECTION_HASH_ENTRIES (1 << RDS_CONNECTION_HASH_BITS) @@ -63,18 +62,7 @@ static struct hlist_head *rds_conn_bucket(__be32 laddr, __be32 faddr) var |= RDS_INFO_CONNECTION_FLAG_##suffix; \ } while (0) -static inline int rds_conn_is_sending(struct rds_connection *conn) -{ - int ret = 0; - - if (!mutex_trylock(&conn->c_send_lock)) - ret = 1; - else - mutex_unlock(&conn->c_send_lock); - - return ret; -} - +/* rcu read lock must be held or the connection spinlock */ static struct rds_connection *rds_conn_lookup(struct hlist_head *head, __be32 laddr, __be32 faddr, struct rds_transport *trans) @@ -82,7 +70,7 @@ static struct rds_connection *rds_conn_lookup(struct hlist_head *head, struct rds_connection *conn, *ret = NULL; struct hlist_node *pos; - hlist_for_each_entry(conn, pos, head, c_hash_node) { + hlist_for_each_entry_rcu(conn, pos, head, c_hash_node) { if (conn->c_faddr == faddr && conn->c_laddr == laddr && conn->c_trans == trans) { ret = conn; @@ -100,7 +88,7 @@ static struct rds_connection *rds_conn_lookup(struct hlist_head *head, * and receiving over this connection again in the future. It is up to * the transport to have serialized this call with its send and recv. */ -void rds_conn_reset(struct rds_connection *conn) +static void rds_conn_reset(struct rds_connection *conn) { rdsdebug("connection %pI4 to %pI4 reset\n", &conn->c_laddr, &conn->c_faddr); @@ -129,10 +117,11 @@ static struct rds_connection *__rds_conn_create(__be32 laddr, __be32 faddr, { struct rds_connection *conn, *parent = NULL; struct hlist_head *head = rds_conn_bucket(laddr, faddr); + struct rds_transport *loop_trans; unsigned long flags; int ret; - spin_lock_irqsave(&rds_conn_lock, flags); + rcu_read_lock(); conn = rds_conn_lookup(head, laddr, faddr, trans); if (conn && conn->c_loopback && conn->c_trans != &rds_loop_transport && !is_outgoing) { @@ -143,12 +132,12 @@ static struct rds_connection *__rds_conn_create(__be32 laddr, __be32 faddr, parent = conn; conn = parent->c_passive; } - spin_unlock_irqrestore(&rds_conn_lock, flags); + rcu_read_unlock(); if (conn) goto out; conn = kmem_cache_zalloc(rds_conn_slab, gfp); - if (conn == NULL) { + if (!conn) { conn = ERR_PTR(-ENOMEM); goto out; } @@ -159,7 +148,7 @@ static struct rds_connection *__rds_conn_create(__be32 laddr, __be32 faddr, spin_lock_init(&conn->c_lock); conn->c_next_tx_seq = 1; - mutex_init(&conn->c_send_lock); + init_waitqueue_head(&conn->c_waitq); INIT_LIST_HEAD(&conn->c_send_queue); INIT_LIST_HEAD(&conn->c_retrans); @@ -175,7 +164,9 @@ static struct rds_connection *__rds_conn_create(__be32 laddr, __be32 faddr, * can bind to the destination address then we'd rather the messages * flow through loopback rather than either transport. */ - if (rds_trans_get_preferred(faddr)) { + loop_trans = rds_trans_get_preferred(faddr); + if (loop_trans) { + rds_trans_put(loop_trans); conn->c_loopback = 1; if (is_outgoing && trans->t_prefer_loopback) { /* "outgoing" connection - and the transport @@ -238,7 +229,7 @@ static struct rds_connection *__rds_conn_create(__be32 laddr, __be32 faddr, kmem_cache_free(rds_conn_slab, conn); conn = found; } else { - hlist_add_head(&conn->c_hash_node, head); + hlist_add_head_rcu(&conn->c_hash_node, head); rds_cong_add_conn(conn); rds_conn_count++; } @@ -263,21 +254,91 @@ struct rds_connection *rds_conn_create_outgoing(__be32 laddr, __be32 faddr, } EXPORT_SYMBOL_GPL(rds_conn_create_outgoing); +void rds_conn_shutdown(struct rds_connection *conn) +{ + /* shut it down unless it's down already */ + if (!rds_conn_transition(conn, RDS_CONN_DOWN, RDS_CONN_DOWN)) { + /* + * Quiesce the connection mgmt handlers before we start tearing + * things down. We don't hold the mutex for the entire + * duration of the shutdown operation, else we may be + * deadlocking with the CM handler. Instead, the CM event + * handler is supposed to check for state DISCONNECTING + */ + mutex_lock(&conn->c_cm_lock); + if (!rds_conn_transition(conn, RDS_CONN_UP, RDS_CONN_DISCONNECTING) + && !rds_conn_transition(conn, RDS_CONN_ERROR, RDS_CONN_DISCONNECTING)) { + rds_conn_error(conn, "shutdown called in state %d\n", + atomic_read(&conn->c_state)); + mutex_unlock(&conn->c_cm_lock); + return; + } + mutex_unlock(&conn->c_cm_lock); + + wait_event(conn->c_waitq, + !test_bit(RDS_IN_XMIT, &conn->c_flags)); + + conn->c_trans->conn_shutdown(conn); + rds_conn_reset(conn); + + if (!rds_conn_transition(conn, RDS_CONN_DISCONNECTING, RDS_CONN_DOWN)) { + /* This can happen - eg when we're in the middle of tearing + * down the connection, and someone unloads the rds module. + * Quite reproduceable with loopback connections. + * Mostly harmless. + */ + rds_conn_error(conn, + "%s: failed to transition to state DOWN, " + "current state is %d\n", + __func__, + atomic_read(&conn->c_state)); + return; + } + } + + /* Then reconnect if it's still live. + * The passive side of an IB loopback connection is never added + * to the conn hash, so we never trigger a reconnect on this + * conn - the reconnect is always triggered by the active peer. */ + cancel_delayed_work_sync(&conn->c_conn_w); + rcu_read_lock(); + if (!hlist_unhashed(&conn->c_hash_node)) { + rcu_read_unlock(); + rds_queue_reconnect(conn); + } else { + rcu_read_unlock(); + } +} + +/* + * Stop and free a connection. + * + * This can only be used in very limited circumstances. It assumes that once + * the conn has been shutdown that no one else is referencing the connection. + * We can only ensure this in the rmmod path in the current code. + */ void rds_conn_destroy(struct rds_connection *conn) { struct rds_message *rm, *rtmp; + unsigned long flags; rdsdebug("freeing conn %p for %pI4 -> " "%pI4\n", conn, &conn->c_laddr, &conn->c_faddr); - hlist_del_init(&conn->c_hash_node); + /* Ensure conn will not be scheduled for reconnect */ + spin_lock_irq(&rds_conn_lock); + hlist_del_init_rcu(&conn->c_hash_node); + spin_unlock_irq(&rds_conn_lock); + synchronize_rcu(); - /* wait for the rds thread to shut it down */ - atomic_set(&conn->c_state, RDS_CONN_ERROR); - cancel_delayed_work(&conn->c_conn_w); - queue_work(rds_wq, &conn->c_down_w); - flush_workqueue(rds_wq); + /* shut the connection down */ + rds_conn_drop(conn); + flush_work(&conn->c_down_w); + + /* make sure lingering queued work won't try to ref the conn */ + cancel_delayed_work_sync(&conn->c_send_w); + cancel_delayed_work_sync(&conn->c_recv_w); /* tear down queued messages */ list_for_each_entry_safe(rm, rtmp, @@ -302,7 +363,9 @@ void rds_conn_destroy(struct rds_connection *conn) BUG_ON(!list_empty(&conn->c_retrans)); kmem_cache_free(rds_conn_slab, conn); + spin_lock_irqsave(&rds_conn_lock, flags); rds_conn_count--; + spin_unlock_irqrestore(&rds_conn_lock, flags); } EXPORT_SYMBOL_GPL(rds_conn_destroy); @@ -316,23 +379,23 @@ static void rds_conn_message_info(struct socket *sock, unsigned int len, struct list_head *list; struct rds_connection *conn; struct rds_message *rm; - unsigned long flags; unsigned int total = 0; + unsigned long flags; size_t i; len /= sizeof(struct rds_info_message); - spin_lock_irqsave(&rds_conn_lock, flags); + rcu_read_lock(); for (i = 0, head = rds_conn_hash; i < ARRAY_SIZE(rds_conn_hash); i++, head++) { - hlist_for_each_entry(conn, pos, head, c_hash_node) { + hlist_for_each_entry_rcu(conn, pos, head, c_hash_node) { if (want_send) list = &conn->c_send_queue; else list = &conn->c_retrans; - spin_lock(&conn->c_lock); + spin_lock_irqsave(&conn->c_lock, flags); /* XXX too lazy to maintain counts.. */ list_for_each_entry(rm, list, m_conn_item) { @@ -343,11 +406,10 @@ static void rds_conn_message_info(struct socket *sock, unsigned int len, conn->c_faddr, 0); } - spin_unlock(&conn->c_lock); + spin_unlock_irqrestore(&conn->c_lock, flags); } } - - spin_unlock_irqrestore(&rds_conn_lock, flags); + rcu_read_unlock(); lens->nr = total; lens->each = sizeof(struct rds_info_message); @@ -377,19 +439,17 @@ void rds_for_each_conn_info(struct socket *sock, unsigned int len, uint64_t buffer[(item_len + 7) / 8]; struct hlist_head *head; struct hlist_node *pos; - struct hlist_node *tmp; struct rds_connection *conn; - unsigned long flags; size_t i; - spin_lock_irqsave(&rds_conn_lock, flags); + rcu_read_lock(); lens->nr = 0; lens->each = item_len; for (i = 0, head = rds_conn_hash; i < ARRAY_SIZE(rds_conn_hash); i++, head++) { - hlist_for_each_entry_safe(conn, pos, tmp, head, c_hash_node) { + hlist_for_each_entry_rcu(conn, pos, head, c_hash_node) { /* XXX no c_lock usage.. */ if (!visitor(conn, buffer)) @@ -405,8 +465,7 @@ void rds_for_each_conn_info(struct socket *sock, unsigned int len, lens->nr++; } } - - spin_unlock_irqrestore(&rds_conn_lock, flags); + rcu_read_unlock(); } EXPORT_SYMBOL_GPL(rds_for_each_conn_info); @@ -423,8 +482,8 @@ static int rds_conn_info_visitor(struct rds_connection *conn, sizeof(cinfo->transport)); cinfo->flags = 0; - rds_conn_info_set(cinfo->flags, - rds_conn_is_sending(conn), SENDING); + rds_conn_info_set(cinfo->flags, test_bit(RDS_IN_XMIT, &conn->c_flags), + SENDING); /* XXX Future: return the state rather than these funky bits */ rds_conn_info_set(cinfo->flags, atomic_read(&conn->c_state) == RDS_CONN_CONNECTING, @@ -444,12 +503,12 @@ static void rds_conn_info(struct socket *sock, unsigned int len, sizeof(struct rds_info_connection)); } -int __init rds_conn_init(void) +int rds_conn_init(void) { rds_conn_slab = kmem_cache_create("rds_connection", sizeof(struct rds_connection), 0, 0, NULL); - if (rds_conn_slab == NULL) + if (!rds_conn_slab) return -ENOMEM; rds_info_register_func(RDS_INFO_CONNECTIONS, rds_conn_info); @@ -487,6 +546,18 @@ void rds_conn_drop(struct rds_connection *conn) EXPORT_SYMBOL_GPL(rds_conn_drop); /* + * If the connection is down, trigger a connect. We may have scheduled a + * delayed reconnect however - in this case we should not interfere. + */ +void rds_conn_connect_if_down(struct rds_connection *conn) +{ + if (rds_conn_state(conn) == RDS_CONN_DOWN && + !test_and_set_bit(RDS_RECONNECT_PENDING, &conn->c_flags)) + queue_delayed_work(rds_wq, &conn->c_conn_w, 0); +} +EXPORT_SYMBOL_GPL(rds_conn_connect_if_down); + +/* * An error occurred on the connection */ void diff --git a/net/rds/ib.c b/net/rds/ib.c index 8f2d6dd7700..4123967d4d6 100644 --- a/net/rds/ib.c +++ b/net/rds/ib.c @@ -42,7 +42,7 @@ #include "rds.h" #include "ib.h" -unsigned int fmr_pool_size = RDS_FMR_POOL_SIZE; +static unsigned int fmr_pool_size = RDS_FMR_POOL_SIZE; unsigned int fmr_message_size = RDS_FMR_SIZE + 1; /* +1 allows for unaligned MRs */ unsigned int rds_ib_retry_count = RDS_IB_DEFAULT_RETRY_COUNT; @@ -53,13 +53,72 @@ MODULE_PARM_DESC(fmr_message_size, " Max size of a RDMA transfer"); module_param(rds_ib_retry_count, int, 0444); MODULE_PARM_DESC(rds_ib_retry_count, " Number of hw retries before reporting an error"); +/* + * we have a clumsy combination of RCU and a rwsem protecting this list + * because it is used both in the get_mr fast path and while blocking in + * the FMR flushing path. + */ +DECLARE_RWSEM(rds_ib_devices_lock); struct list_head rds_ib_devices; /* NOTE: if also grabbing ibdev lock, grab this first */ DEFINE_SPINLOCK(ib_nodev_conns_lock); LIST_HEAD(ib_nodev_conns); -void rds_ib_add_one(struct ib_device *device) +static void rds_ib_nodev_connect(void) +{ + struct rds_ib_connection *ic; + + spin_lock(&ib_nodev_conns_lock); + list_for_each_entry(ic, &ib_nodev_conns, ib_node) + rds_conn_connect_if_down(ic->conn); + spin_unlock(&ib_nodev_conns_lock); +} + +static void rds_ib_dev_shutdown(struct rds_ib_device *rds_ibdev) +{ + struct rds_ib_connection *ic; + unsigned long flags; + + spin_lock_irqsave(&rds_ibdev->spinlock, flags); + list_for_each_entry(ic, &rds_ibdev->conn_list, ib_node) + rds_conn_drop(ic->conn); + spin_unlock_irqrestore(&rds_ibdev->spinlock, flags); +} + +/* + * rds_ib_destroy_mr_pool() blocks on a few things and mrs drop references + * from interrupt context so we push freing off into a work struct in krdsd. + */ +static void rds_ib_dev_free(struct work_struct *work) +{ + struct rds_ib_ipaddr *i_ipaddr, *i_next; + struct rds_ib_device *rds_ibdev = container_of(work, + struct rds_ib_device, free_work); + + if (rds_ibdev->mr_pool) + rds_ib_destroy_mr_pool(rds_ibdev->mr_pool); + if (rds_ibdev->mr) + ib_dereg_mr(rds_ibdev->mr); + if (rds_ibdev->pd) + ib_dealloc_pd(rds_ibdev->pd); + + list_for_each_entry_safe(i_ipaddr, i_next, &rds_ibdev->ipaddr_list, list) { + list_del(&i_ipaddr->list); + kfree(i_ipaddr); + } + + kfree(rds_ibdev); +} + +void rds_ib_dev_put(struct rds_ib_device *rds_ibdev) +{ + BUG_ON(atomic_read(&rds_ibdev->refcount) <= 0); + if (atomic_dec_and_test(&rds_ibdev->refcount)) + queue_work(rds_wq, &rds_ibdev->free_work); +} + +static void rds_ib_add_one(struct ib_device *device) { struct rds_ib_device *rds_ibdev; struct ib_device_attr *dev_attr; @@ -77,11 +136,14 @@ void rds_ib_add_one(struct ib_device *device) goto free_attr; } - rds_ibdev = kmalloc(sizeof *rds_ibdev, GFP_KERNEL); + rds_ibdev = kzalloc_node(sizeof(struct rds_ib_device), GFP_KERNEL, + ibdev_to_node(device)); if (!rds_ibdev) goto free_attr; spin_lock_init(&rds_ibdev->spinlock); + atomic_set(&rds_ibdev->refcount, 1); + INIT_WORK(&rds_ibdev->free_work, rds_ib_dev_free); rds_ibdev->max_wrs = dev_attr->max_qp_wr; rds_ibdev->max_sge = min(dev_attr->max_sge, RDS_IB_MAX_SGE); @@ -91,68 +153,107 @@ void rds_ib_add_one(struct ib_device *device) min_t(unsigned int, dev_attr->max_fmr, fmr_pool_size) : fmr_pool_size; + rds_ibdev->max_initiator_depth = dev_attr->max_qp_init_rd_atom; + rds_ibdev->max_responder_resources = dev_attr->max_qp_rd_atom; + rds_ibdev->dev = device; rds_ibdev->pd = ib_alloc_pd(device); - if (IS_ERR(rds_ibdev->pd)) - goto free_dev; + if (IS_ERR(rds_ibdev->pd)) { + rds_ibdev->pd = NULL; + goto put_dev; + } - rds_ibdev->mr = ib_get_dma_mr(rds_ibdev->pd, - IB_ACCESS_LOCAL_WRITE); - if (IS_ERR(rds_ibdev->mr)) - goto err_pd; + rds_ibdev->mr = ib_get_dma_mr(rds_ibdev->pd, IB_ACCESS_LOCAL_WRITE); + if (IS_ERR(rds_ibdev->mr)) { + rds_ibdev->mr = NULL; + goto put_dev; + } rds_ibdev->mr_pool = rds_ib_create_mr_pool(rds_ibdev); if (IS_ERR(rds_ibdev->mr_pool)) { rds_ibdev->mr_pool = NULL; - goto err_mr; + goto put_dev; } INIT_LIST_HEAD(&rds_ibdev->ipaddr_list); INIT_LIST_HEAD(&rds_ibdev->conn_list); - list_add_tail(&rds_ibdev->list, &rds_ib_devices); + + down_write(&rds_ib_devices_lock); + list_add_tail_rcu(&rds_ibdev->list, &rds_ib_devices); + up_write(&rds_ib_devices_lock); + atomic_inc(&rds_ibdev->refcount); ib_set_client_data(device, &rds_ib_client, rds_ibdev); + atomic_inc(&rds_ibdev->refcount); - goto free_attr; + rds_ib_nodev_connect(); -err_mr: - ib_dereg_mr(rds_ibdev->mr); -err_pd: - ib_dealloc_pd(rds_ibdev->pd); -free_dev: - kfree(rds_ibdev); +put_dev: + rds_ib_dev_put(rds_ibdev); free_attr: kfree(dev_attr); } -void rds_ib_remove_one(struct ib_device *device) +/* + * New connections use this to find the device to associate with the + * connection. It's not in the fast path so we're not concerned about the + * performance of the IB call. (As of this writing, it uses an interrupt + * blocking spinlock to serialize walking a per-device list of all registered + * clients.) + * + * RCU is used to handle incoming connections racing with device teardown. + * Rather than use a lock to serialize removal from the client_data and + * getting a new reference, we use an RCU grace period. The destruction + * path removes the device from client_data and then waits for all RCU + * readers to finish. + * + * A new connection can get NULL from this if its arriving on a + * device that is in the process of being removed. + */ +struct rds_ib_device *rds_ib_get_client_data(struct ib_device *device) { struct rds_ib_device *rds_ibdev; - struct rds_ib_ipaddr *i_ipaddr, *i_next; + rcu_read_lock(); rds_ibdev = ib_get_client_data(device, &rds_ib_client); - if (!rds_ibdev) - return; + if (rds_ibdev) + atomic_inc(&rds_ibdev->refcount); + rcu_read_unlock(); + return rds_ibdev; +} - list_for_each_entry_safe(i_ipaddr, i_next, &rds_ibdev->ipaddr_list, list) { - list_del(&i_ipaddr->list); - kfree(i_ipaddr); - } +/* + * The IB stack is letting us know that a device is going away. This can + * happen if the underlying HCA driver is removed or if PCI hotplug is removing + * the pci function, for example. + * + * This can be called at any time and can be racing with any other RDS path. + */ +static void rds_ib_remove_one(struct ib_device *device) +{ + struct rds_ib_device *rds_ibdev; - rds_ib_destroy_conns(rds_ibdev); + rds_ibdev = ib_get_client_data(device, &rds_ib_client); + if (!rds_ibdev) + return; - if (rds_ibdev->mr_pool) - rds_ib_destroy_mr_pool(rds_ibdev->mr_pool); + rds_ib_dev_shutdown(rds_ibdev); - ib_dereg_mr(rds_ibdev->mr); + /* stop connection attempts from getting a reference to this device. */ + ib_set_client_data(device, &rds_ib_client, NULL); - while (ib_dealloc_pd(rds_ibdev->pd)) { - rdsdebug("Failed to dealloc pd %p\n", rds_ibdev->pd); - msleep(1); - } + down_write(&rds_ib_devices_lock); + list_del_rcu(&rds_ibdev->list); + up_write(&rds_ib_devices_lock); - list_del(&rds_ibdev->list); - kfree(rds_ibdev); + /* + * This synchronize rcu is waiting for readers of both the ib + * client data and the devices list to finish before we drop + * both of those references. + */ + synchronize_rcu(); + rds_ib_dev_put(rds_ibdev); + rds_ib_dev_put(rds_ibdev); } struct ib_client rds_ib_client = { @@ -186,7 +287,7 @@ static int rds_ib_conn_info_visitor(struct rds_connection *conn, rdma_addr_get_sgid(dev_addr, (union ib_gid *) &iinfo->src_gid); rdma_addr_get_dgid(dev_addr, (union ib_gid *) &iinfo->dst_gid); - rds_ibdev = ib_get_client_data(ic->i_cm_id->device, &rds_ib_client); + rds_ibdev = ic->rds_ibdev; iinfo->max_send_wr = ic->i_send_ring.w_nr; iinfo->max_recv_wr = ic->i_recv_ring.w_nr; iinfo->max_send_sge = rds_ibdev->max_sge; @@ -248,29 +349,36 @@ static int rds_ib_laddr_check(__be32 addr) return ret; } +static void rds_ib_unregister_client(void) +{ + ib_unregister_client(&rds_ib_client); + /* wait for rds_ib_dev_free() to complete */ + flush_workqueue(rds_wq); +} + void rds_ib_exit(void) { rds_info_deregister_func(RDS_INFO_IB_CONNECTIONS, rds_ib_ic_info); + rds_ib_unregister_client(); rds_ib_destroy_nodev_conns(); - ib_unregister_client(&rds_ib_client); rds_ib_sysctl_exit(); rds_ib_recv_exit(); rds_trans_unregister(&rds_ib_transport); + rds_ib_fmr_exit(); } struct rds_transport rds_ib_transport = { .laddr_check = rds_ib_laddr_check, .xmit_complete = rds_ib_xmit_complete, .xmit = rds_ib_xmit, - .xmit_cong_map = NULL, .xmit_rdma = rds_ib_xmit_rdma, + .xmit_atomic = rds_ib_xmit_atomic, .recv = rds_ib_recv, .conn_alloc = rds_ib_conn_alloc, .conn_free = rds_ib_conn_free, .conn_connect = rds_ib_conn_connect, .conn_shutdown = rds_ib_conn_shutdown, .inc_copy_to_user = rds_ib_inc_copy_to_user, - .inc_purge = rds_ib_inc_purge, .inc_free = rds_ib_inc_free, .cm_initiate_connect = rds_ib_cm_initiate_connect, .cm_handle_connect = rds_ib_cm_handle_connect, @@ -286,16 +394,20 @@ struct rds_transport rds_ib_transport = { .t_type = RDS_TRANS_IB }; -int __init rds_ib_init(void) +int rds_ib_init(void) { int ret; INIT_LIST_HEAD(&rds_ib_devices); - ret = ib_register_client(&rds_ib_client); + ret = rds_ib_fmr_init(); if (ret) goto out; + ret = ib_register_client(&rds_ib_client); + if (ret) + goto out_fmr_exit; + ret = rds_ib_sysctl_init(); if (ret) goto out_ibreg; @@ -317,7 +429,9 @@ out_recv: out_sysctl: rds_ib_sysctl_exit(); out_ibreg: - ib_unregister_client(&rds_ib_client); + rds_ib_unregister_client(); +out_fmr_exit: + rds_ib_fmr_exit(); out: return ret; } diff --git a/net/rds/ib.h b/net/rds/ib.h index 64df4e79b29..e34ad032b66 100644 --- a/net/rds/ib.h +++ b/net/rds/ib.h @@ -3,11 +3,13 @@ #include <rdma/ib_verbs.h> #include <rdma/rdma_cm.h> +#include <linux/pci.h> +#include <linux/slab.h> #include "rds.h" #include "rdma_transport.h" #define RDS_FMR_SIZE 256 -#define RDS_FMR_POOL_SIZE 4096 +#define RDS_FMR_POOL_SIZE 8192 #define RDS_IB_MAX_SGE 8 #define RDS_IB_RECV_SGE 2 @@ -19,6 +21,9 @@ #define RDS_IB_SUPPORTED_PROTOCOLS 0x00000003 /* minor versions supported */ +#define RDS_IB_RECYCLE_BATCH_COUNT 32 + +extern struct rw_semaphore rds_ib_devices_lock; extern struct list_head rds_ib_devices; /* @@ -26,20 +31,29 @@ extern struct list_head rds_ib_devices; * try and minimize the amount of memory tied up both the device and * socket receive queues. */ -/* page offset of the final full frag that fits in the page */ -#define RDS_PAGE_LAST_OFF (((PAGE_SIZE / RDS_FRAG_SIZE) - 1) * RDS_FRAG_SIZE) struct rds_page_frag { struct list_head f_item; - struct page *f_page; - unsigned long f_offset; - dma_addr_t f_mapped; + struct list_head f_cache_entry; + struct scatterlist f_sg; }; struct rds_ib_incoming { struct list_head ii_frags; + struct list_head ii_cache_entry; struct rds_incoming ii_inc; }; +struct rds_ib_cache_head { + struct list_head *first; + unsigned long count; +}; + +struct rds_ib_refill_cache { + struct rds_ib_cache_head *percpu; + struct list_head *xfer; + struct list_head *ready; +}; + struct rds_ib_connect_private { /* Add new fields at the end, and don't permute existing fields. */ __be32 dp_saddr; @@ -53,8 +67,7 @@ struct rds_ib_connect_private { }; struct rds_ib_send_work { - struct rds_message *s_rm; - struct rds_rdma_op *s_op; + void *s_op; struct ib_send_wr s_wr; struct ib_sge s_sge[RDS_IB_MAX_SGE]; unsigned long s_queued; @@ -92,10 +105,11 @@ struct rds_ib_connection { /* tx */ struct rds_ib_work_ring i_send_ring; - struct rds_message *i_rm; + struct rm_data_op *i_data_op; struct rds_header *i_send_hdrs; u64 i_send_hdrs_dma; struct rds_ib_send_work *i_sends; + atomic_t i_signaled_sends; /* rx */ struct tasklet_struct i_recv_tasklet; @@ -106,8 +120,9 @@ struct rds_ib_connection { struct rds_header *i_recv_hdrs; u64 i_recv_hdrs_dma; struct rds_ib_recv_work *i_recvs; - struct rds_page_frag i_frag; u64 i_ack_recv; /* last ACK received */ + struct rds_ib_refill_cache i_cache_incs; + struct rds_ib_refill_cache i_cache_frags; /* sending acks */ unsigned long i_ack_flags; @@ -138,7 +153,6 @@ struct rds_ib_connection { /* Batched completions */ unsigned int i_unsignaled_wrs; - long i_unsignaled_bytes; }; /* This assumes that atomic_t is at least 32 bits */ @@ -164,9 +178,17 @@ struct rds_ib_device { unsigned int max_fmrs; int max_sge; unsigned int max_wrs; + unsigned int max_initiator_depth; + unsigned int max_responder_resources; spinlock_t spinlock; /* protect the above */ + atomic_t refcount; + struct work_struct free_work; }; +#define pcidev_to_node(pcidev) pcibus_to_node(pcidev->bus) +#define ibdev_to_node(ibdev) pcidev_to_node(to_pci_dev(ibdev->dma_device)) +#define rdsibdev_to_node(rdsibdev) ibdev_to_node(rdsibdev->dev) + /* bits for i_ack_flags */ #define IB_ACK_IN_FLIGHT 0 #define IB_ACK_REQUESTED 1 @@ -202,6 +224,8 @@ struct rds_ib_statistics { uint64_t s_ib_rdma_mr_pool_flush; uint64_t s_ib_rdma_mr_pool_wait; uint64_t s_ib_rdma_mr_pool_depleted; + uint64_t s_ib_atomic_cswp; + uint64_t s_ib_atomic_fadd; }; extern struct workqueue_struct *rds_ib_wq; @@ -241,11 +265,10 @@ static inline void rds_ib_dma_sync_sg_for_device(struct ib_device *dev, /* ib.c */ extern struct rds_transport rds_ib_transport; -extern void rds_ib_add_one(struct ib_device *device); -extern void rds_ib_remove_one(struct ib_device *device); +struct rds_ib_device *rds_ib_get_client_data(struct ib_device *device); +void rds_ib_dev_put(struct rds_ib_device *rds_ibdev); extern struct ib_client rds_ib_client; -extern unsigned int fmr_pool_size; extern unsigned int fmr_message_size; extern unsigned int rds_ib_retry_count; @@ -258,7 +281,7 @@ void rds_ib_conn_free(void *arg); int rds_ib_conn_connect(struct rds_connection *conn); void rds_ib_conn_shutdown(struct rds_connection *conn); void rds_ib_state_change(struct sock *sk); -int __init rds_ib_listen_init(void); +int rds_ib_listen_init(void); void rds_ib_listen_stop(void); void __rds_ib_conn_error(struct rds_connection *conn, const char *, ...); int rds_ib_cm_handle_connect(struct rdma_cm_id *cm_id, @@ -275,15 +298,7 @@ void rds_ib_cm_connect_complete(struct rds_connection *conn, int rds_ib_update_ipaddr(struct rds_ib_device *rds_ibdev, __be32 ipaddr); void rds_ib_add_conn(struct rds_ib_device *rds_ibdev, struct rds_connection *conn); void rds_ib_remove_conn(struct rds_ib_device *rds_ibdev, struct rds_connection *conn); -void __rds_ib_destroy_conns(struct list_head *list, spinlock_t *list_lock); -static inline void rds_ib_destroy_nodev_conns(void) -{ - __rds_ib_destroy_conns(&ib_nodev_conns, &ib_nodev_conns_lock); -} -static inline void rds_ib_destroy_conns(struct rds_ib_device *rds_ibdev) -{ - __rds_ib_destroy_conns(&rds_ibdev->conn_list, &rds_ibdev->spinlock); -} +void rds_ib_destroy_nodev_conns(void); struct rds_ib_mr_pool *rds_ib_create_mr_pool(struct rds_ib_device *); void rds_ib_get_mr_info(struct rds_ib_device *rds_ibdev, struct rds_info_rdma_connection *iinfo); void rds_ib_destroy_mr_pool(struct rds_ib_mr_pool *); @@ -292,14 +307,16 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents, void rds_ib_sync_mr(void *trans_private, int dir); void rds_ib_free_mr(void *trans_private, int invalidate); void rds_ib_flush_mrs(void); +int rds_ib_fmr_init(void); +void rds_ib_fmr_exit(void); /* ib_recv.c */ -int __init rds_ib_recv_init(void); +int rds_ib_recv_init(void); void rds_ib_recv_exit(void); int rds_ib_recv(struct rds_connection *conn); -int rds_ib_recv_refill(struct rds_connection *conn, gfp_t kptr_gfp, - gfp_t page_gfp, int prefill); -void rds_ib_inc_purge(struct rds_incoming *inc); +int rds_ib_recv_alloc_caches(struct rds_ib_connection *ic); +void rds_ib_recv_free_caches(struct rds_ib_connection *ic); +void rds_ib_recv_refill(struct rds_connection *conn, int prefill); void rds_ib_inc_free(struct rds_incoming *inc); int rds_ib_inc_copy_to_user(struct rds_incoming *inc, struct iovec *iov, size_t size); @@ -325,17 +342,19 @@ u32 rds_ib_ring_completed(struct rds_ib_work_ring *ring, u32 wr_id, u32 oldest); extern wait_queue_head_t rds_ib_ring_empty_wait; /* ib_send.c */ +char *rds_ib_wc_status_str(enum ib_wc_status status); void rds_ib_xmit_complete(struct rds_connection *conn); int rds_ib_xmit(struct rds_connection *conn, struct rds_message *rm, unsigned int hdr_off, unsigned int sg, unsigned int off); void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context); void rds_ib_send_init_ring(struct rds_ib_connection *ic); void rds_ib_send_clear_ring(struct rds_ib_connection *ic); -int rds_ib_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op); +int rds_ib_xmit_rdma(struct rds_connection *conn, struct rm_rdma_op *op); void rds_ib_send_add_credits(struct rds_connection *conn, unsigned int credits); void rds_ib_advertise_credits(struct rds_connection *conn, unsigned int posted); int rds_ib_send_grab_credits(struct rds_ib_connection *ic, u32 wanted, u32 *adv_credits, int need_posted, int max_posted); +int rds_ib_xmit_atomic(struct rds_connection *conn, struct rm_atomic_op *op); /* ib_stats.c */ DECLARE_PER_CPU(struct rds_ib_statistics, rds_ib_stats); @@ -344,7 +363,7 @@ unsigned int rds_ib_stats_info_copy(struct rds_info_iterator *iter, unsigned int avail); /* ib_sysctl.c */ -int __init rds_ib_sysctl_init(void); +int rds_ib_sysctl_init(void); void rds_ib_sysctl_exit(void); extern unsigned long rds_ib_sysctl_max_send_wr; extern unsigned long rds_ib_sysctl_max_recv_wr; @@ -352,30 +371,5 @@ extern unsigned long rds_ib_sysctl_max_unsig_wrs; extern unsigned long rds_ib_sysctl_max_unsig_bytes; extern unsigned long rds_ib_sysctl_max_recv_allocation; extern unsigned int rds_ib_sysctl_flow_control; -extern ctl_table rds_ib_sysctl_table[]; - -/* - * Helper functions for getting/setting the header and data SGEs in - * RDS packets (not RDMA) - * - * From version 3.1 onwards, header is in front of data in the sge. - */ -static inline struct ib_sge * -rds_ib_header_sge(struct rds_ib_connection *ic, struct ib_sge *sge) -{ - if (ic->conn->c_version > RDS_PROTOCOL_3_0) - return &sge[0]; - else - return &sge[1]; -} - -static inline struct ib_sge * -rds_ib_data_sge(struct rds_ib_connection *ic, struct ib_sge *sge) -{ - if (ic->conn->c_version > RDS_PROTOCOL_3_0) - return &sge[1]; - else - return &sge[0]; -} #endif diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c index f68832798db..ee369d201a6 100644 --- a/net/rds/ib_cm.c +++ b/net/rds/ib_cm.c @@ -38,6 +38,36 @@ #include "rds.h" #include "ib.h" +static char *rds_ib_event_type_strings[] = { +#define RDS_IB_EVENT_STRING(foo) \ + [IB_EVENT_##foo] = __stringify(IB_EVENT_##foo) + RDS_IB_EVENT_STRING(CQ_ERR), + RDS_IB_EVENT_STRING(QP_FATAL), + RDS_IB_EVENT_STRING(QP_REQ_ERR), + RDS_IB_EVENT_STRING(QP_ACCESS_ERR), + RDS_IB_EVENT_STRING(COMM_EST), + RDS_IB_EVENT_STRING(SQ_DRAINED), + RDS_IB_EVENT_STRING(PATH_MIG), + RDS_IB_EVENT_STRING(PATH_MIG_ERR), + RDS_IB_EVENT_STRING(DEVICE_FATAL), + RDS_IB_EVENT_STRING(PORT_ACTIVE), + RDS_IB_EVENT_STRING(PORT_ERR), + RDS_IB_EVENT_STRING(LID_CHANGE), + RDS_IB_EVENT_STRING(PKEY_CHANGE), + RDS_IB_EVENT_STRING(SM_CHANGE), + RDS_IB_EVENT_STRING(SRQ_ERR), + RDS_IB_EVENT_STRING(SRQ_LIMIT_REACHED), + RDS_IB_EVENT_STRING(QP_LAST_WQE_REACHED), + RDS_IB_EVENT_STRING(CLIENT_REREGISTER), +#undef RDS_IB_EVENT_STRING +}; + +static char *rds_ib_event_str(enum ib_event_type type) +{ + return rds_str_array(rds_ib_event_type_strings, + ARRAY_SIZE(rds_ib_event_type_strings), type); +}; + /* * Set the selected protocol version */ @@ -95,7 +125,6 @@ void rds_ib_cm_connect_complete(struct rds_connection *conn, struct rdma_cm_even { const struct rds_ib_connect_private *dp = NULL; struct rds_ib_connection *ic = conn->c_transport_data; - struct rds_ib_device *rds_ibdev; struct ib_qp_attr qp_attr; int err; @@ -111,11 +140,21 @@ void rds_ib_cm_connect_complete(struct rds_connection *conn, struct rdma_cm_even } } - printk(KERN_NOTICE "RDS/IB: connected to %pI4 version %u.%u%s\n", - &conn->c_faddr, - RDS_PROTOCOL_MAJOR(conn->c_version), - RDS_PROTOCOL_MINOR(conn->c_version), - ic->i_flowctl ? ", flow control" : ""); + if (conn->c_version < RDS_PROTOCOL(3,1)) { + printk(KERN_NOTICE "RDS/IB: Connection to %pI4 version %u.%u failed," + " no longer supported\n", + &conn->c_faddr, + RDS_PROTOCOL_MAJOR(conn->c_version), + RDS_PROTOCOL_MINOR(conn->c_version)); + rds_conn_destroy(conn); + return; + } else { + printk(KERN_NOTICE "RDS/IB: connected to %pI4 version %u.%u%s\n", + &conn->c_faddr, + RDS_PROTOCOL_MAJOR(conn->c_version), + RDS_PROTOCOL_MINOR(conn->c_version), + ic->i_flowctl ? ", flow control" : ""); + } /* * Init rings and fill recv. this needs to wait until protocol negotiation @@ -125,7 +164,7 @@ void rds_ib_cm_connect_complete(struct rds_connection *conn, struct rdma_cm_even rds_ib_recv_init_ring(ic); /* Post receive buffers - as a side effect, this will update * the posted credit count. */ - rds_ib_recv_refill(conn, GFP_KERNEL, GFP_HIGHUSER, 1); + rds_ib_recv_refill(conn, 1); /* Tune RNR behavior */ rds_ib_tune_rnr(ic, &qp_attr); @@ -135,12 +174,11 @@ void rds_ib_cm_connect_complete(struct rds_connection *conn, struct rdma_cm_even if (err) printk(KERN_NOTICE "ib_modify_qp(IB_QP_STATE, RTS): err=%d\n", err); - /* update ib_device with this local ipaddr & conn */ - rds_ibdev = ib_get_client_data(ic->i_cm_id->device, &rds_ib_client); - err = rds_ib_update_ipaddr(rds_ibdev, conn->c_laddr); + /* update ib_device with this local ipaddr */ + err = rds_ib_update_ipaddr(ic->rds_ibdev, conn->c_laddr); if (err) - printk(KERN_ERR "rds_ib_update_ipaddr failed (%d)\n", err); - rds_ib_add_conn(rds_ibdev, conn); + printk(KERN_ERR "rds_ib_update_ipaddr failed (%d)\n", + err); /* If the peer gave us the last packet it saw, process this as if * we had received a regular ACK. */ @@ -153,18 +191,23 @@ void rds_ib_cm_connect_complete(struct rds_connection *conn, struct rdma_cm_even static void rds_ib_cm_fill_conn_param(struct rds_connection *conn, struct rdma_conn_param *conn_param, struct rds_ib_connect_private *dp, - u32 protocol_version) + u32 protocol_version, + u32 max_responder_resources, + u32 max_initiator_depth) { + struct rds_ib_connection *ic = conn->c_transport_data; + struct rds_ib_device *rds_ibdev = ic->rds_ibdev; + memset(conn_param, 0, sizeof(struct rdma_conn_param)); - /* XXX tune these? */ - conn_param->responder_resources = 1; - conn_param->initiator_depth = 1; + + conn_param->responder_resources = + min_t(u32, rds_ibdev->max_responder_resources, max_responder_resources); + conn_param->initiator_depth = + min_t(u32, rds_ibdev->max_initiator_depth, max_initiator_depth); conn_param->retry_count = min_t(unsigned int, rds_ib_retry_count, 7); conn_param->rnr_retry_count = 7; if (dp) { - struct rds_ib_connection *ic = conn->c_transport_data; - memset(dp, 0, sizeof(*dp)); dp->dp_saddr = conn->c_laddr; dp->dp_daddr = conn->c_faddr; @@ -189,7 +232,8 @@ static void rds_ib_cm_fill_conn_param(struct rds_connection *conn, static void rds_ib_cq_event_handler(struct ib_event *event, void *data) { - rdsdebug("event %u data %p\n", event->event, data); + rdsdebug("event %u (%s) data %p\n", + event->event, rds_ib_event_str(event->event), data); } static void rds_ib_qp_event_handler(struct ib_event *event, void *data) @@ -197,16 +241,18 @@ static void rds_ib_qp_event_handler(struct ib_event *event, void *data) struct rds_connection *conn = data; struct rds_ib_connection *ic = conn->c_transport_data; - rdsdebug("conn %p ic %p event %u\n", conn, ic, event->event); + rdsdebug("conn %p ic %p event %u (%s)\n", conn, ic, event->event, + rds_ib_event_str(event->event)); switch (event->event) { case IB_EVENT_COMM_EST: rdma_notify(ic->i_cm_id, IB_EVENT_COMM_EST); break; default: - rdsdebug("Fatal QP Event %u " + rdsdebug("Fatal QP Event %u (%s) " "- connection %pI4->%pI4, reconnecting\n", - event->event, &conn->c_laddr, &conn->c_faddr); + event->event, rds_ib_event_str(event->event), + &conn->c_laddr, &conn->c_faddr); rds_conn_drop(conn); break; } @@ -224,18 +270,16 @@ static int rds_ib_setup_qp(struct rds_connection *conn) struct rds_ib_device *rds_ibdev; int ret; - /* rds_ib_add_one creates a rds_ib_device object per IB device, - * and allocates a protection domain, memory range and FMR pool - * for each. If that fails for any reason, it will not register - * the rds_ibdev at all. + /* + * It's normal to see a null device if an incoming connection races + * with device removal, so we don't print a warning. */ - rds_ibdev = ib_get_client_data(dev, &rds_ib_client); - if (rds_ibdev == NULL) { - if (printk_ratelimit()) - printk(KERN_NOTICE "RDS/IB: No client_data for device %s\n", - dev->name); + rds_ibdev = rds_ib_get_client_data(dev); + if (!rds_ibdev) return -EOPNOTSUPP; - } + + /* add the conn now so that connection establishment has the dev */ + rds_ib_add_conn(rds_ibdev, conn); if (rds_ibdev->max_wrs < ic->i_send_ring.w_nr + 1) rds_ib_ring_resize(&ic->i_send_ring, rds_ibdev->max_wrs - 1); @@ -306,7 +350,7 @@ static int rds_ib_setup_qp(struct rds_connection *conn) ic->i_send_ring.w_nr * sizeof(struct rds_header), &ic->i_send_hdrs_dma, GFP_KERNEL); - if (ic->i_send_hdrs == NULL) { + if (!ic->i_send_hdrs) { ret = -ENOMEM; rdsdebug("ib_dma_alloc_coherent send failed\n"); goto out; @@ -316,7 +360,7 @@ static int rds_ib_setup_qp(struct rds_connection *conn) ic->i_recv_ring.w_nr * sizeof(struct rds_header), &ic->i_recv_hdrs_dma, GFP_KERNEL); - if (ic->i_recv_hdrs == NULL) { + if (!ic->i_recv_hdrs) { ret = -ENOMEM; rdsdebug("ib_dma_alloc_coherent recv failed\n"); goto out; @@ -324,22 +368,24 @@ static int rds_ib_setup_qp(struct rds_connection *conn) ic->i_ack = ib_dma_alloc_coherent(dev, sizeof(struct rds_header), &ic->i_ack_dma, GFP_KERNEL); - if (ic->i_ack == NULL) { + if (!ic->i_ack) { ret = -ENOMEM; rdsdebug("ib_dma_alloc_coherent ack failed\n"); goto out; } - ic->i_sends = vmalloc(ic->i_send_ring.w_nr * sizeof(struct rds_ib_send_work)); - if (ic->i_sends == NULL) { + ic->i_sends = vmalloc_node(ic->i_send_ring.w_nr * sizeof(struct rds_ib_send_work), + ibdev_to_node(dev)); + if (!ic->i_sends) { ret = -ENOMEM; rdsdebug("send allocation failed\n"); goto out; } memset(ic->i_sends, 0, ic->i_send_ring.w_nr * sizeof(struct rds_ib_send_work)); - ic->i_recvs = vmalloc(ic->i_recv_ring.w_nr * sizeof(struct rds_ib_recv_work)); - if (ic->i_recvs == NULL) { + ic->i_recvs = vmalloc_node(ic->i_recv_ring.w_nr * sizeof(struct rds_ib_recv_work), + ibdev_to_node(dev)); + if (!ic->i_recvs) { ret = -ENOMEM; rdsdebug("recv allocation failed\n"); goto out; @@ -352,6 +398,7 @@ static int rds_ib_setup_qp(struct rds_connection *conn) ic->i_send_cq, ic->i_recv_cq); out: + rds_ib_dev_put(rds_ibdev); return ret; } @@ -409,7 +456,7 @@ int rds_ib_cm_handle_connect(struct rdma_cm_id *cm_id, struct rds_ib_connection *ic = NULL; struct rdma_conn_param conn_param; u32 version; - int err, destroy = 1; + int err = 1, destroy = 1; /* Check whether the remote protocol version matches ours. */ version = rds_ib_protocol_compatible(event); @@ -448,7 +495,6 @@ int rds_ib_cm_handle_connect(struct rdma_cm_id *cm_id, /* Wait and see - our connect may still be succeeding */ rds_ib_stats_inc(s_ib_connect_raced); } - mutex_unlock(&conn->c_cm_lock); goto out; } @@ -475,24 +521,23 @@ int rds_ib_cm_handle_connect(struct rdma_cm_id *cm_id, err = rds_ib_setup_qp(conn); if (err) { rds_ib_conn_error(conn, "rds_ib_setup_qp failed (%d)\n", err); - mutex_unlock(&conn->c_cm_lock); goto out; } - rds_ib_cm_fill_conn_param(conn, &conn_param, &dp_rep, version); + rds_ib_cm_fill_conn_param(conn, &conn_param, &dp_rep, version, + event->param.conn.responder_resources, + event->param.conn.initiator_depth); /* rdma_accept() calls rdma_reject() internally if it fails */ err = rdma_accept(cm_id, &conn_param); - mutex_unlock(&conn->c_cm_lock); - if (err) { + if (err) rds_ib_conn_error(conn, "rdma_accept failed (%d)\n", err); - goto out; - } - - return 0; out: - rdma_reject(cm_id, NULL, 0); + if (conn) + mutex_unlock(&conn->c_cm_lock); + if (err) + rdma_reject(cm_id, NULL, 0); return destroy; } @@ -516,8 +561,8 @@ int rds_ib_cm_initiate_connect(struct rdma_cm_id *cm_id) goto out; } - rds_ib_cm_fill_conn_param(conn, &conn_param, &dp, RDS_PROTOCOL_VERSION); - + rds_ib_cm_fill_conn_param(conn, &conn_param, &dp, RDS_PROTOCOL_VERSION, + UINT_MAX, UINT_MAX); ret = rdma_connect(cm_id, &conn_param); if (ret) rds_ib_conn_error(conn, "rdma_connect failed (%d)\n", ret); @@ -601,9 +646,19 @@ void rds_ib_conn_shutdown(struct rds_connection *conn) ic->i_cm_id, err); } + /* + * We want to wait for tx and rx completion to finish + * before we tear down the connection, but we have to be + * careful not to get stuck waiting on a send ring that + * only has unsignaled sends in it. We've shutdown new + * sends before getting here so by waiting for signaled + * sends to complete we're ensured that there will be no + * more tx processing. + */ wait_event(rds_ib_ring_empty_wait, - rds_ib_ring_empty(&ic->i_send_ring) && - rds_ib_ring_empty(&ic->i_recv_ring)); + rds_ib_ring_empty(&ic->i_recv_ring) && + (atomic_read(&ic->i_signaled_sends) == 0)); + tasklet_kill(&ic->i_recv_tasklet); if (ic->i_send_hdrs) ib_dma_free_coherent(dev, @@ -654,9 +709,12 @@ void rds_ib_conn_shutdown(struct rds_connection *conn) BUG_ON(ic->rds_ibdev); /* Clear pending transmit */ - if (ic->i_rm) { - rds_message_put(ic->i_rm); - ic->i_rm = NULL; + if (ic->i_data_op) { + struct rds_message *rm; + + rm = container_of(ic->i_data_op, struct rds_message, data); + rds_message_put(rm); + ic->i_data_op = NULL; } /* Clear the ACK state */ @@ -690,12 +748,19 @@ int rds_ib_conn_alloc(struct rds_connection *conn, gfp_t gfp) { struct rds_ib_connection *ic; unsigned long flags; + int ret; /* XXX too lazy? */ ic = kzalloc(sizeof(struct rds_ib_connection), GFP_KERNEL); - if (ic == NULL) + if (!ic) return -ENOMEM; + ret = rds_ib_recv_alloc_caches(ic); + if (ret) { + kfree(ic); + return ret; + } + INIT_LIST_HEAD(&ic->ib_node); tasklet_init(&ic->i_recv_tasklet, rds_ib_recv_tasklet_fn, (unsigned long) ic); @@ -703,6 +768,7 @@ int rds_ib_conn_alloc(struct rds_connection *conn, gfp_t gfp) #ifndef KERNEL_HAS_ATOMIC64 spin_lock_init(&ic->i_ack_lock); #endif + atomic_set(&ic->i_signaled_sends, 0); /* * rds_ib_conn_shutdown() waits for these to be emptied so they @@ -744,6 +810,8 @@ void rds_ib_conn_free(void *arg) list_del(&ic->ib_node); spin_unlock_irq(lock_ptr); + rds_ib_recv_free_caches(ic); + kfree(ic); } diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c index a54cd63f9e3..18a833c450c 100644 --- a/net/rds/ib_rdma.c +++ b/net/rds/ib_rdma.c @@ -32,11 +32,16 @@ */ #include <linux/kernel.h> #include <linux/slab.h> +#include <linux/rculist.h> #include "rds.h" -#include "rdma.h" #include "ib.h" +#include "xlist.h" +static struct workqueue_struct *rds_ib_fmr_wq; + +static DEFINE_PER_CPU(unsigned long, clean_list_grace); +#define CLEAN_LIST_BUSY_BIT 0 /* * This is stored as mr->r_trans_private. @@ -45,7 +50,11 @@ struct rds_ib_mr { struct rds_ib_device *device; struct rds_ib_mr_pool *pool; struct ib_fmr *fmr; - struct list_head list; + + struct xlist_head xlist; + + /* unmap_list is for freeing */ + struct list_head unmap_list; unsigned int remap_count; struct scatterlist *sg; @@ -59,14 +68,16 @@ struct rds_ib_mr { */ struct rds_ib_mr_pool { struct mutex flush_lock; /* serialize fmr invalidate */ - struct work_struct flush_worker; /* flush worker */ + struct delayed_work flush_worker; /* flush worker */ - spinlock_t list_lock; /* protect variables below */ atomic_t item_count; /* total # of MRs */ atomic_t dirty_count; /* # dirty of MRs */ - struct list_head drop_list; /* MRs that have reached their max_maps limit */ - struct list_head free_list; /* unused MRs */ - struct list_head clean_list; /* unused & unamapped MRs */ + + struct xlist_head drop_list; /* MRs that have reached their max_maps limit */ + struct xlist_head free_list; /* unused MRs */ + struct xlist_head clean_list; /* global unused & unamapped MRs */ + wait_queue_head_t flush_wait; + atomic_t free_pinned; /* memory pinned by free MRs */ unsigned long max_items; unsigned long max_items_soft; @@ -74,7 +85,7 @@ struct rds_ib_mr_pool { struct ib_fmr_attr fmr_attr; }; -static int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool, int free_all); +static int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool, int free_all, struct rds_ib_mr **); static void rds_ib_teardown_mr(struct rds_ib_mr *ibmr); static void rds_ib_mr_pool_flush_worker(struct work_struct *work); @@ -83,16 +94,17 @@ static struct rds_ib_device *rds_ib_get_device(__be32 ipaddr) struct rds_ib_device *rds_ibdev; struct rds_ib_ipaddr *i_ipaddr; - list_for_each_entry(rds_ibdev, &rds_ib_devices, list) { - spin_lock_irq(&rds_ibdev->spinlock); - list_for_each_entry(i_ipaddr, &rds_ibdev->ipaddr_list, list) { + rcu_read_lock(); + list_for_each_entry_rcu(rds_ibdev, &rds_ib_devices, list) { + list_for_each_entry_rcu(i_ipaddr, &rds_ibdev->ipaddr_list, list) { if (i_ipaddr->ipaddr == ipaddr) { - spin_unlock_irq(&rds_ibdev->spinlock); + atomic_inc(&rds_ibdev->refcount); + rcu_read_unlock(); return rds_ibdev; } } - spin_unlock_irq(&rds_ibdev->spinlock); } + rcu_read_unlock(); return NULL; } @@ -108,7 +120,7 @@ static int rds_ib_add_ipaddr(struct rds_ib_device *rds_ibdev, __be32 ipaddr) i_ipaddr->ipaddr = ipaddr; spin_lock_irq(&rds_ibdev->spinlock); - list_add_tail(&i_ipaddr->list, &rds_ibdev->ipaddr_list); + list_add_tail_rcu(&i_ipaddr->list, &rds_ibdev->ipaddr_list); spin_unlock_irq(&rds_ibdev->spinlock); return 0; @@ -116,17 +128,24 @@ static int rds_ib_add_ipaddr(struct rds_ib_device *rds_ibdev, __be32 ipaddr) static void rds_ib_remove_ipaddr(struct rds_ib_device *rds_ibdev, __be32 ipaddr) { - struct rds_ib_ipaddr *i_ipaddr, *next; + struct rds_ib_ipaddr *i_ipaddr; + struct rds_ib_ipaddr *to_free = NULL; + spin_lock_irq(&rds_ibdev->spinlock); - list_for_each_entry_safe(i_ipaddr, next, &rds_ibdev->ipaddr_list, list) { + list_for_each_entry_rcu(i_ipaddr, &rds_ibdev->ipaddr_list, list) { if (i_ipaddr->ipaddr == ipaddr) { - list_del(&i_ipaddr->list); - kfree(i_ipaddr); + list_del_rcu(&i_ipaddr->list); + to_free = i_ipaddr; break; } } spin_unlock_irq(&rds_ibdev->spinlock); + + if (to_free) { + synchronize_rcu(); + kfree(to_free); + } } int rds_ib_update_ipaddr(struct rds_ib_device *rds_ibdev, __be32 ipaddr) @@ -134,8 +153,10 @@ int rds_ib_update_ipaddr(struct rds_ib_device *rds_ibdev, __be32 ipaddr) struct rds_ib_device *rds_ibdev_old; rds_ibdev_old = rds_ib_get_device(ipaddr); - if (rds_ibdev_old) + if (rds_ibdev_old) { rds_ib_remove_ipaddr(rds_ibdev_old, ipaddr); + rds_ib_dev_put(rds_ibdev_old); + } return rds_ib_add_ipaddr(rds_ibdev, ipaddr); } @@ -150,12 +171,13 @@ void rds_ib_add_conn(struct rds_ib_device *rds_ibdev, struct rds_connection *con BUG_ON(list_empty(&ic->ib_node)); list_del(&ic->ib_node); - spin_lock_irq(&rds_ibdev->spinlock); + spin_lock(&rds_ibdev->spinlock); list_add_tail(&ic->ib_node, &rds_ibdev->conn_list); - spin_unlock_irq(&rds_ibdev->spinlock); + spin_unlock(&rds_ibdev->spinlock); spin_unlock_irq(&ib_nodev_conns_lock); ic->rds_ibdev = rds_ibdev; + atomic_inc(&rds_ibdev->refcount); } void rds_ib_remove_conn(struct rds_ib_device *rds_ibdev, struct rds_connection *conn) @@ -175,18 +197,18 @@ void rds_ib_remove_conn(struct rds_ib_device *rds_ibdev, struct rds_connection * spin_unlock(&ib_nodev_conns_lock); ic->rds_ibdev = NULL; + rds_ib_dev_put(rds_ibdev); } -void __rds_ib_destroy_conns(struct list_head *list, spinlock_t *list_lock) +void rds_ib_destroy_nodev_conns(void) { struct rds_ib_connection *ic, *_ic; LIST_HEAD(tmp_list); /* avoid calling conn_destroy with irqs off */ - spin_lock_irq(list_lock); - list_splice(list, &tmp_list); - INIT_LIST_HEAD(list); - spin_unlock_irq(list_lock); + spin_lock_irq(&ib_nodev_conns_lock); + list_splice(&ib_nodev_conns, &tmp_list); + spin_unlock_irq(&ib_nodev_conns_lock); list_for_each_entry_safe(ic, _ic, &tmp_list, ib_node) rds_conn_destroy(ic->conn); @@ -200,12 +222,12 @@ struct rds_ib_mr_pool *rds_ib_create_mr_pool(struct rds_ib_device *rds_ibdev) if (!pool) return ERR_PTR(-ENOMEM); - INIT_LIST_HEAD(&pool->free_list); - INIT_LIST_HEAD(&pool->drop_list); - INIT_LIST_HEAD(&pool->clean_list); + INIT_XLIST_HEAD(&pool->free_list); + INIT_XLIST_HEAD(&pool->drop_list); + INIT_XLIST_HEAD(&pool->clean_list); mutex_init(&pool->flush_lock); - spin_lock_init(&pool->list_lock); - INIT_WORK(&pool->flush_worker, rds_ib_mr_pool_flush_worker); + init_waitqueue_head(&pool->flush_wait); + INIT_DELAYED_WORK(&pool->flush_worker, rds_ib_mr_pool_flush_worker); pool->fmr_attr.max_pages = fmr_message_size; pool->fmr_attr.max_maps = rds_ibdev->fmr_max_remaps; @@ -233,34 +255,60 @@ void rds_ib_get_mr_info(struct rds_ib_device *rds_ibdev, struct rds_info_rdma_co void rds_ib_destroy_mr_pool(struct rds_ib_mr_pool *pool) { - flush_workqueue(rds_wq); - rds_ib_flush_mr_pool(pool, 1); + cancel_delayed_work_sync(&pool->flush_worker); + rds_ib_flush_mr_pool(pool, 1, NULL); WARN_ON(atomic_read(&pool->item_count)); WARN_ON(atomic_read(&pool->free_pinned)); kfree(pool); } +static void refill_local(struct rds_ib_mr_pool *pool, struct xlist_head *xl, + struct rds_ib_mr **ibmr_ret) +{ + struct xlist_head *ibmr_xl; + ibmr_xl = xlist_del_head_fast(xl); + *ibmr_ret = list_entry(ibmr_xl, struct rds_ib_mr, xlist); +} + static inline struct rds_ib_mr *rds_ib_reuse_fmr(struct rds_ib_mr_pool *pool) { struct rds_ib_mr *ibmr = NULL; - unsigned long flags; + struct xlist_head *ret; + unsigned long *flag; - spin_lock_irqsave(&pool->list_lock, flags); - if (!list_empty(&pool->clean_list)) { - ibmr = list_entry(pool->clean_list.next, struct rds_ib_mr, list); - list_del_init(&ibmr->list); - } - spin_unlock_irqrestore(&pool->list_lock, flags); + preempt_disable(); + flag = &__get_cpu_var(clean_list_grace); + set_bit(CLEAN_LIST_BUSY_BIT, flag); + ret = xlist_del_head(&pool->clean_list); + if (ret) + ibmr = list_entry(ret, struct rds_ib_mr, xlist); + clear_bit(CLEAN_LIST_BUSY_BIT, flag); + preempt_enable(); return ibmr; } +static inline void wait_clean_list_grace(void) +{ + int cpu; + unsigned long *flag; + + for_each_online_cpu(cpu) { + flag = &per_cpu(clean_list_grace, cpu); + while (test_bit(CLEAN_LIST_BUSY_BIT, flag)) + cpu_relax(); + } +} + static struct rds_ib_mr *rds_ib_alloc_fmr(struct rds_ib_device *rds_ibdev) { struct rds_ib_mr_pool *pool = rds_ibdev->mr_pool; struct rds_ib_mr *ibmr = NULL; int err = 0, iter = 0; + if (atomic_read(&pool->dirty_count) >= pool->max_items / 10) + queue_delayed_work(rds_ib_fmr_wq, &pool->flush_worker, 10); + while (1) { ibmr = rds_ib_reuse_fmr(pool); if (ibmr) @@ -287,19 +335,24 @@ static struct rds_ib_mr *rds_ib_alloc_fmr(struct rds_ib_device *rds_ibdev) /* We do have some empty MRs. Flush them out. */ rds_ib_stats_inc(s_ib_rdma_mr_pool_wait); - rds_ib_flush_mr_pool(pool, 0); + rds_ib_flush_mr_pool(pool, 0, &ibmr); + if (ibmr) + return ibmr; } - ibmr = kzalloc(sizeof(*ibmr), GFP_KERNEL); + ibmr = kzalloc_node(sizeof(*ibmr), GFP_KERNEL, rdsibdev_to_node(rds_ibdev)); if (!ibmr) { err = -ENOMEM; goto out_no_cigar; } + memset(ibmr, 0, sizeof(*ibmr)); + ibmr->fmr = ib_alloc_fmr(rds_ibdev->pd, (IB_ACCESS_LOCAL_WRITE | IB_ACCESS_REMOTE_READ | - IB_ACCESS_REMOTE_WRITE), + IB_ACCESS_REMOTE_WRITE| + IB_ACCESS_REMOTE_ATOMIC), &pool->fmr_attr); if (IS_ERR(ibmr->fmr)) { err = PTR_ERR(ibmr->fmr); @@ -367,7 +420,8 @@ static int rds_ib_map_fmr(struct rds_ib_device *rds_ibdev, struct rds_ib_mr *ibm if (page_cnt > fmr_message_size) return -EINVAL; - dma_pages = kmalloc(sizeof(u64) * page_cnt, GFP_ATOMIC); + dma_pages = kmalloc_node(sizeof(u64) * page_cnt, GFP_ATOMIC, + rdsibdev_to_node(rds_ibdev)); if (!dma_pages) return -ENOMEM; @@ -441,7 +495,7 @@ static void __rds_ib_teardown_mr(struct rds_ib_mr *ibmr) /* FIXME we need a way to tell a r/w MR * from a r/o MR */ - BUG_ON(in_interrupt()); + BUG_ON(irqs_disabled()); set_page_dirty(page); put_page(page); } @@ -477,33 +531,109 @@ static inline unsigned int rds_ib_flush_goal(struct rds_ib_mr_pool *pool, int fr } /* + * given an xlist of mrs, put them all into the list_head for more processing + */ +static void xlist_append_to_list(struct xlist_head *xlist, struct list_head *list) +{ + struct rds_ib_mr *ibmr; + struct xlist_head splice; + struct xlist_head *cur; + struct xlist_head *next; + + splice.next = NULL; + xlist_splice(xlist, &splice); + cur = splice.next; + while (cur) { + next = cur->next; + ibmr = list_entry(cur, struct rds_ib_mr, xlist); + list_add_tail(&ibmr->unmap_list, list); + cur = next; + } +} + +/* + * this takes a list head of mrs and turns it into an xlist of clusters. + * each cluster has an xlist of MR_CLUSTER_SIZE mrs that are ready for + * reuse. + */ +static void list_append_to_xlist(struct rds_ib_mr_pool *pool, + struct list_head *list, struct xlist_head *xlist, + struct xlist_head **tail_ret) +{ + struct rds_ib_mr *ibmr; + struct xlist_head *cur_mr = xlist; + struct xlist_head *tail_mr = NULL; + + list_for_each_entry(ibmr, list, unmap_list) { + tail_mr = &ibmr->xlist; + tail_mr->next = NULL; + cur_mr->next = tail_mr; + cur_mr = tail_mr; + } + *tail_ret = tail_mr; +} + +/* * Flush our pool of MRs. * At a minimum, all currently unused MRs are unmapped. * If the number of MRs allocated exceeds the limit, we also try * to free as many MRs as needed to get back to this limit. */ -static int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool, int free_all) +static int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool, + int free_all, struct rds_ib_mr **ibmr_ret) { struct rds_ib_mr *ibmr, *next; + struct xlist_head clean_xlist; + struct xlist_head *clean_tail; LIST_HEAD(unmap_list); LIST_HEAD(fmr_list); unsigned long unpinned = 0; - unsigned long flags; unsigned int nfreed = 0, ncleaned = 0, free_goal; int ret = 0; rds_ib_stats_inc(s_ib_rdma_mr_pool_flush); - mutex_lock(&pool->flush_lock); + if (ibmr_ret) { + DEFINE_WAIT(wait); + while(!mutex_trylock(&pool->flush_lock)) { + ibmr = rds_ib_reuse_fmr(pool); + if (ibmr) { + *ibmr_ret = ibmr; + finish_wait(&pool->flush_wait, &wait); + goto out_nolock; + } + + prepare_to_wait(&pool->flush_wait, &wait, + TASK_UNINTERRUPTIBLE); + if (xlist_empty(&pool->clean_list)) + schedule(); + + ibmr = rds_ib_reuse_fmr(pool); + if (ibmr) { + *ibmr_ret = ibmr; + finish_wait(&pool->flush_wait, &wait); + goto out_nolock; + } + } + finish_wait(&pool->flush_wait, &wait); + } else + mutex_lock(&pool->flush_lock); + + if (ibmr_ret) { + ibmr = rds_ib_reuse_fmr(pool); + if (ibmr) { + *ibmr_ret = ibmr; + goto out; + } + } - spin_lock_irqsave(&pool->list_lock, flags); /* Get the list of all MRs to be dropped. Ordering matters - - * we want to put drop_list ahead of free_list. */ - list_splice_init(&pool->free_list, &unmap_list); - list_splice_init(&pool->drop_list, &unmap_list); + * we want to put drop_list ahead of free_list. + */ + xlist_append_to_list(&pool->drop_list, &unmap_list); + xlist_append_to_list(&pool->free_list, &unmap_list); if (free_all) - list_splice_init(&pool->clean_list, &unmap_list); - spin_unlock_irqrestore(&pool->list_lock, flags); + xlist_append_to_list(&pool->clean_list, &unmap_list); free_goal = rds_ib_flush_goal(pool, free_all); @@ -511,19 +641,20 @@ static int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool, int free_all) goto out; /* String all ib_mr's onto one list and hand them to ib_unmap_fmr */ - list_for_each_entry(ibmr, &unmap_list, list) + list_for_each_entry(ibmr, &unmap_list, unmap_list) list_add(&ibmr->fmr->list, &fmr_list); + ret = ib_unmap_fmr(&fmr_list); if (ret) printk(KERN_WARNING "RDS/IB: ib_unmap_fmr failed (err=%d)\n", ret); /* Now we can destroy the DMA mapping and unpin any pages */ - list_for_each_entry_safe(ibmr, next, &unmap_list, list) { + list_for_each_entry_safe(ibmr, next, &unmap_list, unmap_list) { unpinned += ibmr->sg_len; __rds_ib_teardown_mr(ibmr); if (nfreed < free_goal || ibmr->remap_count >= pool->fmr_attr.max_maps) { rds_ib_stats_inc(s_ib_rdma_mr_free); - list_del(&ibmr->list); + list_del(&ibmr->unmap_list); ib_dealloc_fmr(ibmr->fmr); kfree(ibmr); nfreed++; @@ -531,9 +662,27 @@ static int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool, int free_all) ncleaned++; } - spin_lock_irqsave(&pool->list_lock, flags); - list_splice(&unmap_list, &pool->clean_list); - spin_unlock_irqrestore(&pool->list_lock, flags); + if (!list_empty(&unmap_list)) { + /* we have to make sure that none of the things we're about + * to put on the clean list would race with other cpus trying + * to pull items off. The xlist would explode if we managed to + * remove something from the clean list and then add it back again + * while another CPU was spinning on that same item in xlist_del_head. + * + * This is pretty unlikely, but just in case wait for an xlist grace period + * here before adding anything back into the clean list. + */ + wait_clean_list_grace(); + + list_append_to_xlist(pool, &unmap_list, &clean_xlist, &clean_tail); + if (ibmr_ret) + refill_local(pool, &clean_xlist, ibmr_ret); + + /* refill_local may have emptied our list */ + if (!xlist_empty(&clean_xlist)) + xlist_add(clean_xlist.next, clean_tail, &pool->clean_list); + + } atomic_sub(unpinned, &pool->free_pinned); atomic_sub(ncleaned, &pool->dirty_count); @@ -541,14 +690,35 @@ static int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool, int free_all) out: mutex_unlock(&pool->flush_lock); + if (waitqueue_active(&pool->flush_wait)) + wake_up(&pool->flush_wait); +out_nolock: return ret; } +int rds_ib_fmr_init(void) +{ + rds_ib_fmr_wq = create_workqueue("rds_fmr_flushd"); + if (!rds_ib_fmr_wq) + return -ENOMEM; + return 0; +} + +/* + * By the time this is called all the IB devices should have been torn down and + * had their pools freed. As each pool is freed its work struct is waited on, + * so the pool flushing work queue should be idle by the time we get here. + */ +void rds_ib_fmr_exit(void) +{ + destroy_workqueue(rds_ib_fmr_wq); +} + static void rds_ib_mr_pool_flush_worker(struct work_struct *work) { - struct rds_ib_mr_pool *pool = container_of(work, struct rds_ib_mr_pool, flush_worker); + struct rds_ib_mr_pool *pool = container_of(work, struct rds_ib_mr_pool, flush_worker.work); - rds_ib_flush_mr_pool(pool, 0); + rds_ib_flush_mr_pool(pool, 0, NULL); } void rds_ib_free_mr(void *trans_private, int invalidate) @@ -556,47 +726,49 @@ void rds_ib_free_mr(void *trans_private, int invalidate) struct rds_ib_mr *ibmr = trans_private; struct rds_ib_device *rds_ibdev = ibmr->device; struct rds_ib_mr_pool *pool = rds_ibdev->mr_pool; - unsigned long flags; rdsdebug("RDS/IB: free_mr nents %u\n", ibmr->sg_len); /* Return it to the pool's free list */ - spin_lock_irqsave(&pool->list_lock, flags); if (ibmr->remap_count >= pool->fmr_attr.max_maps) - list_add(&ibmr->list, &pool->drop_list); + xlist_add(&ibmr->xlist, &ibmr->xlist, &pool->drop_list); else - list_add(&ibmr->list, &pool->free_list); + xlist_add(&ibmr->xlist, &ibmr->xlist, &pool->free_list); atomic_add(ibmr->sg_len, &pool->free_pinned); atomic_inc(&pool->dirty_count); - spin_unlock_irqrestore(&pool->list_lock, flags); /* If we've pinned too many pages, request a flush */ if (atomic_read(&pool->free_pinned) >= pool->max_free_pinned || atomic_read(&pool->dirty_count) >= pool->max_items / 10) - queue_work(rds_wq, &pool->flush_worker); + queue_delayed_work(rds_ib_fmr_wq, &pool->flush_worker, 10); if (invalidate) { if (likely(!in_interrupt())) { - rds_ib_flush_mr_pool(pool, 0); + rds_ib_flush_mr_pool(pool, 0, NULL); } else { /* We get here if the user created a MR marked * as use_once and invalidate at the same time. */ - queue_work(rds_wq, &pool->flush_worker); + queue_delayed_work(rds_ib_fmr_wq, + &pool->flush_worker, 10); } } + + rds_ib_dev_put(rds_ibdev); } void rds_ib_flush_mrs(void) { struct rds_ib_device *rds_ibdev; + down_read(&rds_ib_devices_lock); list_for_each_entry(rds_ibdev, &rds_ib_devices, list) { struct rds_ib_mr_pool *pool = rds_ibdev->mr_pool; if (pool) - rds_ib_flush_mr_pool(pool, 0); + rds_ib_flush_mr_pool(pool, 0, NULL); } + up_read(&rds_ib_devices_lock); } void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents, @@ -628,6 +800,7 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents, printk(KERN_WARNING "RDS/IB: map_fmr failed (errno=%d)\n", ret); ibmr->device = rds_ibdev; + rds_ibdev = NULL; out: if (ret) { @@ -635,5 +808,8 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents, rds_ib_free_mr(ibmr, 0); ibmr = ERR_PTR(ret); } + if (rds_ibdev) + rds_ib_dev_put(rds_ibdev); return ibmr; } + diff --git a/net/rds/ib_recv.c b/net/rds/ib_recv.c index c74e9904a6b..e29e0ca32f7 100644 --- a/net/rds/ib_recv.c +++ b/net/rds/ib_recv.c @@ -43,42 +43,6 @@ static struct kmem_cache *rds_ib_incoming_slab; static struct kmem_cache *rds_ib_frag_slab; static atomic_t rds_ib_allocation = ATOMIC_INIT(0); -static void rds_ib_frag_drop_page(struct rds_page_frag *frag) -{ - rdsdebug("frag %p page %p\n", frag, frag->f_page); - __free_page(frag->f_page); - frag->f_page = NULL; -} - -static void rds_ib_frag_free(struct rds_page_frag *frag) -{ - rdsdebug("frag %p page %p\n", frag, frag->f_page); - BUG_ON(frag->f_page != NULL); - kmem_cache_free(rds_ib_frag_slab, frag); -} - -/* - * We map a page at a time. Its fragments are posted in order. This - * is called in fragment order as the fragments get send completion events. - * Only the last frag in the page performs the unmapping. - * - * It's OK for ring cleanup to call this in whatever order it likes because - * DMA is not in flight and so we can unmap while other ring entries still - * hold page references in their frags. - */ -static void rds_ib_recv_unmap_page(struct rds_ib_connection *ic, - struct rds_ib_recv_work *recv) -{ - struct rds_page_frag *frag = recv->r_frag; - - rdsdebug("recv %p frag %p page %p\n", recv, frag, frag->f_page); - if (frag->f_mapped) - ib_dma_unmap_page(ic->i_cm_id->device, - frag->f_mapped, - RDS_FRAG_SIZE, DMA_FROM_DEVICE); - frag->f_mapped = 0; -} - void rds_ib_recv_init_ring(struct rds_ib_connection *ic) { struct rds_ib_recv_work *recv; @@ -95,16 +59,161 @@ void rds_ib_recv_init_ring(struct rds_ib_connection *ic) recv->r_wr.sg_list = recv->r_sge; recv->r_wr.num_sge = RDS_IB_RECV_SGE; - sge = rds_ib_data_sge(ic, recv->r_sge); + sge = &recv->r_sge[0]; + sge->addr = ic->i_recv_hdrs_dma + (i * sizeof(struct rds_header)); + sge->length = sizeof(struct rds_header); + sge->lkey = ic->i_mr->lkey; + + sge = &recv->r_sge[1]; sge->addr = 0; sge->length = RDS_FRAG_SIZE; sge->lkey = ic->i_mr->lkey; + } +} - sge = rds_ib_header_sge(ic, recv->r_sge); - sge->addr = ic->i_recv_hdrs_dma + (i * sizeof(struct rds_header)); - sge->length = sizeof(struct rds_header); - sge->lkey = ic->i_mr->lkey; +/* + * The entire 'from' list, including the from element itself, is put on + * to the tail of the 'to' list. + */ +static void list_splice_entire_tail(struct list_head *from, + struct list_head *to) +{ + struct list_head *from_last = from->prev; + + list_splice_tail(from_last, to); + list_add_tail(from_last, to); +} + +static void rds_ib_cache_xfer_to_ready(struct rds_ib_refill_cache *cache) +{ + struct list_head *tmp; + + tmp = xchg(&cache->xfer, NULL); + if (tmp) { + if (cache->ready) + list_splice_entire_tail(tmp, cache->ready); + else + cache->ready = tmp; + } +} + +static int rds_ib_recv_alloc_cache(struct rds_ib_refill_cache *cache) +{ + struct rds_ib_cache_head *head; + int cpu; + + cache->percpu = alloc_percpu(struct rds_ib_cache_head); + if (!cache->percpu) + return -ENOMEM; + + for_each_possible_cpu(cpu) { + head = per_cpu_ptr(cache->percpu, cpu); + head->first = NULL; + head->count = 0; + } + cache->xfer = NULL; + cache->ready = NULL; + + return 0; +} + +int rds_ib_recv_alloc_caches(struct rds_ib_connection *ic) +{ + int ret; + + ret = rds_ib_recv_alloc_cache(&ic->i_cache_incs); + if (!ret) { + ret = rds_ib_recv_alloc_cache(&ic->i_cache_frags); + if (ret) + free_percpu(ic->i_cache_incs.percpu); } + + return ret; +} + +static void rds_ib_cache_splice_all_lists(struct rds_ib_refill_cache *cache, + struct list_head *caller_list) +{ + struct rds_ib_cache_head *head; + int cpu; + + for_each_possible_cpu(cpu) { + head = per_cpu_ptr(cache->percpu, cpu); + if (head->first) { + list_splice_entire_tail(head->first, caller_list); + head->first = NULL; + } + } + + if (cache->ready) { + list_splice_entire_tail(cache->ready, caller_list); + cache->ready = NULL; + } +} + +void rds_ib_recv_free_caches(struct rds_ib_connection *ic) +{ + struct rds_ib_incoming *inc; + struct rds_ib_incoming *inc_tmp; + struct rds_page_frag *frag; + struct rds_page_frag *frag_tmp; + LIST_HEAD(list); + + rds_ib_cache_xfer_to_ready(&ic->i_cache_incs); + rds_ib_cache_splice_all_lists(&ic->i_cache_incs, &list); + free_percpu(ic->i_cache_incs.percpu); + + list_for_each_entry_safe(inc, inc_tmp, &list, ii_cache_entry) { + list_del(&inc->ii_cache_entry); + WARN_ON(!list_empty(&inc->ii_frags)); + kmem_cache_free(rds_ib_incoming_slab, inc); + } + + rds_ib_cache_xfer_to_ready(&ic->i_cache_frags); + rds_ib_cache_splice_all_lists(&ic->i_cache_frags, &list); + free_percpu(ic->i_cache_frags.percpu); + + list_for_each_entry_safe(frag, frag_tmp, &list, f_cache_entry) { + list_del(&frag->f_cache_entry); + WARN_ON(!list_empty(&frag->f_item)); + kmem_cache_free(rds_ib_frag_slab, frag); + } +} + +/* fwd decl */ +static void rds_ib_recv_cache_put(struct list_head *new_item, + struct rds_ib_refill_cache *cache); +static struct list_head *rds_ib_recv_cache_get(struct rds_ib_refill_cache *cache); + + +/* Recycle frag and attached recv buffer f_sg */ +static void rds_ib_frag_free(struct rds_ib_connection *ic, + struct rds_page_frag *frag) +{ + rdsdebug("frag %p page %p\n", frag, sg_page(&frag->f_sg)); + + rds_ib_recv_cache_put(&frag->f_cache_entry, &ic->i_cache_frags); +} + +/* Recycle inc after freeing attached frags */ +void rds_ib_inc_free(struct rds_incoming *inc) +{ + struct rds_ib_incoming *ibinc; + struct rds_page_frag *frag; + struct rds_page_frag *pos; + struct rds_ib_connection *ic = inc->i_conn->c_transport_data; + + ibinc = container_of(inc, struct rds_ib_incoming, ii_inc); + + /* Free attached frags */ + list_for_each_entry_safe(frag, pos, &ibinc->ii_frags, f_item) { + list_del_init(&frag->f_item); + rds_ib_frag_free(ic, frag); + } + BUG_ON(!list_empty(&ibinc->ii_frags)); + + rdsdebug("freeing ibinc %p inc %p\n", ibinc, inc); + rds_ib_recv_cache_put(&ibinc->ii_cache_entry, &ic->i_cache_incs); } static void rds_ib_recv_clear_one(struct rds_ib_connection *ic, @@ -115,10 +224,8 @@ static void rds_ib_recv_clear_one(struct rds_ib_connection *ic, recv->r_ibinc = NULL; } if (recv->r_frag) { - rds_ib_recv_unmap_page(ic, recv); - if (recv->r_frag->f_page) - rds_ib_frag_drop_page(recv->r_frag); - rds_ib_frag_free(recv->r_frag); + ib_dma_unmap_sg(ic->i_cm_id->device, &recv->r_frag->f_sg, 1, DMA_FROM_DEVICE); + rds_ib_frag_free(ic, recv->r_frag); recv->r_frag = NULL; } } @@ -129,84 +236,111 @@ void rds_ib_recv_clear_ring(struct rds_ib_connection *ic) for (i = 0; i < ic->i_recv_ring.w_nr; i++) rds_ib_recv_clear_one(ic, &ic->i_recvs[i]); - - if (ic->i_frag.f_page) - rds_ib_frag_drop_page(&ic->i_frag); } -static int rds_ib_recv_refill_one(struct rds_connection *conn, - struct rds_ib_recv_work *recv, - gfp_t kptr_gfp, gfp_t page_gfp) +static struct rds_ib_incoming *rds_ib_refill_one_inc(struct rds_ib_connection *ic, + gfp_t slab_mask) { - struct rds_ib_connection *ic = conn->c_transport_data; - dma_addr_t dma_addr; - struct ib_sge *sge; - int ret = -ENOMEM; + struct rds_ib_incoming *ibinc; + struct list_head *cache_item; + int avail_allocs; - if (recv->r_ibinc == NULL) { - if (!atomic_add_unless(&rds_ib_allocation, 1, rds_ib_sysctl_max_recv_allocation)) { + cache_item = rds_ib_recv_cache_get(&ic->i_cache_incs); + if (cache_item) { + ibinc = container_of(cache_item, struct rds_ib_incoming, ii_cache_entry); + } else { + avail_allocs = atomic_add_unless(&rds_ib_allocation, + 1, rds_ib_sysctl_max_recv_allocation); + if (!avail_allocs) { rds_ib_stats_inc(s_ib_rx_alloc_limit); - goto out; + return NULL; } - recv->r_ibinc = kmem_cache_alloc(rds_ib_incoming_slab, - kptr_gfp); - if (recv->r_ibinc == NULL) { + ibinc = kmem_cache_alloc(rds_ib_incoming_slab, slab_mask); + if (!ibinc) { atomic_dec(&rds_ib_allocation); - goto out; + return NULL; } - INIT_LIST_HEAD(&recv->r_ibinc->ii_frags); - rds_inc_init(&recv->r_ibinc->ii_inc, conn, conn->c_faddr); } + INIT_LIST_HEAD(&ibinc->ii_frags); + rds_inc_init(&ibinc->ii_inc, ic->conn, ic->conn->c_faddr); - if (recv->r_frag == NULL) { - recv->r_frag = kmem_cache_alloc(rds_ib_frag_slab, kptr_gfp); - if (recv->r_frag == NULL) - goto out; - INIT_LIST_HEAD(&recv->r_frag->f_item); - recv->r_frag->f_page = NULL; + return ibinc; +} + +static struct rds_page_frag *rds_ib_refill_one_frag(struct rds_ib_connection *ic, + gfp_t slab_mask, gfp_t page_mask) +{ + struct rds_page_frag *frag; + struct list_head *cache_item; + int ret; + + cache_item = rds_ib_recv_cache_get(&ic->i_cache_frags); + if (cache_item) { + frag = container_of(cache_item, struct rds_page_frag, f_cache_entry); + } else { + frag = kmem_cache_alloc(rds_ib_frag_slab, slab_mask); + if (!frag) + return NULL; + + sg_init_table(&frag->f_sg, 1); + ret = rds_page_remainder_alloc(&frag->f_sg, + RDS_FRAG_SIZE, page_mask); + if (ret) { + kmem_cache_free(rds_ib_frag_slab, frag); + return NULL; + } } - if (ic->i_frag.f_page == NULL) { - ic->i_frag.f_page = alloc_page(page_gfp); - if (ic->i_frag.f_page == NULL) - goto out; - ic->i_frag.f_offset = 0; + INIT_LIST_HEAD(&frag->f_item); + + return frag; +} + +static int rds_ib_recv_refill_one(struct rds_connection *conn, + struct rds_ib_recv_work *recv, int prefill) +{ + struct rds_ib_connection *ic = conn->c_transport_data; + struct ib_sge *sge; + int ret = -ENOMEM; + gfp_t slab_mask = GFP_NOWAIT; + gfp_t page_mask = GFP_NOWAIT; + + if (prefill) { + slab_mask = GFP_KERNEL; + page_mask = GFP_HIGHUSER; } - dma_addr = ib_dma_map_page(ic->i_cm_id->device, - ic->i_frag.f_page, - ic->i_frag.f_offset, - RDS_FRAG_SIZE, - DMA_FROM_DEVICE); - if (ib_dma_mapping_error(ic->i_cm_id->device, dma_addr)) - goto out; + if (!ic->i_cache_incs.ready) + rds_ib_cache_xfer_to_ready(&ic->i_cache_incs); + if (!ic->i_cache_frags.ready) + rds_ib_cache_xfer_to_ready(&ic->i_cache_frags); /* - * Once we get the RDS_PAGE_LAST_OFF frag then rds_ib_frag_unmap() - * must be called on this recv. This happens as completions hit - * in order or on connection shutdown. + * ibinc was taken from recv if recv contained the start of a message. + * recvs that were continuations will still have this allocated. */ - recv->r_frag->f_page = ic->i_frag.f_page; - recv->r_frag->f_offset = ic->i_frag.f_offset; - recv->r_frag->f_mapped = dma_addr; + if (!recv->r_ibinc) { + recv->r_ibinc = rds_ib_refill_one_inc(ic, slab_mask); + if (!recv->r_ibinc) + goto out; + } - sge = rds_ib_data_sge(ic, recv->r_sge); - sge->addr = dma_addr; - sge->length = RDS_FRAG_SIZE; + WARN_ON(recv->r_frag); /* leak! */ + recv->r_frag = rds_ib_refill_one_frag(ic, slab_mask, page_mask); + if (!recv->r_frag) + goto out; + + ret = ib_dma_map_sg(ic->i_cm_id->device, &recv->r_frag->f_sg, + 1, DMA_FROM_DEVICE); + WARN_ON(ret != 1); - sge = rds_ib_header_sge(ic, recv->r_sge); + sge = &recv->r_sge[0]; sge->addr = ic->i_recv_hdrs_dma + (recv - ic->i_recvs) * sizeof(struct rds_header); sge->length = sizeof(struct rds_header); - get_page(recv->r_frag->f_page); - - if (ic->i_frag.f_offset < RDS_PAGE_LAST_OFF) { - ic->i_frag.f_offset += RDS_FRAG_SIZE; - } else { - put_page(ic->i_frag.f_page); - ic->i_frag.f_page = NULL; - ic->i_frag.f_offset = 0; - } + sge = &recv->r_sge[1]; + sge->addr = sg_dma_address(&recv->r_frag->f_sg); + sge->length = sg_dma_len(&recv->r_frag->f_sg); ret = 0; out: @@ -216,13 +350,11 @@ out: /* * This tries to allocate and post unused work requests after making sure that * they have all the allocations they need to queue received fragments into - * sockets. The i_recv_mutex is held here so that ring_alloc and _unalloc - * pairs don't go unmatched. + * sockets. * * -1 is returned if posting fails due to temporary resource exhaustion. */ -int rds_ib_recv_refill(struct rds_connection *conn, gfp_t kptr_gfp, - gfp_t page_gfp, int prefill) +void rds_ib_recv_refill(struct rds_connection *conn, int prefill) { struct rds_ib_connection *ic = conn->c_transport_data; struct rds_ib_recv_work *recv; @@ -236,28 +368,25 @@ int rds_ib_recv_refill(struct rds_connection *conn, gfp_t kptr_gfp, if (pos >= ic->i_recv_ring.w_nr) { printk(KERN_NOTICE "Argh - ring alloc returned pos=%u\n", pos); - ret = -EINVAL; break; } recv = &ic->i_recvs[pos]; - ret = rds_ib_recv_refill_one(conn, recv, kptr_gfp, page_gfp); + ret = rds_ib_recv_refill_one(conn, recv, prefill); if (ret) { - ret = -1; break; } /* XXX when can this fail? */ ret = ib_post_recv(ic->i_cm_id->qp, &recv->r_wr, &failed_wr); rdsdebug("recv %p ibinc %p page %p addr %lu ret %d\n", recv, - recv->r_ibinc, recv->r_frag->f_page, - (long) recv->r_frag->f_mapped, ret); + recv->r_ibinc, sg_page(&recv->r_frag->f_sg), + (long) sg_dma_address(&recv->r_frag->f_sg), ret); if (ret) { rds_ib_conn_error(conn, "recv post on " "%pI4 returned %d, disconnecting and " "reconnecting\n", &conn->c_faddr, ret); - ret = -1; break; } @@ -270,37 +399,73 @@ int rds_ib_recv_refill(struct rds_connection *conn, gfp_t kptr_gfp, if (ret) rds_ib_ring_unalloc(&ic->i_recv_ring, 1); - return ret; } -void rds_ib_inc_purge(struct rds_incoming *inc) +/* + * We want to recycle several types of recv allocations, like incs and frags. + * To use this, the *_free() function passes in the ptr to a list_head within + * the recyclee, as well as the cache to put it on. + * + * First, we put the memory on a percpu list. When this reaches a certain size, + * We move it to an intermediate non-percpu list in a lockless manner, with some + * xchg/compxchg wizardry. + * + * N.B. Instead of a list_head as the anchor, we use a single pointer, which can + * be NULL and xchg'd. The list is actually empty when the pointer is NULL, and + * list_empty() will return true with one element is actually present. + */ +static void rds_ib_recv_cache_put(struct list_head *new_item, + struct rds_ib_refill_cache *cache) { - struct rds_ib_incoming *ibinc; - struct rds_page_frag *frag; - struct rds_page_frag *pos; + unsigned long flags; + struct rds_ib_cache_head *chp; + struct list_head *old; - ibinc = container_of(inc, struct rds_ib_incoming, ii_inc); - rdsdebug("purging ibinc %p inc %p\n", ibinc, inc); + local_irq_save(flags); - list_for_each_entry_safe(frag, pos, &ibinc->ii_frags, f_item) { - list_del_init(&frag->f_item); - rds_ib_frag_drop_page(frag); - rds_ib_frag_free(frag); - } + chp = per_cpu_ptr(cache->percpu, smp_processor_id()); + if (!chp->first) + INIT_LIST_HEAD(new_item); + else /* put on front */ + list_add_tail(new_item, chp->first); + chp->first = new_item; + chp->count++; + + if (chp->count < RDS_IB_RECYCLE_BATCH_COUNT) + goto end; + + /* + * Return our per-cpu first list to the cache's xfer by atomically + * grabbing the current xfer list, appending it to our per-cpu list, + * and then atomically returning that entire list back to the + * cache's xfer list as long as it's still empty. + */ + do { + old = xchg(&cache->xfer, NULL); + if (old) + list_splice_entire_tail(old, chp->first); + old = cmpxchg(&cache->xfer, NULL, chp->first); + } while (old); + + chp->first = NULL; + chp->count = 0; +end: + local_irq_restore(flags); } -void rds_ib_inc_free(struct rds_incoming *inc) +static struct list_head *rds_ib_recv_cache_get(struct rds_ib_refill_cache *cache) { - struct rds_ib_incoming *ibinc; - - ibinc = container_of(inc, struct rds_ib_incoming, ii_inc); + struct list_head *head = cache->ready; + + if (head) { + if (!list_empty(head)) { + cache->ready = head->next; + list_del_init(head); + } else + cache->ready = NULL; + } - rds_ib_inc_purge(inc); - rdsdebug("freeing ibinc %p inc %p\n", ibinc, inc); - BUG_ON(!list_empty(&ibinc->ii_frags)); - kmem_cache_free(rds_ib_incoming_slab, ibinc); - atomic_dec(&rds_ib_allocation); - BUG_ON(atomic_read(&rds_ib_allocation) < 0); + return head; } int rds_ib_inc_copy_to_user(struct rds_incoming *inc, struct iovec *first_iov, @@ -336,13 +501,13 @@ int rds_ib_inc_copy_to_user(struct rds_incoming *inc, struct iovec *first_iov, to_copy = min_t(unsigned long, to_copy, len - copied); rdsdebug("%lu bytes to user [%p, %zu] + %lu from frag " - "[%p, %lu] + %lu\n", + "[%p, %u] + %lu\n", to_copy, iov->iov_base, iov->iov_len, iov_off, - frag->f_page, frag->f_offset, frag_off); + sg_page(&frag->f_sg), frag->f_sg.offset, frag_off); /* XXX needs + offset for multiple recvs per page */ - ret = rds_page_copy_to_user(frag->f_page, - frag->f_offset + frag_off, + ret = rds_page_copy_to_user(sg_page(&frag->f_sg), + frag->f_sg.offset + frag_off, iov->iov_base + iov_off, to_copy); if (ret) { @@ -557,47 +722,6 @@ u64 rds_ib_piggyb_ack(struct rds_ib_connection *ic) return rds_ib_get_ack(ic); } -static struct rds_header *rds_ib_get_header(struct rds_connection *conn, - struct rds_ib_recv_work *recv, - u32 data_len) -{ - struct rds_ib_connection *ic = conn->c_transport_data; - void *hdr_buff = &ic->i_recv_hdrs[recv - ic->i_recvs]; - void *addr; - u32 misplaced_hdr_bytes; - - /* - * Support header at the front (RDS 3.1+) as well as header-at-end. - * - * Cases: - * 1) header all in header buff (great!) - * 2) header all in data page (copy all to header buff) - * 3) header split across hdr buf + data page - * (move bit in hdr buff to end before copying other bit from data page) - */ - if (conn->c_version > RDS_PROTOCOL_3_0 || data_len == RDS_FRAG_SIZE) - return hdr_buff; - - if (data_len <= (RDS_FRAG_SIZE - sizeof(struct rds_header))) { - addr = kmap_atomic(recv->r_frag->f_page, KM_SOFTIRQ0); - memcpy(hdr_buff, - addr + recv->r_frag->f_offset + data_len, - sizeof(struct rds_header)); - kunmap_atomic(addr, KM_SOFTIRQ0); - return hdr_buff; - } - - misplaced_hdr_bytes = (sizeof(struct rds_header) - (RDS_FRAG_SIZE - data_len)); - - memmove(hdr_buff + misplaced_hdr_bytes, hdr_buff, misplaced_hdr_bytes); - - addr = kmap_atomic(recv->r_frag->f_page, KM_SOFTIRQ0); - memcpy(hdr_buff, addr + recv->r_frag->f_offset + data_len, - sizeof(struct rds_header) - misplaced_hdr_bytes); - kunmap_atomic(addr, KM_SOFTIRQ0); - return hdr_buff; -} - /* * It's kind of lame that we're copying from the posted receive pages into * long-lived bitmaps. We could have posted the bitmaps and rdma written into @@ -639,7 +763,7 @@ static void rds_ib_cong_recv(struct rds_connection *conn, to_copy = min(RDS_FRAG_SIZE - frag_off, PAGE_SIZE - map_off); BUG_ON(to_copy & 7); /* Must be 64bit aligned. */ - addr = kmap_atomic(frag->f_page, KM_SOFTIRQ0); + addr = kmap_atomic(sg_page(&frag->f_sg), KM_SOFTIRQ0); src = addr + frag_off; dst = (void *)map->m_page_addrs[map_page] + map_off; @@ -710,7 +834,7 @@ static void rds_ib_process_recv(struct rds_connection *conn, } data_len -= sizeof(struct rds_header); - ihdr = rds_ib_get_header(conn, recv, data_len); + ihdr = &ic->i_recv_hdrs[recv - ic->i_recvs]; /* Validate the checksum. */ if (!rds_message_verify_checksum(ihdr)) { @@ -742,12 +866,12 @@ static void rds_ib_process_recv(struct rds_connection *conn, * the inc is freed. We don't go that route, so we have to drop the * page ref ourselves. We can't just leave the page on the recv * because that confuses the dma mapping of pages and each recv's use - * of a partial page. We can leave the frag, though, it will be - * reused. + * of a partial page. * * FIXME: Fold this into the code path below. */ - rds_ib_frag_drop_page(recv->r_frag); + rds_ib_frag_free(ic, recv->r_frag); + recv->r_frag = NULL; return; } @@ -757,7 +881,7 @@ static void rds_ib_process_recv(struct rds_connection *conn, * into the inc and save the inc so we can hang upcoming fragments * off its list. */ - if (ibinc == NULL) { + if (!ibinc) { ibinc = recv->r_ibinc; recv->r_ibinc = NULL; ic->i_ibinc = ibinc; @@ -842,32 +966,38 @@ static inline void rds_poll_cq(struct rds_ib_connection *ic, struct rds_ib_recv_work *recv; while (ib_poll_cq(ic->i_recv_cq, 1, &wc) > 0) { - rdsdebug("wc wr_id 0x%llx status %u byte_len %u imm_data %u\n", - (unsigned long long)wc.wr_id, wc.status, wc.byte_len, + rdsdebug("wc wr_id 0x%llx status %u (%s) byte_len %u imm_data %u\n", + (unsigned long long)wc.wr_id, wc.status, + rds_ib_wc_status_str(wc.status), wc.byte_len, be32_to_cpu(wc.ex.imm_data)); rds_ib_stats_inc(s_ib_rx_cq_event); recv = &ic->i_recvs[rds_ib_ring_oldest(&ic->i_recv_ring)]; - rds_ib_recv_unmap_page(ic, recv); + ib_dma_unmap_sg(ic->i_cm_id->device, &recv->r_frag->f_sg, 1, DMA_FROM_DEVICE); /* * Also process recvs in connecting state because it is possible * to get a recv completion _before_ the rdmacm ESTABLISHED * event is processed. */ - if (rds_conn_up(conn) || rds_conn_connecting(conn)) { + if (wc.status == IB_WC_SUCCESS) { + rds_ib_process_recv(conn, recv, wc.byte_len, state); + } else { /* We expect errors as the qp is drained during shutdown */ - if (wc.status == IB_WC_SUCCESS) { - rds_ib_process_recv(conn, recv, wc.byte_len, state); - } else { - rds_ib_conn_error(conn, "recv completion on " - "%pI4 had status %u, disconnecting and " - "reconnecting\n", &conn->c_faddr, - wc.status); - } + if (rds_conn_up(conn) || rds_conn_connecting(conn)) + rds_ib_conn_error(conn, "recv completion on %pI4 had " + "status %u (%s), disconnecting and " + "reconnecting\n", &conn->c_faddr, + wc.status, + rds_ib_wc_status_str(wc.status)); } + /* + * It's very important that we only free this ring entry if we've truly + * freed the resources allocated to the entry. The refilling path can + * leak if we don't. + */ rds_ib_ring_free(&ic->i_recv_ring, 1); } } @@ -897,11 +1027,8 @@ void rds_ib_recv_tasklet_fn(unsigned long data) if (rds_ib_ring_empty(&ic->i_recv_ring)) rds_ib_stats_inc(s_ib_rx_ring_empty); - /* - * If the ring is running low, then schedule the thread to refill. - */ if (rds_ib_ring_low(&ic->i_recv_ring)) - queue_delayed_work(rds_wq, &conn->c_recv_w, 0); + rds_ib_recv_refill(conn, 0); } int rds_ib_recv(struct rds_connection *conn) @@ -910,25 +1037,13 @@ int rds_ib_recv(struct rds_connection *conn) int ret = 0; rdsdebug("conn %p\n", conn); - - /* - * If we get a temporary posting failure in this context then - * we're really low and we want the caller to back off for a bit. - */ - mutex_lock(&ic->i_recv_mutex); - if (rds_ib_recv_refill(conn, GFP_KERNEL, GFP_HIGHUSER, 0)) - ret = -ENOMEM; - else - rds_ib_stats_inc(s_ib_rx_refill_from_thread); - mutex_unlock(&ic->i_recv_mutex); - if (rds_conn_up(conn)) rds_ib_attempt_ack(ic); return ret; } -int __init rds_ib_recv_init(void) +int rds_ib_recv_init(void) { struct sysinfo si; int ret = -ENOMEM; @@ -939,14 +1054,14 @@ int __init rds_ib_recv_init(void) rds_ib_incoming_slab = kmem_cache_create("rds_ib_incoming", sizeof(struct rds_ib_incoming), - 0, 0, NULL); - if (rds_ib_incoming_slab == NULL) + 0, SLAB_HWCACHE_ALIGN, NULL); + if (!rds_ib_incoming_slab) goto out; rds_ib_frag_slab = kmem_cache_create("rds_ib_frag", sizeof(struct rds_page_frag), - 0, 0, NULL); - if (rds_ib_frag_slab == NULL) + 0, SLAB_HWCACHE_ALIGN, NULL); + if (!rds_ib_frag_slab) kmem_cache_destroy(rds_ib_incoming_slab); else ret = 0; diff --git a/net/rds/ib_send.c b/net/rds/ib_send.c index 17fa80803ab..71f373c421b 100644 --- a/net/rds/ib_send.c +++ b/net/rds/ib_send.c @@ -36,11 +36,49 @@ #include <linux/dmapool.h> #include "rds.h" -#include "rdma.h" #include "ib.h" -static void rds_ib_send_rdma_complete(struct rds_message *rm, - int wc_status) +static char *rds_ib_wc_status_strings[] = { +#define RDS_IB_WC_STATUS_STR(foo) \ + [IB_WC_##foo] = __stringify(IB_WC_##foo) + RDS_IB_WC_STATUS_STR(SUCCESS), + RDS_IB_WC_STATUS_STR(LOC_LEN_ERR), + RDS_IB_WC_STATUS_STR(LOC_QP_OP_ERR), + RDS_IB_WC_STATUS_STR(LOC_EEC_OP_ERR), + RDS_IB_WC_STATUS_STR(LOC_PROT_ERR), + RDS_IB_WC_STATUS_STR(WR_FLUSH_ERR), + RDS_IB_WC_STATUS_STR(MW_BIND_ERR), + RDS_IB_WC_STATUS_STR(BAD_RESP_ERR), + RDS_IB_WC_STATUS_STR(LOC_ACCESS_ERR), + RDS_IB_WC_STATUS_STR(REM_INV_REQ_ERR), + RDS_IB_WC_STATUS_STR(REM_ACCESS_ERR), + RDS_IB_WC_STATUS_STR(REM_OP_ERR), + RDS_IB_WC_STATUS_STR(RETRY_EXC_ERR), + RDS_IB_WC_STATUS_STR(RNR_RETRY_EXC_ERR), + RDS_IB_WC_STATUS_STR(LOC_RDD_VIOL_ERR), + RDS_IB_WC_STATUS_STR(REM_INV_RD_REQ_ERR), + RDS_IB_WC_STATUS_STR(REM_ABORT_ERR), + RDS_IB_WC_STATUS_STR(INV_EECN_ERR), + RDS_IB_WC_STATUS_STR(INV_EEC_STATE_ERR), + RDS_IB_WC_STATUS_STR(FATAL_ERR), + RDS_IB_WC_STATUS_STR(RESP_TIMEOUT_ERR), + RDS_IB_WC_STATUS_STR(GENERAL_ERR), +#undef RDS_IB_WC_STATUS_STR +}; + +char *rds_ib_wc_status_str(enum ib_wc_status status) +{ + return rds_str_array(rds_ib_wc_status_strings, + ARRAY_SIZE(rds_ib_wc_status_strings), status); +} + +/* + * Convert IB-specific error message to RDS error message and call core + * completion handler. + */ +static void rds_ib_send_complete(struct rds_message *rm, + int wc_status, + void (*complete)(struct rds_message *rm, int status)) { int notify_status; @@ -60,69 +98,125 @@ static void rds_ib_send_rdma_complete(struct rds_message *rm, notify_status = RDS_RDMA_OTHER_ERROR; break; } - rds_rdma_send_complete(rm, notify_status); + complete(rm, notify_status); +} + +static void rds_ib_send_unmap_data(struct rds_ib_connection *ic, + struct rm_data_op *op, + int wc_status) +{ + if (op->op_nents) + ib_dma_unmap_sg(ic->i_cm_id->device, + op->op_sg, op->op_nents, + DMA_TO_DEVICE); } static void rds_ib_send_unmap_rdma(struct rds_ib_connection *ic, - struct rds_rdma_op *op) + struct rm_rdma_op *op, + int wc_status) { - if (op->r_mapped) { + if (op->op_mapped) { ib_dma_unmap_sg(ic->i_cm_id->device, - op->r_sg, op->r_nents, - op->r_write ? DMA_TO_DEVICE : DMA_FROM_DEVICE); - op->r_mapped = 0; + op->op_sg, op->op_nents, + op->op_write ? DMA_TO_DEVICE : DMA_FROM_DEVICE); + op->op_mapped = 0; } + + /* If the user asked for a completion notification on this + * message, we can implement three different semantics: + * 1. Notify when we received the ACK on the RDS message + * that was queued with the RDMA. This provides reliable + * notification of RDMA status at the expense of a one-way + * packet delay. + * 2. Notify when the IB stack gives us the completion event for + * the RDMA operation. + * 3. Notify when the IB stack gives us the completion event for + * the accompanying RDS messages. + * Here, we implement approach #3. To implement approach #2, + * we would need to take an event for the rdma WR. To implement #1, + * don't call rds_rdma_send_complete at all, and fall back to the notify + * handling in the ACK processing code. + * + * Note: There's no need to explicitly sync any RDMA buffers using + * ib_dma_sync_sg_for_cpu - the completion for the RDMA + * operation itself unmapped the RDMA buffers, which takes care + * of synching. + */ + rds_ib_send_complete(container_of(op, struct rds_message, rdma), + wc_status, rds_rdma_send_complete); + + if (op->op_write) + rds_stats_add(s_send_rdma_bytes, op->op_bytes); + else + rds_stats_add(s_recv_rdma_bytes, op->op_bytes); } -static void rds_ib_send_unmap_rm(struct rds_ib_connection *ic, - struct rds_ib_send_work *send, - int wc_status) +static void rds_ib_send_unmap_atomic(struct rds_ib_connection *ic, + struct rm_atomic_op *op, + int wc_status) { - struct rds_message *rm = send->s_rm; - - rdsdebug("ic %p send %p rm %p\n", ic, send, rm); - - ib_dma_unmap_sg(ic->i_cm_id->device, - rm->m_sg, rm->m_nents, - DMA_TO_DEVICE); - - if (rm->m_rdma_op != NULL) { - rds_ib_send_unmap_rdma(ic, rm->m_rdma_op); - - /* If the user asked for a completion notification on this - * message, we can implement three different semantics: - * 1. Notify when we received the ACK on the RDS message - * that was queued with the RDMA. This provides reliable - * notification of RDMA status at the expense of a one-way - * packet delay. - * 2. Notify when the IB stack gives us the completion event for - * the RDMA operation. - * 3. Notify when the IB stack gives us the completion event for - * the accompanying RDS messages. - * Here, we implement approach #3. To implement approach #2, - * call rds_rdma_send_complete from the cq_handler. To implement #1, - * don't call rds_rdma_send_complete at all, and fall back to the notify - * handling in the ACK processing code. - * - * Note: There's no need to explicitly sync any RDMA buffers using - * ib_dma_sync_sg_for_cpu - the completion for the RDMA - * operation itself unmapped the RDMA buffers, which takes care - * of synching. - */ - rds_ib_send_rdma_complete(rm, wc_status); + /* unmap atomic recvbuf */ + if (op->op_mapped) { + ib_dma_unmap_sg(ic->i_cm_id->device, op->op_sg, 1, + DMA_FROM_DEVICE); + op->op_mapped = 0; + } - if (rm->m_rdma_op->r_write) - rds_stats_add(s_send_rdma_bytes, rm->m_rdma_op->r_bytes); - else - rds_stats_add(s_recv_rdma_bytes, rm->m_rdma_op->r_bytes); + rds_ib_send_complete(container_of(op, struct rds_message, atomic), + wc_status, rds_atomic_send_complete); + + if (op->op_type == RDS_ATOMIC_TYPE_CSWP) + rds_ib_stats_inc(s_ib_atomic_cswp); + else + rds_ib_stats_inc(s_ib_atomic_fadd); +} + +/* + * Unmap the resources associated with a struct send_work. + * + * Returns the rm for no good reason other than it is unobtainable + * other than by switching on wr.opcode, currently, and the caller, + * the event handler, needs it. + */ +static struct rds_message *rds_ib_send_unmap_op(struct rds_ib_connection *ic, + struct rds_ib_send_work *send, + int wc_status) +{ + struct rds_message *rm = NULL; + + /* In the error case, wc.opcode sometimes contains garbage */ + switch (send->s_wr.opcode) { + case IB_WR_SEND: + if (send->s_op) { + rm = container_of(send->s_op, struct rds_message, data); + rds_ib_send_unmap_data(ic, send->s_op, wc_status); + } + break; + case IB_WR_RDMA_WRITE: + case IB_WR_RDMA_READ: + if (send->s_op) { + rm = container_of(send->s_op, struct rds_message, rdma); + rds_ib_send_unmap_rdma(ic, send->s_op, wc_status); + } + break; + case IB_WR_ATOMIC_FETCH_AND_ADD: + case IB_WR_ATOMIC_CMP_AND_SWP: + if (send->s_op) { + rm = container_of(send->s_op, struct rds_message, atomic); + rds_ib_send_unmap_atomic(ic, send->s_op, wc_status); + } + break; + default: + if (printk_ratelimit()) + printk(KERN_NOTICE + "RDS/IB: %s: unexpected opcode 0x%x in WR!\n", + __func__, send->s_wr.opcode); + break; } - /* If anyone waited for this message to get flushed out, wake - * them up now */ - rds_message_unmapped(rm); + send->s_wr.opcode = 0xdead; - rds_message_put(rm); - send->s_rm = NULL; + return rm; } void rds_ib_send_init_ring(struct rds_ib_connection *ic) @@ -133,23 +227,18 @@ void rds_ib_send_init_ring(struct rds_ib_connection *ic) for (i = 0, send = ic->i_sends; i < ic->i_send_ring.w_nr; i++, send++) { struct ib_sge *sge; - send->s_rm = NULL; send->s_op = NULL; send->s_wr.wr_id = i; send->s_wr.sg_list = send->s_sge; - send->s_wr.num_sge = 1; - send->s_wr.opcode = IB_WR_SEND; - send->s_wr.send_flags = 0; send->s_wr.ex.imm_data = 0; - sge = rds_ib_data_sge(ic, send->s_sge); - sge->lkey = ic->i_mr->lkey; - - sge = rds_ib_header_sge(ic, send->s_sge); + sge = &send->s_sge[0]; sge->addr = ic->i_send_hdrs_dma + (i * sizeof(struct rds_header)); sge->length = sizeof(struct rds_header); sge->lkey = ic->i_mr->lkey; + + send->s_sge[1].lkey = ic->i_mr->lkey; } } @@ -159,16 +248,24 @@ void rds_ib_send_clear_ring(struct rds_ib_connection *ic) u32 i; for (i = 0, send = ic->i_sends; i < ic->i_send_ring.w_nr; i++, send++) { - if (send->s_wr.opcode == 0xdead) - continue; - if (send->s_rm) - rds_ib_send_unmap_rm(ic, send, IB_WC_WR_FLUSH_ERR); - if (send->s_op) - rds_ib_send_unmap_rdma(ic, send->s_op); + if (send->s_op && send->s_wr.opcode != 0xdead) + rds_ib_send_unmap_op(ic, send, IB_WC_WR_FLUSH_ERR); } } /* + * The only fast path caller always has a non-zero nr, so we don't + * bother testing nr before performing the atomic sub. + */ +static void rds_ib_sub_signaled(struct rds_ib_connection *ic, int nr) +{ + if ((atomic_sub_return(nr, &ic->i_signaled_sends) == 0) && + waitqueue_active(&rds_ib_ring_empty_wait)) + wake_up(&rds_ib_ring_empty_wait); + BUG_ON(atomic_read(&ic->i_signaled_sends) < 0); +} + +/* * The _oldest/_free ring operations here race cleanly with the alloc/unalloc * operations performed in the send path. As the sender allocs and potentially * unallocs the next free entry in the ring it doesn't alter which is @@ -178,12 +275,14 @@ void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context) { struct rds_connection *conn = context; struct rds_ib_connection *ic = conn->c_transport_data; + struct rds_message *rm = NULL; struct ib_wc wc; struct rds_ib_send_work *send; u32 completed; u32 oldest; u32 i = 0; int ret; + int nr_sig = 0; rdsdebug("cq %p conn %p\n", cq, conn); rds_ib_stats_inc(s_ib_tx_cq_call); @@ -192,8 +291,9 @@ void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context) rdsdebug("ib_req_notify_cq send failed: %d\n", ret); while (ib_poll_cq(cq, 1, &wc) > 0) { - rdsdebug("wc wr_id 0x%llx status %u byte_len %u imm_data %u\n", - (unsigned long long)wc.wr_id, wc.status, wc.byte_len, + rdsdebug("wc wr_id 0x%llx status %u (%s) byte_len %u imm_data %u\n", + (unsigned long long)wc.wr_id, wc.status, + rds_ib_wc_status_str(wc.status), wc.byte_len, be32_to_cpu(wc.ex.imm_data)); rds_ib_stats_inc(s_ib_tx_cq_event); @@ -210,51 +310,30 @@ void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context) for (i = 0; i < completed; i++) { send = &ic->i_sends[oldest]; + if (send->s_wr.send_flags & IB_SEND_SIGNALED) + nr_sig++; - /* In the error case, wc.opcode sometimes contains garbage */ - switch (send->s_wr.opcode) { - case IB_WR_SEND: - if (send->s_rm) - rds_ib_send_unmap_rm(ic, send, wc.status); - break; - case IB_WR_RDMA_WRITE: - case IB_WR_RDMA_READ: - /* Nothing to be done - the SG list will be unmapped - * when the SEND completes. */ - break; - default: - if (printk_ratelimit()) - printk(KERN_NOTICE - "RDS/IB: %s: unexpected opcode 0x%x in WR!\n", - __func__, send->s_wr.opcode); - break; - } + rm = rds_ib_send_unmap_op(ic, send, wc.status); - send->s_wr.opcode = 0xdead; - send->s_wr.num_sge = 1; if (send->s_queued + HZ/2 < jiffies) rds_ib_stats_inc(s_ib_tx_stalled); - /* If a RDMA operation produced an error, signal this right - * away. If we don't, the subsequent SEND that goes with this - * RDMA will be canceled with ERR_WFLUSH, and the application - * never learn that the RDMA failed. */ - if (unlikely(wc.status == IB_WC_REM_ACCESS_ERR && send->s_op)) { - struct rds_message *rm; - - rm = rds_send_get_message(conn, send->s_op); - if (rm) { - if (rm->m_rdma_op) - rds_ib_send_unmap_rdma(ic, rm->m_rdma_op); - rds_ib_send_rdma_complete(rm, wc.status); - rds_message_put(rm); + if (send->s_op) { + if (send->s_op == rm->m_final_op) { + /* If anyone waited for this message to get flushed out, wake + * them up now */ + rds_message_unmapped(rm); } + rds_message_put(rm); + send->s_op = NULL; } oldest = (oldest + 1) % ic->i_send_ring.w_nr; } rds_ib_ring_free(&ic->i_send_ring, completed); + rds_ib_sub_signaled(ic, nr_sig); + nr_sig = 0; if (test_and_clear_bit(RDS_LL_SEND_FULL, &conn->c_flags) || test_bit(0, &conn->c_map_queued)) @@ -262,10 +341,10 @@ void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context) /* We expect errors as the qp is drained during shutdown */ if (wc.status != IB_WC_SUCCESS && rds_conn_up(conn)) { - rds_ib_conn_error(conn, - "send completion on %pI4 " - "had status %u, disconnecting and reconnecting\n", - &conn->c_faddr, wc.status); + rds_ib_conn_error(conn, "send completion on %pI4 had status " + "%u (%s), disconnecting and reconnecting\n", + &conn->c_faddr, wc.status, + rds_ib_wc_status_str(wc.status)); } } } @@ -294,7 +373,7 @@ void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context) * credits (see rds_ib_send_add_credits below). * * The RDS send code is essentially single-threaded; rds_send_xmit - * grabs c_send_lock to ensure exclusive access to the send ring. + * sets RDS_IN_XMIT to ensure exclusive access to the send ring. * However, the ACK sending code is independent and can race with * message SENDs. * @@ -413,40 +492,21 @@ void rds_ib_advertise_credits(struct rds_connection *conn, unsigned int posted) set_bit(IB_ACK_REQUESTED, &ic->i_ack_flags); } -static inline void -rds_ib_xmit_populate_wr(struct rds_ib_connection *ic, - struct rds_ib_send_work *send, unsigned int pos, - unsigned long buffer, unsigned int length, - int send_flags) +static inline int rds_ib_set_wr_signal_state(struct rds_ib_connection *ic, + struct rds_ib_send_work *send, + bool notify) { - struct ib_sge *sge; - - WARN_ON(pos != send - ic->i_sends); - - send->s_wr.send_flags = send_flags; - send->s_wr.opcode = IB_WR_SEND; - send->s_wr.num_sge = 2; - send->s_wr.next = NULL; - send->s_queued = jiffies; - send->s_op = NULL; - - if (length != 0) { - sge = rds_ib_data_sge(ic, send->s_sge); - sge->addr = buffer; - sge->length = length; - sge->lkey = ic->i_mr->lkey; - - sge = rds_ib_header_sge(ic, send->s_sge); - } else { - /* We're sending a packet with no payload. There is only - * one SGE */ - send->s_wr.num_sge = 1; - sge = &send->s_sge[0]; + /* + * We want to delay signaling completions just enough to get + * the batching benefits but not so much that we create dead time + * on the wire. + */ + if (ic->i_unsignaled_wrs-- == 0 || notify) { + ic->i_unsignaled_wrs = rds_ib_sysctl_max_unsig_wrs; + send->s_wr.send_flags |= IB_SEND_SIGNALED; + return 1; } - - sge->addr = ic->i_send_hdrs_dma + (pos * sizeof(struct rds_header)); - sge->length = sizeof(struct rds_header); - sge->lkey = ic->i_mr->lkey; + return 0; } /* @@ -475,13 +535,14 @@ int rds_ib_xmit(struct rds_connection *conn, struct rds_message *rm, u32 pos; u32 i; u32 work_alloc; - u32 credit_alloc; + u32 credit_alloc = 0; u32 posted; u32 adv_credits = 0; int send_flags = 0; - int sent; + int bytes_sent = 0; int ret; int flow_controlled = 0; + int nr_sig = 0; BUG_ON(off % RDS_FRAG_SIZE); BUG_ON(hdr_off != 0 && hdr_off != sizeof(struct rds_header)); @@ -507,14 +568,13 @@ int rds_ib_xmit(struct rds_connection *conn, struct rds_message *rm, goto out; } - credit_alloc = work_alloc; if (ic->i_flowctl) { credit_alloc = rds_ib_send_grab_credits(ic, work_alloc, &posted, 0, RDS_MAX_ADV_CREDIT); adv_credits += posted; if (credit_alloc < work_alloc) { rds_ib_ring_unalloc(&ic->i_send_ring, work_alloc - credit_alloc); work_alloc = credit_alloc; - flow_controlled++; + flow_controlled = 1; } if (work_alloc == 0) { set_bit(RDS_LL_SEND_FULL, &conn->c_flags); @@ -525,31 +585,25 @@ int rds_ib_xmit(struct rds_connection *conn, struct rds_message *rm, } /* map the message the first time we see it */ - if (ic->i_rm == NULL) { - /* - printk(KERN_NOTICE "rds_ib_xmit prep msg dport=%u flags=0x%x len=%d\n", - be16_to_cpu(rm->m_inc.i_hdr.h_dport), - rm->m_inc.i_hdr.h_flags, - be32_to_cpu(rm->m_inc.i_hdr.h_len)); - */ - if (rm->m_nents) { - rm->m_count = ib_dma_map_sg(dev, - rm->m_sg, rm->m_nents, DMA_TO_DEVICE); - rdsdebug("ic %p mapping rm %p: %d\n", ic, rm, rm->m_count); - if (rm->m_count == 0) { + if (!ic->i_data_op) { + if (rm->data.op_nents) { + rm->data.op_count = ib_dma_map_sg(dev, + rm->data.op_sg, + rm->data.op_nents, + DMA_TO_DEVICE); + rdsdebug("ic %p mapping rm %p: %d\n", ic, rm, rm->data.op_count); + if (rm->data.op_count == 0) { rds_ib_stats_inc(s_ib_tx_sg_mapping_failure); rds_ib_ring_unalloc(&ic->i_send_ring, work_alloc); ret = -ENOMEM; /* XXX ? */ goto out; } } else { - rm->m_count = 0; + rm->data.op_count = 0; } - ic->i_unsignaled_wrs = rds_ib_sysctl_max_unsig_wrs; - ic->i_unsignaled_bytes = rds_ib_sysctl_max_unsig_bytes; rds_message_addref(rm); - ic->i_rm = rm; + ic->i_data_op = &rm->data; /* Finalize the header */ if (test_bit(RDS_MSG_ACK_REQUIRED, &rm->m_flags)) @@ -559,10 +613,10 @@ int rds_ib_xmit(struct rds_connection *conn, struct rds_message *rm, /* If it has a RDMA op, tell the peer we did it. This is * used by the peer to release use-once RDMA MRs. */ - if (rm->m_rdma_op) { + if (rm->rdma.op_active) { struct rds_ext_header_rdma ext_hdr; - ext_hdr.h_rdma_rkey = cpu_to_be32(rm->m_rdma_op->r_key); + ext_hdr.h_rdma_rkey = cpu_to_be32(rm->rdma.op_rkey); rds_message_add_extension(&rm->m_inc.i_hdr, RDS_EXTHDR_RDMA, &ext_hdr, sizeof(ext_hdr)); } @@ -582,99 +636,77 @@ int rds_ib_xmit(struct rds_connection *conn, struct rds_message *rm, /* * Update adv_credits since we reset the ACK_REQUIRED bit. */ - rds_ib_send_grab_credits(ic, 0, &posted, 1, RDS_MAX_ADV_CREDIT - adv_credits); - adv_credits += posted; - BUG_ON(adv_credits > 255); + if (ic->i_flowctl) { + rds_ib_send_grab_credits(ic, 0, &posted, 1, RDS_MAX_ADV_CREDIT - adv_credits); + adv_credits += posted; + BUG_ON(adv_credits > 255); + } } - send = &ic->i_sends[pos]; - first = send; - prev = NULL; - scat = &rm->m_sg[sg]; - sent = 0; - i = 0; - /* Sometimes you want to put a fence between an RDMA * READ and the following SEND. * We could either do this all the time * or when requested by the user. Right now, we let * the application choose. */ - if (rm->m_rdma_op && rm->m_rdma_op->r_fence) + if (rm->rdma.op_active && rm->rdma.op_fence) send_flags = IB_SEND_FENCE; - /* - * We could be copying the header into the unused tail of the page. - * That would need to be changed in the future when those pages might - * be mapped userspace pages or page cache pages. So instead we always - * use a second sge and our long-lived ring of mapped headers. We send - * the header after the data so that the data payload can be aligned on - * the receiver. - */ + /* Each frag gets a header. Msgs may be 0 bytes */ + send = &ic->i_sends[pos]; + first = send; + prev = NULL; + scat = &ic->i_data_op->op_sg[sg]; + i = 0; + do { + unsigned int len = 0; - /* handle a 0-len message */ - if (be32_to_cpu(rm->m_inc.i_hdr.h_len) == 0) { - rds_ib_xmit_populate_wr(ic, send, pos, 0, 0, send_flags); - goto add_header; - } + /* Set up the header */ + send->s_wr.send_flags = send_flags; + send->s_wr.opcode = IB_WR_SEND; + send->s_wr.num_sge = 1; + send->s_wr.next = NULL; + send->s_queued = jiffies; + send->s_op = NULL; - /* if there's data reference it with a chain of work reqs */ - for (; i < work_alloc && scat != &rm->m_sg[rm->m_count]; i++) { - unsigned int len; + send->s_sge[0].addr = ic->i_send_hdrs_dma + + (pos * sizeof(struct rds_header)); + send->s_sge[0].length = sizeof(struct rds_header); - send = &ic->i_sends[pos]; + memcpy(&ic->i_send_hdrs[pos], &rm->m_inc.i_hdr, sizeof(struct rds_header)); - len = min(RDS_FRAG_SIZE, ib_sg_dma_len(dev, scat) - off); - rds_ib_xmit_populate_wr(ic, send, pos, - ib_sg_dma_address(dev, scat) + off, len, - send_flags); + /* Set up the data, if present */ + if (i < work_alloc + && scat != &rm->data.op_sg[rm->data.op_count]) { + len = min(RDS_FRAG_SIZE, ib_sg_dma_len(dev, scat) - off); + send->s_wr.num_sge = 2; - /* - * We want to delay signaling completions just enough to get - * the batching benefits but not so much that we create dead time - * on the wire. - */ - if (ic->i_unsignaled_wrs-- == 0) { - ic->i_unsignaled_wrs = rds_ib_sysctl_max_unsig_wrs; - send->s_wr.send_flags |= IB_SEND_SIGNALED | IB_SEND_SOLICITED; - } + send->s_sge[1].addr = ib_sg_dma_address(dev, scat) + off; + send->s_sge[1].length = len; - ic->i_unsignaled_bytes -= len; - if (ic->i_unsignaled_bytes <= 0) { - ic->i_unsignaled_bytes = rds_ib_sysctl_max_unsig_bytes; - send->s_wr.send_flags |= IB_SEND_SIGNALED | IB_SEND_SOLICITED; + bytes_sent += len; + off += len; + if (off == ib_sg_dma_len(dev, scat)) { + scat++; + off = 0; + } } + rds_ib_set_wr_signal_state(ic, send, 0); + /* * Always signal the last one if we're stopping due to flow control. */ - if (flow_controlled && i == (work_alloc-1)) + if (ic->i_flowctl && flow_controlled && i == (work_alloc-1)) send->s_wr.send_flags |= IB_SEND_SIGNALED | IB_SEND_SOLICITED; + if (send->s_wr.send_flags & IB_SEND_SIGNALED) + nr_sig++; + rdsdebug("send %p wr %p num_sge %u next %p\n", send, &send->s_wr, send->s_wr.num_sge, send->s_wr.next); - sent += len; - off += len; - if (off == ib_sg_dma_len(dev, scat)) { - scat++; - off = 0; - } - -add_header: - /* Tack on the header after the data. The header SGE should already - * have been set up to point to the right header buffer. */ - memcpy(&ic->i_send_hdrs[pos], &rm->m_inc.i_hdr, sizeof(struct rds_header)); - - if (0) { - struct rds_header *hdr = &ic->i_send_hdrs[pos]; - - printk(KERN_NOTICE "send WR dport=%u flags=0x%x len=%d\n", - be16_to_cpu(hdr->h_dport), - hdr->h_flags, - be32_to_cpu(hdr->h_len)); - } - if (adv_credits) { + if (ic->i_flowctl && adv_credits) { struct rds_header *hdr = &ic->i_send_hdrs[pos]; /* add credit and redo the header checksum */ @@ -689,20 +721,25 @@ add_header: prev = send; pos = (pos + 1) % ic->i_send_ring.w_nr; - } + send = &ic->i_sends[pos]; + i++; + + } while (i < work_alloc + && scat != &rm->data.op_sg[rm->data.op_count]); /* Account the RDS header in the number of bytes we sent, but just once. * The caller has no concept of fragmentation. */ if (hdr_off == 0) - sent += sizeof(struct rds_header); + bytes_sent += sizeof(struct rds_header); /* if we finished the message then send completion owns it */ - if (scat == &rm->m_sg[rm->m_count]) { - prev->s_rm = ic->i_rm; - prev->s_wr.send_flags |= IB_SEND_SIGNALED | IB_SEND_SOLICITED; - ic->i_rm = NULL; + if (scat == &rm->data.op_sg[rm->data.op_count]) { + prev->s_op = ic->i_data_op; + prev->s_wr.send_flags |= IB_SEND_SOLICITED; + ic->i_data_op = NULL; } + /* Put back wrs & credits we didn't use */ if (i < work_alloc) { rds_ib_ring_unalloc(&ic->i_send_ring, work_alloc - i); work_alloc = i; @@ -710,6 +747,9 @@ add_header: if (ic->i_flowctl && i < credit_alloc) rds_ib_send_add_credits(conn, credit_alloc - i); + if (nr_sig) + atomic_add(nr_sig, &ic->i_signaled_sends); + /* XXX need to worry about failed_wr and partial sends. */ failed_wr = &first->s_wr; ret = ib_post_send(ic->i_cm_id->qp, &first->s_wr, &failed_wr); @@ -720,32 +760,127 @@ add_header: printk(KERN_WARNING "RDS/IB: ib_post_send to %pI4 " "returned %d\n", &conn->c_faddr, ret); rds_ib_ring_unalloc(&ic->i_send_ring, work_alloc); - if (prev->s_rm) { - ic->i_rm = prev->s_rm; - prev->s_rm = NULL; + rds_ib_sub_signaled(ic, nr_sig); + if (prev->s_op) { + ic->i_data_op = prev->s_op; + prev->s_op = NULL; } rds_ib_conn_error(ic->conn, "ib_post_send failed\n"); goto out; } - ret = sent; + ret = bytes_sent; out: BUG_ON(adv_credits); return ret; } -int rds_ib_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) +/* + * Issue atomic operation. + * A simplified version of the rdma case, we always map 1 SG, and + * only 8 bytes, for the return value from the atomic operation. + */ +int rds_ib_xmit_atomic(struct rds_connection *conn, struct rm_atomic_op *op) +{ + struct rds_ib_connection *ic = conn->c_transport_data; + struct rds_ib_send_work *send = NULL; + struct ib_send_wr *failed_wr; + struct rds_ib_device *rds_ibdev; + u32 pos; + u32 work_alloc; + int ret; + int nr_sig = 0; + + rds_ibdev = ib_get_client_data(ic->i_cm_id->device, &rds_ib_client); + + work_alloc = rds_ib_ring_alloc(&ic->i_send_ring, 1, &pos); + if (work_alloc != 1) { + rds_ib_ring_unalloc(&ic->i_send_ring, work_alloc); + rds_ib_stats_inc(s_ib_tx_ring_full); + ret = -ENOMEM; + goto out; + } + + /* address of send request in ring */ + send = &ic->i_sends[pos]; + send->s_queued = jiffies; + + if (op->op_type == RDS_ATOMIC_TYPE_CSWP) { + send->s_wr.opcode = IB_WR_MASKED_ATOMIC_CMP_AND_SWP; + send->s_wr.wr.atomic.compare_add = op->op_m_cswp.compare; + send->s_wr.wr.atomic.swap = op->op_m_cswp.swap; + send->s_wr.wr.atomic.compare_add_mask = op->op_m_cswp.compare_mask; + send->s_wr.wr.atomic.swap_mask = op->op_m_cswp.swap_mask; + } else { /* FADD */ + send->s_wr.opcode = IB_WR_MASKED_ATOMIC_FETCH_AND_ADD; + send->s_wr.wr.atomic.compare_add = op->op_m_fadd.add; + send->s_wr.wr.atomic.swap = 0; + send->s_wr.wr.atomic.compare_add_mask = op->op_m_fadd.nocarry_mask; + send->s_wr.wr.atomic.swap_mask = 0; + } + nr_sig = rds_ib_set_wr_signal_state(ic, send, op->op_notify); + send->s_wr.num_sge = 1; + send->s_wr.next = NULL; + send->s_wr.wr.atomic.remote_addr = op->op_remote_addr; + send->s_wr.wr.atomic.rkey = op->op_rkey; + send->s_op = op; + rds_message_addref(container_of(send->s_op, struct rds_message, atomic)); + + /* map 8 byte retval buffer to the device */ + ret = ib_dma_map_sg(ic->i_cm_id->device, op->op_sg, 1, DMA_FROM_DEVICE); + rdsdebug("ic %p mapping atomic op %p. mapped %d pg\n", ic, op, ret); + if (ret != 1) { + rds_ib_ring_unalloc(&ic->i_send_ring, work_alloc); + rds_ib_stats_inc(s_ib_tx_sg_mapping_failure); + ret = -ENOMEM; /* XXX ? */ + goto out; + } + + /* Convert our struct scatterlist to struct ib_sge */ + send->s_sge[0].addr = ib_sg_dma_address(ic->i_cm_id->device, op->op_sg); + send->s_sge[0].length = ib_sg_dma_len(ic->i_cm_id->device, op->op_sg); + send->s_sge[0].lkey = ic->i_mr->lkey; + + rdsdebug("rva %Lx rpa %Lx len %u\n", op->op_remote_addr, + send->s_sge[0].addr, send->s_sge[0].length); + + if (nr_sig) + atomic_add(nr_sig, &ic->i_signaled_sends); + + failed_wr = &send->s_wr; + ret = ib_post_send(ic->i_cm_id->qp, &send->s_wr, &failed_wr); + rdsdebug("ic %p send %p (wr %p) ret %d wr %p\n", ic, + send, &send->s_wr, ret, failed_wr); + BUG_ON(failed_wr != &send->s_wr); + if (ret) { + printk(KERN_WARNING "RDS/IB: atomic ib_post_send to %pI4 " + "returned %d\n", &conn->c_faddr, ret); + rds_ib_ring_unalloc(&ic->i_send_ring, work_alloc); + rds_ib_sub_signaled(ic, nr_sig); + goto out; + } + + if (unlikely(failed_wr != &send->s_wr)) { + printk(KERN_WARNING "RDS/IB: atomic ib_post_send() rc=%d, but failed_wqe updated!\n", ret); + BUG_ON(failed_wr != &send->s_wr); + } + +out: + return ret; +} + +int rds_ib_xmit_rdma(struct rds_connection *conn, struct rm_rdma_op *op) { struct rds_ib_connection *ic = conn->c_transport_data; struct rds_ib_send_work *send = NULL; struct rds_ib_send_work *first; struct rds_ib_send_work *prev; struct ib_send_wr *failed_wr; - struct rds_ib_device *rds_ibdev; struct scatterlist *scat; unsigned long len; - u64 remote_addr = op->r_remote_addr; + u64 remote_addr = op->op_remote_addr; + u32 max_sge = ic->rds_ibdev->max_sge; u32 pos; u32 work_alloc; u32 i; @@ -753,29 +888,28 @@ int rds_ib_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) int sent; int ret; int num_sge; - - rds_ibdev = ib_get_client_data(ic->i_cm_id->device, &rds_ib_client); - - /* map the message the first time we see it */ - if (!op->r_mapped) { - op->r_count = ib_dma_map_sg(ic->i_cm_id->device, - op->r_sg, op->r_nents, (op->r_write) ? - DMA_TO_DEVICE : DMA_FROM_DEVICE); - rdsdebug("ic %p mapping op %p: %d\n", ic, op, op->r_count); - if (op->r_count == 0) { + int nr_sig = 0; + + /* map the op the first time we see it */ + if (!op->op_mapped) { + op->op_count = ib_dma_map_sg(ic->i_cm_id->device, + op->op_sg, op->op_nents, (op->op_write) ? + DMA_TO_DEVICE : DMA_FROM_DEVICE); + rdsdebug("ic %p mapping op %p: %d\n", ic, op, op->op_count); + if (op->op_count == 0) { rds_ib_stats_inc(s_ib_tx_sg_mapping_failure); ret = -ENOMEM; /* XXX ? */ goto out; } - op->r_mapped = 1; + op->op_mapped = 1; } /* * Instead of knowing how to return a partial rdma read/write we insist that there * be enough work requests to send the entire message. */ - i = ceil(op->r_count, rds_ibdev->max_sge); + i = ceil(op->op_count, max_sge); work_alloc = rds_ib_ring_alloc(&ic->i_send_ring, i, &pos); if (work_alloc != i) { @@ -788,30 +922,24 @@ int rds_ib_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) send = &ic->i_sends[pos]; first = send; prev = NULL; - scat = &op->r_sg[0]; + scat = &op->op_sg[0]; sent = 0; - num_sge = op->r_count; + num_sge = op->op_count; - for (i = 0; i < work_alloc && scat != &op->r_sg[op->r_count]; i++) { + for (i = 0; i < work_alloc && scat != &op->op_sg[op->op_count]; i++) { send->s_wr.send_flags = 0; send->s_queued = jiffies; - /* - * We want to delay signaling completions just enough to get - * the batching benefits but not so much that we create dead time on the wire. - */ - if (ic->i_unsignaled_wrs-- == 0) { - ic->i_unsignaled_wrs = rds_ib_sysctl_max_unsig_wrs; - send->s_wr.send_flags = IB_SEND_SIGNALED; - } + send->s_op = NULL; + + nr_sig += rds_ib_set_wr_signal_state(ic, send, op->op_notify); - send->s_wr.opcode = op->r_write ? IB_WR_RDMA_WRITE : IB_WR_RDMA_READ; + send->s_wr.opcode = op->op_write ? IB_WR_RDMA_WRITE : IB_WR_RDMA_READ; send->s_wr.wr.rdma.remote_addr = remote_addr; - send->s_wr.wr.rdma.rkey = op->r_key; - send->s_op = op; + send->s_wr.wr.rdma.rkey = op->op_rkey; - if (num_sge > rds_ibdev->max_sge) { - send->s_wr.num_sge = rds_ibdev->max_sge; - num_sge -= rds_ibdev->max_sge; + if (num_sge > max_sge) { + send->s_wr.num_sge = max_sge; + num_sge -= max_sge; } else { send->s_wr.num_sge = num_sge; } @@ -821,7 +949,7 @@ int rds_ib_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) if (prev) prev->s_wr.next = &send->s_wr; - for (j = 0; j < send->s_wr.num_sge && scat != &op->r_sg[op->r_count]; j++) { + for (j = 0; j < send->s_wr.num_sge && scat != &op->op_sg[op->op_count]; j++) { len = ib_sg_dma_len(ic->i_cm_id->device, scat); send->s_sge[j].addr = ib_sg_dma_address(ic->i_cm_id->device, scat); @@ -843,15 +971,20 @@ int rds_ib_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) send = ic->i_sends; } - /* if we finished the message then send completion owns it */ - if (scat == &op->r_sg[op->r_count]) - prev->s_wr.send_flags = IB_SEND_SIGNALED; + /* give a reference to the last op */ + if (scat == &op->op_sg[op->op_count]) { + prev->s_op = op; + rds_message_addref(container_of(op, struct rds_message, rdma)); + } if (i < work_alloc) { rds_ib_ring_unalloc(&ic->i_send_ring, work_alloc - i); work_alloc = i; } + if (nr_sig) + atomic_add(nr_sig, &ic->i_signaled_sends); + failed_wr = &first->s_wr; ret = ib_post_send(ic->i_cm_id->qp, &first->s_wr, &failed_wr); rdsdebug("ic %p first %p (wr %p) ret %d wr %p\n", ic, @@ -861,6 +994,7 @@ int rds_ib_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) printk(KERN_WARNING "RDS/IB: rdma ib_post_send to %pI4 " "returned %d\n", &conn->c_faddr, ret); rds_ib_ring_unalloc(&ic->i_send_ring, work_alloc); + rds_ib_sub_signaled(ic, nr_sig); goto out; } diff --git a/net/rds/ib_stats.c b/net/rds/ib_stats.c index d2c904dd6fb..2d5965d6e97 100644 --- a/net/rds/ib_stats.c +++ b/net/rds/ib_stats.c @@ -67,6 +67,8 @@ static const char *const rds_ib_stat_names[] = { "ib_rdma_mr_pool_flush", "ib_rdma_mr_pool_wait", "ib_rdma_mr_pool_depleted", + "ib_atomic_cswp", + "ib_atomic_fadd", }; unsigned int rds_ib_stats_info_copy(struct rds_info_iterator *iter, diff --git a/net/rds/ib_sysctl.c b/net/rds/ib_sysctl.c index 03f01cb4e0f..1253b006efd 100644 --- a/net/rds/ib_sysctl.c +++ b/net/rds/ib_sysctl.c @@ -49,10 +49,6 @@ unsigned long rds_ib_sysctl_max_unsig_wrs = 16; static unsigned long rds_ib_sysctl_max_unsig_wr_min = 1; static unsigned long rds_ib_sysctl_max_unsig_wr_max = 64; -unsigned long rds_ib_sysctl_max_unsig_bytes = (16 << 20); -static unsigned long rds_ib_sysctl_max_unsig_bytes_min = 1; -static unsigned long rds_ib_sysctl_max_unsig_bytes_max = ~0UL; - /* * This sysctl does nothing. * @@ -65,7 +61,7 @@ static unsigned long rds_ib_sysctl_max_unsig_bytes_max = ~0UL; */ unsigned int rds_ib_sysctl_flow_control = 0; -ctl_table rds_ib_sysctl_table[] = { +static ctl_table rds_ib_sysctl_table[] = { { .procname = "max_send_wr", .data = &rds_ib_sysctl_max_send_wr, @@ -94,15 +90,6 @@ ctl_table rds_ib_sysctl_table[] = { .extra2 = &rds_ib_sysctl_max_unsig_wr_max, }, { - .procname = "max_unsignaled_bytes", - .data = &rds_ib_sysctl_max_unsig_bytes, - .maxlen = sizeof(unsigned long), - .mode = 0644, - .proc_handler = proc_doulongvec_minmax, - .extra1 = &rds_ib_sysctl_max_unsig_bytes_min, - .extra2 = &rds_ib_sysctl_max_unsig_bytes_max, - }, - { .procname = "max_recv_allocation", .data = &rds_ib_sysctl_max_recv_allocation, .maxlen = sizeof(unsigned long), @@ -132,10 +119,10 @@ void rds_ib_sysctl_exit(void) unregister_sysctl_table(rds_ib_sysctl_hdr); } -int __init rds_ib_sysctl_init(void) +int rds_ib_sysctl_init(void) { rds_ib_sysctl_hdr = register_sysctl_paths(rds_ib_sysctl_path, rds_ib_sysctl_table); - if (rds_ib_sysctl_hdr == NULL) + if (!rds_ib_sysctl_hdr) return -ENOMEM; return 0; } diff --git a/net/rds/info.c b/net/rds/info.c index c45c4173a44..4fdf1b6e84f 100644 --- a/net/rds/info.c +++ b/net/rds/info.c @@ -76,7 +76,7 @@ void rds_info_register_func(int optname, rds_info_func func) BUG_ON(optname < RDS_INFO_FIRST || optname > RDS_INFO_LAST); spin_lock(&rds_info_lock); - BUG_ON(rds_info_funcs[offset] != NULL); + BUG_ON(rds_info_funcs[offset]); rds_info_funcs[offset] = func; spin_unlock(&rds_info_lock); } @@ -102,7 +102,7 @@ EXPORT_SYMBOL_GPL(rds_info_deregister_func); */ void rds_info_iter_unmap(struct rds_info_iterator *iter) { - if (iter->addr != NULL) { + if (iter->addr) { kunmap_atomic(iter->addr, KM_USER0); iter->addr = NULL; } @@ -117,7 +117,7 @@ void rds_info_copy(struct rds_info_iterator *iter, void *data, unsigned long this; while (bytes) { - if (iter->addr == NULL) + if (!iter->addr) iter->addr = kmap_atomic(*iter->pages, KM_USER0); this = min(bytes, PAGE_SIZE - iter->offset); @@ -188,7 +188,7 @@ int rds_info_getsockopt(struct socket *sock, int optname, char __user *optval, >> PAGE_SHIFT; pages = kmalloc(nr_pages * sizeof(struct page *), GFP_KERNEL); - if (pages == NULL) { + if (!pages) { ret = -ENOMEM; goto out; } @@ -206,7 +206,7 @@ int rds_info_getsockopt(struct socket *sock, int optname, char __user *optval, call_func: func = rds_info_funcs[optname - RDS_INFO_FIRST]; - if (func == NULL) { + if (!func) { ret = -ENOPROTOOPT; goto out; } @@ -234,7 +234,7 @@ call_func: ret = -EFAULT; out: - for (i = 0; pages != NULL && i < nr_pages; i++) + for (i = 0; pages && i < nr_pages; i++) put_page(pages[i]); kfree(pages); diff --git a/net/rds/iw.c b/net/rds/iw.c index c8f3d3525cb..5a9676fe594 100644 --- a/net/rds/iw.c +++ b/net/rds/iw.c @@ -56,7 +56,7 @@ struct list_head rds_iw_devices; DEFINE_SPINLOCK(iw_nodev_conns_lock); LIST_HEAD(iw_nodev_conns); -void rds_iw_add_one(struct ib_device *device) +static void rds_iw_add_one(struct ib_device *device) { struct rds_iw_device *rds_iwdev; struct ib_device_attr *dev_attr; @@ -124,7 +124,7 @@ free_attr: kfree(dev_attr); } -void rds_iw_remove_one(struct ib_device *device) +static void rds_iw_remove_one(struct ib_device *device) { struct rds_iw_device *rds_iwdev; struct rds_iw_cm_id *i_cm_id, *next; @@ -264,7 +264,6 @@ struct rds_transport rds_iw_transport = { .laddr_check = rds_iw_laddr_check, .xmit_complete = rds_iw_xmit_complete, .xmit = rds_iw_xmit, - .xmit_cong_map = NULL, .xmit_rdma = rds_iw_xmit_rdma, .recv = rds_iw_recv, .conn_alloc = rds_iw_conn_alloc, @@ -272,7 +271,6 @@ struct rds_transport rds_iw_transport = { .conn_connect = rds_iw_conn_connect, .conn_shutdown = rds_iw_conn_shutdown, .inc_copy_to_user = rds_iw_inc_copy_to_user, - .inc_purge = rds_iw_inc_purge, .inc_free = rds_iw_inc_free, .cm_initiate_connect = rds_iw_cm_initiate_connect, .cm_handle_connect = rds_iw_cm_handle_connect, @@ -289,7 +287,7 @@ struct rds_transport rds_iw_transport = { .t_prefer_loopback = 1, }; -int __init rds_iw_init(void) +int rds_iw_init(void) { int ret; diff --git a/net/rds/iw.h b/net/rds/iw.h index eef2f0c2847..90151922178 100644 --- a/net/rds/iw.h +++ b/net/rds/iw.h @@ -70,7 +70,7 @@ struct rds_iw_send_work { struct rds_message *s_rm; /* We should really put these into a union: */ - struct rds_rdma_op *s_op; + struct rm_rdma_op *s_op; struct rds_iw_mapping *s_mapping; struct ib_mr *s_mr; struct ib_fast_reg_page_list *s_page_list; @@ -268,8 +268,6 @@ static inline u32 rds_iw_local_dma_lkey(struct rds_iw_connection *ic) /* ib.c */ extern struct rds_transport rds_iw_transport; -extern void rds_iw_add_one(struct ib_device *device); -extern void rds_iw_remove_one(struct ib_device *device); extern struct ib_client rds_iw_client; extern unsigned int fastreg_pool_size; @@ -284,7 +282,7 @@ void rds_iw_conn_free(void *arg); int rds_iw_conn_connect(struct rds_connection *conn); void rds_iw_conn_shutdown(struct rds_connection *conn); void rds_iw_state_change(struct sock *sk); -int __init rds_iw_listen_init(void); +int rds_iw_listen_init(void); void rds_iw_listen_stop(void); void __rds_iw_conn_error(struct rds_connection *conn, const char *, ...); int rds_iw_cm_handle_connect(struct rdma_cm_id *cm_id, @@ -318,15 +316,13 @@ void *rds_iw_get_mr(struct scatterlist *sg, unsigned long nents, void rds_iw_sync_mr(void *trans_private, int dir); void rds_iw_free_mr(void *trans_private, int invalidate); void rds_iw_flush_mrs(void); -void rds_iw_remove_cm_id(struct rds_iw_device *rds_iwdev, struct rdma_cm_id *cm_id); /* ib_recv.c */ -int __init rds_iw_recv_init(void); +int rds_iw_recv_init(void); void rds_iw_recv_exit(void); int rds_iw_recv(struct rds_connection *conn); int rds_iw_recv_refill(struct rds_connection *conn, gfp_t kptr_gfp, gfp_t page_gfp, int prefill); -void rds_iw_inc_purge(struct rds_incoming *inc); void rds_iw_inc_free(struct rds_incoming *inc); int rds_iw_inc_copy_to_user(struct rds_incoming *inc, struct iovec *iov, size_t size); @@ -358,7 +354,7 @@ int rds_iw_xmit(struct rds_connection *conn, struct rds_message *rm, void rds_iw_send_cq_comp_handler(struct ib_cq *cq, void *context); void rds_iw_send_init_ring(struct rds_iw_connection *ic); void rds_iw_send_clear_ring(struct rds_iw_connection *ic); -int rds_iw_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op); +int rds_iw_xmit_rdma(struct rds_connection *conn, struct rm_rdma_op *op); void rds_iw_send_add_credits(struct rds_connection *conn, unsigned int credits); void rds_iw_advertise_credits(struct rds_connection *conn, unsigned int posted); int rds_iw_send_grab_credits(struct rds_iw_connection *ic, u32 wanted, @@ -371,7 +367,7 @@ unsigned int rds_iw_stats_info_copy(struct rds_info_iterator *iter, unsigned int avail); /* ib_sysctl.c */ -int __init rds_iw_sysctl_init(void); +int rds_iw_sysctl_init(void); void rds_iw_sysctl_exit(void); extern unsigned long rds_iw_sysctl_max_send_wr; extern unsigned long rds_iw_sysctl_max_recv_wr; @@ -379,7 +375,6 @@ extern unsigned long rds_iw_sysctl_max_unsig_wrs; extern unsigned long rds_iw_sysctl_max_unsig_bytes; extern unsigned long rds_iw_sysctl_max_recv_allocation; extern unsigned int rds_iw_sysctl_flow_control; -extern ctl_table rds_iw_sysctl_table[]; /* * Helper functions for getting/setting the header and data SGEs in diff --git a/net/rds/iw_cm.c b/net/rds/iw_cm.c index b5dd6ac39be..712cf2d1f28 100644 --- a/net/rds/iw_cm.c +++ b/net/rds/iw_cm.c @@ -257,7 +257,7 @@ static int rds_iw_setup_qp(struct rds_connection *conn) * the rds_iwdev at all. */ rds_iwdev = ib_get_client_data(dev, &rds_iw_client); - if (rds_iwdev == NULL) { + if (!rds_iwdev) { if (printk_ratelimit()) printk(KERN_NOTICE "RDS/IW: No client_data for device %s\n", dev->name); @@ -292,7 +292,7 @@ static int rds_iw_setup_qp(struct rds_connection *conn) ic->i_send_ring.w_nr * sizeof(struct rds_header), &ic->i_send_hdrs_dma, GFP_KERNEL); - if (ic->i_send_hdrs == NULL) { + if (!ic->i_send_hdrs) { ret = -ENOMEM; rdsdebug("ib_dma_alloc_coherent send failed\n"); goto out; @@ -302,7 +302,7 @@ static int rds_iw_setup_qp(struct rds_connection *conn) ic->i_recv_ring.w_nr * sizeof(struct rds_header), &ic->i_recv_hdrs_dma, GFP_KERNEL); - if (ic->i_recv_hdrs == NULL) { + if (!ic->i_recv_hdrs) { ret = -ENOMEM; rdsdebug("ib_dma_alloc_coherent recv failed\n"); goto out; @@ -310,14 +310,14 @@ static int rds_iw_setup_qp(struct rds_connection *conn) ic->i_ack = ib_dma_alloc_coherent(dev, sizeof(struct rds_header), &ic->i_ack_dma, GFP_KERNEL); - if (ic->i_ack == NULL) { + if (!ic->i_ack) { ret = -ENOMEM; rdsdebug("ib_dma_alloc_coherent ack failed\n"); goto out; } ic->i_sends = vmalloc(ic->i_send_ring.w_nr * sizeof(struct rds_iw_send_work)); - if (ic->i_sends == NULL) { + if (!ic->i_sends) { ret = -ENOMEM; rdsdebug("send allocation failed\n"); goto out; @@ -325,7 +325,7 @@ static int rds_iw_setup_qp(struct rds_connection *conn) rds_iw_send_init_ring(ic); ic->i_recvs = vmalloc(ic->i_recv_ring.w_nr * sizeof(struct rds_iw_recv_work)); - if (ic->i_recvs == NULL) { + if (!ic->i_recvs) { ret = -ENOMEM; rdsdebug("recv allocation failed\n"); goto out; @@ -696,7 +696,7 @@ int rds_iw_conn_alloc(struct rds_connection *conn, gfp_t gfp) /* XXX too lazy? */ ic = kzalloc(sizeof(struct rds_iw_connection), GFP_KERNEL); - if (ic == NULL) + if (!ic) return -ENOMEM; INIT_LIST_HEAD(&ic->iw_node); diff --git a/net/rds/iw_rdma.c b/net/rds/iw_rdma.c index 13dc1862d86..59509e9a9e7 100644 --- a/net/rds/iw_rdma.c +++ b/net/rds/iw_rdma.c @@ -34,7 +34,6 @@ #include <linux/slab.h> #include "rds.h" -#include "rdma.h" #include "iw.h" @@ -158,7 +157,8 @@ static int rds_iw_add_cm_id(struct rds_iw_device *rds_iwdev, struct rdma_cm_id * return 0; } -void rds_iw_remove_cm_id(struct rds_iw_device *rds_iwdev, struct rdma_cm_id *cm_id) +static void rds_iw_remove_cm_id(struct rds_iw_device *rds_iwdev, + struct rdma_cm_id *cm_id) { struct rds_iw_cm_id *i_cm_id; @@ -207,9 +207,9 @@ void rds_iw_add_conn(struct rds_iw_device *rds_iwdev, struct rds_connection *con BUG_ON(list_empty(&ic->iw_node)); list_del(&ic->iw_node); - spin_lock_irq(&rds_iwdev->spinlock); + spin_lock(&rds_iwdev->spinlock); list_add_tail(&ic->iw_node, &rds_iwdev->conn_list); - spin_unlock_irq(&rds_iwdev->spinlock); + spin_unlock(&rds_iwdev->spinlock); spin_unlock_irq(&iw_nodev_conns_lock); ic->rds_iwdev = rds_iwdev; diff --git a/net/rds/iw_recv.c b/net/rds/iw_recv.c index 3d479067d54..5e57347f49f 100644 --- a/net/rds/iw_recv.c +++ b/net/rds/iw_recv.c @@ -53,7 +53,7 @@ static void rds_iw_frag_drop_page(struct rds_page_frag *frag) static void rds_iw_frag_free(struct rds_page_frag *frag) { rdsdebug("frag %p page %p\n", frag, frag->f_page); - BUG_ON(frag->f_page != NULL); + BUG_ON(frag->f_page); kmem_cache_free(rds_iw_frag_slab, frag); } @@ -143,14 +143,14 @@ static int rds_iw_recv_refill_one(struct rds_connection *conn, struct ib_sge *sge; int ret = -ENOMEM; - if (recv->r_iwinc == NULL) { + if (!recv->r_iwinc) { if (!atomic_add_unless(&rds_iw_allocation, 1, rds_iw_sysctl_max_recv_allocation)) { rds_iw_stats_inc(s_iw_rx_alloc_limit); goto out; } recv->r_iwinc = kmem_cache_alloc(rds_iw_incoming_slab, kptr_gfp); - if (recv->r_iwinc == NULL) { + if (!recv->r_iwinc) { atomic_dec(&rds_iw_allocation); goto out; } @@ -158,17 +158,17 @@ static int rds_iw_recv_refill_one(struct rds_connection *conn, rds_inc_init(&recv->r_iwinc->ii_inc, conn, conn->c_faddr); } - if (recv->r_frag == NULL) { + if (!recv->r_frag) { recv->r_frag = kmem_cache_alloc(rds_iw_frag_slab, kptr_gfp); - if (recv->r_frag == NULL) + if (!recv->r_frag) goto out; INIT_LIST_HEAD(&recv->r_frag->f_item); recv->r_frag->f_page = NULL; } - if (ic->i_frag.f_page == NULL) { + if (!ic->i_frag.f_page) { ic->i_frag.f_page = alloc_page(page_gfp); - if (ic->i_frag.f_page == NULL) + if (!ic->i_frag.f_page) goto out; ic->i_frag.f_offset = 0; } @@ -273,7 +273,7 @@ int rds_iw_recv_refill(struct rds_connection *conn, gfp_t kptr_gfp, return ret; } -void rds_iw_inc_purge(struct rds_incoming *inc) +static void rds_iw_inc_purge(struct rds_incoming *inc) { struct rds_iw_incoming *iwinc; struct rds_page_frag *frag; @@ -716,7 +716,7 @@ static void rds_iw_process_recv(struct rds_connection *conn, * into the inc and save the inc so we can hang upcoming fragments * off its list. */ - if (iwinc == NULL) { + if (!iwinc) { iwinc = recv->r_iwinc; recv->r_iwinc = NULL; ic->i_iwinc = iwinc; @@ -887,7 +887,7 @@ int rds_iw_recv(struct rds_connection *conn) return ret; } -int __init rds_iw_recv_init(void) +int rds_iw_recv_init(void) { struct sysinfo si; int ret = -ENOMEM; @@ -899,13 +899,13 @@ int __init rds_iw_recv_init(void) rds_iw_incoming_slab = kmem_cache_create("rds_iw_incoming", sizeof(struct rds_iw_incoming), 0, 0, NULL); - if (rds_iw_incoming_slab == NULL) + if (!rds_iw_incoming_slab) goto out; rds_iw_frag_slab = kmem_cache_create("rds_iw_frag", sizeof(struct rds_page_frag), 0, 0, NULL); - if (rds_iw_frag_slab == NULL) + if (!rds_iw_frag_slab) kmem_cache_destroy(rds_iw_incoming_slab); else ret = 0; diff --git a/net/rds/iw_send.c b/net/rds/iw_send.c index 52182ff7519..6280ea020d4 100644 --- a/net/rds/iw_send.c +++ b/net/rds/iw_send.c @@ -36,7 +36,6 @@ #include <linux/dmapool.h> #include "rds.h" -#include "rdma.h" #include "iw.h" static void rds_iw_send_rdma_complete(struct rds_message *rm, @@ -64,13 +63,13 @@ static void rds_iw_send_rdma_complete(struct rds_message *rm, } static void rds_iw_send_unmap_rdma(struct rds_iw_connection *ic, - struct rds_rdma_op *op) + struct rm_rdma_op *op) { - if (op->r_mapped) { + if (op->op_mapped) { ib_dma_unmap_sg(ic->i_cm_id->device, - op->r_sg, op->r_nents, - op->r_write ? DMA_TO_DEVICE : DMA_FROM_DEVICE); - op->r_mapped = 0; + op->op_sg, op->op_nents, + op->op_write ? DMA_TO_DEVICE : DMA_FROM_DEVICE); + op->op_mapped = 0; } } @@ -83,11 +82,11 @@ static void rds_iw_send_unmap_rm(struct rds_iw_connection *ic, rdsdebug("ic %p send %p rm %p\n", ic, send, rm); ib_dma_unmap_sg(ic->i_cm_id->device, - rm->m_sg, rm->m_nents, + rm->data.op_sg, rm->data.op_nents, DMA_TO_DEVICE); - if (rm->m_rdma_op != NULL) { - rds_iw_send_unmap_rdma(ic, rm->m_rdma_op); + if (rm->rdma.op_active) { + rds_iw_send_unmap_rdma(ic, &rm->rdma); /* If the user asked for a completion notification on this * message, we can implement three different semantics: @@ -111,10 +110,10 @@ static void rds_iw_send_unmap_rm(struct rds_iw_connection *ic, */ rds_iw_send_rdma_complete(rm, wc_status); - if (rm->m_rdma_op->r_write) - rds_stats_add(s_send_rdma_bytes, rm->m_rdma_op->r_bytes); + if (rm->rdma.op_write) + rds_stats_add(s_send_rdma_bytes, rm->rdma.op_bytes); else - rds_stats_add(s_recv_rdma_bytes, rm->m_rdma_op->r_bytes); + rds_stats_add(s_recv_rdma_bytes, rm->rdma.op_bytes); } /* If anyone waited for this message to get flushed out, wake @@ -556,25 +555,27 @@ int rds_iw_xmit(struct rds_connection *conn, struct rds_message *rm, } /* map the message the first time we see it */ - if (ic->i_rm == NULL) { + if (!ic->i_rm) { /* printk(KERN_NOTICE "rds_iw_xmit prep msg dport=%u flags=0x%x len=%d\n", be16_to_cpu(rm->m_inc.i_hdr.h_dport), rm->m_inc.i_hdr.h_flags, be32_to_cpu(rm->m_inc.i_hdr.h_len)); */ - if (rm->m_nents) { - rm->m_count = ib_dma_map_sg(dev, - rm->m_sg, rm->m_nents, DMA_TO_DEVICE); - rdsdebug("ic %p mapping rm %p: %d\n", ic, rm, rm->m_count); - if (rm->m_count == 0) { + if (rm->data.op_nents) { + rm->data.op_count = ib_dma_map_sg(dev, + rm->data.op_sg, + rm->data.op_nents, + DMA_TO_DEVICE); + rdsdebug("ic %p mapping rm %p: %d\n", ic, rm, rm->data.op_count); + if (rm->data.op_count == 0) { rds_iw_stats_inc(s_iw_tx_sg_mapping_failure); rds_iw_ring_unalloc(&ic->i_send_ring, work_alloc); ret = -ENOMEM; /* XXX ? */ goto out; } } else { - rm->m_count = 0; + rm->data.op_count = 0; } ic->i_unsignaled_wrs = rds_iw_sysctl_max_unsig_wrs; @@ -590,10 +591,10 @@ int rds_iw_xmit(struct rds_connection *conn, struct rds_message *rm, /* If it has a RDMA op, tell the peer we did it. This is * used by the peer to release use-once RDMA MRs. */ - if (rm->m_rdma_op) { + if (rm->rdma.op_active) { struct rds_ext_header_rdma ext_hdr; - ext_hdr.h_rdma_rkey = cpu_to_be32(rm->m_rdma_op->r_key); + ext_hdr.h_rdma_rkey = cpu_to_be32(rm->rdma.op_rkey); rds_message_add_extension(&rm->m_inc.i_hdr, RDS_EXTHDR_RDMA, &ext_hdr, sizeof(ext_hdr)); } @@ -621,7 +622,7 @@ int rds_iw_xmit(struct rds_connection *conn, struct rds_message *rm, send = &ic->i_sends[pos]; first = send; prev = NULL; - scat = &rm->m_sg[sg]; + scat = &rm->data.op_sg[sg]; sent = 0; i = 0; @@ -631,7 +632,7 @@ int rds_iw_xmit(struct rds_connection *conn, struct rds_message *rm, * or when requested by the user. Right now, we let * the application choose. */ - if (rm->m_rdma_op && rm->m_rdma_op->r_fence) + if (rm->rdma.op_active && rm->rdma.op_fence) send_flags = IB_SEND_FENCE; /* @@ -650,7 +651,7 @@ int rds_iw_xmit(struct rds_connection *conn, struct rds_message *rm, } /* if there's data reference it with a chain of work reqs */ - for (; i < work_alloc && scat != &rm->m_sg[rm->m_count]; i++) { + for (; i < work_alloc && scat != &rm->data.op_sg[rm->data.op_count]; i++) { unsigned int len; send = &ic->i_sends[pos]; @@ -728,7 +729,7 @@ add_header: sent += sizeof(struct rds_header); /* if we finished the message then send completion owns it */ - if (scat == &rm->m_sg[rm->m_count]) { + if (scat == &rm->data.op_sg[rm->data.op_count]) { prev->s_rm = ic->i_rm; prev->s_wr.send_flags |= IB_SEND_SIGNALED | IB_SEND_SOLICITED; ic->i_rm = NULL; @@ -784,7 +785,7 @@ static void rds_iw_build_send_fastreg(struct rds_iw_device *rds_iwdev, struct rd ib_update_fast_reg_key(send->s_mr, send->s_remap_count++); } -int rds_iw_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) +int rds_iw_xmit_rdma(struct rds_connection *conn, struct rm_rdma_op *op) { struct rds_iw_connection *ic = conn->c_transport_data; struct rds_iw_send_work *send = NULL; @@ -794,7 +795,7 @@ int rds_iw_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) struct rds_iw_device *rds_iwdev; struct scatterlist *scat; unsigned long len; - u64 remote_addr = op->r_remote_addr; + u64 remote_addr = op->op_remote_addr; u32 pos, fr_pos; u32 work_alloc; u32 i; @@ -806,21 +807,21 @@ int rds_iw_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) rds_iwdev = ib_get_client_data(ic->i_cm_id->device, &rds_iw_client); /* map the message the first time we see it */ - if (!op->r_mapped) { - op->r_count = ib_dma_map_sg(ic->i_cm_id->device, - op->r_sg, op->r_nents, (op->r_write) ? - DMA_TO_DEVICE : DMA_FROM_DEVICE); - rdsdebug("ic %p mapping op %p: %d\n", ic, op, op->r_count); - if (op->r_count == 0) { + if (!op->op_mapped) { + op->op_count = ib_dma_map_sg(ic->i_cm_id->device, + op->op_sg, op->op_nents, (op->op_write) ? + DMA_TO_DEVICE : DMA_FROM_DEVICE); + rdsdebug("ic %p mapping op %p: %d\n", ic, op, op->op_count); + if (op->op_count == 0) { rds_iw_stats_inc(s_iw_tx_sg_mapping_failure); ret = -ENOMEM; /* XXX ? */ goto out; } - op->r_mapped = 1; + op->op_mapped = 1; } - if (!op->r_write) { + if (!op->op_write) { /* Alloc space on the send queue for the fastreg */ work_alloc = rds_iw_ring_alloc(&ic->i_send_ring, 1, &fr_pos); if (work_alloc != 1) { @@ -835,7 +836,7 @@ int rds_iw_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) * Instead of knowing how to return a partial rdma read/write we insist that there * be enough work requests to send the entire message. */ - i = ceil(op->r_count, rds_iwdev->max_sge); + i = ceil(op->op_count, rds_iwdev->max_sge); work_alloc = rds_iw_ring_alloc(&ic->i_send_ring, i, &pos); if (work_alloc != i) { @@ -846,17 +847,17 @@ int rds_iw_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) } send = &ic->i_sends[pos]; - if (!op->r_write) { + if (!op->op_write) { first = prev = &ic->i_sends[fr_pos]; } else { first = send; prev = NULL; } - scat = &op->r_sg[0]; + scat = &op->op_sg[0]; sent = 0; - num_sge = op->r_count; + num_sge = op->op_count; - for (i = 0; i < work_alloc && scat != &op->r_sg[op->r_count]; i++) { + for (i = 0; i < work_alloc && scat != &op->op_sg[op->op_count]; i++) { send->s_wr.send_flags = 0; send->s_queued = jiffies; @@ -873,13 +874,13 @@ int rds_iw_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) * for local access after RDS is finished with it, using * IB_WR_RDMA_READ_WITH_INV will invalidate it after the read has completed. */ - if (op->r_write) + if (op->op_write) send->s_wr.opcode = IB_WR_RDMA_WRITE; else send->s_wr.opcode = IB_WR_RDMA_READ_WITH_INV; send->s_wr.wr.rdma.remote_addr = remote_addr; - send->s_wr.wr.rdma.rkey = op->r_key; + send->s_wr.wr.rdma.rkey = op->op_rkey; send->s_op = op; if (num_sge > rds_iwdev->max_sge) { @@ -893,7 +894,7 @@ int rds_iw_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) if (prev) prev->s_wr.next = &send->s_wr; - for (j = 0; j < send->s_wr.num_sge && scat != &op->r_sg[op->r_count]; j++) { + for (j = 0; j < send->s_wr.num_sge && scat != &op->op_sg[op->op_count]; j++) { len = ib_sg_dma_len(ic->i_cm_id->device, scat); if (send->s_wr.opcode == IB_WR_RDMA_READ_WITH_INV) @@ -927,7 +928,7 @@ int rds_iw_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) } /* if we finished the message then send completion owns it */ - if (scat == &op->r_sg[op->r_count]) + if (scat == &op->op_sg[op->op_count]) first->s_wr.send_flags = IB_SEND_SIGNALED; if (i < work_alloc) { @@ -941,9 +942,9 @@ int rds_iw_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op) * adapters do not allow using the lkey for this at all. To bypass this use a * fastreg_mr (or possibly a dma_mr) */ - if (!op->r_write) { + if (!op->op_write) { rds_iw_build_send_fastreg(rds_iwdev, ic, &ic->i_sends[fr_pos], - op->r_count, sent, conn->c_xmit_rm->m_rs->rs_user_addr); + op->op_count, sent, conn->c_xmit_rm->m_rs->rs_user_addr); work_alloc++; } diff --git a/net/rds/iw_sysctl.c b/net/rds/iw_sysctl.c index 1c4428a61a0..e2e47176e72 100644 --- a/net/rds/iw_sysctl.c +++ b/net/rds/iw_sysctl.c @@ -55,7 +55,7 @@ static unsigned long rds_iw_sysctl_max_unsig_bytes_max = ~0UL; unsigned int rds_iw_sysctl_flow_control = 1; -ctl_table rds_iw_sysctl_table[] = { +static ctl_table rds_iw_sysctl_table[] = { { .procname = "max_send_wr", .data = &rds_iw_sysctl_max_send_wr, @@ -122,10 +122,10 @@ void rds_iw_sysctl_exit(void) unregister_sysctl_table(rds_iw_sysctl_hdr); } -int __init rds_iw_sysctl_init(void) +int rds_iw_sysctl_init(void) { rds_iw_sysctl_hdr = register_sysctl_paths(rds_iw_sysctl_path, rds_iw_sysctl_table); - if (rds_iw_sysctl_hdr == NULL) + if (!rds_iw_sysctl_hdr) return -ENOMEM; return 0; } diff --git a/net/rds/loop.c b/net/rds/loop.c index dd987937945..c390156b426 100644 --- a/net/rds/loop.c +++ b/net/rds/loop.c @@ -61,10 +61,17 @@ static int rds_loop_xmit(struct rds_connection *conn, struct rds_message *rm, unsigned int hdr_off, unsigned int sg, unsigned int off) { + /* Do not send cong updates to loopback */ + if (rm->m_inc.i_hdr.h_flags & RDS_FLAG_CONG_BITMAP) { + rds_cong_map_updated(conn->c_fcong, ~(u64) 0); + return sizeof(struct rds_header) + RDS_CONG_MAP_BYTES; + } + BUG_ON(hdr_off || sg || off); rds_inc_init(&rm->m_inc, conn, conn->c_laddr); - rds_message_addref(rm); /* for the inc */ + /* For the embedded inc. Matching put is in loop_inc_free() */ + rds_message_addref(rm); rds_recv_incoming(conn, conn->c_laddr, conn->c_faddr, &rm->m_inc, GFP_KERNEL, KM_USER0); @@ -77,16 +84,14 @@ static int rds_loop_xmit(struct rds_connection *conn, struct rds_message *rm, return sizeof(struct rds_header) + be32_to_cpu(rm->m_inc.i_hdr.h_len); } -static int rds_loop_xmit_cong_map(struct rds_connection *conn, - struct rds_cong_map *map, - unsigned long offset) +/* + * See rds_loop_xmit(). Since our inc is embedded in the rm, we + * make sure the rm lives at least until the inc is done. + */ +static void rds_loop_inc_free(struct rds_incoming *inc) { - BUG_ON(offset); - BUG_ON(map != conn->c_lcong); - - rds_cong_map_updated(conn->c_fcong, ~(u64) 0); - - return sizeof(struct rds_header) + RDS_CONG_MAP_BYTES; + struct rds_message *rm = container_of(inc, struct rds_message, m_inc); + rds_message_put(rm); } /* we need to at least give the thread something to succeed */ @@ -112,7 +117,7 @@ static int rds_loop_conn_alloc(struct rds_connection *conn, gfp_t gfp) unsigned long flags; lc = kzalloc(sizeof(struct rds_loop_connection), GFP_KERNEL); - if (lc == NULL) + if (!lc) return -ENOMEM; INIT_LIST_HEAD(&lc->loop_node); @@ -169,14 +174,12 @@ void rds_loop_exit(void) */ struct rds_transport rds_loop_transport = { .xmit = rds_loop_xmit, - .xmit_cong_map = rds_loop_xmit_cong_map, .recv = rds_loop_recv, .conn_alloc = rds_loop_conn_alloc, .conn_free = rds_loop_conn_free, .conn_connect = rds_loop_conn_connect, .conn_shutdown = rds_loop_conn_shutdown, .inc_copy_to_user = rds_message_inc_copy_to_user, - .inc_purge = rds_message_inc_purge, - .inc_free = rds_message_inc_free, + .inc_free = rds_loop_inc_free, .t_name = "loopback", }; diff --git a/net/rds/message.c b/net/rds/message.c index 9a1d67e001b..a84545dae37 100644 --- a/net/rds/message.c +++ b/net/rds/message.c @@ -34,9 +34,6 @@ #include <linux/slab.h> #include "rds.h" -#include "rdma.h" - -static DECLARE_WAIT_QUEUE_HEAD(rds_message_flush_waitq); static unsigned int rds_exthdr_size[__RDS_EXTHDR_MAX] = { [RDS_EXTHDR_NONE] = 0, @@ -63,29 +60,31 @@ static void rds_message_purge(struct rds_message *rm) if (unlikely(test_bit(RDS_MSG_PAGEVEC, &rm->m_flags))) return; - for (i = 0; i < rm->m_nents; i++) { - rdsdebug("putting data page %p\n", (void *)sg_page(&rm->m_sg[i])); + for (i = 0; i < rm->data.op_nents; i++) { + rdsdebug("putting data page %p\n", (void *)sg_page(&rm->data.op_sg[i])); /* XXX will have to put_page for page refs */ - __free_page(sg_page(&rm->m_sg[i])); + __free_page(sg_page(&rm->data.op_sg[i])); } - rm->m_nents = 0; + rm->data.op_nents = 0; - if (rm->m_rdma_op) - rds_rdma_free_op(rm->m_rdma_op); - if (rm->m_rdma_mr) - rds_mr_put(rm->m_rdma_mr); -} + if (rm->rdma.op_active) + rds_rdma_free_op(&rm->rdma); + if (rm->rdma.op_rdma_mr) + rds_mr_put(rm->rdma.op_rdma_mr); -void rds_message_inc_purge(struct rds_incoming *inc) -{ - struct rds_message *rm = container_of(inc, struct rds_message, m_inc); - rds_message_purge(rm); + if (rm->atomic.op_active) + rds_atomic_free_op(&rm->atomic); + if (rm->atomic.op_rdma_mr) + rds_mr_put(rm->atomic.op_rdma_mr); } void rds_message_put(struct rds_message *rm) { rdsdebug("put rm %p ref %d\n", rm, atomic_read(&rm->m_refcount)); - + if (atomic_read(&rm->m_refcount) == 0) { +printk(KERN_CRIT "danger refcount zero on %p\n", rm); +WARN_ON(1); + } if (atomic_dec_and_test(&rm->m_refcount)) { BUG_ON(!list_empty(&rm->m_sock_item)); BUG_ON(!list_empty(&rm->m_conn_item)); @@ -96,12 +95,6 @@ void rds_message_put(struct rds_message *rm) } EXPORT_SYMBOL_GPL(rds_message_put); -void rds_message_inc_free(struct rds_incoming *inc) -{ - struct rds_message *rm = container_of(inc, struct rds_message, m_inc); - rds_message_put(rm); -} - void rds_message_populate_header(struct rds_header *hdr, __be16 sport, __be16 dport, u64 seq) { @@ -113,8 +106,8 @@ void rds_message_populate_header(struct rds_header *hdr, __be16 sport, } EXPORT_SYMBOL_GPL(rds_message_populate_header); -int rds_message_add_extension(struct rds_header *hdr, - unsigned int type, const void *data, unsigned int len) +int rds_message_add_extension(struct rds_header *hdr, unsigned int type, + const void *data, unsigned int len) { unsigned int ext_len = sizeof(u8) + len; unsigned char *dst; @@ -184,26 +177,6 @@ none: return RDS_EXTHDR_NONE; } -int rds_message_add_version_extension(struct rds_header *hdr, unsigned int version) -{ - struct rds_ext_header_version ext_hdr; - - ext_hdr.h_version = cpu_to_be32(version); - return rds_message_add_extension(hdr, RDS_EXTHDR_VERSION, &ext_hdr, sizeof(ext_hdr)); -} - -int rds_message_get_version_extension(struct rds_header *hdr, unsigned int *version) -{ - struct rds_ext_header_version ext_hdr; - unsigned int pos = 0, len = sizeof(ext_hdr); - - /* We assume the version extension is the only one present */ - if (rds_message_next_extension(hdr, &pos, &ext_hdr, &len) != RDS_EXTHDR_VERSION) - return 0; - *version = be32_to_cpu(ext_hdr.h_version); - return 1; -} - int rds_message_add_rdma_dest_extension(struct rds_header *hdr, u32 r_key, u32 offset) { struct rds_ext_header_rdma_dest ext_hdr; @@ -214,41 +187,68 @@ int rds_message_add_rdma_dest_extension(struct rds_header *hdr, u32 r_key, u32 o } EXPORT_SYMBOL_GPL(rds_message_add_rdma_dest_extension); -struct rds_message *rds_message_alloc(unsigned int nents, gfp_t gfp) +/* + * Each rds_message is allocated with extra space for the scatterlist entries + * rds ops will need. This is to minimize memory allocation count. Then, each rds op + * can grab SGs when initializing its part of the rds_message. + */ +struct rds_message *rds_message_alloc(unsigned int extra_len, gfp_t gfp) { struct rds_message *rm; - rm = kzalloc(sizeof(struct rds_message) + - (nents * sizeof(struct scatterlist)), gfp); + rm = kzalloc(sizeof(struct rds_message) + extra_len, gfp); if (!rm) goto out; - if (nents) - sg_init_table(rm->m_sg, nents); + rm->m_used_sgs = 0; + rm->m_total_sgs = extra_len / sizeof(struct scatterlist); + atomic_set(&rm->m_refcount, 1); INIT_LIST_HEAD(&rm->m_sock_item); INIT_LIST_HEAD(&rm->m_conn_item); spin_lock_init(&rm->m_rs_lock); + init_waitqueue_head(&rm->m_flush_wait); out: return rm; } +/* + * RDS ops use this to grab SG entries from the rm's sg pool. + */ +struct scatterlist *rds_message_alloc_sgs(struct rds_message *rm, int nents) +{ + struct scatterlist *sg_first = (struct scatterlist *) &rm[1]; + struct scatterlist *sg_ret; + + WARN_ON(rm->m_used_sgs + nents > rm->m_total_sgs); + WARN_ON(!nents); + + sg_ret = &sg_first[rm->m_used_sgs]; + sg_init_table(sg_ret, nents); + rm->m_used_sgs += nents; + + return sg_ret; +} + struct rds_message *rds_message_map_pages(unsigned long *page_addrs, unsigned int total_len) { struct rds_message *rm; unsigned int i; + int num_sgs = ceil(total_len, PAGE_SIZE); + int extra_bytes = num_sgs * sizeof(struct scatterlist); - rm = rds_message_alloc(ceil(total_len, PAGE_SIZE), GFP_KERNEL); - if (rm == NULL) + rm = rds_message_alloc(extra_bytes, GFP_NOWAIT); + if (!rm) return ERR_PTR(-ENOMEM); set_bit(RDS_MSG_PAGEVEC, &rm->m_flags); rm->m_inc.i_hdr.h_len = cpu_to_be32(total_len); - rm->m_nents = ceil(total_len, PAGE_SIZE); + rm->data.op_nents = ceil(total_len, PAGE_SIZE); + rm->data.op_sg = rds_message_alloc_sgs(rm, num_sgs); - for (i = 0; i < rm->m_nents; ++i) { - sg_set_page(&rm->m_sg[i], + for (i = 0; i < rm->data.op_nents; ++i) { + sg_set_page(&rm->data.op_sg[i], virt_to_page(page_addrs[i]), PAGE_SIZE, 0); } @@ -256,40 +256,33 @@ struct rds_message *rds_message_map_pages(unsigned long *page_addrs, unsigned in return rm; } -struct rds_message *rds_message_copy_from_user(struct iovec *first_iov, +int rds_message_copy_from_user(struct rds_message *rm, struct iovec *first_iov, size_t total_len) { unsigned long to_copy; unsigned long iov_off; unsigned long sg_off; - struct rds_message *rm; struct iovec *iov; struct scatterlist *sg; - int ret; - - rm = rds_message_alloc(ceil(total_len, PAGE_SIZE), GFP_KERNEL); - if (rm == NULL) { - ret = -ENOMEM; - goto out; - } + int ret = 0; rm->m_inc.i_hdr.h_len = cpu_to_be32(total_len); /* * now allocate and copy in the data payload. */ - sg = rm->m_sg; + sg = rm->data.op_sg; iov = first_iov; iov_off = 0; sg_off = 0; /* Dear gcc, sg->page will be null from kzalloc. */ while (total_len) { - if (sg_page(sg) == NULL) { + if (!sg_page(sg)) { ret = rds_page_remainder_alloc(sg, total_len, GFP_HIGHUSER); if (ret) goto out; - rm->m_nents++; + rm->data.op_nents++; sg_off = 0; } @@ -320,14 +313,8 @@ struct rds_message *rds_message_copy_from_user(struct iovec *first_iov, sg++; } - ret = 0; out: - if (ret) { - if (rm) - rds_message_put(rm); - rm = ERR_PTR(ret); - } - return rm; + return ret; } int rds_message_inc_copy_to_user(struct rds_incoming *inc, @@ -348,7 +335,7 @@ int rds_message_inc_copy_to_user(struct rds_incoming *inc, iov = first_iov; iov_off = 0; - sg = rm->m_sg; + sg = rm->data.op_sg; vec_off = 0; copied = 0; @@ -394,15 +381,14 @@ int rds_message_inc_copy_to_user(struct rds_incoming *inc, */ void rds_message_wait(struct rds_message *rm) { - wait_event(rds_message_flush_waitq, + wait_event_interruptible(rm->m_flush_wait, !test_bit(RDS_MSG_MAPPED, &rm->m_flags)); } void rds_message_unmapped(struct rds_message *rm) { clear_bit(RDS_MSG_MAPPED, &rm->m_flags); - if (waitqueue_active(&rds_message_flush_waitq)) - wake_up(&rds_message_flush_waitq); + wake_up_interruptible(&rm->m_flush_wait); } EXPORT_SYMBOL_GPL(rds_message_unmapped); diff --git a/net/rds/page.c b/net/rds/page.c index 595a952d4b1..d8acdebe3c7 100644 --- a/net/rds/page.c +++ b/net/rds/page.c @@ -40,7 +40,8 @@ struct rds_page_remainder { unsigned long r_offset; }; -DEFINE_PER_CPU_SHARED_ALIGNED(struct rds_page_remainder, rds_page_remainders); +static DEFINE_PER_CPU_SHARED_ALIGNED(struct rds_page_remainder, + rds_page_remainders); /* * returns 0 on success or -errno on failure. @@ -57,30 +58,17 @@ int rds_page_copy_user(struct page *page, unsigned long offset, unsigned long ret; void *addr; - if (to_user) + addr = kmap(page); + if (to_user) { rds_stats_add(s_copy_to_user, bytes); - else + ret = copy_to_user(ptr, addr + offset, bytes); + } else { rds_stats_add(s_copy_from_user, bytes); - - addr = kmap_atomic(page, KM_USER0); - if (to_user) - ret = __copy_to_user_inatomic(ptr, addr + offset, bytes); - else - ret = __copy_from_user_inatomic(addr + offset, ptr, bytes); - kunmap_atomic(addr, KM_USER0); - - if (ret) { - addr = kmap(page); - if (to_user) - ret = copy_to_user(ptr, addr + offset, bytes); - else - ret = copy_from_user(addr + offset, ptr, bytes); - kunmap(page); - if (ret) - return -EFAULT; + ret = copy_from_user(addr + offset, ptr, bytes); } + kunmap(page); - return 0; + return ret ? -EFAULT : 0; } EXPORT_SYMBOL_GPL(rds_page_copy_user); @@ -116,7 +104,7 @@ int rds_page_remainder_alloc(struct scatterlist *scat, unsigned long bytes, /* jump straight to allocation if we're trying for a huge page */ if (bytes >= PAGE_SIZE) { page = alloc_page(gfp); - if (page == NULL) { + if (!page) { ret = -ENOMEM; } else { sg_set_page(scat, page, PAGE_SIZE, 0); @@ -162,7 +150,7 @@ int rds_page_remainder_alloc(struct scatterlist *scat, unsigned long bytes, rem = &per_cpu(rds_page_remainders, get_cpu()); local_irq_save(flags); - if (page == NULL) { + if (!page) { ret = -ENOMEM; break; } @@ -186,6 +174,7 @@ out: ret ? 0 : scat->length); return ret; } +EXPORT_SYMBOL_GPL(rds_page_remainder_alloc); static int rds_page_remainder_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) diff --git a/net/rds/rdma.c b/net/rds/rdma.c index 75fd13bb631..1a41debca1c 100644 --- a/net/rds/rdma.c +++ b/net/rds/rdma.c @@ -35,7 +35,7 @@ #include <linux/rbtree.h> #include <linux/dma-mapping.h> /* for DMA_*_DEVICE */ -#include "rdma.h" +#include "rds.h" /* * XXX @@ -130,14 +130,22 @@ void rds_rdma_drop_keys(struct rds_sock *rs) { struct rds_mr *mr; struct rb_node *node; + unsigned long flags; /* Release any MRs associated with this socket */ + spin_lock_irqsave(&rs->rs_rdma_lock, flags); while ((node = rb_first(&rs->rs_rdma_keys))) { mr = container_of(node, struct rds_mr, r_rb_node); if (mr->r_trans == rs->rs_transport) mr->r_invalidate = 0; + rb_erase(&mr->r_rb_node, &rs->rs_rdma_keys); + RB_CLEAR_NODE(&mr->r_rb_node); + spin_unlock_irqrestore(&rs->rs_rdma_lock, flags); + rds_destroy_mr(mr); rds_mr_put(mr); + spin_lock_irqsave(&rs->rs_rdma_lock, flags); } + spin_unlock_irqrestore(&rs->rs_rdma_lock, flags); if (rs->rs_transport && rs->rs_transport->flush_mrs) rs->rs_transport->flush_mrs(); @@ -181,7 +189,7 @@ static int __rds_rdma_map(struct rds_sock *rs, struct rds_get_mr_args *args, goto out; } - if (rs->rs_transport->get_mr == NULL) { + if (!rs->rs_transport->get_mr) { ret = -EOPNOTSUPP; goto out; } @@ -197,13 +205,13 @@ static int __rds_rdma_map(struct rds_sock *rs, struct rds_get_mr_args *args, /* XXX clamp nr_pages to limit the size of this alloc? */ pages = kcalloc(nr_pages, sizeof(struct page *), GFP_KERNEL); - if (pages == NULL) { + if (!pages) { ret = -ENOMEM; goto out; } mr = kzalloc(sizeof(struct rds_mr), GFP_KERNEL); - if (mr == NULL) { + if (!mr) { ret = -ENOMEM; goto out; } @@ -230,13 +238,13 @@ static int __rds_rdma_map(struct rds_sock *rs, struct rds_get_mr_args *args, * r/o or r/w. We need to assume r/w, or we'll do a lot of RDMA to * the zero page. */ - ret = rds_pin_pages(args->vec.addr & PAGE_MASK, nr_pages, pages, 1); + ret = rds_pin_pages(args->vec.addr, nr_pages, pages, 1); if (ret < 0) goto out; nents = ret; sg = kcalloc(nents, sizeof(*sg), GFP_KERNEL); - if (sg == NULL) { + if (!sg) { ret = -ENOMEM; goto out; } @@ -406,68 +414,127 @@ void rds_rdma_unuse(struct rds_sock *rs, u32 r_key, int force) spin_lock_irqsave(&rs->rs_rdma_lock, flags); mr = rds_mr_tree_walk(&rs->rs_rdma_keys, r_key, NULL); - if (mr && (mr->r_use_once || force)) { + if (!mr) { + printk(KERN_ERR "rds: trying to unuse MR with unknown r_key %u!\n", r_key); + spin_unlock_irqrestore(&rs->rs_rdma_lock, flags); + return; + } + + if (mr->r_use_once || force) { rb_erase(&mr->r_rb_node, &rs->rs_rdma_keys); RB_CLEAR_NODE(&mr->r_rb_node); zot_me = 1; - } else if (mr) - atomic_inc(&mr->r_refcount); + } spin_unlock_irqrestore(&rs->rs_rdma_lock, flags); /* May have to issue a dma_sync on this memory region. * Note we could avoid this if the operation was a RDMA READ, * but at this point we can't tell. */ - if (mr != NULL) { - if (mr->r_trans->sync_mr) - mr->r_trans->sync_mr(mr->r_trans_private, DMA_FROM_DEVICE); - - /* If the MR was marked as invalidate, this will - * trigger an async flush. */ - if (zot_me) - rds_destroy_mr(mr); - rds_mr_put(mr); - } + if (mr->r_trans->sync_mr) + mr->r_trans->sync_mr(mr->r_trans_private, DMA_FROM_DEVICE); + + /* If the MR was marked as invalidate, this will + * trigger an async flush. */ + if (zot_me) + rds_destroy_mr(mr); + rds_mr_put(mr); } -void rds_rdma_free_op(struct rds_rdma_op *ro) +void rds_rdma_free_op(struct rm_rdma_op *ro) { unsigned int i; - for (i = 0; i < ro->r_nents; i++) { - struct page *page = sg_page(&ro->r_sg[i]); + for (i = 0; i < ro->op_nents; i++) { + struct page *page = sg_page(&ro->op_sg[i]); /* Mark page dirty if it was possibly modified, which * is the case for a RDMA_READ which copies from remote * to local memory */ - if (!ro->r_write) { - BUG_ON(in_interrupt()); + if (!ro->op_write) { + BUG_ON(irqs_disabled()); set_page_dirty(page); } put_page(page); } - kfree(ro->r_notifier); - kfree(ro); + kfree(ro->op_notifier); + ro->op_notifier = NULL; + ro->op_active = 0; +} + +void rds_atomic_free_op(struct rm_atomic_op *ao) +{ + struct page *page = sg_page(ao->op_sg); + + /* Mark page dirty if it was possibly modified, which + * is the case for a RDMA_READ which copies from remote + * to local memory */ + set_page_dirty(page); + put_page(page); + + kfree(ao->op_notifier); + ao->op_notifier = NULL; + ao->op_active = 0; } + /* - * args is a pointer to an in-kernel copy in the sendmsg cmsg. + * Count the number of pages needed to describe an incoming iovec. */ -static struct rds_rdma_op *rds_rdma_prepare(struct rds_sock *rs, - struct rds_rdma_args *args) +static int rds_rdma_pages(struct rds_rdma_args *args) { struct rds_iovec vec; - struct rds_rdma_op *op = NULL; + struct rds_iovec __user *local_vec; + unsigned int tot_pages = 0; unsigned int nr_pages; - unsigned int max_pages; + unsigned int i; + + local_vec = (struct rds_iovec __user *)(unsigned long) args->local_vec_addr; + + /* figure out the number of pages in the vector */ + for (i = 0; i < args->nr_local; i++) { + if (copy_from_user(&vec, &local_vec[i], + sizeof(struct rds_iovec))) + return -EFAULT; + + nr_pages = rds_pages_in_vec(&vec); + if (nr_pages == 0) + return -EINVAL; + + tot_pages += nr_pages; + } + + return tot_pages; +} + +int rds_rdma_extra_size(struct rds_rdma_args *args) +{ + return rds_rdma_pages(args) * sizeof(struct scatterlist); +} + +/* + * The application asks for a RDMA transfer. + * Extract all arguments and set up the rdma_op + */ +int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm, + struct cmsghdr *cmsg) +{ + struct rds_rdma_args *args; + struct rds_iovec vec; + struct rm_rdma_op *op = &rm->rdma; + int nr_pages; unsigned int nr_bytes; struct page **pages = NULL; struct rds_iovec __user *local_vec; - struct scatterlist *sg; unsigned int nr; unsigned int i, j; - int ret; + int ret = 0; + if (cmsg->cmsg_len < CMSG_LEN(sizeof(struct rds_rdma_args)) + || rm->rdma.op_active) + return -EINVAL; + + args = CMSG_DATA(cmsg); if (rs->rs_bound_addr == 0) { ret = -ENOTCONN; /* XXX not a great errno */ @@ -479,61 +546,38 @@ static struct rds_rdma_op *rds_rdma_prepare(struct rds_sock *rs, goto out; } - nr_pages = 0; - max_pages = 0; - - local_vec = (struct rds_iovec __user *)(unsigned long) args->local_vec_addr; - - /* figure out the number of pages in the vector */ - for (i = 0; i < args->nr_local; i++) { - if (copy_from_user(&vec, &local_vec[i], - sizeof(struct rds_iovec))) { - ret = -EFAULT; - goto out; - } - - nr = rds_pages_in_vec(&vec); - if (nr == 0) { - ret = -EINVAL; - goto out; - } - - max_pages = max(nr, max_pages); - nr_pages += nr; - } - - pages = kcalloc(max_pages, sizeof(struct page *), GFP_KERNEL); - if (pages == NULL) { - ret = -ENOMEM; + nr_pages = rds_rdma_pages(args); + if (nr_pages < 0) goto out; - } - op = kzalloc(offsetof(struct rds_rdma_op, r_sg[nr_pages]), GFP_KERNEL); - if (op == NULL) { + pages = kcalloc(nr_pages, sizeof(struct page *), GFP_KERNEL); + if (!pages) { ret = -ENOMEM; goto out; } - op->r_write = !!(args->flags & RDS_RDMA_READWRITE); - op->r_fence = !!(args->flags & RDS_RDMA_FENCE); - op->r_notify = !!(args->flags & RDS_RDMA_NOTIFY_ME); - op->r_recverr = rs->rs_recverr; + op->op_write = !!(args->flags & RDS_RDMA_READWRITE); + op->op_fence = !!(args->flags & RDS_RDMA_FENCE); + op->op_notify = !!(args->flags & RDS_RDMA_NOTIFY_ME); + op->op_silent = !!(args->flags & RDS_RDMA_SILENT); + op->op_active = 1; + op->op_recverr = rs->rs_recverr; WARN_ON(!nr_pages); - sg_init_table(op->r_sg, nr_pages); + op->op_sg = rds_message_alloc_sgs(rm, nr_pages); - if (op->r_notify || op->r_recverr) { + if (op->op_notify || op->op_recverr) { /* We allocate an uninitialized notifier here, because * we don't want to do that in the completion handler. We * would have to use GFP_ATOMIC there, and don't want to deal * with failed allocations. */ - op->r_notifier = kmalloc(sizeof(struct rds_notifier), GFP_KERNEL); - if (!op->r_notifier) { + op->op_notifier = kmalloc(sizeof(struct rds_notifier), GFP_KERNEL); + if (!op->op_notifier) { ret = -ENOMEM; goto out; } - op->r_notifier->n_user_token = args->user_token; - op->r_notifier->n_status = RDS_RDMA_SUCCESS; + op->op_notifier->n_user_token = args->user_token; + op->op_notifier->n_status = RDS_RDMA_SUCCESS; } /* The cookie contains the R_Key of the remote memory region, and @@ -543,15 +587,17 @@ static struct rds_rdma_op *rds_rdma_prepare(struct rds_sock *rs, * destination address (which is really an offset into the MR) * FIXME: We may want to move this into ib_rdma.c */ - op->r_key = rds_rdma_cookie_key(args->cookie); - op->r_remote_addr = args->remote_vec.addr + rds_rdma_cookie_offset(args->cookie); + op->op_rkey = rds_rdma_cookie_key(args->cookie); + op->op_remote_addr = args->remote_vec.addr + rds_rdma_cookie_offset(args->cookie); nr_bytes = 0; rdsdebug("RDS: rdma prepare nr_local %llu rva %llx rkey %x\n", (unsigned long long)args->nr_local, (unsigned long long)args->remote_vec.addr, - op->r_key); + op->op_rkey); + + local_vec = (struct rds_iovec __user *)(unsigned long) args->local_vec_addr; for (i = 0; i < args->nr_local; i++) { if (copy_from_user(&vec, &local_vec[i], @@ -569,15 +615,10 @@ static struct rds_rdma_op *rds_rdma_prepare(struct rds_sock *rs, rs->rs_user_addr = vec.addr; rs->rs_user_bytes = vec.bytes; - /* did the user change the vec under us? */ - if (nr > max_pages || op->r_nents + nr > nr_pages) { - ret = -EINVAL; - goto out; - } /* If it's a WRITE operation, we want to pin the pages for reading. * If it's a READ operation, we need to pin the pages for writing. */ - ret = rds_pin_pages(vec.addr & PAGE_MASK, nr, pages, !op->r_write); + ret = rds_pin_pages(vec.addr, nr, pages, !op->op_write); if (ret < 0) goto out; @@ -588,8 +629,9 @@ static struct rds_rdma_op *rds_rdma_prepare(struct rds_sock *rs, for (j = 0; j < nr; j++) { unsigned int offset = vec.addr & ~PAGE_MASK; + struct scatterlist *sg; - sg = &op->r_sg[op->r_nents + j]; + sg = &op->op_sg[op->op_nents + j]; sg_set_page(sg, pages[j], min_t(unsigned int, vec.bytes, PAGE_SIZE - offset), offset); @@ -601,10 +643,9 @@ static struct rds_rdma_op *rds_rdma_prepare(struct rds_sock *rs, vec.bytes -= sg->length; } - op->r_nents += nr; + op->op_nents += nr; } - if (nr_bytes > args->remote_vec.bytes) { rdsdebug("RDS nr_bytes %u remote_bytes %u do not match\n", nr_bytes, @@ -612,38 +653,17 @@ static struct rds_rdma_op *rds_rdma_prepare(struct rds_sock *rs, ret = -EINVAL; goto out; } - op->r_bytes = nr_bytes; + op->op_bytes = nr_bytes; ret = 0; out: kfree(pages); - if (ret) { - if (op) - rds_rdma_free_op(op); - op = ERR_PTR(ret); - } - return op; -} - -/* - * The application asks for a RDMA transfer. - * Extract all arguments and set up the rdma_op - */ -int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm, - struct cmsghdr *cmsg) -{ - struct rds_rdma_op *op; - - if (cmsg->cmsg_len < CMSG_LEN(sizeof(struct rds_rdma_args)) || - rm->m_rdma_op != NULL) - return -EINVAL; + if (ret) + rds_rdma_free_op(op); - op = rds_rdma_prepare(rs, CMSG_DATA(cmsg)); - if (IS_ERR(op)) - return PTR_ERR(op); rds_stats_inc(s_send_rdma); - rm->m_rdma_op = op; - return 0; + + return ret; } /* @@ -673,7 +693,7 @@ int rds_cmsg_rdma_dest(struct rds_sock *rs, struct rds_message *rm, spin_lock_irqsave(&rs->rs_rdma_lock, flags); mr = rds_mr_tree_walk(&rs->rs_rdma_keys, r_key, NULL); - if (mr == NULL) + if (!mr) err = -EINVAL; /* invalid r_key */ else atomic_inc(&mr->r_refcount); @@ -681,7 +701,7 @@ int rds_cmsg_rdma_dest(struct rds_sock *rs, struct rds_message *rm, if (mr) { mr->r_trans->sync_mr(mr->r_trans_private, DMA_TO_DEVICE); - rm->m_rdma_mr = mr; + rm->rdma.op_rdma_mr = mr; } return err; } @@ -699,5 +719,98 @@ int rds_cmsg_rdma_map(struct rds_sock *rs, struct rds_message *rm, rm->m_rdma_cookie != 0) return -EINVAL; - return __rds_rdma_map(rs, CMSG_DATA(cmsg), &rm->m_rdma_cookie, &rm->m_rdma_mr); + return __rds_rdma_map(rs, CMSG_DATA(cmsg), &rm->m_rdma_cookie, &rm->rdma.op_rdma_mr); +} + +/* + * Fill in rds_message for an atomic request. + */ +int rds_cmsg_atomic(struct rds_sock *rs, struct rds_message *rm, + struct cmsghdr *cmsg) +{ + struct page *page = NULL; + struct rds_atomic_args *args; + int ret = 0; + + if (cmsg->cmsg_len < CMSG_LEN(sizeof(struct rds_atomic_args)) + || rm->atomic.op_active) + return -EINVAL; + + args = CMSG_DATA(cmsg); + + /* Nonmasked & masked cmsg ops converted to masked hw ops */ + switch (cmsg->cmsg_type) { + case RDS_CMSG_ATOMIC_FADD: + rm->atomic.op_type = RDS_ATOMIC_TYPE_FADD; + rm->atomic.op_m_fadd.add = args->fadd.add; + rm->atomic.op_m_fadd.nocarry_mask = 0; + break; + case RDS_CMSG_MASKED_ATOMIC_FADD: + rm->atomic.op_type = RDS_ATOMIC_TYPE_FADD; + rm->atomic.op_m_fadd.add = args->m_fadd.add; + rm->atomic.op_m_fadd.nocarry_mask = args->m_fadd.nocarry_mask; + break; + case RDS_CMSG_ATOMIC_CSWP: + rm->atomic.op_type = RDS_ATOMIC_TYPE_CSWP; + rm->atomic.op_m_cswp.compare = args->cswp.compare; + rm->atomic.op_m_cswp.swap = args->cswp.swap; + rm->atomic.op_m_cswp.compare_mask = ~0; + rm->atomic.op_m_cswp.swap_mask = ~0; + break; + case RDS_CMSG_MASKED_ATOMIC_CSWP: + rm->atomic.op_type = RDS_ATOMIC_TYPE_CSWP; + rm->atomic.op_m_cswp.compare = args->m_cswp.compare; + rm->atomic.op_m_cswp.swap = args->m_cswp.swap; + rm->atomic.op_m_cswp.compare_mask = args->m_cswp.compare_mask; + rm->atomic.op_m_cswp.swap_mask = args->m_cswp.swap_mask; + break; + default: + BUG(); /* should never happen */ + } + + rm->atomic.op_notify = !!(args->flags & RDS_RDMA_NOTIFY_ME); + rm->atomic.op_silent = !!(args->flags & RDS_RDMA_SILENT); + rm->atomic.op_active = 1; + rm->atomic.op_recverr = rs->rs_recverr; + rm->atomic.op_sg = rds_message_alloc_sgs(rm, 1); + + /* verify 8 byte-aligned */ + if (args->local_addr & 0x7) { + ret = -EFAULT; + goto err; + } + + ret = rds_pin_pages(args->local_addr, 1, &page, 1); + if (ret != 1) + goto err; + ret = 0; + + sg_set_page(rm->atomic.op_sg, page, 8, offset_in_page(args->local_addr)); + + if (rm->atomic.op_notify || rm->atomic.op_recverr) { + /* We allocate an uninitialized notifier here, because + * we don't want to do that in the completion handler. We + * would have to use GFP_ATOMIC there, and don't want to deal + * with failed allocations. + */ + rm->atomic.op_notifier = kmalloc(sizeof(*rm->atomic.op_notifier), GFP_KERNEL); + if (!rm->atomic.op_notifier) { + ret = -ENOMEM; + goto err; + } + + rm->atomic.op_notifier->n_user_token = args->user_token; + rm->atomic.op_notifier->n_status = RDS_RDMA_SUCCESS; + } + + rm->atomic.op_rkey = rds_rdma_cookie_key(args->cookie); + rm->atomic.op_remote_addr = args->remote_addr + rds_rdma_cookie_offset(args->cookie); + + return ret; +err: + if (page) + put_page(page); + kfree(rm->atomic.op_notifier); + + return ret; } diff --git a/net/rds/rdma.h b/net/rds/rdma.h deleted file mode 100644 index 909c39835a5..00000000000 --- a/net/rds/rdma.h +++ /dev/null @@ -1,85 +0,0 @@ -#ifndef _RDS_RDMA_H -#define _RDS_RDMA_H - -#include <linux/rbtree.h> -#include <linux/spinlock.h> -#include <linux/scatterlist.h> - -#include "rds.h" - -struct rds_mr { - struct rb_node r_rb_node; - atomic_t r_refcount; - u32 r_key; - - /* A copy of the creation flags */ - unsigned int r_use_once:1; - unsigned int r_invalidate:1; - unsigned int r_write:1; - - /* This is for RDS_MR_DEAD. - * It would be nice & consistent to make this part of the above - * bit field here, but we need to use test_and_set_bit. - */ - unsigned long r_state; - struct rds_sock *r_sock; /* back pointer to the socket that owns us */ - struct rds_transport *r_trans; - void *r_trans_private; -}; - -/* Flags for mr->r_state */ -#define RDS_MR_DEAD 0 - -struct rds_rdma_op { - u32 r_key; - u64 r_remote_addr; - unsigned int r_write:1; - unsigned int r_fence:1; - unsigned int r_notify:1; - unsigned int r_recverr:1; - unsigned int r_mapped:1; - struct rds_notifier *r_notifier; - unsigned int r_bytes; - unsigned int r_nents; - unsigned int r_count; - struct scatterlist r_sg[0]; -}; - -static inline rds_rdma_cookie_t rds_rdma_make_cookie(u32 r_key, u32 offset) -{ - return r_key | (((u64) offset) << 32); -} - -static inline u32 rds_rdma_cookie_key(rds_rdma_cookie_t cookie) -{ - return cookie; -} - -static inline u32 rds_rdma_cookie_offset(rds_rdma_cookie_t cookie) -{ - return cookie >> 32; -} - -int rds_get_mr(struct rds_sock *rs, char __user *optval, int optlen); -int rds_get_mr_for_dest(struct rds_sock *rs, char __user *optval, int optlen); -int rds_free_mr(struct rds_sock *rs, char __user *optval, int optlen); -void rds_rdma_drop_keys(struct rds_sock *rs); -int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm, - struct cmsghdr *cmsg); -int rds_cmsg_rdma_dest(struct rds_sock *rs, struct rds_message *rm, - struct cmsghdr *cmsg); -int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm, - struct cmsghdr *cmsg); -int rds_cmsg_rdma_map(struct rds_sock *rs, struct rds_message *rm, - struct cmsghdr *cmsg); -void rds_rdma_free_op(struct rds_rdma_op *ro); -void rds_rdma_send_complete(struct rds_message *rm, int); - -extern void __rds_put_mr_final(struct rds_mr *mr); -static inline void rds_mr_put(struct rds_mr *mr) -{ - if (atomic_dec_and_test(&mr->r_refcount)) - __rds_put_mr_final(mr); -} - -#endif diff --git a/net/rds/rdma_transport.c b/net/rds/rdma_transport.c index e599ba2f950..4195a053982 100644 --- a/net/rds/rdma_transport.c +++ b/net/rds/rdma_transport.c @@ -36,6 +36,34 @@ static struct rdma_cm_id *rds_rdma_listen_id; +static char *rds_cm_event_strings[] = { +#define RDS_CM_EVENT_STRING(foo) \ + [RDMA_CM_EVENT_##foo] = __stringify(RDMA_CM_EVENT_##foo) + RDS_CM_EVENT_STRING(ADDR_RESOLVED), + RDS_CM_EVENT_STRING(ADDR_ERROR), + RDS_CM_EVENT_STRING(ROUTE_RESOLVED), + RDS_CM_EVENT_STRING(ROUTE_ERROR), + RDS_CM_EVENT_STRING(CONNECT_REQUEST), + RDS_CM_EVENT_STRING(CONNECT_RESPONSE), + RDS_CM_EVENT_STRING(CONNECT_ERROR), + RDS_CM_EVENT_STRING(UNREACHABLE), + RDS_CM_EVENT_STRING(REJECTED), + RDS_CM_EVENT_STRING(ESTABLISHED), + RDS_CM_EVENT_STRING(DISCONNECTED), + RDS_CM_EVENT_STRING(DEVICE_REMOVAL), + RDS_CM_EVENT_STRING(MULTICAST_JOIN), + RDS_CM_EVENT_STRING(MULTICAST_ERROR), + RDS_CM_EVENT_STRING(ADDR_CHANGE), + RDS_CM_EVENT_STRING(TIMEWAIT_EXIT), +#undef RDS_CM_EVENT_STRING +}; + +static char *rds_cm_event_str(enum rdma_cm_event_type type) +{ + return rds_str_array(rds_cm_event_strings, + ARRAY_SIZE(rds_cm_event_strings), type); +}; + int rds_rdma_cm_event_handler(struct rdma_cm_id *cm_id, struct rdma_cm_event *event) { @@ -44,8 +72,8 @@ int rds_rdma_cm_event_handler(struct rdma_cm_id *cm_id, struct rds_transport *trans; int ret = 0; - rdsdebug("conn %p id %p handling event %u\n", conn, cm_id, - event->event); + rdsdebug("conn %p id %p handling event %u (%s)\n", conn, cm_id, + event->event, rds_cm_event_str(event->event)); if (cm_id->device->node_type == RDMA_NODE_RNIC) trans = &rds_iw_transport; @@ -109,7 +137,8 @@ int rds_rdma_cm_event_handler(struct rdma_cm_id *cm_id, default: /* things like device disconnect? */ - printk(KERN_ERR "RDS: unknown event %u!\n", event->event); + printk(KERN_ERR "RDS: unknown event %u (%s)!\n", + event->event, rds_cm_event_str(event->event)); break; } @@ -117,12 +146,13 @@ out: if (conn) mutex_unlock(&conn->c_cm_lock); - rdsdebug("id %p event %u handling ret %d\n", cm_id, event->event, ret); + rdsdebug("id %p event %u (%s) handling ret %d\n", cm_id, event->event, + rds_cm_event_str(event->event), ret); return ret; } -static int __init rds_rdma_listen_init(void) +static int rds_rdma_listen_init(void) { struct sockaddr_in sin; struct rdma_cm_id *cm_id; @@ -177,7 +207,7 @@ static void rds_rdma_listen_stop(void) } } -int __init rds_rdma_init(void) +static int rds_rdma_init(void) { int ret; @@ -204,7 +234,7 @@ out: } module_init(rds_rdma_init); -void rds_rdma_exit(void) +static void rds_rdma_exit(void) { /* stop listening first to ensure no new connections are attempted */ rds_rdma_listen_stop(); diff --git a/net/rds/rdma_transport.h b/net/rds/rdma_transport.h index 2f2c7d976c2..faba4e38269 100644 --- a/net/rds/rdma_transport.h +++ b/net/rds/rdma_transport.h @@ -11,10 +11,6 @@ int rds_rdma_conn_connect(struct rds_connection *conn); int rds_rdma_cm_event_handler(struct rdma_cm_id *cm_id, struct rdma_cm_event *event); -/* from rdma_transport.c */ -int rds_rdma_init(void); -void rds_rdma_exit(void); - /* from ib.c */ extern struct rds_transport rds_ib_transport; int rds_ib_init(void); diff --git a/net/rds/rds.h b/net/rds/rds.h index c224b5bb3ba..9542449c072 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -80,6 +80,7 @@ enum { /* Bits for c_flags */ #define RDS_LL_SEND_FULL 0 #define RDS_RECONNECT_PENDING 1 +#define RDS_IN_XMIT 2 struct rds_connection { struct hlist_node c_hash_node; @@ -91,12 +92,13 @@ struct rds_connection { struct rds_cong_map *c_lcong; struct rds_cong_map *c_fcong; - struct mutex c_send_lock; /* protect send ring */ struct rds_message *c_xmit_rm; unsigned long c_xmit_sg; unsigned int c_xmit_hdr_off; unsigned int c_xmit_data_off; + unsigned int c_xmit_atomic_sent; unsigned int c_xmit_rdma_sent; + unsigned int c_xmit_data_sent; spinlock_t c_lock; /* protect msg queues */ u64 c_next_tx_seq; @@ -116,11 +118,10 @@ struct rds_connection { struct delayed_work c_conn_w; struct work_struct c_down_w; struct mutex c_cm_lock; /* protect conn state & cm */ + wait_queue_head_t c_waitq; struct list_head c_map_item; unsigned long c_map_queued; - unsigned long c_map_offset; - unsigned long c_map_bytes; unsigned int c_unacked_packets; unsigned int c_unacked_bytes; @@ -206,6 +207,48 @@ struct rds_incoming { rds_rdma_cookie_t i_rdma_cookie; }; +struct rds_mr { + struct rb_node r_rb_node; + atomic_t r_refcount; + u32 r_key; + + /* A copy of the creation flags */ + unsigned int r_use_once:1; + unsigned int r_invalidate:1; + unsigned int r_write:1; + + /* This is for RDS_MR_DEAD. + * It would be nice & consistent to make this part of the above + * bit field here, but we need to use test_and_set_bit. + */ + unsigned long r_state; + struct rds_sock *r_sock; /* back pointer to the socket that owns us */ + struct rds_transport *r_trans; + void *r_trans_private; +}; + +/* Flags for mr->r_state */ +#define RDS_MR_DEAD 0 + +static inline rds_rdma_cookie_t rds_rdma_make_cookie(u32 r_key, u32 offset) +{ + return r_key | (((u64) offset) << 32); +} + +static inline u32 rds_rdma_cookie_key(rds_rdma_cookie_t cookie) +{ + return cookie; +} + +static inline u32 rds_rdma_cookie_offset(rds_rdma_cookie_t cookie) +{ + return cookie >> 32; +} + +/* atomic operation types */ +#define RDS_ATOMIC_TYPE_CSWP 0 +#define RDS_ATOMIC_TYPE_FADD 1 + /* * m_sock_item and m_conn_item are on lists that are serialized under * conn->c_lock. m_sock_item has additional meaning in that once it is empty @@ -258,13 +301,71 @@ struct rds_message { * -> rs->rs_lock */ spinlock_t m_rs_lock; + wait_queue_head_t m_flush_wait; + struct rds_sock *m_rs; - struct rds_rdma_op *m_rdma_op; + + /* cookie to send to remote, in rds header */ rds_rdma_cookie_t m_rdma_cookie; - struct rds_mr *m_rdma_mr; - unsigned int m_nents; - unsigned int m_count; - struct scatterlist m_sg[0]; + + unsigned int m_used_sgs; + unsigned int m_total_sgs; + + void *m_final_op; + + struct { + struct rm_atomic_op { + int op_type; + union { + struct { + uint64_t compare; + uint64_t swap; + uint64_t compare_mask; + uint64_t swap_mask; + } op_m_cswp; + struct { + uint64_t add; + uint64_t nocarry_mask; + } op_m_fadd; + }; + + u32 op_rkey; + u64 op_remote_addr; + unsigned int op_notify:1; + unsigned int op_recverr:1; + unsigned int op_mapped:1; + unsigned int op_silent:1; + unsigned int op_active:1; + struct scatterlist *op_sg; + struct rds_notifier *op_notifier; + + struct rds_mr *op_rdma_mr; + } atomic; + struct rm_rdma_op { + u32 op_rkey; + u64 op_remote_addr; + unsigned int op_write:1; + unsigned int op_fence:1; + unsigned int op_notify:1; + unsigned int op_recverr:1; + unsigned int op_mapped:1; + unsigned int op_silent:1; + unsigned int op_active:1; + unsigned int op_bytes; + unsigned int op_nents; + unsigned int op_count; + struct scatterlist *op_sg; + struct rds_notifier *op_notifier; + + struct rds_mr *op_rdma_mr; + } rdma; + struct rm_data_op { + unsigned int op_active:1; + unsigned int op_nents; + unsigned int op_count; + struct scatterlist *op_sg; + } data; + }; }; /* @@ -305,10 +406,6 @@ struct rds_notifier { * transport is responsible for other serialization, including * rds_recv_incoming(). This is called in process context but * should try hard not to block. - * - * @xmit_cong_map: This asks the transport to send the local bitmap down the - * given connection. XXX get a better story about the bitmap - * flag and header. */ #define RDS_TRANS_IB 0 @@ -332,13 +429,11 @@ struct rds_transport { void (*xmit_complete)(struct rds_connection *conn); int (*xmit)(struct rds_connection *conn, struct rds_message *rm, unsigned int hdr_off, unsigned int sg, unsigned int off); - int (*xmit_cong_map)(struct rds_connection *conn, - struct rds_cong_map *map, unsigned long offset); - int (*xmit_rdma)(struct rds_connection *conn, struct rds_rdma_op *op); + int (*xmit_rdma)(struct rds_connection *conn, struct rm_rdma_op *op); + int (*xmit_atomic)(struct rds_connection *conn, struct rm_atomic_op *op); int (*recv)(struct rds_connection *conn); int (*inc_copy_to_user)(struct rds_incoming *inc, struct iovec *iov, size_t size); - void (*inc_purge)(struct rds_incoming *inc); void (*inc_free)(struct rds_incoming *inc); int (*cm_handle_connect)(struct rdma_cm_id *cm_id, @@ -367,17 +462,11 @@ struct rds_sock { * bound_addr used for both incoming and outgoing, no INADDR_ANY * support. */ - struct rb_node rs_bound_node; + struct hlist_node rs_bound_node; __be32 rs_bound_addr; __be32 rs_conn_addr; __be16 rs_bound_port; __be16 rs_conn_port; - - /* - * This is only used to communicate the transport between bind and - * initiating connections. All other trans use is referenced through - * the connection. - */ struct rds_transport *rs_transport; /* @@ -466,8 +555,8 @@ struct rds_statistics { uint64_t s_recv_ping; uint64_t s_send_queue_empty; uint64_t s_send_queue_full; - uint64_t s_send_sem_contention; - uint64_t s_send_sem_queue_raced; + uint64_t s_send_lock_contention; + uint64_t s_send_lock_queue_raced; uint64_t s_send_immediate_retry; uint64_t s_send_delayed_retry; uint64_t s_send_drop_acked; @@ -487,6 +576,7 @@ struct rds_statistics { }; /* af_rds.c */ +char *rds_str_array(char **array, size_t elements, size_t index); void rds_sock_addref(struct rds_sock *rs); void rds_sock_put(struct rds_sock *rs); void rds_wake_sk_sleep(struct rds_sock *rs); @@ -521,15 +611,16 @@ void rds_cong_exit(void); struct rds_message *rds_cong_update_alloc(struct rds_connection *conn); /* conn.c */ -int __init rds_conn_init(void); +int rds_conn_init(void); void rds_conn_exit(void); struct rds_connection *rds_conn_create(__be32 laddr, __be32 faddr, struct rds_transport *trans, gfp_t gfp); struct rds_connection *rds_conn_create_outgoing(__be32 laddr, __be32 faddr, struct rds_transport *trans, gfp_t gfp); +void rds_conn_shutdown(struct rds_connection *conn); void rds_conn_destroy(struct rds_connection *conn); -void rds_conn_reset(struct rds_connection *conn); void rds_conn_drop(struct rds_connection *conn); +void rds_conn_connect_if_down(struct rds_connection *conn); void rds_for_each_conn_info(struct socket *sock, unsigned int len, struct rds_info_iterator *iter, struct rds_info_lengths *lens, @@ -566,7 +657,8 @@ rds_conn_connecting(struct rds_connection *conn) /* message.c */ struct rds_message *rds_message_alloc(unsigned int nents, gfp_t gfp); -struct rds_message *rds_message_copy_from_user(struct iovec *first_iov, +struct scatterlist *rds_message_alloc_sgs(struct rds_message *rm, int nents); +int rds_message_copy_from_user(struct rds_message *rm, struct iovec *first_iov, size_t total_len); struct rds_message *rds_message_map_pages(unsigned long *page_addrs, unsigned int total_len); void rds_message_populate_header(struct rds_header *hdr, __be16 sport, @@ -575,12 +667,9 @@ int rds_message_add_extension(struct rds_header *hdr, unsigned int type, const void *data, unsigned int len); int rds_message_next_extension(struct rds_header *hdr, unsigned int *pos, void *buf, unsigned int *buflen); -int rds_message_add_version_extension(struct rds_header *hdr, unsigned int version); -int rds_message_get_version_extension(struct rds_header *hdr, unsigned int *version); int rds_message_add_rdma_dest_extension(struct rds_header *hdr, u32 r_key, u32 offset); int rds_message_inc_copy_to_user(struct rds_incoming *inc, struct iovec *first_iov, size_t size); -void rds_message_inc_purge(struct rds_incoming *inc); void rds_message_inc_free(struct rds_incoming *inc); void rds_message_addref(struct rds_message *rm); void rds_message_put(struct rds_message *rm); @@ -614,7 +703,6 @@ void rds_page_exit(void); /* recv.c */ void rds_inc_init(struct rds_incoming *inc, struct rds_connection *conn, __be32 saddr); -void rds_inc_addref(struct rds_incoming *inc); void rds_inc_put(struct rds_incoming *inc); void rds_recv_incoming(struct rds_connection *conn, __be32 saddr, __be32 daddr, struct rds_incoming *inc, gfp_t gfp, enum km_type km); @@ -636,14 +724,38 @@ void rds_send_drop_to(struct rds_sock *rs, struct sockaddr_in *dest); typedef int (*is_acked_func)(struct rds_message *rm, uint64_t ack); void rds_send_drop_acked(struct rds_connection *conn, u64 ack, is_acked_func is_acked); -int rds_send_acked_before(struct rds_connection *conn, u64 seq); -void rds_send_remove_from_sock(struct list_head *messages, int status); int rds_send_pong(struct rds_connection *conn, __be16 dport); struct rds_message *rds_send_get_message(struct rds_connection *, - struct rds_rdma_op *); + struct rm_rdma_op *); /* rdma.c */ void rds_rdma_unuse(struct rds_sock *rs, u32 r_key, int force); +int rds_get_mr(struct rds_sock *rs, char __user *optval, int optlen); +int rds_get_mr_for_dest(struct rds_sock *rs, char __user *optval, int optlen); +int rds_free_mr(struct rds_sock *rs, char __user *optval, int optlen); +void rds_rdma_drop_keys(struct rds_sock *rs); +int rds_rdma_extra_size(struct rds_rdma_args *args); +int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm, + struct cmsghdr *cmsg); +int rds_cmsg_rdma_dest(struct rds_sock *rs, struct rds_message *rm, + struct cmsghdr *cmsg); +int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm, + struct cmsghdr *cmsg); +int rds_cmsg_rdma_map(struct rds_sock *rs, struct rds_message *rm, + struct cmsghdr *cmsg); +void rds_rdma_free_op(struct rm_rdma_op *ro); +void rds_atomic_free_op(struct rm_atomic_op *ao); +void rds_rdma_send_complete(struct rds_message *rm, int wc_status); +void rds_atomic_send_complete(struct rds_message *rm, int wc_status); +int rds_cmsg_atomic(struct rds_sock *rs, struct rds_message *rm, + struct cmsghdr *cmsg); + +extern void __rds_put_mr_final(struct rds_mr *mr); +static inline void rds_mr_put(struct rds_mr *mr) +{ + if (atomic_dec_and_test(&mr->r_refcount)) + __rds_put_mr_final(mr); +} /* stats.c */ DECLARE_PER_CPU_SHARED_ALIGNED(struct rds_statistics, rds_stats); @@ -657,14 +769,14 @@ DECLARE_PER_CPU_SHARED_ALIGNED(struct rds_statistics, rds_stats); put_cpu(); \ } while (0) #define rds_stats_add(member, count) rds_stats_add_which(rds_stats, member, count) -int __init rds_stats_init(void); +int rds_stats_init(void); void rds_stats_exit(void); void rds_stats_info_copy(struct rds_info_iterator *iter, uint64_t *values, const char *const *names, size_t nr); /* sysctl.c */ -int __init rds_sysctl_init(void); +int rds_sysctl_init(void); void rds_sysctl_exit(void); extern unsigned long rds_sysctl_sndbuf_min; extern unsigned long rds_sysctl_sndbuf_default; @@ -678,9 +790,10 @@ extern unsigned long rds_sysctl_trace_flags; extern unsigned int rds_sysctl_trace_level; /* threads.c */ -int __init rds_threads_init(void); +int rds_threads_init(void); void rds_threads_exit(void); extern struct workqueue_struct *rds_wq; +void rds_queue_reconnect(struct rds_connection *conn); void rds_connect_worker(struct work_struct *); void rds_shutdown_worker(struct work_struct *); void rds_send_worker(struct work_struct *); @@ -691,9 +804,10 @@ void rds_connect_complete(struct rds_connection *conn); int rds_trans_register(struct rds_transport *trans); void rds_trans_unregister(struct rds_transport *trans); struct rds_transport *rds_trans_get_preferred(__be32 addr); +void rds_trans_put(struct rds_transport *trans); unsigned int rds_trans_stats_info_copy(struct rds_info_iterator *iter, unsigned int avail); -int __init rds_trans_init(void); +int rds_trans_init(void); void rds_trans_exit(void); #endif diff --git a/net/rds/recv.c b/net/rds/recv.c index 795a00b7f2c..596689e5927 100644 --- a/net/rds/recv.c +++ b/net/rds/recv.c @@ -36,7 +36,6 @@ #include <linux/in.h> #include "rds.h" -#include "rdma.h" void rds_inc_init(struct rds_incoming *inc, struct rds_connection *conn, __be32 saddr) @@ -49,12 +48,11 @@ void rds_inc_init(struct rds_incoming *inc, struct rds_connection *conn, } EXPORT_SYMBOL_GPL(rds_inc_init); -void rds_inc_addref(struct rds_incoming *inc) +static void rds_inc_addref(struct rds_incoming *inc) { rdsdebug("addref inc %p ref %d\n", inc, atomic_read(&inc->i_refcount)); atomic_inc(&inc->i_refcount); } -EXPORT_SYMBOL_GPL(rds_inc_addref); void rds_inc_put(struct rds_incoming *inc) { @@ -210,7 +208,7 @@ void rds_recv_incoming(struct rds_connection *conn, __be32 saddr, __be32 daddr, } rs = rds_find_bound(daddr, inc->i_hdr.h_dport); - if (rs == NULL) { + if (!rs) { rds_stats_inc(s_recv_drop_no_sock); goto out; } @@ -251,7 +249,7 @@ static int rds_next_incoming(struct rds_sock *rs, struct rds_incoming **inc) { unsigned long flags; - if (*inc == NULL) { + if (!*inc) { read_lock_irqsave(&rs->rs_recv_lock, flags); if (!list_empty(&rs->rs_recv_queue)) { *inc = list_entry(rs->rs_recv_queue.next, @@ -297,7 +295,7 @@ static int rds_still_queued(struct rds_sock *rs, struct rds_incoming *inc, int rds_notify_queue_get(struct rds_sock *rs, struct msghdr *msghdr) { struct rds_notifier *notifier; - struct rds_rdma_notify cmsg; + struct rds_rdma_notify cmsg = { 0 }; /* fill holes with zero */ unsigned int count = 0, max_messages = ~0U; unsigned long flags; LIST_HEAD(copy); @@ -334,10 +332,10 @@ int rds_notify_queue_get(struct rds_sock *rs, struct msghdr *msghdr) if (msghdr) { cmsg.user_token = notifier->n_user_token; - cmsg.status = notifier->n_status; + cmsg.status = notifier->n_status; err = put_cmsg(msghdr, SOL_RDS, RDS_CMSG_RDMA_STATUS, - sizeof(cmsg), &cmsg); + sizeof(cmsg), &cmsg); if (err) break; } diff --git a/net/rds/send.c b/net/rds/send.c index 9c1c6bcaa6c..0bc9db17a87 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -37,7 +37,6 @@ #include <linux/list.h> #include "rds.h" -#include "rdma.h" /* When transmitting messages in rds_send_xmit, we need to emerge from * time to time and briefly release the CPU. Otherwise the softlock watchdog @@ -53,8 +52,11 @@ static int send_batch_count = 64; module_param(send_batch_count, int, 0444); MODULE_PARM_DESC(send_batch_count, " batch factor when working the send queue"); +static void rds_send_remove_from_sock(struct list_head *messages, int status); + /* - * Reset the send state. Caller must hold c_send_lock when calling here. + * Reset the send state. Callers must ensure that this doesn't race with + * rds_send_xmit(). */ void rds_send_reset(struct rds_connection *conn) { @@ -62,18 +64,22 @@ void rds_send_reset(struct rds_connection *conn) unsigned long flags; if (conn->c_xmit_rm) { + rm = conn->c_xmit_rm; + conn->c_xmit_rm = NULL; /* Tell the user the RDMA op is no longer mapped by the * transport. This isn't entirely true (it's flushed out * independently) but as the connection is down, there's * no ongoing RDMA to/from that memory */ - rds_message_unmapped(conn->c_xmit_rm); - rds_message_put(conn->c_xmit_rm); - conn->c_xmit_rm = NULL; + rds_message_unmapped(rm); + rds_message_put(rm); } + conn->c_xmit_sg = 0; conn->c_xmit_hdr_off = 0; conn->c_xmit_data_off = 0; + conn->c_xmit_atomic_sent = 0; conn->c_xmit_rdma_sent = 0; + conn->c_xmit_data_sent = 0; conn->c_map_queued = 0; @@ -90,6 +96,25 @@ void rds_send_reset(struct rds_connection *conn) spin_unlock_irqrestore(&conn->c_lock, flags); } +static int acquire_in_xmit(struct rds_connection *conn) +{ + return test_and_set_bit(RDS_IN_XMIT, &conn->c_flags) == 0; +} + +static void release_in_xmit(struct rds_connection *conn) +{ + clear_bit(RDS_IN_XMIT, &conn->c_flags); + smp_mb__after_clear_bit(); + /* + * We don't use wait_on_bit()/wake_up_bit() because our waking is in a + * hot path and finding waiters is very rare. We don't want to walk + * the system-wide hashed waitqueue buckets in the fast path only to + * almost never find waiters. + */ + if (waitqueue_active(&conn->c_waitq)) + wake_up_all(&conn->c_waitq); +} + /* * We're making the concious trade-off here to only send one message * down the connection at a time. @@ -109,102 +134,69 @@ int rds_send_xmit(struct rds_connection *conn) struct rds_message *rm; unsigned long flags; unsigned int tmp; - unsigned int send_quota = send_batch_count; struct scatterlist *sg; int ret = 0; - int was_empty = 0; LIST_HEAD(to_be_dropped); +restart: + /* * sendmsg calls here after having queued its message on the send * queue. We only have one task feeding the connection at a time. If * another thread is already feeding the queue then we back off. This * avoids blocking the caller and trading per-connection data between * caches per message. - * - * The sem holder will issue a retry if they notice that someone queued - * a message after they stopped walking the send queue but before they - * dropped the sem. */ - if (!mutex_trylock(&conn->c_send_lock)) { - rds_stats_inc(s_send_sem_contention); + if (!acquire_in_xmit(conn)) { + rds_stats_inc(s_send_lock_contention); ret = -ENOMEM; goto out; } + /* + * rds_conn_shutdown() sets the conn state and then tests RDS_IN_XMIT, + * we do the opposite to avoid races. + */ + if (!rds_conn_up(conn)) { + release_in_xmit(conn); + ret = 0; + goto out; + } + if (conn->c_trans->xmit_prepare) conn->c_trans->xmit_prepare(conn); /* * spin trying to push headers and data down the connection until - * the connection doens't make forward progress. + * the connection doesn't make forward progress. */ - while (--send_quota) { - /* - * See if need to send a congestion map update if we're - * between sending messages. The send_sem protects our sole - * use of c_map_offset and _bytes. - * Note this is used only by transports that define a special - * xmit_cong_map function. For all others, we create allocate - * a cong_map message and treat it just like any other send. - */ - if (conn->c_map_bytes) { - ret = conn->c_trans->xmit_cong_map(conn, conn->c_lcong, - conn->c_map_offset); - if (ret <= 0) - break; + while (1) { - conn->c_map_offset += ret; - conn->c_map_bytes -= ret; - if (conn->c_map_bytes) - continue; - } - - /* If we're done sending the current message, clear the - * offset and S/G temporaries. - */ rm = conn->c_xmit_rm; - if (rm != NULL && - conn->c_xmit_hdr_off == sizeof(struct rds_header) && - conn->c_xmit_sg == rm->m_nents) { - conn->c_xmit_rm = NULL; - conn->c_xmit_sg = 0; - conn->c_xmit_hdr_off = 0; - conn->c_xmit_data_off = 0; - conn->c_xmit_rdma_sent = 0; - - /* Release the reference to the previous message. */ - rds_message_put(rm); - rm = NULL; - } - /* If we're asked to send a cong map update, do so. + /* + * If between sending messages, we can send a pending congestion + * map update. */ - if (rm == NULL && test_and_clear_bit(0, &conn->c_map_queued)) { - if (conn->c_trans->xmit_cong_map != NULL) { - conn->c_map_offset = 0; - conn->c_map_bytes = sizeof(struct rds_header) + - RDS_CONG_MAP_BYTES; - continue; - } - + if (!rm && test_and_clear_bit(0, &conn->c_map_queued)) { rm = rds_cong_update_alloc(conn); if (IS_ERR(rm)) { ret = PTR_ERR(rm); break; } + rm->data.op_active = 1; conn->c_xmit_rm = rm; } /* - * Grab the next message from the send queue, if there is one. + * If not already working on one, grab the next message. * * c_xmit_rm holds a ref while we're sending this message down * the connction. We can use this ref while holding the * send_sem.. rds_send_reset() is serialized with it. */ - if (rm == NULL) { + if (!rm) { unsigned int len; spin_lock_irqsave(&conn->c_lock, flags); @@ -224,10 +216,8 @@ int rds_send_xmit(struct rds_connection *conn) spin_unlock_irqrestore(&conn->c_lock, flags); - if (rm == NULL) { - was_empty = 1; + if (!rm) break; - } /* Unfortunately, the way Infiniband deals with * RDMA to a bad MR key is by moving the entire @@ -236,13 +226,12 @@ int rds_send_xmit(struct rds_connection *conn) * connection. * Therefore, we never retransmit messages with RDMA ops. */ - if (rm->m_rdma_op && + if (rm->rdma.op_active && test_bit(RDS_MSG_RETRANSMITTED, &rm->m_flags)) { spin_lock_irqsave(&conn->c_lock, flags); if (test_and_clear_bit(RDS_MSG_ON_CONN, &rm->m_flags)) list_move(&rm->m_conn_item, &to_be_dropped); spin_unlock_irqrestore(&conn->c_lock, flags); - rds_message_put(rm); continue; } @@ -263,23 +252,55 @@ int rds_send_xmit(struct rds_connection *conn) conn->c_xmit_rm = rm; } - /* - * Try and send an rdma message. Let's see if we can - * keep this simple and require that the transport either - * send the whole rdma or none of it. - */ - if (rm->m_rdma_op && !conn->c_xmit_rdma_sent) { - ret = conn->c_trans->xmit_rdma(conn, rm->m_rdma_op); + /* The transport either sends the whole rdma or none of it */ + if (rm->rdma.op_active && !conn->c_xmit_rdma_sent) { + rm->m_final_op = &rm->rdma; + ret = conn->c_trans->xmit_rdma(conn, &rm->rdma); if (ret) break; conn->c_xmit_rdma_sent = 1; + /* The transport owns the mapped memory for now. * You can't unmap it while it's on the send queue */ set_bit(RDS_MSG_MAPPED, &rm->m_flags); } - if (conn->c_xmit_hdr_off < sizeof(struct rds_header) || - conn->c_xmit_sg < rm->m_nents) { + if (rm->atomic.op_active && !conn->c_xmit_atomic_sent) { + rm->m_final_op = &rm->atomic; + ret = conn->c_trans->xmit_atomic(conn, &rm->atomic); + if (ret) + break; + conn->c_xmit_atomic_sent = 1; + + /* The transport owns the mapped memory for now. + * You can't unmap it while it's on the send queue */ + set_bit(RDS_MSG_MAPPED, &rm->m_flags); + } + + /* + * A number of cases require an RDS header to be sent + * even if there is no data. + * We permit 0-byte sends; rds-ping depends on this. + * However, if there are exclusively attached silent ops, + * we skip the hdr/data send, to enable silent operation. + */ + if (rm->data.op_nents == 0) { + int ops_present; + int all_ops_are_silent = 1; + + ops_present = (rm->atomic.op_active || rm->rdma.op_active); + if (rm->atomic.op_active && !rm->atomic.op_silent) + all_ops_are_silent = 0; + if (rm->rdma.op_active && !rm->rdma.op_silent) + all_ops_are_silent = 0; + + if (ops_present && all_ops_are_silent + && !rm->m_rdma_cookie) + rm->data.op_active = 0; + } + + if (rm->data.op_active && !conn->c_xmit_data_sent) { + rm->m_final_op = &rm->data; ret = conn->c_trans->xmit(conn, rm, conn->c_xmit_hdr_off, conn->c_xmit_sg, @@ -295,7 +316,7 @@ int rds_send_xmit(struct rds_connection *conn) ret -= tmp; } - sg = &rm->m_sg[conn->c_xmit_sg]; + sg = &rm->data.op_sg[conn->c_xmit_sg]; while (ret) { tmp = min_t(int, ret, sg->length - conn->c_xmit_data_off); @@ -306,49 +327,63 @@ int rds_send_xmit(struct rds_connection *conn) sg++; conn->c_xmit_sg++; BUG_ON(ret != 0 && - conn->c_xmit_sg == rm->m_nents); + conn->c_xmit_sg == rm->data.op_nents); } } + + if (conn->c_xmit_hdr_off == sizeof(struct rds_header) && + (conn->c_xmit_sg == rm->data.op_nents)) + conn->c_xmit_data_sent = 1; } - } - /* Nuke any messages we decided not to retransmit. */ - if (!list_empty(&to_be_dropped)) - rds_send_remove_from_sock(&to_be_dropped, RDS_RDMA_DROPPED); + /* + * A rm will only take multiple times through this loop + * if there is a data op. Thus, if the data is sent (or there was + * none), then we're done with the rm. + */ + if (!rm->data.op_active || conn->c_xmit_data_sent) { + conn->c_xmit_rm = NULL; + conn->c_xmit_sg = 0; + conn->c_xmit_hdr_off = 0; + conn->c_xmit_data_off = 0; + conn->c_xmit_rdma_sent = 0; + conn->c_xmit_atomic_sent = 0; + conn->c_xmit_data_sent = 0; + + rds_message_put(rm); + } + } if (conn->c_trans->xmit_complete) conn->c_trans->xmit_complete(conn); - /* - * We might be racing with another sender who queued a message but - * backed off on noticing that we held the c_send_lock. If we check - * for queued messages after dropping the sem then either we'll - * see the queued message or the queuer will get the sem. If we - * notice the queued message then we trigger an immediate retry. - * - * We need to be careful only to do this when we stopped processing - * the send queue because it was empty. It's the only way we - * stop processing the loop when the transport hasn't taken - * responsibility for forward progress. - */ - mutex_unlock(&conn->c_send_lock); + release_in_xmit(conn); - if (conn->c_map_bytes || (send_quota == 0 && !was_empty)) { - /* We exhausted the send quota, but there's work left to - * do. Return and (re-)schedule the send worker. - */ - ret = -EAGAIN; + /* Nuke any messages we decided not to retransmit. */ + if (!list_empty(&to_be_dropped)) { + /* irqs on here, so we can put(), unlike above */ + list_for_each_entry(rm, &to_be_dropped, m_conn_item) + rds_message_put(rm); + rds_send_remove_from_sock(&to_be_dropped, RDS_RDMA_DROPPED); } - if (ret == 0 && was_empty) { - /* A simple bit test would be way faster than taking the - * spin lock */ - spin_lock_irqsave(&conn->c_lock, flags); + /* + * Other senders can queue a message after we last test the send queue + * but before we clear RDS_IN_XMIT. In that case they'd back off and + * not try and send their newly queued message. We need to check the + * send queue after having cleared RDS_IN_XMIT so that their message + * doesn't get stuck on the send queue. + * + * If the transport cannot continue (i.e ret != 0), then it must + * call us when more room is available, such as from the tx + * completion handler. + */ + if (ret == 0) { + smp_mb(); if (!list_empty(&conn->c_send_queue)) { - rds_stats_inc(s_send_sem_queue_raced); - ret = -EAGAIN; + rds_stats_inc(s_send_lock_queue_raced); + goto restart; } - spin_unlock_irqrestore(&conn->c_lock, flags); } out: return ret; @@ -376,52 +411,60 @@ static inline int rds_send_is_acked(struct rds_message *rm, u64 ack, } /* - * Returns true if there are no messages on the send and retransmit queues - * which have a sequence number greater than or equal to the given sequence - * number. + * This is pretty similar to what happens below in the ACK + * handling code - except that we call here as soon as we get + * the IB send completion on the RDMA op and the accompanying + * message. */ -int rds_send_acked_before(struct rds_connection *conn, u64 seq) +void rds_rdma_send_complete(struct rds_message *rm, int status) { - struct rds_message *rm, *tmp; - int ret = 1; + struct rds_sock *rs = NULL; + struct rm_rdma_op *ro; + struct rds_notifier *notifier; + unsigned long flags; - spin_lock(&conn->c_lock); + spin_lock_irqsave(&rm->m_rs_lock, flags); - list_for_each_entry_safe(rm, tmp, &conn->c_retrans, m_conn_item) { - if (be64_to_cpu(rm->m_inc.i_hdr.h_sequence) < seq) - ret = 0; - break; - } + ro = &rm->rdma; + if (test_bit(RDS_MSG_ON_SOCK, &rm->m_flags) && + ro->op_active && ro->op_notify && ro->op_notifier) { + notifier = ro->op_notifier; + rs = rm->m_rs; + sock_hold(rds_rs_to_sk(rs)); - list_for_each_entry_safe(rm, tmp, &conn->c_send_queue, m_conn_item) { - if (be64_to_cpu(rm->m_inc.i_hdr.h_sequence) < seq) - ret = 0; - break; + notifier->n_status = status; + spin_lock(&rs->rs_lock); + list_add_tail(¬ifier->n_list, &rs->rs_notify_queue); + spin_unlock(&rs->rs_lock); + + ro->op_notifier = NULL; } - spin_unlock(&conn->c_lock); + spin_unlock_irqrestore(&rm->m_rs_lock, flags); - return ret; + if (rs) { + rds_wake_sk_sleep(rs); + sock_put(rds_rs_to_sk(rs)); + } } +EXPORT_SYMBOL_GPL(rds_rdma_send_complete); /* - * This is pretty similar to what happens below in the ACK - * handling code - except that we call here as soon as we get - * the IB send completion on the RDMA op and the accompanying - * message. + * Just like above, except looks at atomic op */ -void rds_rdma_send_complete(struct rds_message *rm, int status) +void rds_atomic_send_complete(struct rds_message *rm, int status) { struct rds_sock *rs = NULL; - struct rds_rdma_op *ro; + struct rm_atomic_op *ao; struct rds_notifier *notifier; + unsigned long flags; - spin_lock(&rm->m_rs_lock); + spin_lock_irqsave(&rm->m_rs_lock, flags); - ro = rm->m_rdma_op; - if (test_bit(RDS_MSG_ON_SOCK, &rm->m_flags) && - ro && ro->r_notify && ro->r_notifier) { - notifier = ro->r_notifier; + ao = &rm->atomic; + if (test_bit(RDS_MSG_ON_SOCK, &rm->m_flags) + && ao->op_active && ao->op_notify && ao->op_notifier) { + notifier = ao->op_notifier; rs = rm->m_rs; sock_hold(rds_rs_to_sk(rs)); @@ -430,17 +473,17 @@ void rds_rdma_send_complete(struct rds_message *rm, int status) list_add_tail(¬ifier->n_list, &rs->rs_notify_queue); spin_unlock(&rs->rs_lock); - ro->r_notifier = NULL; + ao->op_notifier = NULL; } - spin_unlock(&rm->m_rs_lock); + spin_unlock_irqrestore(&rm->m_rs_lock, flags); if (rs) { rds_wake_sk_sleep(rs); sock_put(rds_rs_to_sk(rs)); } } -EXPORT_SYMBOL_GPL(rds_rdma_send_complete); +EXPORT_SYMBOL_GPL(rds_atomic_send_complete); /* * This is the same as rds_rdma_send_complete except we @@ -448,15 +491,23 @@ EXPORT_SYMBOL_GPL(rds_rdma_send_complete); * socket, socket lock) and can just move the notifier. */ static inline void -__rds_rdma_send_complete(struct rds_sock *rs, struct rds_message *rm, int status) +__rds_send_complete(struct rds_sock *rs, struct rds_message *rm, int status) { - struct rds_rdma_op *ro; + struct rm_rdma_op *ro; + struct rm_atomic_op *ao; + + ro = &rm->rdma; + if (ro->op_active && ro->op_notify && ro->op_notifier) { + ro->op_notifier->n_status = status; + list_add_tail(&ro->op_notifier->n_list, &rs->rs_notify_queue); + ro->op_notifier = NULL; + } - ro = rm->m_rdma_op; - if (ro && ro->r_notify && ro->r_notifier) { - ro->r_notifier->n_status = status; - list_add_tail(&ro->r_notifier->n_list, &rs->rs_notify_queue); - ro->r_notifier = NULL; + ao = &rm->atomic; + if (ao->op_active && ao->op_notify && ao->op_notifier) { + ao->op_notifier->n_status = status; + list_add_tail(&ao->op_notifier->n_list, &rs->rs_notify_queue); + ao->op_notifier = NULL; } /* No need to wake the app - caller does this */ @@ -468,7 +519,7 @@ __rds_rdma_send_complete(struct rds_sock *rs, struct rds_message *rm, int status * So speed is not an issue here. */ struct rds_message *rds_send_get_message(struct rds_connection *conn, - struct rds_rdma_op *op) + struct rm_rdma_op *op) { struct rds_message *rm, *tmp, *found = NULL; unsigned long flags; @@ -476,7 +527,7 @@ struct rds_message *rds_send_get_message(struct rds_connection *conn, spin_lock_irqsave(&conn->c_lock, flags); list_for_each_entry_safe(rm, tmp, &conn->c_retrans, m_conn_item) { - if (rm->m_rdma_op == op) { + if (&rm->rdma == op) { atomic_inc(&rm->m_refcount); found = rm; goto out; @@ -484,7 +535,7 @@ struct rds_message *rds_send_get_message(struct rds_connection *conn, } list_for_each_entry_safe(rm, tmp, &conn->c_send_queue, m_conn_item) { - if (rm->m_rdma_op == op) { + if (&rm->rdma == op) { atomic_inc(&rm->m_refcount); found = rm; break; @@ -506,7 +557,7 @@ EXPORT_SYMBOL_GPL(rds_send_get_message); * removing the messages from the 'messages' list regardless of if it found * the messages on the socket list or not. */ -void rds_send_remove_from_sock(struct list_head *messages, int status) +static void rds_send_remove_from_sock(struct list_head *messages, int status) { unsigned long flags; struct rds_sock *rs = NULL; @@ -544,19 +595,20 @@ void rds_send_remove_from_sock(struct list_head *messages, int status) spin_lock(&rs->rs_lock); if (test_and_clear_bit(RDS_MSG_ON_SOCK, &rm->m_flags)) { - struct rds_rdma_op *ro = rm->m_rdma_op; + struct rm_rdma_op *ro = &rm->rdma; struct rds_notifier *notifier; list_del_init(&rm->m_sock_item); rds_send_sndbuf_remove(rs, rm); - if (ro && ro->r_notifier && (status || ro->r_notify)) { - notifier = ro->r_notifier; + if (ro->op_active && ro->op_notifier && + (ro->op_notify || (ro->op_recverr && status))) { + notifier = ro->op_notifier; list_add_tail(¬ifier->n_list, &rs->rs_notify_queue); if (!notifier->n_status) notifier->n_status = status; - rm->m_rdma_op->r_notifier = NULL; + rm->rdma.op_notifier = NULL; } was_on_sock = 1; rm->m_rs = NULL; @@ -619,9 +671,8 @@ void rds_send_drop_to(struct rds_sock *rs, struct sockaddr_in *dest) { struct rds_message *rm, *tmp; struct rds_connection *conn; - unsigned long flags, flags2; + unsigned long flags; LIST_HEAD(list); - int wake = 0; /* get all the messages we're dropping under the rs lock */ spin_lock_irqsave(&rs->rs_lock, flags); @@ -631,59 +682,54 @@ void rds_send_drop_to(struct rds_sock *rs, struct sockaddr_in *dest) dest->sin_port != rm->m_inc.i_hdr.h_dport)) continue; - wake = 1; list_move(&rm->m_sock_item, &list); rds_send_sndbuf_remove(rs, rm); clear_bit(RDS_MSG_ON_SOCK, &rm->m_flags); } /* order flag updates with the rs lock */ - if (wake) - smp_mb__after_clear_bit(); + smp_mb__after_clear_bit(); spin_unlock_irqrestore(&rs->rs_lock, flags); - conn = NULL; + if (list_empty(&list)) + return; - /* now remove the messages from the conn list as needed */ + /* Remove the messages from the conn */ list_for_each_entry(rm, &list, m_sock_item) { - /* We do this here rather than in the loop above, so that - * we don't have to nest m_rs_lock under rs->rs_lock */ - spin_lock_irqsave(&rm->m_rs_lock, flags2); - /* If this is a RDMA operation, notify the app. */ - spin_lock(&rs->rs_lock); - __rds_rdma_send_complete(rs, rm, RDS_RDMA_CANCELED); - spin_unlock(&rs->rs_lock); - rm->m_rs = NULL; - spin_unlock_irqrestore(&rm->m_rs_lock, flags2); + conn = rm->m_inc.i_conn; + + spin_lock_irqsave(&conn->c_lock, flags); /* - * If we see this flag cleared then we're *sure* that someone - * else beat us to removing it from the conn. If we race - * with their flag update we'll get the lock and then really - * see that the flag has been cleared. + * Maybe someone else beat us to removing rm from the conn. + * If we race with their flag update we'll get the lock and + * then really see that the flag has been cleared. */ - if (!test_bit(RDS_MSG_ON_CONN, &rm->m_flags)) + if (!test_and_clear_bit(RDS_MSG_ON_CONN, &rm->m_flags)) { + spin_unlock_irqrestore(&conn->c_lock, flags); continue; - - if (conn != rm->m_inc.i_conn) { - if (conn) - spin_unlock_irqrestore(&conn->c_lock, flags); - conn = rm->m_inc.i_conn; - spin_lock_irqsave(&conn->c_lock, flags); } + list_del_init(&rm->m_conn_item); + spin_unlock_irqrestore(&conn->c_lock, flags); - if (test_and_clear_bit(RDS_MSG_ON_CONN, &rm->m_flags)) { - list_del_init(&rm->m_conn_item); - rds_message_put(rm); - } - } + /* + * Couldn't grab m_rs_lock in top loop (lock ordering), + * but we can now. + */ + spin_lock_irqsave(&rm->m_rs_lock, flags); - if (conn) - spin_unlock_irqrestore(&conn->c_lock, flags); + spin_lock(&rs->rs_lock); + __rds_send_complete(rs, rm, RDS_RDMA_CANCELED); + spin_unlock(&rs->rs_lock); - if (wake) - rds_wake_sk_sleep(rs); + rm->m_rs = NULL; + spin_unlock_irqrestore(&rm->m_rs_lock, flags); + + rds_message_put(rm); + } + + rds_wake_sk_sleep(rs); while (!list_empty(&list)) { rm = list_entry(list.next, struct rds_message, m_sock_item); @@ -763,6 +809,63 @@ out: return *queued; } +/* + * rds_message is getting to be quite complicated, and we'd like to allocate + * it all in one go. This figures out how big it needs to be up front. + */ +static int rds_rm_size(struct msghdr *msg, int data_len) +{ + struct cmsghdr *cmsg; + int size = 0; + int cmsg_groups = 0; + int retval; + + for (cmsg = CMSG_FIRSTHDR(msg); cmsg; cmsg = CMSG_NXTHDR(msg, cmsg)) { + if (!CMSG_OK(msg, cmsg)) + return -EINVAL; + + if (cmsg->cmsg_level != SOL_RDS) + continue; + + switch (cmsg->cmsg_type) { + case RDS_CMSG_RDMA_ARGS: + cmsg_groups |= 1; + retval = rds_rdma_extra_size(CMSG_DATA(cmsg)); + if (retval < 0) + return retval; + size += retval; + + break; + + case RDS_CMSG_RDMA_DEST: + case RDS_CMSG_RDMA_MAP: + cmsg_groups |= 2; + /* these are valid but do no add any size */ + break; + + case RDS_CMSG_ATOMIC_CSWP: + case RDS_CMSG_ATOMIC_FADD: + case RDS_CMSG_MASKED_ATOMIC_CSWP: + case RDS_CMSG_MASKED_ATOMIC_FADD: + cmsg_groups |= 1; + size += sizeof(struct scatterlist); + break; + + default: + return -EINVAL; + } + + } + + size += ceil(data_len, PAGE_SIZE) * sizeof(struct scatterlist); + + /* Ensure (DEST, MAP) are never used with (ARGS, ATOMIC) */ + if (cmsg_groups == 3) + return -EINVAL; + + return size; +} + static int rds_cmsg_send(struct rds_sock *rs, struct rds_message *rm, struct msghdr *msg, int *allocated_mr) { @@ -777,7 +880,7 @@ static int rds_cmsg_send(struct rds_sock *rs, struct rds_message *rm, continue; /* As a side effect, RDMA_DEST and RDMA_MAP will set - * rm->m_rdma_cookie and rm->m_rdma_mr. + * rm->rdma.m_rdma_cookie and rm->rdma.m_rdma_mr. */ switch (cmsg->cmsg_type) { case RDS_CMSG_RDMA_ARGS: @@ -793,6 +896,12 @@ static int rds_cmsg_send(struct rds_sock *rs, struct rds_message *rm, if (!ret) *allocated_mr = 1; break; + case RDS_CMSG_ATOMIC_CSWP: + case RDS_CMSG_ATOMIC_FADD: + case RDS_CMSG_MASKED_ATOMIC_CSWP: + case RDS_CMSG_MASKED_ATOMIC_FADD: + ret = rds_cmsg_atomic(rs, rm, cmsg); + break; default: return -EINVAL; @@ -850,13 +959,26 @@ int rds_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg, goto out; } - rm = rds_message_copy_from_user(msg->msg_iov, payload_len); - if (IS_ERR(rm)) { - ret = PTR_ERR(rm); - rm = NULL; + /* size of rm including all sgs */ + ret = rds_rm_size(msg, payload_len); + if (ret < 0) + goto out; + + rm = rds_message_alloc(ret, GFP_KERNEL); + if (!rm) { + ret = -ENOMEM; goto out; } + /* Attach data to the rm */ + if (payload_len) { + rm->data.op_sg = rds_message_alloc_sgs(rm, ceil(payload_len, PAGE_SIZE)); + ret = rds_message_copy_from_user(rm, msg->msg_iov, payload_len); + if (ret) + goto out; + } + rm->data.op_active = 1; + rm->m_daddr = daddr; /* rds_conn_create has a spinlock that runs with IRQ off. @@ -879,22 +1001,23 @@ int rds_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg, if (ret) goto out; - if ((rm->m_rdma_cookie || rm->m_rdma_op) && - conn->c_trans->xmit_rdma == NULL) { + if (rm->rdma.op_active && !conn->c_trans->xmit_rdma) { if (printk_ratelimit()) printk(KERN_NOTICE "rdma_op %p conn xmit_rdma %p\n", - rm->m_rdma_op, conn->c_trans->xmit_rdma); + &rm->rdma, conn->c_trans->xmit_rdma); ret = -EOPNOTSUPP; goto out; } - /* If the connection is down, trigger a connect. We may - * have scheduled a delayed reconnect however - in this case - * we should not interfere. - */ - if (rds_conn_state(conn) == RDS_CONN_DOWN && - !test_and_set_bit(RDS_RECONNECT_PENDING, &conn->c_flags)) - queue_delayed_work(rds_wq, &conn->c_conn_w, 0); + if (rm->atomic.op_active && !conn->c_trans->xmit_atomic) { + if (printk_ratelimit()) + printk(KERN_NOTICE "atomic_op %p conn xmit_atomic %p\n", + &rm->atomic, conn->c_trans->xmit_atomic); + ret = -EOPNOTSUPP; + goto out; + } + + rds_conn_connect_if_down(conn); ret = rds_cong_wait(conn->c_fcong, dport, nonblock, rs); if (ret) { @@ -938,7 +1061,7 @@ int rds_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg, rds_stats_inc(s_send_queued); if (!test_bit(RDS_LL_SEND_FULL, &conn->c_flags)) - rds_send_worker(&conn->c_send_w.work); + rds_send_xmit(conn); rds_message_put(rm); return payload_len; @@ -966,20 +1089,15 @@ rds_send_pong(struct rds_connection *conn, __be16 dport) int ret = 0; rm = rds_message_alloc(0, GFP_ATOMIC); - if (rm == NULL) { + if (!rm) { ret = -ENOMEM; goto out; } rm->m_daddr = conn->c_faddr; + rm->data.op_active = 1; - /* If the connection is down, trigger a connect. We may - * have scheduled a delayed reconnect however - in this case - * we should not interfere. - */ - if (rds_conn_state(conn) == RDS_CONN_DOWN && - !test_and_set_bit(RDS_RECONNECT_PENDING, &conn->c_flags)) - queue_delayed_work(rds_wq, &conn->c_conn_w, 0); + rds_conn_connect_if_down(conn); ret = rds_cong_wait(conn->c_fcong, dport, 1, NULL); if (ret) @@ -999,7 +1117,9 @@ rds_send_pong(struct rds_connection *conn, __be16 dport) rds_stats_inc(s_send_queued); rds_stats_inc(s_send_pong); - queue_delayed_work(rds_wq, &conn->c_send_w, 0); + if (!test_bit(RDS_LL_SEND_FULL, &conn->c_flags)) + rds_send_xmit(conn); + rds_message_put(rm); return 0; diff --git a/net/rds/stats.c b/net/rds/stats.c index 7598eb07cfb..10c759ccac0 100644 --- a/net/rds/stats.c +++ b/net/rds/stats.c @@ -57,8 +57,8 @@ static const char *const rds_stat_names[] = { "recv_ping", "send_queue_empty", "send_queue_full", - "send_sem_contention", - "send_sem_queue_raced", + "send_lock_contention", + "send_lock_queue_raced", "send_immediate_retry", "send_delayed_retry", "send_drop_acked", @@ -143,7 +143,7 @@ void rds_stats_exit(void) rds_info_deregister_func(RDS_INFO_COUNTERS, rds_stats_info); } -int __init rds_stats_init(void) +int rds_stats_init(void) { rds_info_register_func(RDS_INFO_COUNTERS, rds_stats_info); return 0; diff --git a/net/rds/sysctl.c b/net/rds/sysctl.c index 7829a20325d..25ad0c77a26 100644 --- a/net/rds/sysctl.c +++ b/net/rds/sysctl.c @@ -105,13 +105,13 @@ void rds_sysctl_exit(void) unregister_sysctl_table(rds_sysctl_reg_table); } -int __init rds_sysctl_init(void) +int rds_sysctl_init(void) { rds_sysctl_reconnect_min = msecs_to_jiffies(1); rds_sysctl_reconnect_min_jiffies = rds_sysctl_reconnect_min; rds_sysctl_reg_table = register_sysctl_paths(rds_sysctl_path, rds_sysctl_rds_table); - if (rds_sysctl_reg_table == NULL) + if (!rds_sysctl_reg_table) return -ENOMEM; return 0; } diff --git a/net/rds/tcp.c b/net/rds/tcp.c index babf4577ff7..08a8c6cf2d1 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -41,7 +41,7 @@ /* only for info exporting */ static DEFINE_SPINLOCK(rds_tcp_tc_list_lock); static LIST_HEAD(rds_tcp_tc_list); -unsigned int rds_tcp_tc_count; +static unsigned int rds_tcp_tc_count; /* Track rds_tcp_connection structs so they can be cleaned up */ static DEFINE_SPINLOCK(rds_tcp_conn_lock); @@ -200,7 +200,7 @@ static int rds_tcp_conn_alloc(struct rds_connection *conn, gfp_t gfp) struct rds_tcp_connection *tc; tc = kmem_cache_alloc(rds_tcp_conn_slab, gfp); - if (tc == NULL) + if (!tc) return -ENOMEM; tc->t_sock = NULL; @@ -243,7 +243,7 @@ static void rds_tcp_destroy_conns(void) } } -void rds_tcp_exit(void) +static void rds_tcp_exit(void) { rds_info_deregister_func(RDS_INFO_TCP_SOCKETS, rds_tcp_tc_info); rds_tcp_listen_stop(); @@ -258,7 +258,6 @@ struct rds_transport rds_tcp_transport = { .laddr_check = rds_tcp_laddr_check, .xmit_prepare = rds_tcp_xmit_prepare, .xmit_complete = rds_tcp_xmit_complete, - .xmit_cong_map = rds_tcp_xmit_cong_map, .xmit = rds_tcp_xmit, .recv = rds_tcp_recv, .conn_alloc = rds_tcp_conn_alloc, @@ -266,7 +265,6 @@ struct rds_transport rds_tcp_transport = { .conn_connect = rds_tcp_conn_connect, .conn_shutdown = rds_tcp_conn_shutdown, .inc_copy_to_user = rds_tcp_inc_copy_to_user, - .inc_purge = rds_tcp_inc_purge, .inc_free = rds_tcp_inc_free, .stats_info_copy = rds_tcp_stats_info_copy, .exit = rds_tcp_exit, @@ -276,14 +274,14 @@ struct rds_transport rds_tcp_transport = { .t_prefer_loopback = 1, }; -int __init rds_tcp_init(void) +static int rds_tcp_init(void) { int ret; rds_tcp_conn_slab = kmem_cache_create("rds_tcp_connection", sizeof(struct rds_tcp_connection), 0, 0, NULL); - if (rds_tcp_conn_slab == NULL) { + if (!rds_tcp_conn_slab) { ret = -ENOMEM; goto out; } diff --git a/net/rds/tcp.h b/net/rds/tcp.h index 844fa6b9cf5..9cf2927d002 100644 --- a/net/rds/tcp.h +++ b/net/rds/tcp.h @@ -43,8 +43,6 @@ struct rds_tcp_statistics { }; /* tcp.c */ -int __init rds_tcp_init(void); -void rds_tcp_exit(void); void rds_tcp_tune(struct socket *sock); void rds_tcp_nonagle(struct socket *sock); void rds_tcp_set_callbacks(struct socket *sock, struct rds_connection *conn); @@ -61,16 +59,15 @@ void rds_tcp_conn_shutdown(struct rds_connection *conn); void rds_tcp_state_change(struct sock *sk); /* tcp_listen.c */ -int __init rds_tcp_listen_init(void); +int rds_tcp_listen_init(void); void rds_tcp_listen_stop(void); void rds_tcp_listen_data_ready(struct sock *sk, int bytes); /* tcp_recv.c */ -int __init rds_tcp_recv_init(void); +int rds_tcp_recv_init(void); void rds_tcp_recv_exit(void); void rds_tcp_data_ready(struct sock *sk, int bytes); int rds_tcp_recv(struct rds_connection *conn); -void rds_tcp_inc_purge(struct rds_incoming *inc); void rds_tcp_inc_free(struct rds_incoming *inc); int rds_tcp_inc_copy_to_user(struct rds_incoming *inc, struct iovec *iov, size_t size); @@ -81,8 +78,6 @@ void rds_tcp_xmit_complete(struct rds_connection *conn); int rds_tcp_xmit(struct rds_connection *conn, struct rds_message *rm, unsigned int hdr_off, unsigned int sg, unsigned int off); void rds_tcp_write_space(struct sock *sk); -int rds_tcp_xmit_cong_map(struct rds_connection *conn, - struct rds_cong_map *map, unsigned long offset); /* tcp_stats.c */ DECLARE_PER_CPU(struct rds_tcp_statistics, rds_tcp_stats); diff --git a/net/rds/tcp_connect.c b/net/rds/tcp_connect.c index c397524c039..af95c8e058f 100644 --- a/net/rds/tcp_connect.c +++ b/net/rds/tcp_connect.c @@ -43,9 +43,9 @@ void rds_tcp_state_change(struct sock *sk) struct rds_connection *conn; struct rds_tcp_connection *tc; - read_lock(&sk->sk_callback_lock); + read_lock_bh(&sk->sk_callback_lock); conn = sk->sk_user_data; - if (conn == NULL) { + if (!conn) { state_change = sk->sk_state_change; goto out; } @@ -68,7 +68,7 @@ void rds_tcp_state_change(struct sock *sk) break; } out: - read_unlock(&sk->sk_callback_lock); + read_unlock_bh(&sk->sk_callback_lock); state_change(sk); } diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c index 975183fe695..8b5cc4aa886 100644 --- a/net/rds/tcp_listen.c +++ b/net/rds/tcp_listen.c @@ -114,9 +114,9 @@ void rds_tcp_listen_data_ready(struct sock *sk, int bytes) rdsdebug("listen data ready sk %p\n", sk); - read_lock(&sk->sk_callback_lock); + read_lock_bh(&sk->sk_callback_lock); ready = sk->sk_user_data; - if (ready == NULL) { /* check for teardown race */ + if (!ready) { /* check for teardown race */ ready = sk->sk_data_ready; goto out; } @@ -131,11 +131,11 @@ void rds_tcp_listen_data_ready(struct sock *sk, int bytes) queue_work(rds_wq, &rds_tcp_listen_work); out: - read_unlock(&sk->sk_callback_lock); + read_unlock_bh(&sk->sk_callback_lock); ready(sk, bytes); } -int __init rds_tcp_listen_init(void) +int rds_tcp_listen_init(void) { struct sockaddr_in sin; struct socket *sock = NULL; @@ -178,7 +178,7 @@ void rds_tcp_listen_stop(void) struct socket *sock = rds_tcp_listen_sock; struct sock *sk; - if (sock == NULL) + if (!sock) return; sk = sock->sk; diff --git a/net/rds/tcp_recv.c b/net/rds/tcp_recv.c index 1aba6878fa5..78205e25500 100644 --- a/net/rds/tcp_recv.c +++ b/net/rds/tcp_recv.c @@ -39,7 +39,7 @@ static struct kmem_cache *rds_tcp_incoming_slab; -void rds_tcp_inc_purge(struct rds_incoming *inc) +static void rds_tcp_inc_purge(struct rds_incoming *inc) { struct rds_tcp_incoming *tinc; tinc = container_of(inc, struct rds_tcp_incoming, ti_inc); @@ -190,10 +190,10 @@ static int rds_tcp_data_recv(read_descriptor_t *desc, struct sk_buff *skb, * processing. */ while (left) { - if (tinc == NULL) { + if (!tinc) { tinc = kmem_cache_alloc(rds_tcp_incoming_slab, arg->gfp); - if (tinc == NULL) { + if (!tinc) { desc->error = -ENOMEM; goto out; } @@ -229,7 +229,7 @@ static int rds_tcp_data_recv(read_descriptor_t *desc, struct sk_buff *skb, if (left && tc->t_tinc_data_rem) { clone = skb_clone(skb, arg->gfp); - if (clone == NULL) { + if (!clone) { desc->error = -ENOMEM; goto out; } @@ -272,7 +272,8 @@ out: } /* the caller has to hold the sock lock */ -int rds_tcp_read_sock(struct rds_connection *conn, gfp_t gfp, enum km_type km) +static int rds_tcp_read_sock(struct rds_connection *conn, gfp_t gfp, + enum km_type km) { struct rds_tcp_connection *tc = conn->c_transport_data; struct socket *sock = tc->t_sock; @@ -324,9 +325,9 @@ void rds_tcp_data_ready(struct sock *sk, int bytes) rdsdebug("data ready sk %p bytes %d\n", sk, bytes); - read_lock(&sk->sk_callback_lock); + read_lock_bh(&sk->sk_callback_lock); conn = sk->sk_user_data; - if (conn == NULL) { /* check for teardown race */ + if (!conn) { /* check for teardown race */ ready = sk->sk_data_ready; goto out; } @@ -338,16 +339,16 @@ void rds_tcp_data_ready(struct sock *sk, int bytes) if (rds_tcp_read_sock(conn, GFP_ATOMIC, KM_SOFTIRQ0) == -ENOMEM) queue_delayed_work(rds_wq, &conn->c_recv_w, 0); out: - read_unlock(&sk->sk_callback_lock); + read_unlock_bh(&sk->sk_callback_lock); ready(sk, bytes); } -int __init rds_tcp_recv_init(void) +int rds_tcp_recv_init(void) { rds_tcp_incoming_slab = kmem_cache_create("rds_tcp_incoming", sizeof(struct rds_tcp_incoming), 0, 0, NULL); - if (rds_tcp_incoming_slab == NULL) + if (!rds_tcp_incoming_slab) return -ENOMEM; return 0; } diff --git a/net/rds/tcp_send.c b/net/rds/tcp_send.c index a28b895ff0d..1b4fd68f0c7 100644 --- a/net/rds/tcp_send.c +++ b/net/rds/tcp_send.c @@ -63,7 +63,7 @@ void rds_tcp_xmit_complete(struct rds_connection *conn) } /* the core send_sem serializes this with other xmit and shutdown */ -int rds_tcp_sendmsg(struct socket *sock, void *data, unsigned int len) +static int rds_tcp_sendmsg(struct socket *sock, void *data, unsigned int len) { struct kvec vec = { .iov_base = data, @@ -77,56 +77,6 @@ int rds_tcp_sendmsg(struct socket *sock, void *data, unsigned int len) } /* the core send_sem serializes this with other xmit and shutdown */ -int rds_tcp_xmit_cong_map(struct rds_connection *conn, - struct rds_cong_map *map, unsigned long offset) -{ - static struct rds_header rds_tcp_map_header = { - .h_flags = RDS_FLAG_CONG_BITMAP, - }; - struct rds_tcp_connection *tc = conn->c_transport_data; - unsigned long i; - int ret; - int copied = 0; - - /* Some problem claims cpu_to_be32(constant) isn't a constant. */ - rds_tcp_map_header.h_len = cpu_to_be32(RDS_CONG_MAP_BYTES); - - if (offset < sizeof(struct rds_header)) { - ret = rds_tcp_sendmsg(tc->t_sock, - (void *)&rds_tcp_map_header + offset, - sizeof(struct rds_header) - offset); - if (ret <= 0) - return ret; - offset += ret; - copied = ret; - if (offset < sizeof(struct rds_header)) - return ret; - } - - offset -= sizeof(struct rds_header); - i = offset / PAGE_SIZE; - offset = offset % PAGE_SIZE; - BUG_ON(i >= RDS_CONG_MAP_PAGES); - - do { - ret = tc->t_sock->ops->sendpage(tc->t_sock, - virt_to_page(map->m_page_addrs[i]), - offset, PAGE_SIZE - offset, - MSG_DONTWAIT); - if (ret <= 0) - break; - copied += ret; - offset += ret; - if (offset == PAGE_SIZE) { - offset = 0; - i++; - } - } while (i < RDS_CONG_MAP_PAGES); - - return copied ? copied : ret; -} - -/* the core send_sem serializes this with other xmit and shutdown */ int rds_tcp_xmit(struct rds_connection *conn, struct rds_message *rm, unsigned int hdr_off, unsigned int sg, unsigned int off) { @@ -166,21 +116,21 @@ int rds_tcp_xmit(struct rds_connection *conn, struct rds_message *rm, goto out; } - while (sg < rm->m_nents) { + while (sg < rm->data.op_nents) { ret = tc->t_sock->ops->sendpage(tc->t_sock, - sg_page(&rm->m_sg[sg]), - rm->m_sg[sg].offset + off, - rm->m_sg[sg].length - off, + sg_page(&rm->data.op_sg[sg]), + rm->data.op_sg[sg].offset + off, + rm->data.op_sg[sg].length - off, MSG_DONTWAIT|MSG_NOSIGNAL); - rdsdebug("tcp sendpage %p:%u:%u ret %d\n", (void *)sg_page(&rm->m_sg[sg]), - rm->m_sg[sg].offset + off, rm->m_sg[sg].length - off, + rdsdebug("tcp sendpage %p:%u:%u ret %d\n", (void *)sg_page(&rm->data.op_sg[sg]), + rm->data.op_sg[sg].offset + off, rm->data.op_sg[sg].length - off, ret); if (ret <= 0) break; off += ret; done += ret; - if (off == rm->m_sg[sg].length) { + if (off == rm->data.op_sg[sg].length) { off = 0; sg++; } @@ -224,9 +174,9 @@ void rds_tcp_write_space(struct sock *sk) struct rds_connection *conn; struct rds_tcp_connection *tc; - read_lock(&sk->sk_callback_lock); + read_lock_bh(&sk->sk_callback_lock); conn = sk->sk_user_data; - if (conn == NULL) { + if (!conn) { write_space = sk->sk_write_space; goto out; } @@ -244,7 +194,7 @@ void rds_tcp_write_space(struct sock *sk) queue_delayed_work(rds_wq, &conn->c_send_w, 0); out: - read_unlock(&sk->sk_callback_lock); + read_unlock_bh(&sk->sk_callback_lock); /* * write_space is only called when data leaves tcp's send queue if diff --git a/net/rds/threads.c b/net/rds/threads.c index 786c20eaaf5..0fd90f8c5f5 100644 --- a/net/rds/threads.c +++ b/net/rds/threads.c @@ -61,7 +61,7 @@ * * Transition to state DISCONNECTING/DOWN: * - Inside the shutdown worker; synchronizes with xmit path - * through c_send_lock, and with connection management callbacks + * through RDS_IN_XMIT, and with connection management callbacks * via c_cm_lock. * * For receive callbacks, we rely on the underlying transport @@ -110,7 +110,7 @@ EXPORT_SYMBOL_GPL(rds_connect_complete); * We should *always* start with a random backoff; otherwise a broken connection * will always take several iterations to be re-established. */ -static void rds_queue_reconnect(struct rds_connection *conn) +void rds_queue_reconnect(struct rds_connection *conn) { unsigned long rand; @@ -156,58 +156,6 @@ void rds_connect_worker(struct work_struct *work) } } -void rds_shutdown_worker(struct work_struct *work) -{ - struct rds_connection *conn = container_of(work, struct rds_connection, c_down_w); - - /* shut it down unless it's down already */ - if (!rds_conn_transition(conn, RDS_CONN_DOWN, RDS_CONN_DOWN)) { - /* - * Quiesce the connection mgmt handlers before we start tearing - * things down. We don't hold the mutex for the entire - * duration of the shutdown operation, else we may be - * deadlocking with the CM handler. Instead, the CM event - * handler is supposed to check for state DISCONNECTING - */ - mutex_lock(&conn->c_cm_lock); - if (!rds_conn_transition(conn, RDS_CONN_UP, RDS_CONN_DISCONNECTING) && - !rds_conn_transition(conn, RDS_CONN_ERROR, RDS_CONN_DISCONNECTING)) { - rds_conn_error(conn, "shutdown called in state %d\n", - atomic_read(&conn->c_state)); - mutex_unlock(&conn->c_cm_lock); - return; - } - mutex_unlock(&conn->c_cm_lock); - - mutex_lock(&conn->c_send_lock); - conn->c_trans->conn_shutdown(conn); - rds_conn_reset(conn); - mutex_unlock(&conn->c_send_lock); - - if (!rds_conn_transition(conn, RDS_CONN_DISCONNECTING, RDS_CONN_DOWN)) { - /* This can happen - eg when we're in the middle of tearing - * down the connection, and someone unloads the rds module. - * Quite reproduceable with loopback connections. - * Mostly harmless. - */ - rds_conn_error(conn, - "%s: failed to transition to state DOWN, " - "current state is %d\n", - __func__, - atomic_read(&conn->c_state)); - return; - } - } - - /* Then reconnect if it's still live. - * The passive side of an IB loopback connection is never added - * to the conn hash, so we never trigger a reconnect on this - * conn - the reconnect is always triggered by the active peer. */ - cancel_delayed_work(&conn->c_conn_w); - if (!hlist_unhashed(&conn->c_hash_node)) - rds_queue_reconnect(conn); -} - void rds_send_worker(struct work_struct *work) { struct rds_connection *conn = container_of(work, struct rds_connection, c_send_w.work); @@ -252,15 +200,22 @@ void rds_recv_worker(struct work_struct *work) } } +void rds_shutdown_worker(struct work_struct *work) +{ + struct rds_connection *conn = container_of(work, struct rds_connection, c_down_w); + + rds_conn_shutdown(conn); +} + void rds_threads_exit(void) { destroy_workqueue(rds_wq); } -int __init rds_threads_init(void) +int rds_threads_init(void) { - rds_wq = create_workqueue("krdsd"); - if (rds_wq == NULL) + rds_wq = create_singlethread_workqueue("krdsd"); + if (!rds_wq) return -ENOMEM; return 0; diff --git a/net/rds/transport.c b/net/rds/transport.c index 7e106790135..7f2ac4fec36 100644 --- a/net/rds/transport.c +++ b/net/rds/transport.c @@ -71,19 +71,28 @@ void rds_trans_unregister(struct rds_transport *trans) } EXPORT_SYMBOL_GPL(rds_trans_unregister); +void rds_trans_put(struct rds_transport *trans) +{ + if (trans && trans->t_owner) + module_put(trans->t_owner); +} + struct rds_transport *rds_trans_get_preferred(__be32 addr) { struct rds_transport *ret = NULL; - int i; + struct rds_transport *trans; + unsigned int i; if (IN_LOOPBACK(ntohl(addr))) return &rds_loop_transport; down_read(&rds_trans_sem); - for (i = 0; i < RDS_TRANS_COUNT; i++) - { - if (transports[i] && (transports[i]->laddr_check(addr) == 0)) { - ret = transports[i]; + for (i = 0; i < RDS_TRANS_COUNT; i++) { + trans = transports[i]; + + if (trans && (trans->laddr_check(addr) == 0) && + (!trans->t_owner || try_module_get(trans->t_owner))) { + ret = trans; break; } } diff --git a/net/rds/xlist.h b/net/rds/xlist.h new file mode 100644 index 00000000000..e6b5190dadd --- /dev/null +++ b/net/rds/xlist.h @@ -0,0 +1,80 @@ +#ifndef _LINUX_XLIST_H +#define _LINUX_XLIST_H + +#include <linux/stddef.h> +#include <linux/poison.h> +#include <linux/prefetch.h> +#include <asm/system.h> + +struct xlist_head { + struct xlist_head *next; +}; + +static inline void INIT_XLIST_HEAD(struct xlist_head *list) +{ + list->next = NULL; +} + +static inline int xlist_empty(struct xlist_head *head) +{ + return head->next == NULL; +} + +static inline void xlist_add(struct xlist_head *new, struct xlist_head *tail, + struct xlist_head *head) +{ + struct xlist_head *cur; + struct xlist_head *check; + + while (1) { + cur = head->next; + tail->next = cur; + check = cmpxchg(&head->next, cur, new); + if (check == cur) + break; + } +} + +static inline struct xlist_head *xlist_del_head(struct xlist_head *head) +{ + struct xlist_head *cur; + struct xlist_head *check; + struct xlist_head *next; + + while (1) { + cur = head->next; + if (!cur) + goto out; + + next = cur->next; + check = cmpxchg(&head->next, cur, next); + if (check == cur) + goto out; + } +out: + return cur; +} + +static inline struct xlist_head *xlist_del_head_fast(struct xlist_head *head) +{ + struct xlist_head *cur; + + cur = head->next; + if (!cur) + return NULL; + + head->next = cur->next; + return cur; +} + +static inline void xlist_splice(struct xlist_head *list, + struct xlist_head *head) +{ + struct xlist_head *cur; + + WARN_ON(head->next); + cur = xchg(&list->next, NULL); + head->next = cur; +} + +#endif diff --git a/net/rfkill/core.c b/net/rfkill/core.c index 51875a0c5d4..04f599089e6 100644 --- a/net/rfkill/core.c +++ b/net/rfkill/core.c @@ -1241,6 +1241,7 @@ static const struct file_operations rfkill_fops = { .unlocked_ioctl = rfkill_fop_ioctl, .compat_ioctl = rfkill_fop_ioctl, #endif + .llseek = no_llseek, }; static struct miscdevice rfkill_miscdev = { diff --git a/net/rfkill/input.c b/net/rfkill/input.c index 3713d7ecab9..1bca6d49ec9 100644 --- a/net/rfkill/input.c +++ b/net/rfkill/input.c @@ -142,7 +142,7 @@ static unsigned long rfkill_last_scheduled; static unsigned long rfkill_ratelimit(const unsigned long last) { const unsigned long delay = msecs_to_jiffies(RFKILL_OPS_DELAY); - return (time_after(jiffies, last + delay)) ? 0 : delay; + return time_after(jiffies, last + delay) ? 0 : delay; } static void rfkill_schedule_ratelimited(void) diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c index 8e45e76a95f..d952e7eac18 100644 --- a/net/rose/af_rose.c +++ b/net/rose/af_rose.c @@ -679,7 +679,7 @@ static int rose_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) if (addr_len == sizeof(struct sockaddr_rose) && addr->srose_ndigis > 1) return -EINVAL; - if (addr->srose_ndigis > ROSE_MAX_DIGIS) + if ((unsigned int) addr->srose_ndigis > ROSE_MAX_DIGIS) return -EINVAL; if ((dev = rose_dev_get(&addr->srose_addr)) == NULL) { @@ -739,7 +739,7 @@ static int rose_connect(struct socket *sock, struct sockaddr *uaddr, int addr_le if (addr_len == sizeof(struct sockaddr_rose) && addr->srose_ndigis > 1) return -EINVAL; - if (addr->srose_ndigis > ROSE_MAX_DIGIS) + if ((unsigned int) addr->srose_ndigis > ROSE_MAX_DIGIS) return -EINVAL; /* Source + Destination digis should not exceed ROSE_MAX_DIGIS */ diff --git a/net/rose/rose_link.c b/net/rose/rose_link.c index a750a28e022..fa5f5641a2c 100644 --- a/net/rose/rose_link.c +++ b/net/rose/rose_link.c @@ -114,7 +114,7 @@ static int rose_send_frame(struct sk_buff *skb, struct rose_neigh *neigh) if (ax25s) ax25_cb_put(ax25s); - return (neigh->ax25 != NULL); + return neigh->ax25 != NULL; } /* @@ -137,7 +137,7 @@ static int rose_link_up(struct rose_neigh *neigh) if (ax25s) ax25_cb_put(ax25s); - return (neigh->ax25 != NULL); + return neigh->ax25 != NULL; } /* diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 7043b294bb6..8e22bd345e7 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -597,12 +597,6 @@ extern unsigned rxrpc_debug; #define dbgprintk(FMT,...) \ printk("[%-6.6s] "FMT"\n", current->comm ,##__VA_ARGS__) -/* make sure we maintain the format strings, even when debugging is disabled */ -static inline __attribute__((format(printf,1,2))) -void _dbprintk(const char *fmt, ...) -{ -} - #define kenter(FMT,...) dbgprintk("==> %s("FMT")",__func__ ,##__VA_ARGS__) #define kleave(FMT,...) dbgprintk("<== %s()"FMT"",__func__ ,##__VA_ARGS__) #define kdebug(FMT,...) dbgprintk(" "FMT ,##__VA_ARGS__) @@ -655,11 +649,11 @@ do { \ } while (0) #else -#define _enter(FMT,...) _dbprintk("==> %s("FMT")",__func__ ,##__VA_ARGS__) -#define _leave(FMT,...) _dbprintk("<== %s()"FMT"",__func__ ,##__VA_ARGS__) -#define _debug(FMT,...) _dbprintk(" "FMT ,##__VA_ARGS__) -#define _proto(FMT,...) _dbprintk("### "FMT ,##__VA_ARGS__) -#define _net(FMT,...) _dbprintk("@@@ "FMT ,##__VA_ARGS__) +#define _enter(FMT,...) no_printk("==> %s("FMT")",__func__ ,##__VA_ARGS__) +#define _leave(FMT,...) no_printk("<== %s()"FMT"",__func__ ,##__VA_ARGS__) +#define _debug(FMT,...) no_printk(" "FMT ,##__VA_ARGS__) +#define _proto(FMT,...) no_printk("### "FMT ,##__VA_ARGS__) +#define _net(FMT,...) no_printk("@@@ "FMT ,##__VA_ARGS__) #endif /* diff --git a/net/sched/Kconfig b/net/sched/Kconfig index 2f691fb180d..a36270a994d 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -518,6 +518,16 @@ config NET_ACT_SKBEDIT To compile this code as a module, choose M here: the module will be called act_skbedit. +config NET_ACT_CSUM + tristate "Checksum Updating" + depends on NET_CLS_ACT && INET + ---help--- + Say Y here to update some common checksum after some direct + packet alterations. + + To compile this code as a module, choose M here: the + module will be called act_csum. + config NET_CLS_IND bool "Incoming device classification" depends on NET_CLS_U32 || NET_CLS_FW diff --git a/net/sched/Makefile b/net/sched/Makefile index f14e71bfa58..960f5dba630 100644 --- a/net/sched/Makefile +++ b/net/sched/Makefile @@ -15,6 +15,7 @@ obj-$(CONFIG_NET_ACT_NAT) += act_nat.o obj-$(CONFIG_NET_ACT_PEDIT) += act_pedit.o obj-$(CONFIG_NET_ACT_SIMP) += act_simple.o obj-$(CONFIG_NET_ACT_SKBEDIT) += act_skbedit.o +obj-$(CONFIG_NET_ACT_CSUM) += act_csum.o obj-$(CONFIG_NET_SCH_FIFO) += sch_fifo.o obj-$(CONFIG_NET_SCH_CBQ) += sch_cbq.o obj-$(CONFIG_NET_SCH_HTB) += sch_htb.o diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c new file mode 100644 index 00000000000..67dc7ce9b63 --- /dev/null +++ b/net/sched/act_csum.c @@ -0,0 +1,595 @@ +/* + * Checksum updating actions + * + * Copyright (c) 2010 Gregoire Baron <baronchon@n7mm.org> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * + */ + +#include <linux/types.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/spinlock.h> + +#include <linux/netlink.h> +#include <net/netlink.h> +#include <linux/rtnetlink.h> + +#include <linux/skbuff.h> + +#include <net/ip.h> +#include <net/ipv6.h> +#include <net/icmp.h> +#include <linux/icmpv6.h> +#include <linux/igmp.h> +#include <net/tcp.h> +#include <net/udp.h> +#include <net/ip6_checksum.h> + +#include <net/act_api.h> + +#include <linux/tc_act/tc_csum.h> +#include <net/tc_act/tc_csum.h> + +#define CSUM_TAB_MASK 15 +static struct tcf_common *tcf_csum_ht[CSUM_TAB_MASK + 1]; +static u32 csum_idx_gen; +static DEFINE_RWLOCK(csum_lock); + +static struct tcf_hashinfo csum_hash_info = { + .htab = tcf_csum_ht, + .hmask = CSUM_TAB_MASK, + .lock = &csum_lock, +}; + +static const struct nla_policy csum_policy[TCA_CSUM_MAX + 1] = { + [TCA_CSUM_PARMS] = { .len = sizeof(struct tc_csum), }, +}; + +static int tcf_csum_init(struct nlattr *nla, struct nlattr *est, + struct tc_action *a, int ovr, int bind) +{ + struct nlattr *tb[TCA_CSUM_MAX + 1]; + struct tc_csum *parm; + struct tcf_common *pc; + struct tcf_csum *p; + int ret = 0, err; + + if (nla == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_CSUM_MAX, nla,csum_policy); + if (err < 0) + return err; + + if (tb[TCA_CSUM_PARMS] == NULL) + return -EINVAL; + parm = nla_data(tb[TCA_CSUM_PARMS]); + + pc = tcf_hash_check(parm->index, a, bind, &csum_hash_info); + if (!pc) { + pc = tcf_hash_create(parm->index, est, a, sizeof(*p), bind, + &csum_idx_gen, &csum_hash_info); + if (IS_ERR(pc)) + return PTR_ERR(pc); + p = to_tcf_csum(pc); + ret = ACT_P_CREATED; + } else { + p = to_tcf_csum(pc); + if (!ovr) { + tcf_hash_release(pc, bind, &csum_hash_info); + return -EEXIST; + } + } + + spin_lock_bh(&p->tcf_lock); + p->tcf_action = parm->action; + p->update_flags = parm->update_flags; + spin_unlock_bh(&p->tcf_lock); + + if (ret == ACT_P_CREATED) + tcf_hash_insert(pc, &csum_hash_info); + + return ret; +} + +static int tcf_csum_cleanup(struct tc_action *a, int bind) +{ + struct tcf_csum *p = a->priv; + return tcf_hash_release(&p->common, bind, &csum_hash_info); +} + +/** + * tcf_csum_skb_nextlayer - Get next layer pointer + * @skb: sk_buff to use + * @ihl: previous summed headers length + * @ipl: complete packet length + * @jhl: next header length + * + * Check the expected next layer availability in the specified sk_buff. + * Return the next layer pointer if pass, NULL otherwise. + */ +static void *tcf_csum_skb_nextlayer(struct sk_buff *skb, + unsigned int ihl, unsigned int ipl, + unsigned int jhl) +{ + int ntkoff = skb_network_offset(skb); + int hl = ihl + jhl; + + if (!pskb_may_pull(skb, ipl + ntkoff) || (ipl < hl) || + (skb_cloned(skb) && + !skb_clone_writable(skb, hl + ntkoff) && + pskb_expand_head(skb, 0, 0, GFP_ATOMIC))) + return NULL; + else + return (void *)(skb_network_header(skb) + ihl); +} + +static int tcf_csum_ipv4_icmp(struct sk_buff *skb, + unsigned int ihl, unsigned int ipl) +{ + struct icmphdr *icmph; + + icmph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*icmph)); + if (icmph == NULL) + return 0; + + icmph->checksum = 0; + skb->csum = csum_partial(icmph, ipl - ihl, 0); + icmph->checksum = csum_fold(skb->csum); + + skb->ip_summed = CHECKSUM_NONE; + + return 1; +} + +static int tcf_csum_ipv4_igmp(struct sk_buff *skb, + unsigned int ihl, unsigned int ipl) +{ + struct igmphdr *igmph; + + igmph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*igmph)); + if (igmph == NULL) + return 0; + + igmph->csum = 0; + skb->csum = csum_partial(igmph, ipl - ihl, 0); + igmph->csum = csum_fold(skb->csum); + + skb->ip_summed = CHECKSUM_NONE; + + return 1; +} + +static int tcf_csum_ipv6_icmp(struct sk_buff *skb, struct ipv6hdr *ip6h, + unsigned int ihl, unsigned int ipl) +{ + struct icmp6hdr *icmp6h; + + icmp6h = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*icmp6h)); + if (icmp6h == NULL) + return 0; + + icmp6h->icmp6_cksum = 0; + skb->csum = csum_partial(icmp6h, ipl - ihl, 0); + icmp6h->icmp6_cksum = csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, + ipl - ihl, IPPROTO_ICMPV6, + skb->csum); + + skb->ip_summed = CHECKSUM_NONE; + + return 1; +} + +static int tcf_csum_ipv4_tcp(struct sk_buff *skb, struct iphdr *iph, + unsigned int ihl, unsigned int ipl) +{ + struct tcphdr *tcph; + + tcph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*tcph)); + if (tcph == NULL) + return 0; + + tcph->check = 0; + skb->csum = csum_partial(tcph, ipl - ihl, 0); + tcph->check = tcp_v4_check(ipl - ihl, + iph->saddr, iph->daddr, skb->csum); + + skb->ip_summed = CHECKSUM_NONE; + + return 1; +} + +static int tcf_csum_ipv6_tcp(struct sk_buff *skb, struct ipv6hdr *ip6h, + unsigned int ihl, unsigned int ipl) +{ + struct tcphdr *tcph; + + tcph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*tcph)); + if (tcph == NULL) + return 0; + + tcph->check = 0; + skb->csum = csum_partial(tcph, ipl - ihl, 0); + tcph->check = csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, + ipl - ihl, IPPROTO_TCP, + skb->csum); + + skb->ip_summed = CHECKSUM_NONE; + + return 1; +} + +static int tcf_csum_ipv4_udp(struct sk_buff *skb, struct iphdr *iph, + unsigned int ihl, unsigned int ipl, int udplite) +{ + struct udphdr *udph; + u16 ul; + + /* + * Support both UDP and UDPLITE checksum algorithms, Don't use + * udph->len to get the real length without any protocol check, + * UDPLITE uses udph->len for another thing, + * Use iph->tot_len, or just ipl. + */ + + udph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*udph)); + if (udph == NULL) + return 0; + + ul = ntohs(udph->len); + + if (udplite || udph->check) { + + udph->check = 0; + + if (udplite) { + if (ul == 0) + skb->csum = csum_partial(udph, ipl - ihl, 0); + else if ((ul >= sizeof(*udph)) && (ul <= ipl - ihl)) + skb->csum = csum_partial(udph, ul, 0); + else + goto ignore_obscure_skb; + } else { + if (ul != ipl - ihl) + goto ignore_obscure_skb; + + skb->csum = csum_partial(udph, ul, 0); + } + + udph->check = csum_tcpudp_magic(iph->saddr, iph->daddr, + ul, iph->protocol, + skb->csum); + + if (!udph->check) + udph->check = CSUM_MANGLED_0; + } + + skb->ip_summed = CHECKSUM_NONE; + +ignore_obscure_skb: + return 1; +} + +static int tcf_csum_ipv6_udp(struct sk_buff *skb, struct ipv6hdr *ip6h, + unsigned int ihl, unsigned int ipl, int udplite) +{ + struct udphdr *udph; + u16 ul; + + /* + * Support both UDP and UDPLITE checksum algorithms, Don't use + * udph->len to get the real length without any protocol check, + * UDPLITE uses udph->len for another thing, + * Use ip6h->payload_len + sizeof(*ip6h) ... , or just ipl. + */ + + udph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*udph)); + if (udph == NULL) + return 0; + + ul = ntohs(udph->len); + + udph->check = 0; + + if (udplite) { + if (ul == 0) + skb->csum = csum_partial(udph, ipl - ihl, 0); + + else if ((ul >= sizeof(*udph)) && (ul <= ipl - ihl)) + skb->csum = csum_partial(udph, ul, 0); + + else + goto ignore_obscure_skb; + } else { + if (ul != ipl - ihl) + goto ignore_obscure_skb; + + skb->csum = csum_partial(udph, ul, 0); + } + + udph->check = csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, ul, + udplite ? IPPROTO_UDPLITE : IPPROTO_UDP, + skb->csum); + + if (!udph->check) + udph->check = CSUM_MANGLED_0; + + skb->ip_summed = CHECKSUM_NONE; + +ignore_obscure_skb: + return 1; +} + +static int tcf_csum_ipv4(struct sk_buff *skb, u32 update_flags) +{ + struct iphdr *iph; + int ntkoff; + + ntkoff = skb_network_offset(skb); + + if (!pskb_may_pull(skb, sizeof(*iph) + ntkoff)) + goto fail; + + iph = ip_hdr(skb); + + switch (iph->frag_off & htons(IP_OFFSET) ? 0 : iph->protocol) { + case IPPROTO_ICMP: + if (update_flags & TCA_CSUM_UPDATE_FLAG_ICMP) + if (!tcf_csum_ipv4_icmp(skb, iph->ihl * 4, + ntohs(iph->tot_len))) + goto fail; + break; + case IPPROTO_IGMP: + if (update_flags & TCA_CSUM_UPDATE_FLAG_IGMP) + if (!tcf_csum_ipv4_igmp(skb, iph->ihl * 4, + ntohs(iph->tot_len))) + goto fail; + break; + case IPPROTO_TCP: + if (update_flags & TCA_CSUM_UPDATE_FLAG_TCP) + if (!tcf_csum_ipv4_tcp(skb, iph, iph->ihl * 4, + ntohs(iph->tot_len))) + goto fail; + break; + case IPPROTO_UDP: + if (update_flags & TCA_CSUM_UPDATE_FLAG_UDP) + if (!tcf_csum_ipv4_udp(skb, iph, iph->ihl * 4, + ntohs(iph->tot_len), 0)) + goto fail; + break; + case IPPROTO_UDPLITE: + if (update_flags & TCA_CSUM_UPDATE_FLAG_UDPLITE) + if (!tcf_csum_ipv4_udp(skb, iph, iph->ihl * 4, + ntohs(iph->tot_len), 1)) + goto fail; + break; + } + + if (update_flags & TCA_CSUM_UPDATE_FLAG_IPV4HDR) { + if (skb_cloned(skb) && + !skb_clone_writable(skb, sizeof(*iph) + ntkoff) && + pskb_expand_head(skb, 0, 0, GFP_ATOMIC)) + goto fail; + + ip_send_check(iph); + } + + return 1; + +fail: + return 0; +} + +static int tcf_csum_ipv6_hopopts(struct ipv6_opt_hdr *ip6xh, + unsigned int ixhl, unsigned int *pl) +{ + int off, len, optlen; + unsigned char *xh = (void *)ip6xh; + + off = sizeof(*ip6xh); + len = ixhl - off; + + while (len > 1) { + switch (xh[off]) { + case IPV6_TLV_PAD0: + optlen = 1; + break; + case IPV6_TLV_JUMBO: + optlen = xh[off + 1] + 2; + if (optlen != 6 || len < 6 || (off & 3) != 2) + /* wrong jumbo option length/alignment */ + return 0; + *pl = ntohl(*(__be32 *)(xh + off + 2)); + goto done; + default: + optlen = xh[off + 1] + 2; + if (optlen > len) + /* ignore obscure options */ + goto done; + break; + } + off += optlen; + len -= optlen; + } + +done: + return 1; +} + +static int tcf_csum_ipv6(struct sk_buff *skb, u32 update_flags) +{ + struct ipv6hdr *ip6h; + struct ipv6_opt_hdr *ip6xh; + unsigned int hl, ixhl; + unsigned int pl; + int ntkoff; + u8 nexthdr; + + ntkoff = skb_network_offset(skb); + + hl = sizeof(*ip6h); + + if (!pskb_may_pull(skb, hl + ntkoff)) + goto fail; + + ip6h = ipv6_hdr(skb); + + pl = ntohs(ip6h->payload_len); + nexthdr = ip6h->nexthdr; + + do { + switch (nexthdr) { + case NEXTHDR_FRAGMENT: + goto ignore_skb; + case NEXTHDR_ROUTING: + case NEXTHDR_HOP: + case NEXTHDR_DEST: + if (!pskb_may_pull(skb, hl + sizeof(*ip6xh) + ntkoff)) + goto fail; + ip6xh = (void *)(skb_network_header(skb) + hl); + ixhl = ipv6_optlen(ip6xh); + if (!pskb_may_pull(skb, hl + ixhl + ntkoff)) + goto fail; + if ((nexthdr == NEXTHDR_HOP) && + !(tcf_csum_ipv6_hopopts(ip6xh, ixhl, &pl))) + goto fail; + nexthdr = ip6xh->nexthdr; + hl += ixhl; + break; + case IPPROTO_ICMPV6: + if (update_flags & TCA_CSUM_UPDATE_FLAG_ICMP) + if (!tcf_csum_ipv6_icmp(skb, ip6h, + hl, pl + sizeof(*ip6h))) + goto fail; + goto done; + case IPPROTO_TCP: + if (update_flags & TCA_CSUM_UPDATE_FLAG_TCP) + if (!tcf_csum_ipv6_tcp(skb, ip6h, + hl, pl + sizeof(*ip6h))) + goto fail; + goto done; + case IPPROTO_UDP: + if (update_flags & TCA_CSUM_UPDATE_FLAG_UDP) + if (!tcf_csum_ipv6_udp(skb, ip6h, hl, + pl + sizeof(*ip6h), 0)) + goto fail; + goto done; + case IPPROTO_UDPLITE: + if (update_flags & TCA_CSUM_UPDATE_FLAG_UDPLITE) + if (!tcf_csum_ipv6_udp(skb, ip6h, hl, + pl + sizeof(*ip6h), 1)) + goto fail; + goto done; + default: + goto ignore_skb; + } + } while (pskb_may_pull(skb, hl + 1 + ntkoff)); + +done: +ignore_skb: + return 1; + +fail: + return 0; +} + +static int tcf_csum(struct sk_buff *skb, + struct tc_action *a, struct tcf_result *res) +{ + struct tcf_csum *p = a->priv; + int action; + u32 update_flags; + + spin_lock(&p->tcf_lock); + p->tcf_tm.lastuse = jiffies; + p->tcf_bstats.bytes += qdisc_pkt_len(skb); + p->tcf_bstats.packets++; + action = p->tcf_action; + update_flags = p->update_flags; + spin_unlock(&p->tcf_lock); + + if (unlikely(action == TC_ACT_SHOT)) + goto drop; + + switch (skb->protocol) { + case cpu_to_be16(ETH_P_IP): + if (!tcf_csum_ipv4(skb, update_flags)) + goto drop; + break; + case cpu_to_be16(ETH_P_IPV6): + if (!tcf_csum_ipv6(skb, update_flags)) + goto drop; + break; + } + + return action; + +drop: + spin_lock(&p->tcf_lock); + p->tcf_qstats.drops++; + spin_unlock(&p->tcf_lock); + return TC_ACT_SHOT; +} + +static int tcf_csum_dump(struct sk_buff *skb, + struct tc_action *a, int bind, int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_csum *p = a->priv; + struct tc_csum opt = { + .update_flags = p->update_flags, + .index = p->tcf_index, + .action = p->tcf_action, + .refcnt = p->tcf_refcnt - ref, + .bindcnt = p->tcf_bindcnt - bind, + }; + struct tcf_t t; + + NLA_PUT(skb, TCA_CSUM_PARMS, sizeof(opt), &opt); + t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install); + t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse); + t.expires = jiffies_to_clock_t(p->tcf_tm.expires); + NLA_PUT(skb, TCA_CSUM_TM, sizeof(t), &t); + + return skb->len; + +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static struct tc_action_ops act_csum_ops = { + .kind = "csum", + .hinfo = &csum_hash_info, + .type = TCA_ACT_CSUM, + .capab = TCA_CAP_NONE, + .owner = THIS_MODULE, + .act = tcf_csum, + .dump = tcf_csum_dump, + .cleanup = tcf_csum_cleanup, + .lookup = tcf_hash_search, + .init = tcf_csum_init, + .walk = tcf_generic_walker +}; + +MODULE_DESCRIPTION("Checksum updating actions"); +MODULE_LICENSE("GPL"); + +static int __init csum_init_module(void) +{ + return tcf_register_action(&act_csum_ops); +} + +static void __exit csum_cleanup_module(void) +{ + tcf_unregister_action(&act_csum_ops); +} + +module_init(csum_init_module); +module_exit(csum_cleanup_module); diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c index 8406c665499..c2ed90a4c0b 100644 --- a/net/sched/act_gact.c +++ b/net/sched/act_gact.c @@ -152,21 +152,24 @@ static int tcf_gact(struct sk_buff *skb, struct tc_action *a, struct tcf_result static int tcf_gact_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); - struct tc_gact opt; struct tcf_gact *gact = a->priv; + struct tc_gact opt = { + .index = gact->tcf_index, + .refcnt = gact->tcf_refcnt - ref, + .bindcnt = gact->tcf_bindcnt - bind, + .action = gact->tcf_action, + }; struct tcf_t t; - opt.index = gact->tcf_index; - opt.refcnt = gact->tcf_refcnt - ref; - opt.bindcnt = gact->tcf_bindcnt - bind; - opt.action = gact->tcf_action; NLA_PUT(skb, TCA_GACT_PARMS, sizeof(opt), &opt); #ifdef CONFIG_GACT_PROB if (gact->tcfg_ptype) { - struct tc_gact_p p_opt; - p_opt.paction = gact->tcfg_paction; - p_opt.pval = gact->tcfg_pval; - p_opt.ptype = gact->tcfg_ptype; + struct tc_gact_p p_opt = { + .paction = gact->tcfg_paction, + .pval = gact->tcfg_pval, + .ptype = gact->tcfg_ptype, + }; + NLA_PUT(skb, TCA_GACT_PROB, sizeof(p_opt), &p_opt); } #endif diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index c7e59e6ec34..8daef963225 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -39,7 +39,7 @@ static struct tcf_hashinfo ipt_hash_info = { .lock = &ipt_lock, }; -static int ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int hook) +static int ipt_init_target(struct xt_entry_target *t, char *table, unsigned int hook) { struct xt_tgchk_param par; struct xt_target *target; @@ -66,7 +66,7 @@ static int ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int return 0; } -static void ipt_destroy_target(struct ipt_entry_target *t) +static void ipt_destroy_target(struct xt_entry_target *t) { struct xt_tgdtor_param par = { .target = t->u.kernel.target, @@ -99,7 +99,7 @@ static const struct nla_policy ipt_policy[TCA_IPT_MAX + 1] = { [TCA_IPT_TABLE] = { .type = NLA_STRING, .len = IFNAMSIZ }, [TCA_IPT_HOOK] = { .type = NLA_U32 }, [TCA_IPT_INDEX] = { .type = NLA_U32 }, - [TCA_IPT_TARG] = { .len = sizeof(struct ipt_entry_target) }, + [TCA_IPT_TARG] = { .len = sizeof(struct xt_entry_target) }, }; static int tcf_ipt_init(struct nlattr *nla, struct nlattr *est, @@ -108,7 +108,7 @@ static int tcf_ipt_init(struct nlattr *nla, struct nlattr *est, struct nlattr *tb[TCA_IPT_MAX + 1]; struct tcf_ipt *ipt; struct tcf_common *pc; - struct ipt_entry_target *td, *t; + struct xt_entry_target *td, *t; char *tname; int ret = 0, err; u32 hook = 0; @@ -126,7 +126,7 @@ static int tcf_ipt_init(struct nlattr *nla, struct nlattr *est, if (tb[TCA_IPT_TARG] == NULL) return -EINVAL; - td = (struct ipt_entry_target *)nla_data(tb[TCA_IPT_TARG]); + td = (struct xt_entry_target *)nla_data(tb[TCA_IPT_TARG]); if (nla_len(tb[TCA_IPT_TARG]) < td->u.target_size) return -EINVAL; @@ -230,7 +230,7 @@ static int tcf_ipt(struct sk_buff *skb, struct tc_action *a, result = TC_ACT_SHOT; ipt->tcf_qstats.drops++; break; - case IPT_CONTINUE: + case XT_CONTINUE: result = TC_ACT_PIPE; break; default: @@ -249,7 +249,7 @@ static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, int { unsigned char *b = skb_tail_pointer(skb); struct tcf_ipt *ipt = a->priv; - struct ipt_entry_target *t; + struct xt_entry_target *t; struct tcf_t tm; struct tc_cnt c; diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index 11f195af2da..0c311be9282 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -219,15 +219,16 @@ static int tcf_mirred_dump(struct sk_buff *skb, struct tc_action *a, int bind, i { unsigned char *b = skb_tail_pointer(skb); struct tcf_mirred *m = a->priv; - struct tc_mirred opt; + struct tc_mirred opt = { + .index = m->tcf_index, + .action = m->tcf_action, + .refcnt = m->tcf_refcnt - ref, + .bindcnt = m->tcf_bindcnt - bind, + .eaction = m->tcfm_eaction, + .ifindex = m->tcfm_ifindex, + }; struct tcf_t t; - opt.index = m->tcf_index; - opt.action = m->tcf_action; - opt.refcnt = m->tcf_refcnt - ref; - opt.bindcnt = m->tcf_bindcnt - bind; - opt.eaction = m->tcfm_eaction; - opt.ifindex = m->tcfm_ifindex; NLA_PUT(skb, TCA_MIRRED_PARMS, sizeof(opt), &opt); t.install = jiffies_to_clock_t(jiffies - m->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - m->tcf_tm.lastuse); diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c index 509a2d53a99..186eb837e60 100644 --- a/net/sched/act_nat.c +++ b/net/sched/act_nat.c @@ -272,19 +272,19 @@ static int tcf_nat_dump(struct sk_buff *skb, struct tc_action *a, { unsigned char *b = skb_tail_pointer(skb); struct tcf_nat *p = a->priv; - struct tc_nat opt; + struct tc_nat opt = { + .old_addr = p->old_addr, + .new_addr = p->new_addr, + .mask = p->mask, + .flags = p->flags, + + .index = p->tcf_index, + .action = p->tcf_action, + .refcnt = p->tcf_refcnt - ref, + .bindcnt = p->tcf_bindcnt - bind, + }; struct tcf_t t; - opt.old_addr = p->old_addr; - opt.new_addr = p->new_addr; - opt.mask = p->mask; - opt.flags = p->flags; - - opt.index = p->tcf_index; - opt.action = p->tcf_action; - opt.refcnt = p->tcf_refcnt - ref; - opt.bindcnt = p->tcf_bindcnt - bind; - NLA_PUT(skb, TCA_NAT_PARMS, sizeof(opt), &opt); t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse); diff --git a/net/sched/act_police.c b/net/sched/act_police.c index 537a48732e9..7ebf7439b47 100644 --- a/net/sched/act_police.c +++ b/net/sched/act_police.c @@ -350,22 +350,19 @@ tcf_act_police_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); struct tcf_police *police = a->priv; - struct tc_police opt; - - opt.index = police->tcf_index; - opt.action = police->tcf_action; - opt.mtu = police->tcfp_mtu; - opt.burst = police->tcfp_burst; - opt.refcnt = police->tcf_refcnt - ref; - opt.bindcnt = police->tcf_bindcnt - bind; + struct tc_police opt = { + .index = police->tcf_index, + .action = police->tcf_action, + .mtu = police->tcfp_mtu, + .burst = police->tcfp_burst, + .refcnt = police->tcf_refcnt - ref, + .bindcnt = police->tcf_bindcnt - bind, + }; + if (police->tcfp_R_tab) opt.rate = police->tcfp_R_tab->rate; - else - memset(&opt.rate, 0, sizeof(opt.rate)); if (police->tcfp_P_tab) opt.peakrate = police->tcfp_P_tab->rate; - else - memset(&opt.peakrate, 0, sizeof(opt.peakrate)); NLA_PUT(skb, TCA_POLICE_TBF, sizeof(opt), &opt); if (police->tcfp_result) NLA_PUT_U32(skb, TCA_POLICE_RESULT, police->tcfp_result); diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c index 4a1d640b0cf..97e84f3ee77 100644 --- a/net/sched/act_simple.c +++ b/net/sched/act_simple.c @@ -164,13 +164,14 @@ static inline int tcf_simp_dump(struct sk_buff *skb, struct tc_action *a, { unsigned char *b = skb_tail_pointer(skb); struct tcf_defact *d = a->priv; - struct tc_defact opt; + struct tc_defact opt = { + .index = d->tcf_index, + .refcnt = d->tcf_refcnt - ref, + .bindcnt = d->tcf_bindcnt - bind, + .action = d->tcf_action, + }; struct tcf_t t; - opt.index = d->tcf_index; - opt.refcnt = d->tcf_refcnt - ref; - opt.bindcnt = d->tcf_bindcnt - bind; - opt.action = d->tcf_action; NLA_PUT(skb, TCA_DEF_PARMS, sizeof(opt), &opt); NLA_PUT_STRING(skb, TCA_DEF_DATA, d->tcfd_defdata); t.install = jiffies_to_clock_t(jiffies - d->tcf_tm.install); diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c index e9607fe55b5..66cbf4eb885 100644 --- a/net/sched/act_skbedit.c +++ b/net/sched/act_skbedit.c @@ -159,13 +159,14 @@ static inline int tcf_skbedit_dump(struct sk_buff *skb, struct tc_action *a, { unsigned char *b = skb_tail_pointer(skb); struct tcf_skbedit *d = a->priv; - struct tc_skbedit opt; + struct tc_skbedit opt = { + .index = d->tcf_index, + .refcnt = d->tcf_refcnt - ref, + .bindcnt = d->tcf_bindcnt - bind, + .action = d->tcf_action, + }; struct tcf_t t; - opt.index = d->tcf_index; - opt.refcnt = d->tcf_refcnt - ref; - opt.bindcnt = d->tcf_bindcnt - bind; - opt.action = d->tcf_action; NLA_PUT(skb, TCA_SKBEDIT_PARMS, sizeof(opt), &opt); if (d->flags & SKBEDIT_F_PRIORITY) NLA_PUT(skb, TCA_SKBEDIT_PRIORITY, sizeof(d->priority), diff --git a/net/sched/cls_cgroup.c b/net/sched/cls_cgroup.c index 78ef2c5e130..37dff78e9cb 100644 --- a/net/sched/cls_cgroup.c +++ b/net/sched/cls_cgroup.c @@ -123,7 +123,7 @@ static int cls_cgroup_classify(struct sk_buff *skb, struct tcf_proto *tp, * calls by looking at the number of nested bh disable calls because * softirqs always disables bh. */ - if (softirq_count() != SOFTIRQ_OFFSET) { + if (in_serving_softirq()) { /* If there is an sk_classid we'll use that. */ if (!skb->sk) return -1; diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c index e17096e3913..5b271a18bc3 100644 --- a/net/sched/cls_flow.c +++ b/net/sched/cls_flow.c @@ -111,44 +111,41 @@ static u32 flow_get_proto(struct sk_buff *skb) } } -static int has_ports(u8 protocol) -{ - switch (protocol) { - case IPPROTO_TCP: - case IPPROTO_UDP: - case IPPROTO_UDPLITE: - case IPPROTO_SCTP: - case IPPROTO_DCCP: - case IPPROTO_ESP: - return 1; - default: - return 0; - } -} - static u32 flow_get_proto_src(struct sk_buff *skb) { switch (skb->protocol) { case htons(ETH_P_IP): { struct iphdr *iph; + int poff; if (!pskb_network_may_pull(skb, sizeof(*iph))) break; iph = ip_hdr(skb); - if (!(iph->frag_off&htons(IP_MF|IP_OFFSET)) && - has_ports(iph->protocol) && - pskb_network_may_pull(skb, iph->ihl * 4 + 2)) - return ntohs(*(__be16 *)((void *)iph + iph->ihl * 4)); + if (iph->frag_off & htons(IP_MF|IP_OFFSET)) + break; + poff = proto_ports_offset(iph->protocol); + if (poff >= 0 && + pskb_network_may_pull(skb, iph->ihl * 4 + 2 + poff)) { + iph = ip_hdr(skb); + return ntohs(*(__be16 *)((void *)iph + iph->ihl * 4 + + poff)); + } break; } case htons(ETH_P_IPV6): { struct ipv6hdr *iph; + int poff; - if (!pskb_network_may_pull(skb, sizeof(*iph) + 2)) + if (!pskb_network_may_pull(skb, sizeof(*iph))) break; iph = ipv6_hdr(skb); - if (has_ports(iph->nexthdr)) - return ntohs(*(__be16 *)&iph[1]); + poff = proto_ports_offset(iph->nexthdr); + if (poff >= 0 && + pskb_network_may_pull(skb, sizeof(*iph) + poff + 2)) { + iph = ipv6_hdr(skb); + return ntohs(*(__be16 *)((void *)iph + sizeof(*iph) + + poff)); + } break; } } @@ -161,24 +158,36 @@ static u32 flow_get_proto_dst(struct sk_buff *skb) switch (skb->protocol) { case htons(ETH_P_IP): { struct iphdr *iph; + int poff; if (!pskb_network_may_pull(skb, sizeof(*iph))) break; iph = ip_hdr(skb); - if (!(iph->frag_off&htons(IP_MF|IP_OFFSET)) && - has_ports(iph->protocol) && - pskb_network_may_pull(skb, iph->ihl * 4 + 4)) - return ntohs(*(__be16 *)((void *)iph + iph->ihl * 4 + 2)); + if (iph->frag_off & htons(IP_MF|IP_OFFSET)) + break; + poff = proto_ports_offset(iph->protocol); + if (poff >= 0 && + pskb_network_may_pull(skb, iph->ihl * 4 + 4 + poff)) { + iph = ip_hdr(skb); + return ntohs(*(__be16 *)((void *)iph + iph->ihl * 4 + + 2 + poff)); + } break; } case htons(ETH_P_IPV6): { struct ipv6hdr *iph; + int poff; - if (!pskb_network_may_pull(skb, sizeof(*iph) + 4)) + if (!pskb_network_may_pull(skb, sizeof(*iph))) break; iph = ipv6_hdr(skb); - if (has_ports(iph->nexthdr)) - return ntohs(*(__be16 *)((void *)&iph[1] + 2)); + poff = proto_ports_offset(iph->nexthdr); + if (poff >= 0 && + pskb_network_may_pull(skb, sizeof(*iph) + poff + 4)) { + iph = ipv6_hdr(skb); + return ntohs(*(__be16 *)((void *)iph + sizeof(*iph) + + poff + 2)); + } break; } } @@ -297,6 +306,11 @@ static u32 flow_get_vlan_tag(const struct sk_buff *skb) return tag & VLAN_VID_MASK; } +static u32 flow_get_rxhash(struct sk_buff *skb) +{ + return skb_get_rxhash(skb); +} + static u32 flow_key_get(struct sk_buff *skb, int key) { switch (key) { @@ -334,6 +348,8 @@ static u32 flow_key_get(struct sk_buff *skb, int key) return flow_get_skgid(skb); case FLOW_KEY_VLAN_TAG: return flow_get_vlan_tag(skb); + case FLOW_KEY_RXHASH: + return flow_get_rxhash(skb); default: WARN_ON(1); return 0; diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c index 7416a5c73b2..b0c2a82178a 100644 --- a/net/sched/cls_u32.c +++ b/net/sched/cls_u32.c @@ -137,7 +137,7 @@ next_knode: int toff = off + key->off + (off2 & key->offmask); __be32 *data, _data; - if (skb_headroom(skb) + toff < 0) + if (skb_headroom(skb) + toff > INT_MAX) goto out; data = skb_header_pointer(skb, toff, 4, &_data); diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index 3bcac8aa333..34da5e29ea1 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -223,6 +223,11 @@ META_COLLECTOR(int_maclen) dst->value = skb->mac_len; } +META_COLLECTOR(int_rxhash) +{ + dst->value = skb_get_rxhash(skb); +} + /************************************************************************** * Netfilter **************************************************************************/ @@ -541,6 +546,7 @@ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = { [META_ID(SK_SENDMSG_OFF)] = META_FUNC(int_sk_sendmsg_off), [META_ID(SK_WRITE_PENDING)] = META_FUNC(int_sk_write_pend), [META_ID(VLAN_TAG)] = META_FUNC(int_vlan_tag), + [META_ID(RXHASH)] = META_FUNC(int_rxhash), } }; diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index b9e8c3b7d40..b22ca2d1ceb 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -150,22 +150,34 @@ int register_qdisc(struct Qdisc_ops *qops) if (qops->enqueue == NULL) qops->enqueue = noop_qdisc_ops.enqueue; if (qops->peek == NULL) { - if (qops->dequeue == NULL) { + if (qops->dequeue == NULL) qops->peek = noop_qdisc_ops.peek; - } else { - rc = -EINVAL; - goto out; - } + else + goto out_einval; } if (qops->dequeue == NULL) qops->dequeue = noop_qdisc_ops.dequeue; + if (qops->cl_ops) { + const struct Qdisc_class_ops *cops = qops->cl_ops; + + if (!(cops->get && cops->put && cops->walk && cops->leaf)) + goto out_einval; + + if (cops->tcf_chain && !(cops->bind_tcf && cops->unbind_tcf)) + goto out_einval; + } + qops->next = NULL; *qp = qops; rc = 0; out: write_unlock(&qdisc_mod_lock); return rc; + +out_einval: + rc = -EINVAL; + goto out; } EXPORT_SYMBOL(register_qdisc); @@ -228,7 +240,10 @@ struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle) if (q) goto out; - q = qdisc_match_from_root(dev->rx_queue.qdisc_sleeping, handle); + if (dev_ingress_queue(dev)) + q = qdisc_match_from_root( + dev_ingress_queue(dev)->qdisc_sleeping, + handle); out: return q; } @@ -348,7 +363,7 @@ static struct qdisc_size_table *qdisc_get_stab(struct nlattr *opt) tsize = nla_len(tb[TCA_STAB_DATA]) / sizeof(u16); } - if (!s || tsize != s->tsize || (!tab && tsize > 0)) + if (tsize != s->tsize || (!tab && tsize > 0)) return ERR_PTR(-EINVAL); spin_lock(&qdisc_stab_lock); @@ -678,6 +693,8 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent, (new && new->flags & TCQ_F_INGRESS)) { num_q = 1; ingress = 1; + if (!dev_ingress_queue(dev)) + return -ENOENT; } if (dev->flags & IFF_UP) @@ -689,7 +706,7 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent, } for (i = 0; i < num_q; i++) { - struct netdev_queue *dev_queue = &dev->rx_queue; + struct netdev_queue *dev_queue = dev_ingress_queue(dev); if (!ingress) dev_queue = netdev_get_tx_queue(dev, i); @@ -967,7 +984,8 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg) return -ENOENT; q = qdisc_leaf(p, clid); } else { /* ingress */ - q = dev->rx_queue.qdisc_sleeping; + if (dev_ingress_queue(dev)) + q = dev_ingress_queue(dev)->qdisc_sleeping; } } else { q = dev->qdisc; @@ -1031,8 +1049,9 @@ replay: if ((p = qdisc_lookup(dev, TC_H_MAJ(clid))) == NULL) return -ENOENT; q = qdisc_leaf(p, clid); - } else { /*ingress */ - q = dev->rx_queue.qdisc_sleeping; + } else { /* ingress */ + if (dev_ingress_queue_create(dev)) + q = dev_ingress_queue(dev)->qdisc_sleeping; } } else { q = dev->qdisc; @@ -1111,11 +1130,14 @@ replay: create_n_graft: if (!(n->nlmsg_flags&NLM_F_CREATE)) return -ENOENT; - if (clid == TC_H_INGRESS) - q = qdisc_create(dev, &dev->rx_queue, p, - tcm->tcm_parent, tcm->tcm_parent, - tca, &err); - else { + if (clid == TC_H_INGRESS) { + if (dev_ingress_queue(dev)) + q = qdisc_create(dev, dev_ingress_queue(dev), p, + tcm->tcm_parent, tcm->tcm_parent, + tca, &err); + else + err = -ENOENT; + } else { struct netdev_queue *dev_queue; if (p && p->ops->cl_ops && p->ops->cl_ops->select_queue) @@ -1292,8 +1314,10 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb) if (tc_dump_qdisc_root(dev->qdisc, skb, cb, &q_idx, s_q_idx) < 0) goto done; - dev_queue = &dev->rx_queue; - if (tc_dump_qdisc_root(dev_queue->qdisc_sleeping, skb, cb, &q_idx, s_q_idx) < 0) + dev_queue = dev_ingress_queue(dev); + if (dev_queue && + tc_dump_qdisc_root(dev_queue->qdisc_sleeping, skb, cb, + &q_idx, s_q_idx) < 0) goto done; cont: @@ -1583,8 +1607,10 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb) if (tc_dump_tclass_root(dev->qdisc, skb, tcm, cb, &t, s_t) < 0) goto done; - dev_queue = &dev->rx_queue; - if (tc_dump_tclass_root(dev_queue->qdisc_sleeping, skb, tcm, cb, &t, s_t) < 0) + dev_queue = dev_ingress_queue(dev); + if (dev_queue && + tc_dump_tclass_root(dev_queue->qdisc_sleeping, skb, tcm, cb, + &t, s_t) < 0) goto done; done: diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index e114f23d5ea..282540778aa 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -255,10 +255,6 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent, error = -EINVAL; goto err_out; } - if (!list_empty(&flow->list)) { - error = -EEXIST; - goto err_out; - } } else { int i; unsigned long cl; @@ -279,8 +275,7 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent, goto err_out; } flow->filter_list = NULL; - flow->q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, - &pfifo_qdisc_ops, classid); + flow->q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, classid); if (!flow->q) flow->q = &noop_qdisc; pr_debug("atm_tc_change: qdisc %p\n", flow->q); @@ -418,7 +413,7 @@ static int atm_tc_enqueue(struct sk_buff *skb, struct Qdisc *sch) } ret = qdisc_enqueue(skb, flow->q); - if (ret != 0) { + if (ret != NET_XMIT_SUCCESS) { drop: __maybe_unused if (net_xmit_drop_count(ret)) { sch->qstats.drops++; @@ -442,7 +437,7 @@ drop: __maybe_unused */ if (flow == &p->link) { sch->q.qlen++; - return 0; + return NET_XMIT_SUCCESS; } tasklet_schedule(&p->task); return NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; @@ -547,7 +542,7 @@ static int atm_tc_init(struct Qdisc *sch, struct nlattr *opt) INIT_LIST_HEAD(&p->flows); INIT_LIST_HEAD(&p->link.list); list_add(&p->link.list, &p->flows); - p->link.q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, + p->link.q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, sch->handle); if (!p->link.q) p->link.q = &noop_qdisc; diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index 28c01ef5abc..eb763159086 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -1379,9 +1379,9 @@ static int cbq_init(struct Qdisc *sch, struct nlattr *opt) q->link.sibling = &q->link; q->link.common.classid = sch->handle; q->link.qdisc = sch; - if (!(q->link.q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, - &pfifo_qdisc_ops, - sch->handle))) + q->link.q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + sch->handle); + if (!q->link.q) q->link.q = &noop_qdisc; q->link.priority = TC_CBQ_MAXPRIO-1; @@ -1623,7 +1623,7 @@ static int cbq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, struct cbq_class *cl = (struct cbq_class*)arg; if (new == NULL) { - new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, + new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, cl->common.classid); if (new == NULL) return -ENOBUFS; @@ -1874,8 +1874,8 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t cl->R_tab = rtab; rtab = NULL; cl->refcnt = 1; - if (!(cl->q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, - &pfifo_qdisc_ops, classid))) + cl->q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, classid); + if (!cl->q) cl->q = &noop_qdisc; cl->common.classid = classid; cl->tparent = parent; diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c index b74046a9539..aa8b5313f8c 100644 --- a/net/sched/sch_drr.c +++ b/net/sched/sch_drr.c @@ -110,7 +110,7 @@ static int drr_change_class(struct Qdisc *sch, u32 classid, u32 parentid, cl->refcnt = 1; cl->common.classid = classid; cl->quantum = quantum; - cl->qdisc = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, + cl->qdisc = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, classid); if (cl->qdisc == NULL) cl->qdisc = &noop_qdisc; @@ -218,7 +218,7 @@ static int drr_graft_class(struct Qdisc *sch, unsigned long arg, struct drr_class *cl = (struct drr_class *)arg; if (new == NULL) { - new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, + new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, cl->common.classid); if (new == NULL) new = &noop_qdisc; diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c index 63d41f86679..1d295d62bb5 100644 --- a/net/sched/sch_dsmark.c +++ b/net/sched/sch_dsmark.c @@ -61,8 +61,7 @@ static int dsmark_graft(struct Qdisc *sch, unsigned long arg, sch, p, new, old); if (new == NULL) { - new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, - &pfifo_qdisc_ops, + new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, sch->handle); if (new == NULL) new = &noop_qdisc; @@ -384,8 +383,7 @@ static int dsmark_init(struct Qdisc *sch, struct nlattr *opt) p->default_index = default_index; p->set_tc_index = nla_get_flag(tb[TCA_DSMARK_SET_TC_INDEX]); - p->q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, - &pfifo_qdisc_ops, sch->handle); + p->q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, sch->handle); if (p->q == NULL) p->q = &noop_qdisc; diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c index 5948bafa8ce..4dfecb0cba3 100644 --- a/net/sched/sch_fifo.c +++ b/net/sched/sch_fifo.c @@ -172,8 +172,7 @@ struct Qdisc *fifo_create_dflt(struct Qdisc *sch, struct Qdisc_ops *ops, struct Qdisc *q; int err = -ENOMEM; - q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, - ops, TC_H_MAKE(sch->handle, 1)); + q = qdisc_create_dflt(sch->dev_queue, ops, TC_H_MAKE(sch->handle, 1)); if (q) { err = fifo_set_limit(q, limit); if (err < 0) { diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 2aeb3a4386a..5dbb3cd96e5 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -383,6 +383,7 @@ struct Qdisc noop_qdisc = { .list = LIST_HEAD_INIT(noop_qdisc.list), .q.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock), .dev_queue = &noop_netdev_queue, + .busylock = __SPIN_LOCK_UNLOCKED(noop_qdisc.busylock), }; EXPORT_SYMBOL(noop_qdisc); @@ -409,6 +410,7 @@ static struct Qdisc noqueue_qdisc = { .list = LIST_HEAD_INIT(noqueue_qdisc.list), .q.lock = __SPIN_LOCK_UNLOCKED(noqueue_qdisc.q.lock), .dev_queue = &noqueue_netdev_queue, + .busylock = __SPIN_LOCK_UNLOCKED(noqueue_qdisc.busylock), }; @@ -574,10 +576,8 @@ errout: return ERR_PTR(err); } -struct Qdisc * qdisc_create_dflt(struct net_device *dev, - struct netdev_queue *dev_queue, - struct Qdisc_ops *ops, - unsigned int parentid) +struct Qdisc *qdisc_create_dflt(struct netdev_queue *dev_queue, + struct Qdisc_ops *ops, unsigned int parentid) { struct Qdisc *sch; @@ -682,7 +682,7 @@ static void attach_one_default_qdisc(struct net_device *dev, struct Qdisc *qdisc; if (dev->tx_queue_len) { - qdisc = qdisc_create_dflt(dev, dev_queue, + qdisc = qdisc_create_dflt(dev_queue, &pfifo_fast_ops, TC_H_ROOT); if (!qdisc) { printk(KERN_INFO "%s: activation failed\n", dev->name); @@ -709,7 +709,7 @@ static void attach_default_qdiscs(struct net_device *dev) dev->qdisc = txq->qdisc_sleeping; atomic_inc(&dev->qdisc->refcnt); } else { - qdisc = qdisc_create_dflt(dev, txq, &mq_qdisc_ops, TC_H_ROOT); + qdisc = qdisc_create_dflt(txq, &mq_qdisc_ops, TC_H_ROOT); if (qdisc) { qdisc->ops->attach(qdisc); dev->qdisc = qdisc; @@ -753,7 +753,8 @@ void dev_activate(struct net_device *dev) need_watchdog = 0; netdev_for_each_tx_queue(dev, transition_one_qdisc, &need_watchdog); - transition_one_qdisc(dev, &dev->rx_queue, NULL); + if (dev_ingress_queue(dev)) + transition_one_qdisc(dev, dev_ingress_queue(dev), NULL); if (need_watchdog) { dev->trans_start = jiffies; @@ -812,7 +813,8 @@ static bool some_qdisc_is_busy(struct net_device *dev) void dev_deactivate(struct net_device *dev) { netdev_for_each_tx_queue(dev, dev_deactivate_queue, &noop_qdisc); - dev_deactivate_queue(dev, &dev->rx_queue, &noop_qdisc); + if (dev_ingress_queue(dev)) + dev_deactivate_queue(dev, dev_ingress_queue(dev), &noop_qdisc); dev_watchdog_down(dev); @@ -838,7 +840,8 @@ void dev_init_scheduler(struct net_device *dev) { dev->qdisc = &noop_qdisc; netdev_for_each_tx_queue(dev, dev_init_scheduler_queue, &noop_qdisc); - dev_init_scheduler_queue(dev, &dev->rx_queue, &noop_qdisc); + if (dev_ingress_queue(dev)) + dev_init_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc); setup_timer(&dev->watchdog_timer, dev_watchdog, (unsigned long)dev); } @@ -861,7 +864,8 @@ static void shutdown_scheduler_queue(struct net_device *dev, void dev_shutdown(struct net_device *dev) { netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc); - shutdown_scheduler_queue(dev, &dev->rx_queue, &noop_qdisc); + if (dev_ingress_queue(dev)) + shutdown_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc); qdisc_destroy(dev->qdisc); dev->qdisc = &noop_qdisc; diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index abd904be428..069c62b7bb3 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -761,8 +761,8 @@ init_vf(struct hfsc_class *cl, unsigned int len) if (f != cl->cl_f) { cl->cl_f = f; cftree_update(cl); - update_cfmin(cl->cl_parent); } + update_cfmin(cl->cl_parent); } } @@ -1088,7 +1088,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid, cl->refcnt = 1; cl->sched = q; cl->cl_parent = parent; - cl->qdisc = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, + cl->qdisc = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, classid); if (cl->qdisc == NULL) cl->qdisc = &noop_qdisc; @@ -1209,8 +1209,7 @@ hfsc_graft_class(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, if (cl->level > 0) return -EINVAL; if (new == NULL) { - new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, - &pfifo_qdisc_ops, + new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, cl->cl_common.classid); if (new == NULL) new = &noop_qdisc; @@ -1452,8 +1451,7 @@ hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt) q->root.cl_common.classid = sch->handle; q->root.refcnt = 1; q->root.sched = q; - q->root.qdisc = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, - &pfifo_qdisc_ops, + q->root.qdisc = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, sch->handle); if (q->root.qdisc == NULL) q->root.qdisc = &noop_qdisc; diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 4be8d04b262..01b519d6c52 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -1121,8 +1121,7 @@ static int htb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, if (cl->level) return -EINVAL; if (new == NULL && - (new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, - &pfifo_qdisc_ops, + (new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, cl->common.classid)) == NULL) return -ENOBUFS; @@ -1247,8 +1246,7 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg) return -EBUSY; if (!cl->level && htb_parent_last_child(cl)) { - new_q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, - &pfifo_qdisc_ops, + new_q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, cl->parent->common.classid); last_child = 1; } @@ -1302,14 +1300,14 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, struct htb_class *cl = (struct htb_class *)*arg, *parent; struct nlattr *opt = tca[TCA_OPTIONS]; struct qdisc_rate_table *rtab = NULL, *ctab = NULL; - struct nlattr *tb[TCA_HTB_RTAB + 1]; + struct nlattr *tb[__TCA_HTB_MAX]; struct tc_htb_opt *hopt; /* extract all subattrs from opt attr */ if (!opt) goto failure; - err = nla_parse_nested(tb, TCA_HTB_RTAB, opt, htb_policy); + err = nla_parse_nested(tb, TCA_HTB_MAX, opt, htb_policy); if (err < 0) goto failure; @@ -1377,7 +1375,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, /* create leaf qdisc early because it uses kmalloc(GFP_KERNEL) so that can't be used inside of sch_tree_lock -- thanks to Karlis Peisenieks */ - new_q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, + new_q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, classid); sch_tree_lock(sch); if (parent && !parent->level) { diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c index fe91e50f9d9..ecc302f4d2a 100644 --- a/net/sched/sch_mq.c +++ b/net/sched/sch_mq.c @@ -56,7 +56,7 @@ static int mq_init(struct Qdisc *sch, struct nlattr *opt) for (ntx = 0; ntx < dev->num_tx_queues; ntx++) { dev_queue = netdev_get_tx_queue(dev, ntx); - qdisc = qdisc_create_dflt(dev, dev_queue, &pfifo_fast_ops, + qdisc = qdisc_create_dflt(dev_queue, &pfifo_fast_ops, TC_H_MAKE(TC_H_MAJ(sch->handle), TC_H_MIN(ntx + 1))); if (qdisc == NULL) diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c index 6ae251279fc..32690deab5d 100644 --- a/net/sched/sch_multiq.c +++ b/net/sched/sch_multiq.c @@ -227,8 +227,7 @@ static int multiq_tune(struct Qdisc *sch, struct nlattr *opt) for (i = 0; i < q->bands; i++) { if (q->queues[i] == &noop_qdisc) { struct Qdisc *child, *old; - child = qdisc_create_dflt(qdisc_dev(sch), - sch->dev_queue, + child = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, TC_H_MAKE(sch->handle, i + 1)); diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 4714ff162bb..e5593c083a7 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -538,8 +538,7 @@ static int netem_init(struct Qdisc *sch, struct nlattr *opt) qdisc_watchdog_init(&q->watchdog, sch); - q->qdisc = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, - &tfifo_qdisc_ops, + q->qdisc = qdisc_create_dflt(sch->dev_queue, &tfifo_qdisc_ops, TC_H_MAKE(sch->handle, 1)); if (!q->qdisc) { pr_debug("netem: qdisc create failed\n"); diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index 0748fb1e3a4..b1c95bce33c 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -200,7 +200,7 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt) for (i=0; i<q->bands; i++) { if (q->queues[i] == &noop_qdisc) { struct Qdisc *child, *old; - child = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, + child = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, TC_H_MAKE(sch->handle, i + 1)); if (child) { diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 534f33231c1..3cf478d012d 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -123,40 +123,39 @@ static unsigned sfq_hash(struct sfq_sched_data *q, struct sk_buff *skb) case htons(ETH_P_IP): { const struct iphdr *iph; + int poff; if (!pskb_network_may_pull(skb, sizeof(*iph))) goto err; iph = ip_hdr(skb); h = (__force u32)iph->daddr; h2 = (__force u32)iph->saddr ^ iph->protocol; - if (!(iph->frag_off&htons(IP_MF|IP_OFFSET)) && - (iph->protocol == IPPROTO_TCP || - iph->protocol == IPPROTO_UDP || - iph->protocol == IPPROTO_UDPLITE || - iph->protocol == IPPROTO_SCTP || - iph->protocol == IPPROTO_DCCP || - iph->protocol == IPPROTO_ESP) && - pskb_network_may_pull(skb, iph->ihl * 4 + 4)) - h2 ^= *(((u32*)iph) + iph->ihl); + if (iph->frag_off & htons(IP_MF|IP_OFFSET)) + break; + poff = proto_ports_offset(iph->protocol); + if (poff >= 0 && + pskb_network_may_pull(skb, iph->ihl * 4 + 4 + poff)) { + iph = ip_hdr(skb); + h2 ^= *(u32*)((void *)iph + iph->ihl * 4 + poff); + } break; } case htons(ETH_P_IPV6): { struct ipv6hdr *iph; + int poff; if (!pskb_network_may_pull(skb, sizeof(*iph))) goto err; iph = ipv6_hdr(skb); h = (__force u32)iph->daddr.s6_addr32[3]; h2 = (__force u32)iph->saddr.s6_addr32[3] ^ iph->nexthdr; - if ((iph->nexthdr == IPPROTO_TCP || - iph->nexthdr == IPPROTO_UDP || - iph->nexthdr == IPPROTO_UDPLITE || - iph->nexthdr == IPPROTO_SCTP || - iph->nexthdr == IPPROTO_DCCP || - iph->nexthdr == IPPROTO_ESP) && - pskb_network_may_pull(skb, sizeof(*iph) + 4)) - h2 ^= *(u32*)&iph[1]; + poff = proto_ports_offset(iph->nexthdr); + if (poff >= 0 && + pskb_network_may_pull(skb, sizeof(*iph) + 4 + poff)) { + iph = ipv6_hdr(skb); + h2 ^= *(u32*)((void *)iph + sizeof(*iph) + poff); + } break; } default: @@ -334,7 +333,7 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (++sch->q.qlen <= q->limit) { sch->bstats.bytes += qdisc_pkt_len(skb); sch->bstats.packets++; - return 0; + return NET_XMIT_SUCCESS; } sfq_drop(sch); @@ -508,6 +507,11 @@ nla_put_failure: return -1; } +static struct Qdisc *sfq_leaf(struct Qdisc *sch, unsigned long arg) +{ + return NULL; +} + static unsigned long sfq_get(struct Qdisc *sch, u32 classid) { return 0; @@ -519,6 +523,10 @@ static unsigned long sfq_bind(struct Qdisc *sch, unsigned long parent, return 0; } +static void sfq_put(struct Qdisc *q, unsigned long cl) +{ +} + static struct tcf_proto **sfq_find_tcf(struct Qdisc *sch, unsigned long cl) { struct sfq_sched_data *q = qdisc_priv(sch); @@ -571,9 +579,12 @@ static void sfq_walk(struct Qdisc *sch, struct qdisc_walker *arg) } static const struct Qdisc_class_ops sfq_class_ops = { + .leaf = sfq_leaf, .get = sfq_get, + .put = sfq_put, .tcf_chain = sfq_find_tcf, .bind_tcf = sfq_bind, + .unbind_tcf = sfq_put, .dump = sfq_dump_class, .dump_stats = sfq_dump_class_stats, .walk = sfq_walk, diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c index 0991c640cd3..641a30d6463 100644 --- a/net/sched/sch_tbf.c +++ b/net/sched/sch_tbf.c @@ -127,7 +127,7 @@ static int tbf_enqueue(struct sk_buff *skb, struct Qdisc* sch) return qdisc_reshape_fail(skb, sch); ret = qdisc_enqueue(skb, q->qdisc); - if (ret != 0) { + if (ret != NET_XMIT_SUCCESS) { if (net_xmit_drop_count(ret)) sch->qstats.drops++; return ret; @@ -136,7 +136,7 @@ static int tbf_enqueue(struct sk_buff *skb, struct Qdisc* sch) sch->q.qlen++; sch->bstats.bytes += qdisc_pkt_len(skb); sch->bstats.packets++; - return 0; + return NET_XMIT_SUCCESS; } static unsigned int tbf_drop(struct Qdisc* sch) diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c index 807643bdcba..401af959670 100644 --- a/net/sched/sch_teql.c +++ b/net/sched/sch_teql.c @@ -85,7 +85,7 @@ teql_enqueue(struct sk_buff *skb, struct Qdisc* sch) __skb_queue_tail(&q->q, skb); sch->bstats.bytes += qdisc_pkt_len(skb); sch->bstats.packets++; - return 0; + return NET_XMIT_SUCCESS; } kfree_skb(skb); @@ -241,11 +241,11 @@ __teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res, struct net_device * } if (neigh_event_send(n, skb_res) == 0) { int err; + char haddr[MAX_ADDR_LEN]; - read_lock(&n->lock); - err = dev_hard_header(skb, dev, ntohs(skb->protocol), - n->ha, NULL, skb->len); - read_unlock(&n->lock); + neigh_ha_snapshot(haddr, n, dev); + err = dev_hard_header(skb, dev, ntohs(skb->protocol), haddr, + NULL, skb->len); if (err < 0) { neigh_release(n); diff --git a/net/sctp/associola.c b/net/sctp/associola.c index 0b85e525643..5f1fb8bd862 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -48,6 +48,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/types.h> #include <linux/fcntl.h> #include <linux/poll.h> diff --git a/net/sctp/auth.c b/net/sctp/auth.c index 86366390038..ddbbf7c81fa 100644 --- a/net/sctp/auth.c +++ b/net/sctp/auth.c @@ -543,16 +543,20 @@ struct sctp_hmac *sctp_auth_asoc_get_hmac(const struct sctp_association *asoc) id = ntohs(hmacs->hmac_ids[i]); /* Check the id is in the supported range */ - if (id > SCTP_AUTH_HMAC_ID_MAX) + if (id > SCTP_AUTH_HMAC_ID_MAX) { + id = 0; continue; + } /* See is we support the id. Supported IDs have name and * length fields set, so that we can allocated and use * them. We can safely just check for name, for without the * name, we can't allocate the TFM. */ - if (!sctp_hmac_list[id].hmac_name) + if (!sctp_hmac_list[id].hmac_name) { + id = 0; continue; + } break; } diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c index 476caaf100e..6c8556459a7 100644 --- a/net/sctp/chunk.c +++ b/net/sctp/chunk.c @@ -37,6 +37,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/types.h> #include <linux/kernel.h> #include <linux/net.h> diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c index ccb6dc48d15..397296fb156 100644 --- a/net/sctp/inqueue.c +++ b/net/sctp/inqueue.c @@ -43,6 +43,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <net/sctp/sctp.h> #include <net/sctp/sm.h> #include <linux/interrupt.h> diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 732689140fb..95e0c8eda1a 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -47,6 +47,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/module.h> #include <linux/errno.h> #include <linux/types.h> @@ -336,7 +338,7 @@ static void sctp_v6_get_saddr(struct sctp_sock *sk, memcpy(saddr, baddr, sizeof(union sctp_addr)); SCTP_DEBUG_PRINTK("saddr: %pI6\n", &saddr->v6.sin6_addr); } else { - printk(KERN_ERR "%s: asoc:%p Could not find a valid source " + pr_err("%s: asoc:%p Could not find a valid source " "address for the dest:%pI6\n", __func__, asoc, &daddr->v6.sin6_addr); } diff --git a/net/sctp/objcnt.c b/net/sctp/objcnt.c index f73ec0ea93b..8ef8e7d9eb6 100644 --- a/net/sctp/objcnt.c +++ b/net/sctp/objcnt.c @@ -38,6 +38,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/kernel.h> #include <net/sctp/sctp.h> @@ -134,8 +136,7 @@ void sctp_dbg_objcnt_init(void) ent = proc_create("sctp_dbg_objcnt", 0, proc_net_sctp, &sctp_objcnt_ops); if (!ent) - printk(KERN_WARNING - "sctp_dbg_objcnt: Unable to create /proc entry.\n"); + pr_warn("sctp_dbg_objcnt: Unable to create /proc entry.\n"); } /* Cleanup the objcount entry in the proc filesystem. */ diff --git a/net/sctp/output.c b/net/sctp/output.c index a646681f5ac..60600d337a3 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -41,6 +41,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/types.h> #include <linux/kernel.h> #include <linux/wait.h> @@ -92,7 +94,6 @@ struct sctp_packet *sctp_packet_config(struct sctp_packet *packet, SCTP_DEBUG_PRINTK("%s: packet:%p vtag:0x%x\n", __func__, packet, vtag); - sctp_packet_reset(packet); packet->vtag = vtag; if (ecn_capable && sctp_packet_empty(packet)) { diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c index c04b2eb5918..8c6d379b4bb 100644 --- a/net/sctp/outqueue.c +++ b/net/sctp/outqueue.c @@ -46,6 +46,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/types.h> #include <linux/list.h> /* For struct list_head */ #include <linux/socket.h> @@ -1463,23 +1465,23 @@ static void sctp_check_transmitted(struct sctp_outq *q, /* Display the end of the * current range. */ - SCTP_DEBUG_PRINTK("-%08x", - dbg_last_ack_tsn); + SCTP_DEBUG_PRINTK_CONT("-%08x", + dbg_last_ack_tsn); } /* Start a new range. */ - SCTP_DEBUG_PRINTK(",%08x", tsn); + SCTP_DEBUG_PRINTK_CONT(",%08x", tsn); dbg_ack_tsn = tsn; break; case 1: /* The last TSN was NOT ACKed. */ if (dbg_last_kept_tsn != dbg_kept_tsn) { /* Display the end of current range. */ - SCTP_DEBUG_PRINTK("-%08x", - dbg_last_kept_tsn); + SCTP_DEBUG_PRINTK_CONT("-%08x", + dbg_last_kept_tsn); } - SCTP_DEBUG_PRINTK("\n"); + SCTP_DEBUG_PRINTK_CONT("\n"); /* FALL THROUGH... */ default: @@ -1526,18 +1528,18 @@ static void sctp_check_transmitted(struct sctp_outq *q, break; if (dbg_last_kept_tsn != dbg_kept_tsn) - SCTP_DEBUG_PRINTK("-%08x", - dbg_last_kept_tsn); + SCTP_DEBUG_PRINTK_CONT("-%08x", + dbg_last_kept_tsn); - SCTP_DEBUG_PRINTK(",%08x", tsn); + SCTP_DEBUG_PRINTK_CONT(",%08x", tsn); dbg_kept_tsn = tsn; break; case 0: if (dbg_last_ack_tsn != dbg_ack_tsn) - SCTP_DEBUG_PRINTK("-%08x", - dbg_last_ack_tsn); - SCTP_DEBUG_PRINTK("\n"); + SCTP_DEBUG_PRINTK_CONT("-%08x", + dbg_last_ack_tsn); + SCTP_DEBUG_PRINTK_CONT("\n"); /* FALL THROUGH... */ default: @@ -1556,17 +1558,17 @@ static void sctp_check_transmitted(struct sctp_outq *q, switch (dbg_prt_state) { case 0: if (dbg_last_ack_tsn != dbg_ack_tsn) { - SCTP_DEBUG_PRINTK("-%08x\n", dbg_last_ack_tsn); + SCTP_DEBUG_PRINTK_CONT("-%08x\n", dbg_last_ack_tsn); } else { - SCTP_DEBUG_PRINTK("\n"); + SCTP_DEBUG_PRINTK_CONT("\n"); } break; case 1: if (dbg_last_kept_tsn != dbg_kept_tsn) { - SCTP_DEBUG_PRINTK("-%08x\n", dbg_last_kept_tsn); + SCTP_DEBUG_PRINTK_CONT("-%08x\n", dbg_last_kept_tsn); } else { - SCTP_DEBUG_PRINTK("\n"); + SCTP_DEBUG_PRINTK_CONT("\n"); } } #endif /* SCTP_DEBUG */ diff --git a/net/sctp/probe.c b/net/sctp/probe.c index db3a42b8b34..bc6cd75cc1d 100644 --- a/net/sctp/probe.c +++ b/net/sctp/probe.c @@ -22,6 +22,8 @@ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/kernel.h> #include <linux/kprobes.h> #include <linux/socket.h> @@ -117,6 +119,7 @@ static const struct file_operations sctpprobe_fops = { .owner = THIS_MODULE, .open = sctpprobe_open, .read = sctpprobe_read, + .llseek = noop_llseek, }; sctp_disposition_t jsctp_sf_eat_sack(const struct sctp_endpoint *ep, @@ -192,7 +195,7 @@ static __init int sctpprobe_init(void) if (ret) goto remove_proc; - pr_info("SCTP probe registered (port=%d)\n", port); + pr_info("probe registered (port=%d)\n", port); return 0; diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 5027b83f1cc..1ef29c74d85 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -46,6 +46,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/module.h> #include <linux/init.h> #include <linux/netdevice.h> @@ -707,8 +709,7 @@ static int sctp_ctl_sock_init(void) &init_net); if (err < 0) { - printk(KERN_ERR - "SCTP: Failed to create the SCTP control socket.\n"); + pr_err("Failed to create the SCTP control socket\n"); return err; } return 0; @@ -798,7 +799,7 @@ static void sctp_inet_skb_msgname(struct sk_buff *skb, char *msgname, int *len) static int sctp_inet_af_supported(sa_family_t family, struct sctp_sock *sp) { /* PF_INET only supports AF_INET addresses. */ - return (AF_INET == family); + return AF_INET == family; } /* Address matching with wildcards allowed. */ @@ -1206,7 +1207,7 @@ SCTP_STATIC __init int sctp_init(void) __get_free_pages(GFP_ATOMIC, order); } while (!sctp_assoc_hashtable && --order > 0); if (!sctp_assoc_hashtable) { - printk(KERN_ERR "SCTP: Failed association hash alloc.\n"); + pr_err("Failed association hash alloc\n"); status = -ENOMEM; goto err_ahash_alloc; } @@ -1220,7 +1221,7 @@ SCTP_STATIC __init int sctp_init(void) sctp_ep_hashtable = (struct sctp_hashbucket *) kmalloc(64 * sizeof(struct sctp_hashbucket), GFP_KERNEL); if (!sctp_ep_hashtable) { - printk(KERN_ERR "SCTP: Failed endpoint_hash alloc.\n"); + pr_err("Failed endpoint_hash alloc\n"); status = -ENOMEM; goto err_ehash_alloc; } @@ -1239,7 +1240,7 @@ SCTP_STATIC __init int sctp_init(void) __get_free_pages(GFP_ATOMIC, order); } while (!sctp_port_hashtable && --order > 0); if (!sctp_port_hashtable) { - printk(KERN_ERR "SCTP: Failed bind hash alloc."); + pr_err("Failed bind hash alloc\n"); status = -ENOMEM; goto err_bhash_alloc; } @@ -1248,8 +1249,7 @@ SCTP_STATIC __init int sctp_init(void) INIT_HLIST_HEAD(&sctp_port_hashtable[i].chain); } - printk(KERN_INFO "SCTP: Hash tables configured " - "(established %d bind %d)\n", + pr_info("Hash tables configured (established %d bind %d)\n", sctp_assoc_hashsize, sctp_port_hashsize); /* Disable ADDIP by default. */ @@ -1290,8 +1290,7 @@ SCTP_STATIC __init int sctp_init(void) /* Initialize the control inode/socket for handling OOTB packets. */ if ((status = sctp_ctl_sock_init())) { - printk (KERN_ERR - "SCTP: Failed to initialize the SCTP control sock.\n"); + pr_err("Failed to initialize the SCTP control sock\n"); goto err_ctl_sock_init; } diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 246f9292465..2cc46f0962c 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -50,6 +50,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/types.h> #include <linux/kernel.h> #include <linux/ip.h> diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index f5e5e27cac5..b21b218d564 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -47,6 +47,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/skbuff.h> #include <linux/types.h> #include <linux/socket.h> @@ -1146,26 +1148,23 @@ static int sctp_side_effects(sctp_event_t event_type, sctp_subtype_t subtype, case SCTP_DISPOSITION_VIOLATION: if (net_ratelimit()) - printk(KERN_ERR "sctp protocol violation state %d " - "chunkid %d\n", state, subtype.chunk); + pr_err("protocol violation state %d chunkid %d\n", + state, subtype.chunk); break; case SCTP_DISPOSITION_NOT_IMPL: - printk(KERN_WARNING "sctp unimplemented feature in state %d, " - "event_type %d, event_id %d\n", - state, event_type, subtype.chunk); + pr_warn("unimplemented feature in state %d, event_type %d, event_id %d\n", + state, event_type, subtype.chunk); break; case SCTP_DISPOSITION_BUG: - printk(KERN_ERR "sctp bug in state %d, " - "event_type %d, event_id %d\n", + pr_err("bug in state %d, event_type %d, event_id %d\n", state, event_type, subtype.chunk); BUG(); break; default: - printk(KERN_ERR "sctp impossible disposition %d " - "in state %d, event_type %d, event_id %d\n", + pr_err("impossible disposition %d in state %d, event_type %d, event_id %d\n", status, state, event_type, subtype.chunk); BUG(); break; @@ -1679,8 +1678,8 @@ static int sctp_cmd_interpreter(sctp_event_t event_type, sctp_cmd_send_asconf(asoc); break; default: - printk(KERN_WARNING "Impossible command: %u, %p\n", - cmd->verb, cmd->obj.ptr); + pr_warn("Impossible command: %u, %p\n", + cmd->verb, cmd->obj.ptr); break; } diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 24b2cd55563..4b4eb7c96bb 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -50,6 +50,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/types.h> #include <linux/kernel.h> #include <linux/ip.h> @@ -1138,18 +1140,16 @@ sctp_disposition_t sctp_sf_backbeat_8_3(const struct sctp_endpoint *ep, if (unlikely(!link)) { if (from_addr.sa.sa_family == AF_INET6) { if (net_ratelimit()) - printk(KERN_WARNING - "%s association %p could not find address %pI6\n", - __func__, - asoc, - &from_addr.v6.sin6_addr); + pr_warn("%s association %p could not find address %pI6\n", + __func__, + asoc, + &from_addr.v6.sin6_addr); } else { if (net_ratelimit()) - printk(KERN_WARNING - "%s association %p could not find address %pI4\n", - __func__, - asoc, - &from_addr.v4.sin_addr.s_addr); + pr_warn("%s association %p could not find address %pI4\n", + __func__, + asoc, + &from_addr.v4.sin_addr.s_addr); } return SCTP_DISPOSITION_DISCARD; } @@ -1232,6 +1232,18 @@ out: return 0; } +static bool list_has_sctp_addr(const struct list_head *list, + union sctp_addr *ipaddr) +{ + struct sctp_transport *addr; + + list_for_each_entry(addr, list, transports) { + if (sctp_cmp_addr_exact(ipaddr, &addr->ipaddr)) + return true; + } + + return false; +} /* A restart is occurring, check to make sure no new addresses * are being added as we may be under a takeover attack. */ @@ -1240,10 +1252,10 @@ static int sctp_sf_check_restart_addrs(const struct sctp_association *new_asoc, struct sctp_chunk *init, sctp_cmd_seq_t *commands) { - struct sctp_transport *new_addr, *addr; - int found; + struct sctp_transport *new_addr; + int ret = 1; - /* Implementor's Guide - Sectin 5.2.2 + /* Implementor's Guide - Section 5.2.2 * ... * Before responding the endpoint MUST check to see if the * unexpected INIT adds new addresses to the association. If new @@ -1254,31 +1266,19 @@ static int sctp_sf_check_restart_addrs(const struct sctp_association *new_asoc, /* Search through all current addresses and make sure * we aren't adding any new ones. */ - new_addr = NULL; - found = 0; - list_for_each_entry(new_addr, &new_asoc->peer.transport_addr_list, - transports) { - found = 0; - list_for_each_entry(addr, &asoc->peer.transport_addr_list, - transports) { - if (sctp_cmp_addr_exact(&new_addr->ipaddr, - &addr->ipaddr)) { - found = 1; - break; - } - } - if (!found) + transports) { + if (!list_has_sctp_addr(&asoc->peer.transport_addr_list, + &new_addr->ipaddr)) { + sctp_sf_send_restart_abort(&new_addr->ipaddr, init, + commands); + ret = 0; break; - } - - /* If a new address was added, ABORT the sender. */ - if (!found && new_addr) { - sctp_sf_send_restart_abort(&new_addr->ipaddr, init, commands); + } } /* Return success if all addresses were found. */ - return found; + return ret; } /* Populate the verification/tie tags based on overlapping INIT diff --git a/net/sctp/sm_statetable.c b/net/sctp/sm_statetable.c index 6d9b3aafcc5..546d4387fb3 100644 --- a/net/sctp/sm_statetable.c +++ b/net/sctp/sm_statetable.c @@ -46,6 +46,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/skbuff.h> #include <net/sctp/sctp.h> #include <net/sctp/sm.h> @@ -66,15 +68,19 @@ static const sctp_sm_table_entry_t bug = { .name = "sctp_sf_bug" }; -#define DO_LOOKUP(_max, _type, _table) \ - if ((event_subtype._type > (_max))) { \ - printk(KERN_WARNING \ - "sctp table %p possible attack:" \ - " event %d exceeds max %d\n", \ - _table, event_subtype._type, _max); \ - return &bug; \ - } \ - return &_table[event_subtype._type][(int)state]; +#define DO_LOOKUP(_max, _type, _table) \ +({ \ + const sctp_sm_table_entry_t *rtn; \ + \ + if ((event_subtype._type > (_max))) { \ + pr_warn("table %p possible attack: event %d exceeds max %d\n", \ + _table, event_subtype._type, _max); \ + rtn = &bug; \ + } else \ + rtn = &_table[event_subtype._type][(int)state]; \ + \ + rtn; \ +}) const sctp_sm_table_entry_t *sctp_sm_lookup_event(sctp_event_t event_type, sctp_state_t state, @@ -83,21 +89,15 @@ const sctp_sm_table_entry_t *sctp_sm_lookup_event(sctp_event_t event_type, switch (event_type) { case SCTP_EVENT_T_CHUNK: return sctp_chunk_event_lookup(event_subtype.chunk, state); - break; case SCTP_EVENT_T_TIMEOUT: - DO_LOOKUP(SCTP_EVENT_TIMEOUT_MAX, timeout, - timeout_event_table); - break; - + return DO_LOOKUP(SCTP_EVENT_TIMEOUT_MAX, timeout, + timeout_event_table); case SCTP_EVENT_T_OTHER: - DO_LOOKUP(SCTP_EVENT_OTHER_MAX, other, other_event_table); - break; - + return DO_LOOKUP(SCTP_EVENT_OTHER_MAX, other, + other_event_table); case SCTP_EVENT_T_PRIMITIVE: - DO_LOOKUP(SCTP_EVENT_PRIMITIVE_MAX, primitive, - primitive_event_table); - break; - + return DO_LOOKUP(SCTP_EVENT_PRIMITIVE_MAX, primitive, + primitive_event_table); default: /* Yikes! We got an illegal event type. */ return &bug; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index ca44917872d..e34ca9cc116 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -57,6 +57,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/types.h> #include <linux/kernel.h> #include <linux/wait.h> @@ -916,6 +918,11 @@ SCTP_STATIC int sctp_setsockopt_bindx(struct sock* sk, /* Walk through the addrs buffer and count the number of addresses. */ addr_buf = kaddrs; while (walk_size < addrs_size) { + if (walk_size + sizeof(sa_family_t) > addrs_size) { + kfree(kaddrs); + return -EINVAL; + } + sa_addr = (struct sockaddr *)addr_buf; af = sctp_get_af_specific(sa_addr->sa_family); @@ -1002,9 +1009,13 @@ static int __sctp_connect(struct sock* sk, /* Walk through the addrs buffer and count the number of addresses. */ addr_buf = kaddrs; while (walk_size < addrs_size) { + if (walk_size + sizeof(sa_family_t) > addrs_size) { + err = -EINVAL; + goto out_free; + } + sa_addr = (union sctp_addr *)addr_buf; af = sctp_get_af_specific(sa_addr->sa.sa_family); - port = ntohs(sa_addr->v4.sin_port); /* If the address family is not supported or if this address * causes the address buffer to overflow return EINVAL. @@ -1014,6 +1025,8 @@ static int __sctp_connect(struct sock* sk, goto out_free; } + port = ntohs(sa_addr->v4.sin_port); + /* Save current address so we can work with it */ memcpy(&to, sa_addr, af->sockaddr_len); @@ -2458,9 +2471,8 @@ static int sctp_setsockopt_delayed_ack(struct sock *sk, if (params.sack_delay == 0 && params.sack_freq == 0) return 0; } else if (optlen == sizeof(struct sctp_assoc_value)) { - printk(KERN_WARNING "SCTP: Use of struct sctp_assoc_value " - "in delayed_ack socket option deprecated\n"); - printk(KERN_WARNING "SCTP: Use struct sctp_sack_info instead\n"); + pr_warn("Use of struct sctp_assoc_value in delayed_ack socket option deprecated\n"); + pr_warn("Use struct sctp_sack_info instead\n"); if (copy_from_user(¶ms, optval, optlen)) return -EFAULT; @@ -2868,10 +2880,8 @@ static int sctp_setsockopt_maxseg(struct sock *sk, char __user *optval, unsigned int val; if (optlen == sizeof(int)) { - printk(KERN_WARNING - "SCTP: Use of int in maxseg socket option deprecated\n"); - printk(KERN_WARNING - "SCTP: Use struct sctp_assoc_value instead\n"); + pr_warn("Use of int in maxseg socket option deprecated\n"); + pr_warn("Use struct sctp_assoc_value instead\n"); if (copy_from_user(&val, optval, optlen)) return -EFAULT; params.assoc_id = 0; @@ -3121,10 +3131,8 @@ static int sctp_setsockopt_maxburst(struct sock *sk, int assoc_id = 0; if (optlen == sizeof(int)) { - printk(KERN_WARNING - "SCTP: Use of int in max_burst socket option deprecated\n"); - printk(KERN_WARNING - "SCTP: Use struct sctp_assoc_value instead\n"); + pr_warn("Use of int in max_burst socket option deprecated\n"); + pr_warn("Use struct sctp_assoc_value instead\n"); if (copy_from_user(&val, optval, optlen)) return -EFAULT; } else if (optlen == sizeof(struct sctp_assoc_value)) { @@ -3595,7 +3603,40 @@ out: /* The SCTP ioctl handler. */ SCTP_STATIC int sctp_ioctl(struct sock *sk, int cmd, unsigned long arg) { - return -ENOIOCTLCMD; + int rc = -ENOTCONN; + + sctp_lock_sock(sk); + + /* + * SEQPACKET-style sockets in LISTENING state are valid, for + * SCTP, so only discard TCP-style sockets in LISTENING state. + */ + if (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING)) + goto out; + + switch (cmd) { + case SIOCINQ: { + struct sk_buff *skb; + unsigned int amount = 0; + + skb = skb_peek(&sk->sk_receive_queue); + if (skb != NULL) { + /* + * We will only return the amount of this packet since + * that is all that will be read. + */ + amount = skb->len; + } + rc = put_user(amount, (int __user *)arg); + break; + } + default: + rc = -ENOIOCTLCMD; + break; + } +out: + sctp_release_sock(sk); + return rc; } /* This is the function which gets called during socket creation to @@ -3854,7 +3895,7 @@ static int sctp_getsockopt_sctp_status(struct sock *sk, int len, } out: - return (retval); + return retval; } @@ -3910,7 +3951,7 @@ static int sctp_getsockopt_peer_addr_info(struct sock *sk, int len, } out: - return (retval); + return retval; } /* 7.1.12 Enable/Disable message fragmentation (SCTP_DISABLE_FRAGMENTS) @@ -4281,9 +4322,8 @@ static int sctp_getsockopt_delayed_ack(struct sock *sk, int len, if (copy_from_user(¶ms, optval, len)) return -EFAULT; } else if (len == sizeof(struct sctp_assoc_value)) { - printk(KERN_WARNING "SCTP: Use of struct sctp_assoc_value " - "in delayed_ack socket option deprecated\n"); - printk(KERN_WARNING "SCTP: Use struct sctp_sack_info instead\n"); + pr_warn("Use of struct sctp_assoc_value in delayed_ack socket option deprecated\n"); + pr_warn("Use struct sctp_sack_info instead\n"); if (copy_from_user(¶ms, optval, len)) return -EFAULT; } else @@ -4929,10 +4969,8 @@ static int sctp_getsockopt_maxseg(struct sock *sk, int len, struct sctp_association *asoc; if (len == sizeof(int)) { - printk(KERN_WARNING - "SCTP: Use of int in maxseg socket option deprecated\n"); - printk(KERN_WARNING - "SCTP: Use struct sctp_assoc_value instead\n"); + pr_warn("Use of int in maxseg socket option deprecated\n"); + pr_warn("Use struct sctp_assoc_value instead\n"); params.assoc_id = 0; } else if (len >= sizeof(struct sctp_assoc_value)) { len = sizeof(struct sctp_assoc_value); @@ -5023,10 +5061,8 @@ static int sctp_getsockopt_maxburst(struct sock *sk, int len, struct sctp_association *asoc; if (len == sizeof(int)) { - printk(KERN_WARNING - "SCTP: Use of int in max_burst socket option deprecated\n"); - printk(KERN_WARNING - "SCTP: Use struct sctp_assoc_value instead\n"); + pr_warn("Use of int in max_burst socket option deprecated\n"); + pr_warn("Use struct sctp_assoc_value instead\n"); params.assoc_id = 0; } else if (len >= sizeof(struct sctp_assoc_value)) { len = sizeof(struct sctp_assoc_value); @@ -5569,7 +5605,7 @@ static int sctp_get_port(struct sock *sk, unsigned short snum) /* Note: sk->sk_num gets filled in if ephemeral port request. */ ret = sctp_get_port_local(sk, &addr); - return (ret ? 1 : 0); + return ret ? 1 : 0; } /* @@ -5586,8 +5622,7 @@ SCTP_STATIC int sctp_listen_start(struct sock *sk, int backlog) tfm = crypto_alloc_hash(sctp_hmac_alg, 0, CRYPTO_ALG_ASYNC); if (IS_ERR(tfm)) { if (net_ratelimit()) { - printk(KERN_INFO - "SCTP: failed to load transform for %s: %ld\n", + pr_info("failed to load transform for %s: %ld\n", sctp_hmac_alg, PTR_ERR(tfm)); } return -ENOSYS; @@ -5716,13 +5751,12 @@ unsigned int sctp_poll(struct file *file, struct socket *sock, poll_table *wait) if (sk->sk_err || !skb_queue_empty(&sk->sk_error_queue)) mask |= POLLERR; if (sk->sk_shutdown & RCV_SHUTDOWN) - mask |= POLLRDHUP; + mask |= POLLRDHUP | POLLIN | POLLRDNORM; if (sk->sk_shutdown == SHUTDOWN_MASK) mask |= POLLHUP; /* Is it readable? Reconsider this code with TCP-style support. */ - if (!skb_queue_empty(&sk->sk_receive_queue) || - (sk->sk_shutdown & RCV_SHUTDOWN)) + if (!skb_queue_empty(&sk->sk_receive_queue)) mask |= POLLIN | POLLRDNORM; /* The association is either gone or not ready. */ diff --git a/net/sctp/transport.c b/net/sctp/transport.c index 132046cb82f..d3ae493d234 100644 --- a/net/sctp/transport.c +++ b/net/sctp/transport.c @@ -48,6 +48,8 @@ * be incorporated into the next SCTP release. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/slab.h> #include <linux/types.h> #include <linux/random.h> @@ -244,10 +246,9 @@ void sctp_transport_update_pmtu(struct sctp_transport *t, u32 pmtu) struct dst_entry *dst; if (unlikely(pmtu < SCTP_DEFAULT_MINSEGMENT)) { - printk(KERN_WARNING "%s: Reported pmtu %d too low, " - "using default minimum of %d\n", - __func__, pmtu, - SCTP_DEFAULT_MINSEGMENT); + pr_warn("%s: Reported pmtu %d too low, using default minimum of %d\n", + __func__, pmtu, + SCTP_DEFAULT_MINSEGMENT); /* Use default minimum segment size and disable * pmtu discovery on this transport. */ diff --git a/net/socket.c b/net/socket.c index 2270b941bcc..abf3e256152 100644 --- a/net/socket.c +++ b/net/socket.c @@ -209,8 +209,8 @@ int move_addr_to_kernel(void __user *uaddr, int ulen, struct sockaddr *kaddr) * specified. Zero is returned for a success. */ -int move_addr_to_user(struct sockaddr *kaddr, int klen, void __user *uaddr, - int __user *ulen) +static int move_addr_to_user(struct sockaddr *kaddr, int klen, + void __user *uaddr, int __user *ulen) { int err; int len; @@ -502,6 +502,7 @@ static int sock_no_open(struct inode *irrelevant, struct file *dontcare) const struct file_operations bad_sock_fops = { .owner = THIS_MODULE, .open = sock_no_open, + .llseek = noop_llseek, }; /** @@ -535,14 +536,13 @@ void sock_release(struct socket *sock) } EXPORT_SYMBOL(sock_release); -int sock_tx_timestamp(struct msghdr *msg, struct sock *sk, - union skb_shared_tx *shtx) +int sock_tx_timestamp(struct sock *sk, __u8 *tx_flags) { - shtx->flags = 0; + *tx_flags = 0; if (sock_flag(sk, SOCK_TIMESTAMPING_TX_HARDWARE)) - shtx->hardware = 1; + *tx_flags |= SKBTX_HW_TSTAMP; if (sock_flag(sk, SOCK_TIMESTAMPING_TX_SOFTWARE)) - shtx->software = 1; + *tx_flags |= SKBTX_SW_TSTAMP; return 0; } EXPORT_SYMBOL(sock_tx_timestamp); @@ -662,7 +662,8 @@ void __sock_recv_timestamp(struct msghdr *msg, struct sock *sk, } EXPORT_SYMBOL_GPL(__sock_recv_timestamp); -inline void sock_recv_drops(struct msghdr *msg, struct sock *sk, struct sk_buff *skb) +static inline void sock_recv_drops(struct msghdr *msg, struct sock *sk, + struct sk_buff *skb) { if (sock_flag(sk, SOCK_RXQ_OVFL) && skb && skb->dropcount) put_cmsg(msg, SOL_SOCKET, SO_RXQ_OVFL, @@ -1919,7 +1920,8 @@ SYSCALL_DEFINE3(sendmsg, int, fd, struct msghdr __user *, msg, unsigned, flags) * Afterwards, it will be a kernel pointer. Thus the compiler-assisted * checking falls down on this. */ - if (copy_from_user(ctl_buf, (void __user *)msg_sys.msg_control, + if (copy_from_user(ctl_buf, + (void __user __force *)msg_sys.msg_control, ctl_len)) goto out_freectl; msg_sys.msg_control = ctl_buf; @@ -3054,14 +3056,19 @@ int kernel_getsockopt(struct socket *sock, int level, int optname, char *optval, int *optlen) { mm_segment_t oldfs = get_fs(); + char __user *uoptval; + int __user *uoptlen; int err; + uoptval = (char __user __force *) optval; + uoptlen = (int __user __force *) optlen; + set_fs(KERNEL_DS); if (level == SOL_SOCKET) - err = sock_getsockopt(sock, level, optname, optval, optlen); + err = sock_getsockopt(sock, level, optname, uoptval, uoptlen); else - err = sock->ops->getsockopt(sock, level, optname, optval, - optlen); + err = sock->ops->getsockopt(sock, level, optname, uoptval, + uoptlen); set_fs(oldfs); return err; } @@ -3071,13 +3078,16 @@ int kernel_setsockopt(struct socket *sock, int level, int optname, char *optval, unsigned int optlen) { mm_segment_t oldfs = get_fs(); + char __user *uoptval; int err; + uoptval = (char __user __force *) optval; + set_fs(KERNEL_DS); if (level == SOL_SOCKET) - err = sock_setsockopt(sock, level, optname, optval, optlen); + err = sock_setsockopt(sock, level, optname, uoptval, optlen); else - err = sock->ops->setsockopt(sock, level, optname, optval, + err = sock->ops->setsockopt(sock, level, optname, uoptval, optlen); set_fs(oldfs); return err; diff --git a/net/sunrpc/Kconfig b/net/sunrpc/Kconfig index 443c161eb8b..3376d765718 100644 --- a/net/sunrpc/Kconfig +++ b/net/sunrpc/Kconfig @@ -18,10 +18,11 @@ config SUNRPC_XPRT_RDMA If unsure, say N. config RPCSEC_GSS_KRB5 - tristate "Secure RPC: Kerberos V mechanism (EXPERIMENTAL)" - depends on SUNRPC && EXPERIMENTAL + tristate + depends on SUNRPC && CRYPTO + prompt "Secure RPC: Kerberos V mechanism" if !(NFS_V4 || NFSD_V4) + default y select SUNRPC_GSS - select CRYPTO select CRYPTO_MD5 select CRYPTO_DES select CRYPTO_CBC @@ -34,7 +35,7 @@ config RPCSEC_GSS_KRB5 available from http://linux-nfs.org/. In addition, user-space Kerberos support should be installed. - If unsure, say N. + If unsure, say Y. config RPCSEC_GSS_SPKM3 tristate "Secure RPC: SPKM3 mechanism (EXPERIMENTAL)" diff --git a/net/sunrpc/auth.c b/net/sunrpc/auth.c index 880d0de3f50..e9eaaf7d43c 100644 --- a/net/sunrpc/auth.c +++ b/net/sunrpc/auth.c @@ -38,8 +38,8 @@ static const struct rpc_authops *auth_flavors[RPC_AUTH_MAXFLAVOR] = { static LIST_HEAD(cred_unused); static unsigned long number_cred_unused; -#define MAX_HASHTABLE_BITS (10) -static int param_set_hashtbl_sz(const char *val, struct kernel_param *kp) +#define MAX_HASHTABLE_BITS (14) +static int param_set_hashtbl_sz(const char *val, const struct kernel_param *kp) { unsigned long num; unsigned int nbits; @@ -61,7 +61,7 @@ out_inval: return -EINVAL; } -static int param_get_hashtbl_sz(char *buffer, struct kernel_param *kp) +static int param_get_hashtbl_sz(char *buffer, const struct kernel_param *kp) { unsigned int nbits; @@ -71,6 +71,11 @@ static int param_get_hashtbl_sz(char *buffer, struct kernel_param *kp) #define param_check_hashtbl_sz(name, p) __param_check(name, p, unsigned int); +static struct kernel_param_ops param_ops_hashtbl_sz = { + .set = param_set_hashtbl_sz, + .get = param_get_hashtbl_sz, +}; + module_param_named(auth_hashtable_size, auth_hashbits, hashtbl_sz, 0644); MODULE_PARM_DESC(auth_hashtable_size, "RPC credential cache hashtable size"); diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c index dcfc66bab2b..3835ce35e22 100644 --- a/net/sunrpc/auth_gss/auth_gss.c +++ b/net/sunrpc/auth_gss/auth_gss.c @@ -745,17 +745,18 @@ gss_pipe_release(struct inode *inode) struct rpc_inode *rpci = RPC_I(inode); struct gss_upcall_msg *gss_msg; +restart: spin_lock(&inode->i_lock); - while (!list_empty(&rpci->in_downcall)) { + list_for_each_entry(gss_msg, &rpci->in_downcall, list) { - gss_msg = list_entry(rpci->in_downcall.next, - struct gss_upcall_msg, list); + if (!list_empty(&gss_msg->msg.list)) + continue; gss_msg->msg.errno = -EPIPE; atomic_inc(&gss_msg->count); __gss_unhash_msg(gss_msg); spin_unlock(&inode->i_lock); gss_release_msg(gss_msg); - spin_lock(&inode->i_lock); + goto restart; } spin_unlock(&inode->i_lock); @@ -1049,7 +1050,7 @@ gss_match(struct auth_cred *acred, struct rpc_cred *rc, int flags) out: if (acred->machine_cred != gss_cred->gc_machine_cred) return 0; - return (rc->cr_uid == acred->uid); + return rc->cr_uid == acred->uid; } /* diff --git a/net/sunrpc/auth_gss/gss_generic_token.c b/net/sunrpc/auth_gss/gss_generic_token.c index 310b78e9945..c586e92bcf7 100644 --- a/net/sunrpc/auth_gss/gss_generic_token.c +++ b/net/sunrpc/auth_gss/gss_generic_token.c @@ -76,19 +76,19 @@ static int der_length_size( int length) { if (length < (1<<7)) - return(1); + return 1; else if (length < (1<<8)) - return(2); + return 2; #if (SIZEOF_INT == 2) else - return(3); + return 3; #else else if (length < (1<<16)) - return(3); + return 3; else if (length < (1<<24)) - return(4); + return 4; else - return(5); + return 5; #endif } @@ -121,14 +121,14 @@ der_read_length(unsigned char **buf, int *bufsize) int ret; if (*bufsize < 1) - return(-1); + return -1; sf = *(*buf)++; (*bufsize)--; if (sf & 0x80) { if ((sf &= 0x7f) > ((*bufsize)-1)) - return(-1); + return -1; if (sf > SIZEOF_INT) - return (-1); + return -1; ret = 0; for (; sf; sf--) { ret = (ret<<8) + (*(*buf)++); @@ -138,7 +138,7 @@ der_read_length(unsigned char **buf, int *bufsize) ret = sf; } - return(ret); + return ret; } /* returns the length of a token, given the mech oid and the body size */ @@ -148,7 +148,7 @@ g_token_size(struct xdr_netobj *mech, unsigned int body_size) { /* set body_size to sequence contents size */ body_size += 2 + (int) mech->len; /* NEED overflow check */ - return(1 + der_length_size(body_size) + body_size); + return 1 + der_length_size(body_size) + body_size; } EXPORT_SYMBOL_GPL(g_token_size); @@ -186,27 +186,27 @@ g_verify_token_header(struct xdr_netobj *mech, int *body_size, int ret = 0; if ((toksize-=1) < 0) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if (*buf++ != 0x60) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if ((seqsize = der_read_length(&buf, &toksize)) < 0) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if (seqsize != toksize) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if ((toksize-=1) < 0) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if (*buf++ != 0x06) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if ((toksize-=1) < 0) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; toid.len = *buf++; if ((toksize-=toid.len) < 0) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; toid.data = buf; buf+=toid.len; @@ -217,17 +217,17 @@ g_verify_token_header(struct xdr_netobj *mech, int *body_size, to return G_BAD_TOK_HEADER if the token header is in fact bad */ if ((toksize-=2) < 0) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if (ret) - return(ret); + return ret; if (!ret) { *buf_in = buf; *body_size = toksize; } - return(ret); + return ret; } EXPORT_SYMBOL_GPL(g_verify_token_header); diff --git a/net/sunrpc/auth_gss/gss_krb5_mech.c b/net/sunrpc/auth_gss/gss_krb5_mech.c index 03264461052..778e5dfc514 100644 --- a/net/sunrpc/auth_gss/gss_krb5_mech.c +++ b/net/sunrpc/auth_gss/gss_krb5_mech.c @@ -237,6 +237,7 @@ get_key(const void *p, const void *end, if (!supported_gss_krb5_enctype(alg)) { printk(KERN_WARNING "gss_kerberos_mech: unsupported " "encryption key algorithm %d\n", alg); + p = ERR_PTR(-EINVAL); goto out_err; } p = simple_get_netobj(p, end, &key); @@ -282,15 +283,19 @@ gss_import_v1_context(const void *p, const void *end, struct krb5_ctx *ctx) ctx->enctype = ENCTYPE_DES_CBC_RAW; ctx->gk5e = get_gss_krb5_enctype(ctx->enctype); - if (ctx->gk5e == NULL) + if (ctx->gk5e == NULL) { + p = ERR_PTR(-EINVAL); goto out_err; + } /* The downcall format was designed before we completely understood * the uses of the context fields; so it includes some stuff we * just give some minimal sanity-checking, and some we ignore * completely (like the next twenty bytes): */ - if (unlikely(p + 20 > end || p + 20 < p)) + if (unlikely(p + 20 > end || p + 20 < p)) { + p = ERR_PTR(-EFAULT); goto out_err; + } p += 20; p = simple_get_bytes(p, end, &tmp, sizeof(tmp)); if (IS_ERR(p)) @@ -619,6 +624,7 @@ gss_import_v2_context(const void *p, const void *end, struct krb5_ctx *ctx, if (ctx->seq_send64 != ctx->seq_send) { dprintk("%s: seq_send64 %lx, seq_send %x overflow?\n", __func__, (long unsigned)ctx->seq_send64, ctx->seq_send); + p = ERR_PTR(-EINVAL); goto out_err; } p = simple_get_bytes(p, end, &ctx->enctype, sizeof(ctx->enctype)); diff --git a/net/sunrpc/auth_gss/gss_krb5_seqnum.c b/net/sunrpc/auth_gss/gss_krb5_seqnum.c index 415c013ba38..62ac90c62cb 100644 --- a/net/sunrpc/auth_gss/gss_krb5_seqnum.c +++ b/net/sunrpc/auth_gss/gss_krb5_seqnum.c @@ -162,5 +162,5 @@ krb5_get_seq_num(struct krb5_ctx *kctx, *seqnum = ((plain[0]) | (plain[1] << 8) | (plain[2] << 16) | (plain[3] << 24)); - return (0); + return 0; } diff --git a/net/sunrpc/auth_gss/gss_mech_switch.c b/net/sunrpc/auth_gss/gss_mech_switch.c index 2689de39dc7..8b4061049d7 100644 --- a/net/sunrpc/auth_gss/gss_mech_switch.c +++ b/net/sunrpc/auth_gss/gss_mech_switch.c @@ -331,7 +331,7 @@ gss_delete_sec_context(struct gss_ctx **context_handle) *context_handle); if (!*context_handle) - return(GSS_S_NO_CONTEXT); + return GSS_S_NO_CONTEXT; if ((*context_handle)->internal_ctx_id) (*context_handle)->mech_type->gm_ops ->gss_delete_sec_context((*context_handle) diff --git a/net/sunrpc/auth_gss/gss_spkm3_mech.c b/net/sunrpc/auth_gss/gss_spkm3_mech.c index dc3f1f5ed86..adade3d313f 100644 --- a/net/sunrpc/auth_gss/gss_spkm3_mech.c +++ b/net/sunrpc/auth_gss/gss_spkm3_mech.c @@ -100,6 +100,7 @@ gss_import_sec_context_spkm3(const void *p, size_t len, if (version != 1) { dprintk("RPC: unknown spkm3 token format: " "obsolete nfs-utils?\n"); + p = ERR_PTR(-EINVAL); goto out_err_free_ctx; } @@ -135,8 +136,10 @@ gss_import_sec_context_spkm3(const void *p, size_t len, if (IS_ERR(p)) goto out_err_free_intg_alg; - if (p != end) + if (p != end) { + p = ERR_PTR(-EFAULT); goto out_err_free_intg_key; + } ctx_id->internal_ctx_id = ctx; diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index 2b06410e584..7dce81a926c 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -28,7 +28,6 @@ #include <linux/workqueue.h> #include <linux/mutex.h> #include <linux/pagemap.h> -#include <linux/smp_lock.h> #include <asm/ioctls.h> #include <linux/sunrpc/types.h> #include <linux/sunrpc/cache.h> @@ -1348,15 +1347,10 @@ static unsigned int cache_poll_procfs(struct file *filp, poll_table *wait) static long cache_ioctl_procfs(struct file *filp, unsigned int cmd, unsigned long arg) { - long ret; struct inode *inode = filp->f_path.dentry->d_inode; struct cache_detail *cd = PDE(inode)->data; - lock_kernel(); - ret = cache_ioctl(inode, filp, cmd, arg, cd); - unlock_kernel(); - - return ret; + return cache_ioctl(inode, filp, cmd, arg, cd); } static int cache_open_procfs(struct inode *inode, struct file *filp) @@ -1441,6 +1435,7 @@ static const struct file_operations cache_flush_operations_procfs = { .read = read_flush_procfs, .write = write_flush_procfs, .release = release_flush_procfs, + .llseek = no_llseek, }; static void remove_cache_proc_entries(struct cache_detail *cd) @@ -1555,13 +1550,8 @@ static long cache_ioctl_pipefs(struct file *filp, { struct inode *inode = filp->f_dentry->d_inode; struct cache_detail *cd = RPC_I(inode)->private; - long ret; - lock_kernel(); - ret = cache_ioctl(inode, filp, cmd, arg, cd); - unlock_kernel(); - - return ret; + return cache_ioctl(inode, filp, cmd, arg, cd); } static int cache_open_pipefs(struct inode *inode, struct file *filp) @@ -1646,6 +1636,7 @@ const struct file_operations cache_flush_operations_pipefs = { .read = read_flush_pipefs, .write = write_flush_pipefs, .release = release_flush_pipefs, + .llseek = no_llseek, }; int sunrpc_cache_register_pipefs(struct dentry *parent, diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 2388d83b68f..fa5549079d7 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -226,7 +226,7 @@ static struct rpc_clnt * rpc_new_client(const struct rpc_create_args *args, stru goto out_no_principal; } - kref_init(&clnt->cl_kref); + atomic_set(&clnt->cl_count, 1); err = rpc_setup_pipedir(clnt, program->pipe_dir_name); if (err < 0) @@ -390,14 +390,14 @@ rpc_clone_client(struct rpc_clnt *clnt) if (new->cl_principal == NULL) goto out_no_principal; } - kref_init(&new->cl_kref); + atomic_set(&new->cl_count, 1); err = rpc_setup_pipedir(new, clnt->cl_program->pipe_dir_name); if (err != 0) goto out_no_path; if (new->cl_auth) atomic_inc(&new->cl_auth->au_count); xprt_get(clnt->cl_xprt); - kref_get(&clnt->cl_kref); + atomic_inc(&clnt->cl_count); rpc_register_client(new); rpciod_up(); return new; @@ -465,10 +465,8 @@ EXPORT_SYMBOL_GPL(rpc_shutdown_client); * Free an RPC client */ static void -rpc_free_client(struct kref *kref) +rpc_free_client(struct rpc_clnt *clnt) { - struct rpc_clnt *clnt = container_of(kref, struct rpc_clnt, cl_kref); - dprintk("RPC: destroying %s client for %s\n", clnt->cl_protname, clnt->cl_server); if (!IS_ERR(clnt->cl_path.dentry)) { @@ -495,12 +493,10 @@ out_free: * Free an RPC client */ static void -rpc_free_auth(struct kref *kref) +rpc_free_auth(struct rpc_clnt *clnt) { - struct rpc_clnt *clnt = container_of(kref, struct rpc_clnt, cl_kref); - if (clnt->cl_auth == NULL) { - rpc_free_client(kref); + rpc_free_client(clnt); return; } @@ -509,10 +505,11 @@ rpc_free_auth(struct kref *kref) * release remaining GSS contexts. This mechanism ensures * that it can do so safely. */ - kref_init(kref); + atomic_inc(&clnt->cl_count); rpcauth_release(clnt->cl_auth); clnt->cl_auth = NULL; - kref_put(kref, rpc_free_client); + if (atomic_dec_and_test(&clnt->cl_count)) + rpc_free_client(clnt); } /* @@ -525,7 +522,8 @@ rpc_release_client(struct rpc_clnt *clnt) if (list_empty(&clnt->cl_tasks)) wake_up(&destroy_wait); - kref_put(&clnt->cl_kref, rpc_free_auth); + if (atomic_dec_and_test(&clnt->cl_count)) + rpc_free_auth(clnt); } /** @@ -588,7 +586,7 @@ void rpc_task_set_client(struct rpc_task *task, struct rpc_clnt *clnt) if (clnt != NULL) { rpc_task_release_client(task); task->tk_client = clnt; - kref_get(&clnt->cl_kref); + atomic_inc(&clnt->cl_count); if (clnt->cl_softrtry) task->tk_flags |= RPC_TASK_SOFT; /* Add to the client's list of all tasks */ @@ -931,7 +929,7 @@ call_reserveresult(struct rpc_task *task) task->tk_status = 0; if (status >= 0) { if (task->tk_rqstp) { - task->tk_action = call_allocate; + task->tk_action = call_refresh; return; } @@ -966,13 +964,54 @@ call_reserveresult(struct rpc_task *task) } /* - * 2. Allocate the buffer. For details, see sched.c:rpc_malloc. + * 2. Bind and/or refresh the credentials + */ +static void +call_refresh(struct rpc_task *task) +{ + dprint_status(task); + + task->tk_action = call_refreshresult; + task->tk_status = 0; + task->tk_client->cl_stats->rpcauthrefresh++; + rpcauth_refreshcred(task); +} + +/* + * 2a. Process the results of a credential refresh + */ +static void +call_refreshresult(struct rpc_task *task) +{ + int status = task->tk_status; + + dprint_status(task); + + task->tk_status = 0; + task->tk_action = call_allocate; + if (status >= 0 && rpcauth_uptodatecred(task)) + return; + switch (status) { + case -EACCES: + rpc_exit(task, -EACCES); + return; + case -ENOMEM: + rpc_exit(task, -ENOMEM); + return; + case -ETIMEDOUT: + rpc_delay(task, 3*HZ); + } + task->tk_action = call_refresh; +} + +/* + * 2b. Allocate the buffer. For details, see sched.c:rpc_malloc. * (Note: buffer memory is freed in xprt_release). */ static void call_allocate(struct rpc_task *task) { - unsigned int slack = task->tk_client->cl_auth->au_cslack; + unsigned int slack = task->tk_rqstp->rq_cred->cr_auth->au_cslack; struct rpc_rqst *req = task->tk_rqstp; struct rpc_xprt *xprt = task->tk_xprt; struct rpc_procinfo *proc = task->tk_msg.rpc_proc; @@ -980,7 +1019,7 @@ call_allocate(struct rpc_task *task) dprint_status(task); task->tk_status = 0; - task->tk_action = call_refresh; + task->tk_action = call_bind; if (req->rq_buffer) return; @@ -1017,47 +1056,6 @@ call_allocate(struct rpc_task *task) rpc_exit(task, -ERESTARTSYS); } -/* - * 2a. Bind and/or refresh the credentials - */ -static void -call_refresh(struct rpc_task *task) -{ - dprint_status(task); - - task->tk_action = call_refreshresult; - task->tk_status = 0; - task->tk_client->cl_stats->rpcauthrefresh++; - rpcauth_refreshcred(task); -} - -/* - * 2b. Process the results of a credential refresh - */ -static void -call_refreshresult(struct rpc_task *task) -{ - int status = task->tk_status; - - dprint_status(task); - - task->tk_status = 0; - task->tk_action = call_bind; - if (status >= 0 && rpcauth_uptodatecred(task)) - return; - switch (status) { - case -EACCES: - rpc_exit(task, -EACCES); - return; - case -ENOMEM: - rpc_exit(task, -ENOMEM); - return; - case -ETIMEDOUT: - rpc_delay(task, 3*HZ); - } - task->tk_action = call_refresh; -} - static inline int rpc_task_need_encode(struct rpc_task *task) { diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index 1f7fc502a50..52f25243214 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -27,7 +27,6 @@ #include <linux/workqueue.h> #include <linux/sunrpc/rpc_pipe_fs.h> #include <linux/sunrpc/cache.h> -#include <linux/smp_lock.h> static struct vfsmount *rpc_mount __read_mostly; static int rpc_mount_count; @@ -48,7 +47,7 @@ static void rpc_purge_list(struct rpc_inode *rpci, struct list_head *head, return; do { msg = list_entry(head->next, struct rpc_pipe_msg, list); - list_del(&msg->list); + list_del_init(&msg->list); msg->errno = err; destroy_msg(msg); } while (!list_empty(head)); @@ -208,7 +207,7 @@ rpc_pipe_release(struct inode *inode, struct file *filp) if (msg != NULL) { spin_lock(&inode->i_lock); msg->errno = -EAGAIN; - list_del(&msg->list); + list_del_init(&msg->list); spin_unlock(&inode->i_lock); rpci->ops->destroy_msg(msg); } @@ -268,7 +267,7 @@ rpc_pipe_read(struct file *filp, char __user *buf, size_t len, loff_t *offset) if (res < 0 || msg->len == msg->copied) { filp->private_data = NULL; spin_lock(&inode->i_lock); - list_del(&msg->list); + list_del_init(&msg->list); spin_unlock(&inode->i_lock); rpci->ops->destroy_msg(msg); } @@ -309,40 +308,33 @@ rpc_pipe_poll(struct file *filp, struct poll_table_struct *wait) return mask; } -static int -rpc_pipe_ioctl_unlocked(struct file *filp, unsigned int cmd, unsigned long arg) +static long +rpc_pipe_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { - struct rpc_inode *rpci = RPC_I(filp->f_path.dentry->d_inode); + struct inode *inode = filp->f_path.dentry->d_inode; + struct rpc_inode *rpci = RPC_I(inode); int len; switch (cmd) { case FIONREAD: - if (rpci->ops == NULL) + spin_lock(&inode->i_lock); + if (rpci->ops == NULL) { + spin_unlock(&inode->i_lock); return -EPIPE; + } len = rpci->pipelen; if (filp->private_data) { struct rpc_pipe_msg *msg; msg = filp->private_data; len += msg->len - msg->copied; } + spin_unlock(&inode->i_lock); return put_user(len, (int __user *)arg); default: return -EINVAL; } } -static long -rpc_pipe_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) -{ - long ret; - - lock_kernel(); - ret = rpc_pipe_ioctl_unlocked(filp, cmd, arg); - unlock_kernel(); - - return ret; -} - static const struct file_operations rpc_pipe_fops = { .owner = THIS_MODULE, .llseek = no_llseek, @@ -371,21 +363,23 @@ rpc_show_info(struct seq_file *m, void *v) static int rpc_info_open(struct inode *inode, struct file *file) { - struct rpc_clnt *clnt; + struct rpc_clnt *clnt = NULL; int ret = single_open(file, rpc_show_info, NULL); if (!ret) { struct seq_file *m = file->private_data; - mutex_lock(&inode->i_mutex); - clnt = RPC_I(inode)->private; - if (clnt) { - kref_get(&clnt->cl_kref); + + spin_lock(&file->f_path.dentry->d_lock); + if (!d_unhashed(file->f_path.dentry)) + clnt = RPC_I(inode)->private; + if (clnt != NULL && atomic_inc_not_zero(&clnt->cl_count)) { + spin_unlock(&file->f_path.dentry->d_lock); m->private = clnt; } else { + spin_unlock(&file->f_path.dentry->d_lock); single_release(inode, file); ret = -EINVAL; } - mutex_unlock(&inode->i_mutex); } return ret; } diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index cace6049e4a..aa5dbda6608 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -376,7 +376,7 @@ int rpc_queue_empty(struct rpc_wait_queue *queue) spin_lock_bh(&queue->lock); res = queue->qlen; spin_unlock_bh(&queue->lock); - return (res == 0); + return res == 0; } EXPORT_SYMBOL_GPL(rpc_queue_empty); diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c index e5e28d1946a..2ac3f6e8adf 100644 --- a/net/sunrpc/xprtrdma/rpc_rdma.c +++ b/net/sunrpc/xprtrdma/rpc_rdma.c @@ -249,6 +249,8 @@ rpcrdma_create_chunks(struct rpc_rqst *rqst, struct xdr_buf *target, req->rl_nchunks = nchunks; BUG_ON(nchunks == 0); + BUG_ON((r_xprt->rx_ia.ri_memreg_strategy == RPCRDMA_FRMR) + && (nchunks > 3)); /* * finish off header. If write, marshal discrim and nchunks. diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index 27015c6d8eb..5f4c7b3bc71 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -650,10 +650,22 @@ rpcrdma_ep_create(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia, ep->rep_attr.cap.max_send_wr = cdata->max_requests; switch (ia->ri_memreg_strategy) { case RPCRDMA_FRMR: - /* Add room for frmr register and invalidate WRs */ - ep->rep_attr.cap.max_send_wr *= 3; - if (ep->rep_attr.cap.max_send_wr > devattr.max_qp_wr) - return -EINVAL; + /* Add room for frmr register and invalidate WRs. + * 1. FRMR reg WR for head + * 2. FRMR invalidate WR for head + * 3. FRMR reg WR for pagelist + * 4. FRMR invalidate WR for pagelist + * 5. FRMR reg WR for tail + * 6. FRMR invalidate WR for tail + * 7. The RDMA_SEND WR + */ + ep->rep_attr.cap.max_send_wr *= 7; + if (ep->rep_attr.cap.max_send_wr > devattr.max_qp_wr) { + cdata->max_requests = devattr.max_qp_wr / 7; + if (!cdata->max_requests) + return -EINVAL; + ep->rep_attr.cap.max_send_wr = cdata->max_requests * 7; + } break; case RPCRDMA_MEMWINDOWS_ASYNC: case RPCRDMA_MEMWINDOWS: @@ -1490,7 +1502,7 @@ rpcrdma_register_frmr_external(struct rpcrdma_mr_seg *seg, memset(&frmr_wr, 0, sizeof frmr_wr); frmr_wr.opcode = IB_WR_FAST_REG_MR; frmr_wr.send_flags = 0; /* unsignaled */ - frmr_wr.wr.fast_reg.iova_start = (unsigned long)seg1->mr_dma; + frmr_wr.wr.fast_reg.iova_start = seg1->mr_dma; frmr_wr.wr.fast_reg.page_list = seg1->mr_chunk.rl_mw->r.frmr.fr_pgl; frmr_wr.wr.fast_reg.page_list_len = i; frmr_wr.wr.fast_reg.page_shift = PAGE_SHIFT; diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 7ca65c7005e..fe9306bf10c 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -800,7 +800,7 @@ static void xs_udp_data_ready(struct sock *sk, int len) u32 _xid; __be32 *xp; - read_lock(&sk->sk_callback_lock); + read_lock_bh(&sk->sk_callback_lock); dprintk("RPC: xs_udp_data_ready...\n"); if (!(xprt = xprt_from_sock(sk))) goto out; @@ -852,7 +852,7 @@ static void xs_udp_data_ready(struct sock *sk, int len) dropit: skb_free_datagram(sk, skb); out: - read_unlock(&sk->sk_callback_lock); + read_unlock_bh(&sk->sk_callback_lock); } static inline void xs_tcp_read_fraghdr(struct rpc_xprt *xprt, struct xdr_skb_reader *desc) @@ -1229,7 +1229,7 @@ static void xs_tcp_data_ready(struct sock *sk, int bytes) dprintk("RPC: xs_tcp_data_ready...\n"); - read_lock(&sk->sk_callback_lock); + read_lock_bh(&sk->sk_callback_lock); if (!(xprt = xprt_from_sock(sk))) goto out; if (xprt->shutdown) @@ -1248,7 +1248,7 @@ static void xs_tcp_data_ready(struct sock *sk, int bytes) read = tcp_read_sock(sk, &rd_desc, xs_tcp_data_recv); } while (read > 0); out: - read_unlock(&sk->sk_callback_lock); + read_unlock_bh(&sk->sk_callback_lock); } /* @@ -1301,18 +1301,19 @@ static void xs_tcp_state_change(struct sock *sk) { struct rpc_xprt *xprt; - read_lock(&sk->sk_callback_lock); + read_lock_bh(&sk->sk_callback_lock); if (!(xprt = xprt_from_sock(sk))) goto out; dprintk("RPC: xs_tcp_state_change client %p...\n", xprt); - dprintk("RPC: state %x conn %d dead %d zapped %d\n", + dprintk("RPC: state %x conn %d dead %d zapped %d sk_shutdown %d\n", sk->sk_state, xprt_connected(xprt), sock_flag(sk, SOCK_DEAD), - sock_flag(sk, SOCK_ZAPPED)); + sock_flag(sk, SOCK_ZAPPED), + sk->sk_shutdown); switch (sk->sk_state) { case TCP_ESTABLISHED: - spin_lock_bh(&xprt->transport_lock); + spin_lock(&xprt->transport_lock); if (!xprt_test_and_set_connected(xprt)) { struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); @@ -1326,7 +1327,7 @@ static void xs_tcp_state_change(struct sock *sk) xprt_wake_pending_tasks(xprt, -EAGAIN); } - spin_unlock_bh(&xprt->transport_lock); + spin_unlock(&xprt->transport_lock); break; case TCP_FIN_WAIT1: /* The client initiated a shutdown of the socket */ @@ -1364,7 +1365,7 @@ static void xs_tcp_state_change(struct sock *sk) xs_sock_mark_closed(xprt); } out: - read_unlock(&sk->sk_callback_lock); + read_unlock_bh(&sk->sk_callback_lock); } /** @@ -1375,7 +1376,7 @@ static void xs_error_report(struct sock *sk) { struct rpc_xprt *xprt; - read_lock(&sk->sk_callback_lock); + read_lock_bh(&sk->sk_callback_lock); if (!(xprt = xprt_from_sock(sk))) goto out; dprintk("RPC: %s client %p...\n" @@ -1383,7 +1384,7 @@ static void xs_error_report(struct sock *sk) __func__, xprt, sk->sk_err); xprt_wake_pending_tasks(xprt, -EAGAIN); out: - read_unlock(&sk->sk_callback_lock); + read_unlock_bh(&sk->sk_callback_lock); } static void xs_write_space(struct sock *sk) @@ -1415,13 +1416,13 @@ static void xs_write_space(struct sock *sk) */ static void xs_udp_write_space(struct sock *sk) { - read_lock(&sk->sk_callback_lock); + read_lock_bh(&sk->sk_callback_lock); /* from net/core/sock.c:sock_def_write_space */ if (sock_writeable(sk)) xs_write_space(sk); - read_unlock(&sk->sk_callback_lock); + read_unlock_bh(&sk->sk_callback_lock); } /** @@ -1436,13 +1437,13 @@ static void xs_udp_write_space(struct sock *sk) */ static void xs_tcp_write_space(struct sock *sk) { - read_lock(&sk->sk_callback_lock); + read_lock_bh(&sk->sk_callback_lock); /* from net/core/stream.c:sk_stream_write_space */ if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk)) xs_write_space(sk); - read_unlock(&sk->sk_callback_lock); + read_unlock_bh(&sk->sk_callback_lock); } static void xs_udp_do_set_buffer_size(struct rpc_xprt *xprt) @@ -1779,10 +1780,25 @@ static void xs_tcp_reuse_connection(struct rpc_xprt *xprt, struct sock_xprt *tra { unsigned int state = transport->inet->sk_state; - if (state == TCP_CLOSE && transport->sock->state == SS_UNCONNECTED) - return; - if ((1 << state) & (TCPF_ESTABLISHED|TCPF_SYN_SENT)) - return; + if (state == TCP_CLOSE && transport->sock->state == SS_UNCONNECTED) { + /* we don't need to abort the connection if the socket + * hasn't undergone a shutdown + */ + if (transport->inet->sk_shutdown == 0) + return; + dprintk("RPC: %s: TCP_CLOSEd and sk_shutdown set to %d\n", + __func__, transport->inet->sk_shutdown); + } + if ((1 << state) & (TCPF_ESTABLISHED|TCPF_SYN_SENT)) { + /* we don't need to abort the connection if the socket + * hasn't undergone a shutdown + */ + if (transport->inet->sk_shutdown == 0) + return; + dprintk("RPC: %s: ESTABLISHED/SYN_SENT " + "sk_shutdown set to %d\n", + __func__, transport->inet->sk_shutdown); + } xs_abort_connection(xprt, transport); } @@ -2577,7 +2593,8 @@ void cleanup_socket_xprt(void) xprt_unregister_transport(&xs_bc_tcp_transport); } -static int param_set_uint_minmax(const char *val, struct kernel_param *kp, +static int param_set_uint_minmax(const char *val, + const struct kernel_param *kp, unsigned int min, unsigned int max) { unsigned long num; @@ -2592,34 +2609,37 @@ static int param_set_uint_minmax(const char *val, struct kernel_param *kp, return 0; } -static int param_set_portnr(const char *val, struct kernel_param *kp) +static int param_set_portnr(const char *val, const struct kernel_param *kp) { return param_set_uint_minmax(val, kp, RPC_MIN_RESVPORT, RPC_MAX_RESVPORT); } -static int param_get_portnr(char *buffer, struct kernel_param *kp) -{ - return param_get_uint(buffer, kp); -} +static struct kernel_param_ops param_ops_portnr = { + .set = param_set_portnr, + .get = param_get_uint, +}; + #define param_check_portnr(name, p) \ __param_check(name, p, unsigned int); module_param_named(min_resvport, xprt_min_resvport, portnr, 0644); module_param_named(max_resvport, xprt_max_resvport, portnr, 0644); -static int param_set_slot_table_size(const char *val, struct kernel_param *kp) +static int param_set_slot_table_size(const char *val, + const struct kernel_param *kp) { return param_set_uint_minmax(val, kp, RPC_MIN_SLOT_TABLE, RPC_MAX_SLOT_TABLE); } -static int param_get_slot_table_size(char *buffer, struct kernel_param *kp) -{ - return param_get_uint(buffer, kp); -} +static struct kernel_param_ops param_ops_slot_table_size = { + .set = param_set_slot_table_size, + .get = param_get_uint, +}; + #define param_check_slot_table_size(name, p) \ __param_check(name, p, unsigned int); diff --git a/net/tipc/addr.c b/net/tipc/addr.c index c048543ffbe..8a2e89bffde 100644 --- a/net/tipc/addr.c +++ b/net/tipc/addr.c @@ -41,11 +41,6 @@ #include "cluster.h" #include "net.h" -u32 tipc_get_addr(void) -{ - return tipc_own_addr; -} - /** * tipc_addr_domain_valid - validates a network domain address * @@ -89,7 +84,7 @@ int tipc_addr_domain_valid(u32 addr) int tipc_addr_node_valid(u32 addr) { - return (tipc_addr_domain_valid(addr) && tipc_node(addr)); + return tipc_addr_domain_valid(addr) && tipc_node(addr); } int tipc_in_scope(u32 domain, u32 addr) diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c index a008c668930..22a60fc9839 100644 --- a/net/tipc/bcast.c +++ b/net/tipc/bcast.c @@ -121,6 +121,9 @@ static DEFINE_SPINLOCK(bc_lock); const char tipc_bclink_name[] = "broadcast-link"; +static void tipc_nmap_diff(struct tipc_node_map *nm_a, + struct tipc_node_map *nm_b, + struct tipc_node_map *nm_diff); static u32 buf_seqno(struct sk_buff *buf) { @@ -143,6 +146,19 @@ static void bcbuf_decr_acks(struct sk_buff *buf) } +static void bclink_set_last_sent(void) +{ + if (bcl->next_out) + bcl->fsm_msg_cnt = mod(buf_seqno(bcl->next_out) - 1); + else + bcl->fsm_msg_cnt = mod(bcl->next_out_no - 1); +} + +u32 tipc_bclink_get_last_sent(void) +{ + return bcl->fsm_msg_cnt; +} + /** * bclink_set_gap - set gap according to contents of current deferred pkt queue * @@ -171,7 +187,7 @@ static void bclink_set_gap(struct tipc_node *n_ptr) static int bclink_ack_allowed(u32 n) { - return((n % TIPC_MIN_LINK_WIN) == tipc_own_tag); + return (n % TIPC_MIN_LINK_WIN) == tipc_own_tag; } @@ -237,8 +253,10 @@ void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked) /* Try resolving broadcast link congestion, if necessary */ - if (unlikely(bcl->next_out)) + if (unlikely(bcl->next_out)) { tipc_link_push_queue(bcl); + bclink_set_last_sent(); + } if (unlikely(released && !list_empty(&bcl->waiting_ports))) tipc_link_wakeup_ports(bcl, 0); spin_unlock_bh(&bc_lock); @@ -272,7 +290,7 @@ static void bclink_send_nack(struct tipc_node *n_ptr) if (!less(n_ptr->bclink.gap_after, n_ptr->bclink.gap_to)) return; - buf = buf_acquire(INT_H_SIZE); + buf = tipc_buf_acquire(INT_H_SIZE); if (buf) { msg = buf_msg(buf); tipc_msg_init(msg, BCAST_PROTOCOL, STATE_MSG, @@ -395,7 +413,7 @@ int tipc_bclink_send_msg(struct sk_buff *buf) if (unlikely(res == -ELINKCONG)) buf_discard(buf); else - bcl->stats.sent_info++; + bclink_set_last_sent(); if (bcl->out_queue_size > bcl->stats.max_queue_sz) bcl->stats.max_queue_sz = bcl->out_queue_size; @@ -529,15 +547,6 @@ receive: tipc_node_unlock(node); } -u32 tipc_bclink_get_last_sent(void) -{ - u32 last_sent = mod(bcl->next_out_no - 1); - - if (bcl->next_out) - last_sent = mod(buf_seqno(bcl->next_out) - 1); - return last_sent; -} - u32 tipc_bclink_acks_missing(struct tipc_node *n_ptr) { return (n_ptr->bclink.supported && @@ -570,6 +579,7 @@ static int tipc_bcbearer_send(struct sk_buff *buf, msg = buf_msg(buf); msg_set_non_seq(msg, 1); msg_set_mc_netid(msg, tipc_net_id); + bcl->stats.sent_info++; } /* Send buffer over bearers until all targets reached */ @@ -609,11 +619,13 @@ static int tipc_bcbearer_send(struct sk_buff *buf, bcbearer->remains = bcbearer->remains_new; } - /* Unable to reach all targets */ + /* + * Unable to reach all targets (indicate success, since currently + * there isn't code in place to properly block & unblock the + * pseudo-bearer used by the broadcast link) + */ - bcbearer->bearer.publ.blocked = 1; - bcl->stats.bearer_congs++; - return 1; + return TIPC_OK; } /** @@ -862,8 +874,9 @@ void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node) * @nm_diff: output node map A-B (i.e. nodes of A that are not in B) */ -void tipc_nmap_diff(struct tipc_node_map *nm_a, struct tipc_node_map *nm_b, - struct tipc_node_map *nm_diff) +static void tipc_nmap_diff(struct tipc_node_map *nm_a, + struct tipc_node_map *nm_b, + struct tipc_node_map *nm_diff) { int stop = ARRAY_SIZE(nm_a->map); int w; diff --git a/net/tipc/bcast.h b/net/tipc/bcast.h index e8c2b81658c..011c03f0a4a 100644 --- a/net/tipc/bcast.h +++ b/net/tipc/bcast.h @@ -84,9 +84,6 @@ static inline int tipc_nmap_equal(struct tipc_node_map *nm_a, struct tipc_node_m return !memcmp(nm_a, nm_b, sizeof(*nm_a)); } -void tipc_nmap_diff(struct tipc_node_map *nm_a, struct tipc_node_map *nm_b, - struct tipc_node_map *nm_diff); - void tipc_port_list_add(struct port_list *pl_ptr, u32 port); void tipc_port_list_free(struct port_list *pl_ptr); diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 52ae17b2583..9927d1d56c4 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -63,7 +63,7 @@ static int media_name_valid(const char *name) len = strlen(name); if ((len + 1) > TIPC_MAX_MEDIA_NAME) return 0; - return (strspn(name, tipc_alphabet) == len); + return strspn(name, tipc_alphabet) == len; } /** @@ -288,9 +288,6 @@ static struct bearer *bearer_find(const char *name) struct bearer *b_ptr; u32 i; - if (tipc_mode != TIPC_NET_MODE) - return NULL; - for (i = 0, b_ptr = tipc_bearers; i < MAX_BEARERS; i++, b_ptr++) { if (b_ptr->active && (!strcmp(b_ptr->publ.name, name))) return b_ptr; @@ -559,8 +556,6 @@ restart: } b_ptr = &tipc_bearers[bearer_id]; - memset(b_ptr, 0, sizeof(struct bearer)); - strcpy(b_ptr->publ.name, name); res = m_ptr->enable_bearer(&b_ptr->publ); if (res) { @@ -630,30 +625,17 @@ int tipc_block_bearer(const char *name) * Note: This routine assumes caller holds tipc_net_lock. */ -static int bearer_disable(const char *name) +static int bearer_disable(struct bearer *b_ptr) { - struct bearer *b_ptr; struct link *l_ptr; struct link *temp_l_ptr; - b_ptr = bearer_find(name); - if (!b_ptr) { - warn("Attempt to disable unknown bearer <%s>\n", name); - return -EINVAL; - } - - info("Disabling bearer <%s>\n", name); + info("Disabling bearer <%s>\n", b_ptr->publ.name); tipc_disc_stop_link_req(b_ptr->link_req); spin_lock_bh(&b_ptr->publ.lock); b_ptr->link_req = NULL; b_ptr->publ.blocked = 1; - if (b_ptr->media->disable_bearer) { - spin_unlock_bh(&b_ptr->publ.lock); - write_unlock_bh(&tipc_net_lock); - b_ptr->media->disable_bearer(&b_ptr->publ); - write_lock_bh(&tipc_net_lock); - spin_lock_bh(&b_ptr->publ.lock); - } + b_ptr->media->disable_bearer(&b_ptr->publ); list_for_each_entry_safe(l_ptr, temp_l_ptr, &b_ptr->links, link_list) { tipc_link_delete(l_ptr); } @@ -664,10 +646,16 @@ static int bearer_disable(const char *name) int tipc_disable_bearer(const char *name) { + struct bearer *b_ptr; int res; write_lock_bh(&tipc_net_lock); - res = bearer_disable(name); + b_ptr = bearer_find(name); + if (b_ptr == NULL) { + warn("Attempt to disable unknown bearer <%s>\n", name); + res = -EINVAL; + } else + res = bearer_disable(b_ptr); write_unlock_bh(&tipc_net_lock); return res; } @@ -680,13 +668,7 @@ void tipc_bearer_stop(void) for (i = 0; i < MAX_BEARERS; i++) { if (tipc_bearers[i].active) - tipc_bearers[i].publ.blocked = 1; - } - for (i = 0; i < MAX_BEARERS; i++) { - if (tipc_bearers[i].active) - bearer_disable(tipc_bearers[i].publ.name); + bearer_disable(&tipc_bearers[i]); } media_count = 0; } - - diff --git a/net/tipc/cluster.c b/net/tipc/cluster.c index e68f705381b..7fea14b98b9 100644 --- a/net/tipc/cluster.c +++ b/net/tipc/cluster.c @@ -113,25 +113,6 @@ void tipc_cltr_delete(struct cluster *c_ptr) kfree(c_ptr); } -u32 tipc_cltr_next_node(struct cluster *c_ptr, u32 addr) -{ - struct tipc_node *n_ptr; - u32 n_num = tipc_node(addr) + 1; - - if (!c_ptr) - return addr; - for (; n_num <= c_ptr->highest_node; n_num++) { - n_ptr = c_ptr->nodes[n_num]; - if (n_ptr && tipc_node_has_active_links(n_ptr)) - return n_ptr->addr; - } - for (n_num = 1; n_num < tipc_node(addr); n_num++) { - n_ptr = c_ptr->nodes[n_num]; - if (n_ptr && tipc_node_has_active_links(n_ptr)) - return n_ptr->addr; - } - return 0; -} void tipc_cltr_attach_node(struct cluster *c_ptr, struct tipc_node *n_ptr) { @@ -232,7 +213,7 @@ struct tipc_node *tipc_cltr_select_node(struct cluster *c_ptr, u32 selector) static struct sk_buff *tipc_cltr_prepare_routing_msg(u32 data_size, u32 dest) { u32 size = INT_H_SIZE + data_size; - struct sk_buff *buf = buf_acquire(size); + struct sk_buff *buf = tipc_buf_acquire(size); struct tipc_msg *msg; if (buf) { diff --git a/net/tipc/cluster.h b/net/tipc/cluster.h index 333efb0b9c4..32636d98c9c 100644 --- a/net/tipc/cluster.h +++ b/net/tipc/cluster.h @@ -75,7 +75,7 @@ void tipc_cltr_attach_node(struct cluster *c_ptr, struct tipc_node *n_ptr); void tipc_cltr_send_slave_routes(struct cluster *c_ptr, u32 dest); void tipc_cltr_broadcast(struct sk_buff *buf); int tipc_cltr_init(void); -u32 tipc_cltr_next_node(struct cluster *c_ptr, u32 addr); + void tipc_cltr_bcast_new_route(struct cluster *c_ptr, u32 dest, u32 lo, u32 hi); void tipc_cltr_send_local_routes(struct cluster *c_ptr, u32 dest); void tipc_cltr_bcast_lost_route(struct cluster *c_ptr, u32 dest, u32 lo, u32 hi); diff --git a/net/tipc/config.c b/net/tipc/config.c index 961d1b09714..50a6133a366 100644 --- a/net/tipc/config.c +++ b/net/tipc/config.c @@ -95,7 +95,7 @@ int tipc_cfg_append_tlv(struct sk_buff *buf, int tlv_type, return 1; } -struct sk_buff *tipc_cfg_reply_unsigned_type(u16 tlv_type, u32 value) +static struct sk_buff *tipc_cfg_reply_unsigned_type(u16 tlv_type, u32 value) { struct sk_buff *buf; __be32 value_net; @@ -109,6 +109,11 @@ struct sk_buff *tipc_cfg_reply_unsigned_type(u16 tlv_type, u32 value) return buf; } +static struct sk_buff *tipc_cfg_reply_unsigned(u32 value) +{ + return tipc_cfg_reply_unsigned_type(TIPC_TLV_UNSIGNED, value); +} + struct sk_buff *tipc_cfg_reply_string_type(u16 tlv_type, char *string) { struct sk_buff *buf; @@ -120,139 +125,6 @@ struct sk_buff *tipc_cfg_reply_string_type(u16 tlv_type, char *string) return buf; } - -#if 0 - -/* Now obsolete code for handling commands not yet implemented the new way */ - -/* - * Some of this code assumed that the manager structure contains two added - * fields: - * u32 link_subscriptions; - * struct list_head link_subscribers; - * which are currently not present. These fields may need to be re-introduced - * if and when support for link subscriptions is added. - */ - -void tipc_cfg_link_event(u32 addr, char *name, int up) -{ - /* TIPC DOESN'T HANDLE LINK EVENT SUBSCRIPTIONS AT THE MOMENT */ -} - -int tipc_cfg_cmd(const struct tipc_cmd_msg * msg, - char *data, - u32 sz, - u32 *ret_size, - struct tipc_portid *orig) -{ - int rv = -EINVAL; - u32 cmd = msg->cmd; - - *ret_size = 0; - switch (cmd) { - case TIPC_REMOVE_LINK: - case TIPC_CMD_BLOCK_LINK: - case TIPC_CMD_UNBLOCK_LINK: - if (!cfg_check_connection(orig)) - rv = link_control(msg->argv.link_name, msg->cmd, 0); - break; - case TIPC_ESTABLISH: - { - int connected; - - tipc_isconnected(mng.conn_port_ref, &connected); - if (connected || !orig) { - rv = TIPC_FAILURE; - break; - } - rv = tipc_connect2port(mng.conn_port_ref, orig); - if (rv == TIPC_OK) - orig = 0; - break; - } - case TIPC_GET_PEER_ADDRESS: - *ret_size = link_peer_addr(msg->argv.link_name, data, sz); - break; - case TIPC_GET_ROUTES: - rv = TIPC_OK; - break; - default: {} - } - if (*ret_size) - rv = TIPC_OK; - return rv; -} - -static void cfg_cmd_event(struct tipc_cmd_msg *msg, - char *data, - u32 sz, - struct tipc_portid const *orig) -{ - int rv = -EINVAL; - struct tipc_cmd_result_msg rmsg; - struct iovec msg_sect[2]; - int *arg; - - msg->cmd = ntohl(msg->cmd); - - cfg_prepare_res_msg(msg->cmd, msg->usr_handle, rv, &rmsg, msg_sect, - data, 0); - if (ntohl(msg->magic) != TIPC_MAGIC) - goto exit; - - switch (msg->cmd) { - case TIPC_CREATE_LINK: - if (!cfg_check_connection(orig)) - rv = disc_create_link(&msg->argv.create_link); - break; - case TIPC_LINK_SUBSCRIBE: - { - struct subscr_data *sub; - - if (mng.link_subscriptions > 64) - break; - sub = kmalloc(sizeof(*sub), - GFP_ATOMIC); - if (sub == NULL) { - warn("Memory squeeze; dropped remote link subscription\n"); - break; - } - INIT_LIST_HEAD(&sub->subd_list); - tipc_createport(mng.user_ref, - (void *)sub, - TIPC_HIGH_IMPORTANCE, - 0, - 0, - (tipc_conn_shutdown_event)cfg_linksubscr_cancel, - 0, - 0, - (tipc_conn_msg_event)cfg_linksubscr_cancel, - 0, - &sub->port_ref); - if (!sub->port_ref) { - kfree(sub); - break; - } - memcpy(sub->usr_handle,msg->usr_handle, - sizeof(sub->usr_handle)); - sub->domain = msg->argv.domain; - list_add_tail(&sub->subd_list, &mng.link_subscribers); - tipc_connect2port(sub->port_ref, orig); - rmsg.retval = TIPC_OK; - tipc_send(sub->port_ref, 2u, msg_sect); - mng.link_subscriptions++; - return; - } - default: - rv = tipc_cfg_cmd(msg, data, sz, (u32 *)&msg_sect[1].iov_len, orig); - } -exit: - rmsg.result_len = htonl(msg_sect[1].iov_len); - rmsg.retval = htonl(rv); - tipc_cfg_respond(msg_sect, 2u, orig); -} -#endif - #define MAX_STATS_INFO 2000 static struct sk_buff *tipc_show_stats(void) @@ -557,14 +429,6 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area case TIPC_CMD_SHOW_PORTS: rep_tlv_buf = tipc_port_get_ports(); break; -#if 0 - case TIPC_CMD_SHOW_PORT_STATS: - rep_tlv_buf = port_show_stats(req_tlv_area, req_tlv_space); - break; - case TIPC_CMD_RESET_PORT_STATS: - rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED); - break; -#endif case TIPC_CMD_SET_LOG_SIZE: rep_tlv_buf = tipc_log_resize_cmd(req_tlv_area, req_tlv_space); break; diff --git a/net/tipc/config.h b/net/tipc/config.h index 5cd7cc56c54..481e12ece71 100644 --- a/net/tipc/config.h +++ b/net/tipc/config.h @@ -45,7 +45,6 @@ struct sk_buff *tipc_cfg_reply_alloc(int payload_size); int tipc_cfg_append_tlv(struct sk_buff *buf, int tlv_type, void *tlv_data, int tlv_data_size); -struct sk_buff *tipc_cfg_reply_unsigned_type(u16 tlv_type, u32 value); struct sk_buff *tipc_cfg_reply_string_type(u16 tlv_type, char *string); static inline struct sk_buff *tipc_cfg_reply_none(void) @@ -53,11 +52,6 @@ static inline struct sk_buff *tipc_cfg_reply_none(void) return tipc_cfg_reply_alloc(0); } -static inline struct sk_buff *tipc_cfg_reply_unsigned(u32 value) -{ - return tipc_cfg_reply_unsigned_type(TIPC_TLV_UNSIGNED, value); -} - static inline struct sk_buff *tipc_cfg_reply_error_string(char *string) { return tipc_cfg_reply_string_type(TIPC_TLV_ERROR_STRING, string); diff --git a/net/tipc/core.c b/net/tipc/core.c index 69646811798..e2a09eb8efd 100644 --- a/net/tipc/core.c +++ b/net/tipc/core.c @@ -96,13 +96,8 @@ int tipc_net_id; int tipc_remote_management; -int tipc_get_mode(void) -{ - return tipc_mode; -} - /** - * buf_acquire - creates a TIPC message buffer + * tipc_buf_acquire - creates a TIPC message buffer * @size: message size (including TIPC header) * * Returns a new buffer with data pointers set to the specified size. @@ -111,7 +106,7 @@ int tipc_get_mode(void) * There may also be unrequested tailroom present at the buffer's end. */ -struct sk_buff *buf_acquire(u32 size) +struct sk_buff *tipc_buf_acquire(u32 size) { struct sk_buff *skb; unsigned int buf_size = (BUF_HEADROOM + size + 3) & ~3u; @@ -129,7 +124,7 @@ struct sk_buff *buf_acquire(u32 size) * tipc_core_stop_net - shut down TIPC networking sub-systems */ -void tipc_core_stop_net(void) +static void tipc_core_stop_net(void) { tipc_eth_media_stop(); tipc_net_stop(); @@ -154,7 +149,7 @@ int tipc_core_start_net(unsigned long addr) * tipc_core_stop - switch TIPC from SINGLE NODE to NOT RUNNING mode */ -void tipc_core_stop(void) +static void tipc_core_stop(void) { if (tipc_mode != TIPC_NODE_MODE) return; @@ -169,13 +164,14 @@ void tipc_core_stop(void) tipc_nametbl_stop(); tipc_ref_table_stop(); tipc_socket_stop(); + tipc_log_resize(0); } /** * tipc_core_start - switch TIPC from NOT RUNNING to SINGLE NODE mode */ -int tipc_core_start(void) +static int tipc_core_start(void) { int res; @@ -203,7 +199,9 @@ static int __init tipc_init(void) { int res; - tipc_log_resize(CONFIG_TIPC_LOG); + if (tipc_log_resize(CONFIG_TIPC_LOG) != 0) + warn("Unable to create log buffer\n"); + info("Activated (version " TIPC_MOD_VER " compiled " __DATE__ " " __TIME__ ")\n"); @@ -230,7 +228,6 @@ static void __exit tipc_exit(void) tipc_core_stop_net(); tipc_core_stop(); info("Deactivated\n"); - tipc_log_resize(0); } module_init(tipc_init); @@ -244,8 +241,6 @@ MODULE_VERSION(TIPC_MOD_VER); EXPORT_SYMBOL(tipc_attach); EXPORT_SYMBOL(tipc_detach); -EXPORT_SYMBOL(tipc_get_addr); -EXPORT_SYMBOL(tipc_get_mode); EXPORT_SYMBOL(tipc_createport); EXPORT_SYMBOL(tipc_deleteport); EXPORT_SYMBOL(tipc_ownidentity); @@ -260,23 +255,10 @@ EXPORT_SYMBOL(tipc_withdraw); EXPORT_SYMBOL(tipc_connect2port); EXPORT_SYMBOL(tipc_disconnect); EXPORT_SYMBOL(tipc_shutdown); -EXPORT_SYMBOL(tipc_isconnected); -EXPORT_SYMBOL(tipc_peer); -EXPORT_SYMBOL(tipc_ref_valid); EXPORT_SYMBOL(tipc_send); -EXPORT_SYMBOL(tipc_send_buf); EXPORT_SYMBOL(tipc_send2name); -EXPORT_SYMBOL(tipc_forward2name); -EXPORT_SYMBOL(tipc_send_buf2name); -EXPORT_SYMBOL(tipc_forward_buf2name); EXPORT_SYMBOL(tipc_send2port); -EXPORT_SYMBOL(tipc_forward2port); -EXPORT_SYMBOL(tipc_send_buf2port); -EXPORT_SYMBOL(tipc_forward_buf2port); EXPORT_SYMBOL(tipc_multicast); -/* EXPORT_SYMBOL(tipc_multicast_buf); not available yet */ -EXPORT_SYMBOL(tipc_ispublished); -EXPORT_SYMBOL(tipc_available_nodes); /* TIPC API for external bearers (see tipc_bearer.h) */ @@ -293,6 +275,4 @@ EXPORT_SYMBOL(tipc_createport_raw); EXPORT_SYMBOL(tipc_reject_msg); EXPORT_SYMBOL(tipc_send_buf_fast); EXPORT_SYMBOL(tipc_acknowledge); -EXPORT_SYMBOL(tipc_get_port); -EXPORT_SYMBOL(tipc_get_handle); diff --git a/net/tipc/core.h b/net/tipc/core.h index 188799017ab..e19389e5722 100644 --- a/net/tipc/core.h +++ b/net/tipc/core.h @@ -83,9 +83,7 @@ * Note: TIPC_LOG is configured to echo its output to the system console; * user-defined buffers can be configured to do the same thing. */ - extern struct print_buf *const TIPC_NULL; -extern struct print_buf *const TIPC_CONS; extern struct print_buf *const TIPC_LOG; void tipc_printf(struct print_buf *, const char *fmt, ...); @@ -204,10 +202,7 @@ extern atomic_t tipc_user_count; * Routines available to privileged subsystems */ -extern int tipc_core_start(void); -extern void tipc_core_stop(void); -extern int tipc_core_start_net(unsigned long addr); -extern void tipc_core_stop_net(void); +extern int tipc_core_start_net(unsigned long); extern int tipc_handler_start(void); extern void tipc_handler_stop(void); extern int tipc_netlink_start(void); @@ -328,7 +323,7 @@ static inline struct tipc_msg *buf_msg(struct sk_buff *skb) return (struct tipc_msg *)skb->data; } -extern struct sk_buff *buf_acquire(u32 size); +extern struct sk_buff *tipc_buf_acquire(u32 size); /** * buf_discard - frees a TIPC message buffer diff --git a/net/tipc/dbg.c b/net/tipc/dbg.c index 1885a7edb0c..46f51d208e5 100644 --- a/net/tipc/dbg.c +++ b/net/tipc/dbg.c @@ -52,7 +52,7 @@ static struct print_buf null_buf = { NULL, 0, NULL, 0 }; struct print_buf *const TIPC_NULL = &null_buf; static struct print_buf cons_buf = { NULL, 0, NULL, 1 }; -struct print_buf *const TIPC_CONS = &cons_buf; +static struct print_buf *const TIPC_CONS = &cons_buf; static struct print_buf log_buf = { NULL, 0, NULL, 1 }; struct print_buf *const TIPC_LOG = &log_buf; @@ -76,6 +76,10 @@ struct print_buf *const TIPC_LOG = &log_buf; static char print_string[TIPC_PB_MAX_STR]; static DEFINE_SPINLOCK(print_lock); +static void tipc_printbuf_reset(struct print_buf *pb); +static int tipc_printbuf_empty(struct print_buf *pb); +static void tipc_printbuf_move(struct print_buf *pb_to, + struct print_buf *pb_from); #define FORMAT(PTR,LEN,FMT) \ {\ @@ -116,7 +120,7 @@ void tipc_printbuf_init(struct print_buf *pb, char *raw, u32 size) * @pb: pointer to print buffer structure */ -void tipc_printbuf_reset(struct print_buf *pb) +static void tipc_printbuf_reset(struct print_buf *pb) { if (pb->buf) { pb->crs = pb->buf; @@ -132,9 +136,9 @@ void tipc_printbuf_reset(struct print_buf *pb) * Returns non-zero if print buffer is empty. */ -int tipc_printbuf_empty(struct print_buf *pb) +static int tipc_printbuf_empty(struct print_buf *pb) { - return (!pb->buf || (pb->crs == pb->buf)); + return !pb->buf || (pb->crs == pb->buf); } /** @@ -169,7 +173,7 @@ int tipc_printbuf_validate(struct print_buf *pb) tipc_printf(pb, err); } } - return (pb->crs - pb->buf + 1); + return pb->crs - pb->buf + 1; } /** @@ -181,7 +185,8 @@ int tipc_printbuf_validate(struct print_buf *pb) * Source print buffer becomes empty if a successful move occurs. */ -void tipc_printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from) +static void tipc_printbuf_move(struct print_buf *pb_to, + struct print_buf *pb_from) { int len; diff --git a/net/tipc/dbg.h b/net/tipc/dbg.h index 5ef1bc8f64e..3ba6ba8b434 100644 --- a/net/tipc/dbg.h +++ b/net/tipc/dbg.h @@ -56,10 +56,7 @@ struct print_buf { #define TIPC_PB_MAX_STR 512 /* max printable string (with trailing NUL) */ void tipc_printbuf_init(struct print_buf *pb, char *buf, u32 size); -void tipc_printbuf_reset(struct print_buf *pb); -int tipc_printbuf_empty(struct print_buf *pb); int tipc_printbuf_validate(struct print_buf *pb); -void tipc_printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from); int tipc_log_resize(int log_size); diff --git a/net/tipc/discover.c b/net/tipc/discover.c index fc1fcf5e6b5..4a7cd3719b7 100644 --- a/net/tipc/discover.c +++ b/net/tipc/discover.c @@ -46,16 +46,6 @@ #define TIPC_LINK_REQ_FAST 2000 /* normal delay if bearer has no links */ #define TIPC_LINK_REQ_SLOW 600000 /* normal delay if bearer has links */ -#if 0 -#define GET_NODE_INFO 300 -#define GET_NODE_INFO_RESULT 301 -#define FORWARD_LINK_PROBE 302 -#define LINK_REQUEST_REJECTED 303 -#define LINK_REQUEST_ACCEPTED 304 -#define DROP_LINK_REQUEST 305 -#define CHECK_LINK_COUNT 306 -#endif - /* * TODO: Most of the inter-cluster setup stuff should be * rewritten, and be made conformant with specification. @@ -78,30 +68,6 @@ struct link_req { unsigned int timer_intv; }; - -#if 0 -int disc_create_link(const struct tipc_link_create *argv) -{ - /* - * Code for inter cluster link setup here - */ - return TIPC_OK; -} -#endif - -/* - * disc_lost_link(): A link has lost contact - */ - -void tipc_disc_link_event(u32 addr, char *name, int up) -{ - if (in_own_cluster(addr)) - return; - /* - * Code for inter cluster link setup here - */ -} - /** * tipc_disc_init_msg - initialize a link setup message * @type: message type (request or response) @@ -115,7 +81,7 @@ static struct sk_buff *tipc_disc_init_msg(u32 type, u32 dest_domain, struct bearer *b_ptr) { - struct sk_buff *buf = buf_acquire(DSC_H_SIZE); + struct sk_buff *buf = tipc_buf_acquire(DSC_H_SIZE); struct tipc_msg *msg; if (buf) { @@ -203,6 +169,14 @@ void tipc_disc_recv_msg(struct sk_buff *buf, struct bearer *b_ptr) return; } spin_lock_bh(&n_ptr->lock); + + /* Don't talk to neighbor during cleanup after last session */ + + if (n_ptr->cleanup_required) { + spin_unlock_bh(&n_ptr->lock); + return; + } + link = n_ptr->links[b_ptr->identity]; if (!link) { dbg("creating link\n"); diff --git a/net/tipc/discover.h b/net/tipc/discover.h index c36eaeb7d5d..f8e75063612 100644 --- a/net/tipc/discover.h +++ b/net/tipc/discover.h @@ -50,9 +50,4 @@ void tipc_disc_stop_link_req(struct link_req *req); void tipc_disc_recv_msg(struct sk_buff *buf, struct bearer *b_ptr); -void tipc_disc_link_event(u32 addr, char *name, int up); -#if 0 -int disc_create_link(const struct tipc_link_create *argv); -#endif - #endif diff --git a/net/tipc/eth_media.c b/net/tipc/eth_media.c index 6230d16020c..6e988ba485f 100644 --- a/net/tipc/eth_media.c +++ b/net/tipc/eth_media.c @@ -72,17 +72,26 @@ static int send_msg(struct sk_buff *buf, struct tipc_bearer *tb_ptr, { struct sk_buff *clone; struct net_device *dev; + int delta; clone = skb_clone(buf, GFP_ATOMIC); - if (clone) { - skb_reset_network_header(clone); - dev = ((struct eth_bearer *)(tb_ptr->usr_handle))->dev; - clone->dev = dev; - dev_hard_header(clone, dev, ETH_P_TIPC, - &dest->dev_addr.eth_addr, - dev->dev_addr, clone->len); - dev_queue_xmit(clone); + if (!clone) + return 0; + + dev = ((struct eth_bearer *)(tb_ptr->usr_handle))->dev; + delta = dev->hard_header_len - skb_headroom(buf); + + if ((delta > 0) && + pskb_expand_head(clone, SKB_DATA_ALIGN(delta), 0, GFP_ATOMIC)) { + kfree_skb(clone); + return 0; } + + skb_reset_network_header(clone); + clone->dev = dev; + dev_hard_header(clone, dev, ETH_P_TIPC, &dest->dev_addr.eth_addr, + dev->dev_addr, clone->len); + dev_queue_xmit(clone); return 0; } @@ -92,15 +101,12 @@ static int send_msg(struct sk_buff *buf, struct tipc_bearer *tb_ptr, * Accept only packets explicitly sent to this node, or broadcast packets; * ignores packets sent using Ethernet multicast, and traffic sent to other * nodes (which can happen if interface is running in promiscuous mode). - * Routine truncates any Ethernet padding/CRC appended to the message, - * and ensures message size matches actual length */ static int recv_msg(struct sk_buff *buf, struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev) { struct eth_bearer *eb_ptr = (struct eth_bearer *)pt->af_packet_priv; - u32 size; if (!net_eq(dev_net(dev), &init_net)) { kfree_skb(buf); @@ -109,13 +115,9 @@ static int recv_msg(struct sk_buff *buf, struct net_device *dev, if (likely(eb_ptr->bearer)) { if (likely(buf->pkt_type <= PACKET_BROADCAST)) { - size = msg_size((struct tipc_msg *)buf->data); - skb_trim(buf, size); - if (likely(buf->len == size)) { - buf->next = NULL; - tipc_recv_msg(buf, eb_ptr->bearer); - return 0; - } + buf->next = NULL; + tipc_recv_msg(buf, eb_ptr->bearer); + return 0; } } kfree_skb(buf); @@ -133,6 +135,16 @@ static int enable_bearer(struct tipc_bearer *tb_ptr) struct eth_bearer *eb_ptr = ð_bearers[0]; struct eth_bearer *stop = ð_bearers[MAX_ETH_BEARERS]; char *driver_name = strchr((const char *)tb_ptr->name, ':') + 1; + int pending_dev = 0; + + /* Find unused Ethernet bearer structure */ + + while (eb_ptr->dev) { + if (!eb_ptr->bearer) + pending_dev++; + if (++eb_ptr == stop) + return pending_dev ? -EAGAIN : -EDQUOT; + } /* Find device with specified name */ diff --git a/net/tipc/link.c b/net/tipc/link.c index a3616b99529..b31992ccd5d 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -99,23 +99,6 @@ struct link_name { char if_peer[TIPC_MAX_IF_NAME]; }; -#if 0 - -/* LINK EVENT CODE IS NOT SUPPORTED AT PRESENT */ - -/** - * struct link_event - link up/down event notification - */ - -struct link_event { - u32 addr; - int up; - void (*fcn)(u32, char *, int); - char name[TIPC_MAX_LINK_NAME]; -}; - -#endif - static void link_handle_out_of_seq_msg(struct link *l_ptr, struct sk_buff *buf); static void link_recv_proto_msg(struct link *l_ptr, struct sk_buff *buf); @@ -129,6 +112,9 @@ static void link_state_event(struct link *l_ptr, u32 event); static void link_reset_statistics(struct link *l_ptr); static void link_print(struct link *l_ptr, struct print_buf *buf, const char *str); +static void link_start(struct link *l_ptr); +static int link_send_long_buf(struct link *l_ptr, struct sk_buff *buf); + /* * Debugging code used by link routines only @@ -239,13 +225,13 @@ int tipc_link_is_up(struct link *l_ptr) { if (!l_ptr) return 0; - return (link_working_working(l_ptr) || link_working_unknown(l_ptr)); + return link_working_working(l_ptr) || link_working_unknown(l_ptr); } int tipc_link_is_active(struct link *l_ptr) { - return ((l_ptr->owner->active_links[0] == l_ptr) || - (l_ptr->owner->active_links[1] == l_ptr)); + return (l_ptr->owner->active_links[0] == l_ptr) || + (l_ptr->owner->active_links[1] == l_ptr); } /** @@ -459,7 +445,7 @@ struct link *tipc_link_create(struct bearer *b_ptr, const u32 peer, k_init_timer(&l_ptr->timer, (Handler)link_timeout, (unsigned long)l_ptr); list_add_tail(&l_ptr->link_list, &b_ptr->links); - tipc_k_signal((Handler)tipc_link_start, (unsigned long)l_ptr); + tipc_k_signal((Handler)link_start, (unsigned long)l_ptr); dbg("tipc_link_create(): tolerance = %u,cont intv = %u, abort_limit = %u\n", l_ptr->tolerance, l_ptr->continuity_interval, l_ptr->abort_limit); @@ -499,9 +485,9 @@ void tipc_link_delete(struct link *l_ptr) kfree(l_ptr); } -void tipc_link_start(struct link *l_ptr) +static void link_start(struct link *l_ptr) { - dbg("tipc_link_start %x\n", l_ptr); + dbg("link_start %x\n", l_ptr); link_state_event(l_ptr, STARTING_EVT); } @@ -634,39 +620,9 @@ void tipc_link_stop(struct link *l_ptr) l_ptr->proto_msg_queue = NULL; } -#if 0 - /* LINK EVENT CODE IS NOT SUPPORTED AT PRESENT */ - -static void link_recv_event(struct link_event *ev) -{ - ev->fcn(ev->addr, ev->name, ev->up); - kfree(ev); -} - -static void link_send_event(void (*fcn)(u32 a, char *n, int up), - struct link *l_ptr, int up) -{ - struct link_event *ev; - - ev = kmalloc(sizeof(*ev), GFP_ATOMIC); - if (!ev) { - warn("Link event allocation failure\n"); - return; - } - ev->addr = l_ptr->addr; - ev->up = up; - ev->fcn = fcn; - memcpy(ev->name, l_ptr->name, TIPC_MAX_LINK_NAME); - tipc_k_signal((Handler)link_recv_event, (unsigned long)ev); -} - -#else - #define link_send_event(fcn, l_ptr, up) do { } while (0) -#endif - void tipc_link_reset(struct link *l_ptr) { struct sk_buff *buf; @@ -690,10 +646,7 @@ void tipc_link_reset(struct link *l_ptr) tipc_node_link_down(l_ptr->owner, l_ptr); tipc_bearer_remove_dest(l_ptr->b_ptr, l_ptr->addr); -#if 0 - tipc_printf(TIPC_CONS, "\nReset link <%s>\n", l_ptr->name); - dbg_link_dump(); -#endif + if (was_active_link && tipc_node_has_active_links(l_ptr->owner) && l_ptr->owner->permit_changeover) { l_ptr->reset_checkpoint = checkpoint; @@ -1050,7 +1003,7 @@ int tipc_link_send_buf(struct link *l_ptr, struct sk_buff *buf) /* Fragmentation needed ? */ if (size > max_packet) - return tipc_link_send_long_buf(l_ptr, buf); + return link_send_long_buf(l_ptr, buf); /* Packet can be queued or sent: */ @@ -1086,7 +1039,7 @@ int tipc_link_send_buf(struct link *l_ptr, struct sk_buff *buf) /* Try creating a new bundle */ if (size <= max_packet * 2 / 3) { - struct sk_buff *bundler = buf_acquire(max_packet); + struct sk_buff *bundler = tipc_buf_acquire(max_packet); struct tipc_msg bundler_hdr; if (bundler) { @@ -1362,7 +1315,7 @@ again: /* Prepare header of first fragment: */ - buf_chain = buf = buf_acquire(max_pkt); + buf_chain = buf = tipc_buf_acquire(max_pkt); if (!buf) return -ENOMEM; buf->next = NULL; @@ -1419,7 +1372,7 @@ error: msg_set_size(&fragm_hdr, fragm_sz + INT_H_SIZE); msg_set_fragm_no(&fragm_hdr, ++fragm_no); prev = buf; - buf = buf_acquire(fragm_sz + INT_H_SIZE); + buf = tipc_buf_acquire(fragm_sz + INT_H_SIZE); if (!buf) goto error; @@ -1802,6 +1755,15 @@ static int link_recv_buf_validate(struct sk_buff *buf) return pskb_may_pull(buf, hdr_size); } +/** + * tipc_recv_msg - process TIPC messages arriving from off-node + * @head: pointer to message buffer chain + * @tb_ptr: pointer to bearer message arrived on + * + * Invoked with no locks held. Bearer pointer must point to a valid bearer + * structure (i.e. cannot be NULL), but bearer can be inactive. + */ + void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *tb_ptr) { read_lock_bh(&tipc_net_lock); @@ -1819,6 +1781,11 @@ void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *tb_ptr) head = head->next; + /* Ensure bearer is still enabled */ + + if (unlikely(!b_ptr->active)) + goto cont; + /* Ensure message is well-formed */ if (unlikely(!link_recv_buf_validate(buf))) @@ -1855,13 +1822,22 @@ void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *tb_ptr) goto cont; } - /* Locate unicast link endpoint that should handle message */ + /* Locate neighboring node that sent message */ n_ptr = tipc_node_find(msg_prevnode(msg)); if (unlikely(!n_ptr)) goto cont; tipc_node_lock(n_ptr); + /* Don't talk to neighbor during cleanup after last session */ + + if (n_ptr->cleanup_required) { + tipc_node_unlock(n_ptr); + goto cont; + } + + /* Locate unicast link endpoint that should handle message */ + l_ptr = n_ptr->links[b_ptr->identity]; if (unlikely(!l_ptr)) { tipc_node_unlock(n_ptr); @@ -2172,7 +2148,7 @@ void tipc_link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int probe_msg, if (tipc_bearer_congested(l_ptr->b_ptr, l_ptr)) { if (!l_ptr->proto_msg_queue) { l_ptr->proto_msg_queue = - buf_acquire(sizeof(l_ptr->proto_msg)); + tipc_buf_acquire(sizeof(l_ptr->proto_msg)); } buf = l_ptr->proto_msg_queue; if (!buf) @@ -2186,7 +2162,7 @@ void tipc_link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int probe_msg, msg_dbg(msg, ">>"); - buf = buf_acquire(msg_size); + buf = tipc_buf_acquire(msg_size); if (!buf) return; @@ -2345,10 +2321,10 @@ exit: * tipc_link_tunnel(): Send one message via a link belonging to * another bearer. Owner node is locked. */ -void tipc_link_tunnel(struct link *l_ptr, - struct tipc_msg *tunnel_hdr, - struct tipc_msg *msg, - u32 selector) +static void tipc_link_tunnel(struct link *l_ptr, + struct tipc_msg *tunnel_hdr, + struct tipc_msg *msg, + u32 selector) { struct link *tunnel; struct sk_buff *buf; @@ -2361,7 +2337,7 @@ void tipc_link_tunnel(struct link *l_ptr, return; } msg_set_size(tunnel_hdr, length + INT_H_SIZE); - buf = buf_acquire(length + INT_H_SIZE); + buf = tipc_buf_acquire(length + INT_H_SIZE); if (!buf) { warn("Link changeover error, " "unable to send tunnel msg\n"); @@ -2407,7 +2383,7 @@ void tipc_link_changeover(struct link *l_ptr) if (!l_ptr->first_out) { struct sk_buff *buf; - buf = buf_acquire(INT_H_SIZE); + buf = tipc_buf_acquire(INT_H_SIZE); if (buf) { skb_copy_to_linear_data(buf, &tunnel_hdr, INT_H_SIZE); msg_set_size(&tunnel_hdr, INT_H_SIZE); @@ -2468,7 +2444,7 @@ void tipc_link_send_duplicate(struct link *l_ptr, struct link *tunnel) msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); /* Update */ msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); msg_set_size(&tunnel_hdr, length + INT_H_SIZE); - outbuf = buf_acquire(length + INT_H_SIZE); + outbuf = tipc_buf_acquire(length + INT_H_SIZE); if (outbuf == NULL) { warn("Link changeover error, " "unable to send duplicate msg\n"); @@ -2504,7 +2480,7 @@ static struct sk_buff *buf_extract(struct sk_buff *skb, u32 from_pos) u32 size = msg_size(msg); struct sk_buff *eb; - eb = buf_acquire(size); + eb = tipc_buf_acquire(size); if (eb) skb_copy_to_linear_data(eb, msg, size); return eb; @@ -2632,11 +2608,11 @@ void tipc_link_recv_bundle(struct sk_buff *buf) /* - * tipc_link_send_long_buf: Entry for buffers needing fragmentation. + * link_send_long_buf: Entry for buffers needing fragmentation. * The buffer is complete, inclusive total message length. * Returns user data length. */ -int tipc_link_send_long_buf(struct link *l_ptr, struct sk_buff *buf) +static int link_send_long_buf(struct link *l_ptr, struct sk_buff *buf) { struct tipc_msg *inmsg = buf_msg(buf); struct tipc_msg fragm_hdr; @@ -2675,7 +2651,7 @@ int tipc_link_send_long_buf(struct link *l_ptr, struct sk_buff *buf) fragm_sz = rest; msg_set_type(&fragm_hdr, LAST_FRAGMENT); } - fragm = buf_acquire(fragm_sz + INT_H_SIZE); + fragm = tipc_buf_acquire(fragm_sz + INT_H_SIZE); if (fragm == NULL) { warn("Link unable to fragment message\n"); dsz = -ENOMEM; @@ -2780,7 +2756,7 @@ int tipc_link_recv_fragment(struct sk_buff **pending, struct sk_buff **fb, buf_discard(fbuf); return 0; } - pbuf = buf_acquire(msg_size(imsg)); + pbuf = tipc_buf_acquire(msg_size(imsg)); if (pbuf != NULL) { pbuf->next = *pending; *pending = pbuf; @@ -3174,44 +3150,6 @@ struct sk_buff *tipc_link_cmd_show_stats(const void *req_tlv_area, int req_tlv_s return buf; } -#if 0 -int link_control(const char *name, u32 op, u32 val) -{ - int res = -EINVAL; - struct link *l_ptr; - u32 bearer_id; - struct tipc_node * node; - u32 a; - - a = link_name2addr(name, &bearer_id); - read_lock_bh(&tipc_net_lock); - node = tipc_node_find(a); - if (node) { - tipc_node_lock(node); - l_ptr = node->links[bearer_id]; - if (l_ptr) { - if (op == TIPC_REMOVE_LINK) { - struct bearer *b_ptr = l_ptr->b_ptr; - spin_lock_bh(&b_ptr->publ.lock); - tipc_link_delete(l_ptr); - spin_unlock_bh(&b_ptr->publ.lock); - } - if (op == TIPC_CMD_BLOCK_LINK) { - tipc_link_reset(l_ptr); - l_ptr->blocked = 1; - } - if (op == TIPC_CMD_UNBLOCK_LINK) { - l_ptr->blocked = 0; - } - res = 0; - } - tipc_node_unlock(node); - } - read_unlock_bh(&tipc_net_lock); - return res; -} -#endif - /** * tipc_link_get_max_pkt - get maximum packet size to use when sending to destination * @dest: network address of destination node @@ -3242,28 +3180,6 @@ u32 tipc_link_get_max_pkt(u32 dest, u32 selector) return res; } -#if 0 -static void link_dump_rec_queue(struct link *l_ptr) -{ - struct sk_buff *crs; - - if (!l_ptr->oldest_deferred_in) { - info("Reception queue empty\n"); - return; - } - info("Contents of Reception queue:\n"); - crs = l_ptr->oldest_deferred_in; - while (crs) { - if (crs->data == (void *)0x0000a3a3) { - info("buffer %x invalid\n", crs); - return; - } - msg_dbg(buf_msg(crs), "In rec queue:\n"); - crs = crs->next; - } -} -#endif - static void link_dump_send_queue(struct link *l_ptr) { if (l_ptr->next_out) { diff --git a/net/tipc/link.h b/net/tipc/link.h index 2e5385c47d3..f98bc613de6 100644 --- a/net/tipc/link.h +++ b/net/tipc/link.h @@ -210,10 +210,6 @@ struct link { u32 msg_length_counts; u32 msg_lengths_total; u32 msg_length_profile[7]; -#if 0 - u32 sent_tunneled; - u32 recv_tunneled; -#endif } stats; struct print_buf print_buf; @@ -229,7 +225,6 @@ void tipc_link_send_duplicate(struct link *l_ptr, struct link *dest); void tipc_link_reset_fragments(struct link *l_ptr); int tipc_link_is_up(struct link *l_ptr); int tipc_link_is_active(struct link *l_ptr); -void tipc_link_start(struct link *l_ptr); u32 tipc_link_push_packet(struct link *l_ptr); void tipc_link_stop(struct link *l_ptr); struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, int req_tlv_space, u16 cmd); @@ -243,9 +238,6 @@ int tipc_link_send_sections_fast(struct port* sender, struct iovec const *msg_sect, const u32 num_sect, u32 destnode); -int tipc_link_send_long_buf(struct link *l_ptr, struct sk_buff *buf); -void tipc_link_tunnel(struct link *l_ptr, struct tipc_msg *tnl_hdr, - struct tipc_msg *msg, u32 selector); void tipc_link_recv_bundle(struct sk_buff *buf); int tipc_link_recv_fragment(struct sk_buff **pending, struct sk_buff **fb, @@ -279,12 +271,12 @@ static inline int between(u32 lower, u32 upper, u32 n) static inline int less_eq(u32 left, u32 right) { - return (mod(right - left) < 32768u); + return mod(right - left) < 32768u; } static inline int less(u32 left, u32 right) { - return (less_eq(left, right) && (mod(right) != mod(left))); + return less_eq(left, right) && (mod(right) != mod(left)); } static inline u32 lesser(u32 left, u32 right) @@ -299,32 +291,32 @@ static inline u32 lesser(u32 left, u32 right) static inline int link_working_working(struct link *l_ptr) { - return (l_ptr->state == WORKING_WORKING); + return l_ptr->state == WORKING_WORKING; } static inline int link_working_unknown(struct link *l_ptr) { - return (l_ptr->state == WORKING_UNKNOWN); + return l_ptr->state == WORKING_UNKNOWN; } static inline int link_reset_unknown(struct link *l_ptr) { - return (l_ptr->state == RESET_UNKNOWN); + return l_ptr->state == RESET_UNKNOWN; } static inline int link_reset_reset(struct link *l_ptr) { - return (l_ptr->state == RESET_RESET); + return l_ptr->state == RESET_RESET; } static inline int link_blocked(struct link *l_ptr) { - return (l_ptr->exp_msg_count || l_ptr->blocked); + return l_ptr->exp_msg_count || l_ptr->blocked; } static inline int link_congested(struct link *l_ptr) { - return (l_ptr->out_queue_size >= l_ptr->queue_limit[0]); + return l_ptr->out_queue_size >= l_ptr->queue_limit[0]; } #endif diff --git a/net/tipc/msg.c b/net/tipc/msg.c index 381063817b4..ecb532fb035 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -112,7 +112,7 @@ int tipc_msg_build(struct tipc_msg *hdr, return dsz; } - *buf = buf_acquire(sz); + *buf = tipc_buf_acquire(sz); if (!(*buf)) return -ENOMEM; skb_copy_to_linear_data(*buf, hdr, hsz); diff --git a/net/tipc/msg.h b/net/tipc/msg.h index 995d2da35b0..031aad18efc 100644 --- a/net/tipc/msg.h +++ b/net/tipc/msg.h @@ -104,7 +104,7 @@ static inline u32 msg_user(struct tipc_msg *m) static inline u32 msg_isdata(struct tipc_msg *m) { - return (msg_user(m) <= TIPC_CRITICAL_IMPORTANCE); + return msg_user(m) <= TIPC_CRITICAL_IMPORTANCE; } static inline void msg_set_user(struct tipc_msg *m, u32 n) @@ -289,7 +289,7 @@ static inline void msg_set_destnode(struct tipc_msg *m, u32 a) static inline int msg_is_dest(struct tipc_msg *m, u32 d) { - return(msg_short(m) || (msg_destnode(m) == d)); + return msg_short(m) || (msg_destnode(m) == d); } static inline u32 msg_routed(struct tipc_msg *m) @@ -632,7 +632,7 @@ static inline void msg_set_bcast_tag(struct tipc_msg *m, u32 n) static inline u32 msg_max_pkt(struct tipc_msg *m) { - return (msg_bits(m, 9, 16, 0xffff) * 4); + return msg_bits(m, 9, 16, 0xffff) * 4; } static inline void msg_set_max_pkt(struct tipc_msg *m, u32 n) diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c index 6ac3c543250..7b907171f87 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -98,7 +98,7 @@ static void publ_to_item(struct distr_item *i, struct publication *p) static struct sk_buff *named_prepare_buf(u32 type, u32 size, u32 dest) { - struct sk_buff *buf = buf_acquire(LONG_H_SIZE + size); + struct sk_buff *buf = tipc_buf_acquire(LONG_H_SIZE + size); struct tipc_msg *msg; if (buf != NULL) { diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c index 8ba79620db3..3a8de4334da 100644 --- a/net/tipc/name_table.c +++ b/net/tipc/name_table.c @@ -116,7 +116,7 @@ DEFINE_RWLOCK(tipc_nametbl_lock); static int hash(int x) { - return(x & (tipc_nametbl_size - 1)); + return x & (tipc_nametbl_size - 1); } /** @@ -613,8 +613,7 @@ struct publication *tipc_nametbl_remove_publ(u32 type, u32 lower, } /* - * tipc_nametbl_translate(): Translate tipc_name -> tipc_portid. - * Very time-critical. + * tipc_nametbl_translate - translate name to port id * * Note: on entry 'destnode' is the search domain used during translation; * on exit it passes back the node address of the matching port (if any) @@ -685,7 +684,6 @@ found: } spin_unlock_bh(&seq->lock); not_found: - *destnode = 0; read_unlock_bh(&tipc_nametbl_lock); return 0; } @@ -877,7 +875,7 @@ static void subseq_list(struct sub_seq *sseq, struct print_buf *buf, u32 depth, u32 index) { char portIdStr[27]; - char *scopeStr; + const char *scope_str[] = {"", " zone", " cluster", " node"}; struct publication *publ = sseq->zone_list; tipc_printf(buf, "%-10u %-10u ", sseq->lower, sseq->upper); @@ -893,15 +891,8 @@ static void subseq_list(struct sub_seq *sseq, struct print_buf *buf, u32 depth, tipc_node(publ->node), publ->ref); tipc_printf(buf, "%-26s ", portIdStr); if (depth > 3) { - if (publ->node != tipc_own_addr) - scopeStr = ""; - else if (publ->scope == TIPC_NODE_SCOPE) - scopeStr = "node"; - else if (publ->scope == TIPC_CLUSTER_SCOPE) - scopeStr = "cluster"; - else - scopeStr = "zone"; - tipc_printf(buf, "%-10u %s", publ->key, scopeStr); + tipc_printf(buf, "%-10u %s", publ->key, + scope_str[publ->scope]); } publ = publ->zone_list_next; @@ -951,24 +942,19 @@ static void nameseq_list(struct name_seq *seq, struct print_buf *buf, u32 depth, static void nametbl_header(struct print_buf *buf, u32 depth) { - tipc_printf(buf, "Type "); - - if (depth > 1) - tipc_printf(buf, "Lower Upper "); - if (depth > 2) - tipc_printf(buf, "Port Identity "); - if (depth > 3) - tipc_printf(buf, "Publication"); - - tipc_printf(buf, "\n-----------"); - - if (depth > 1) - tipc_printf(buf, "--------------------- "); - if (depth > 2) - tipc_printf(buf, "-------------------------- "); - if (depth > 3) - tipc_printf(buf, "------------------"); - + const char *header[] = { + "Type ", + "Lower Upper ", + "Port Identity ", + "Publication Scope" + }; + + int i; + + if (depth > 4) + depth = 4; + for (i = 0; i < depth; i++) + tipc_printf(buf, header[i]); tipc_printf(buf, "\n"); } @@ -1023,16 +1009,6 @@ static void nametbl_list(struct print_buf *buf, u32 depth_info, } } -#if 0 -void tipc_nametbl_print(struct print_buf *buf, const char *str) -{ - tipc_printf(buf, str); - read_lock_bh(&tipc_nametbl_lock); - nametbl_list(buf, 0, 0, 0, 0); - read_unlock_bh(&tipc_nametbl_lock); -} -#endif - #define MAX_NAME_TBL_QUERY 32768 struct sk_buff *tipc_nametbl_get(const void *req_tlv_area, int req_tlv_space) @@ -1065,13 +1041,6 @@ struct sk_buff *tipc_nametbl_get(const void *req_tlv_area, int req_tlv_space) return buf; } -#if 0 -void tipc_nametbl_dump(void) -{ - nametbl_list(TIPC_CONS, 0, 0, 0, 0); -} -#endif - int tipc_nametbl_init(void) { table.types = kcalloc(tipc_nametbl_size, sizeof(struct hlist_head), diff --git a/net/tipc/net.c b/net/tipc/net.c index f61b7694138..1a621cfd660 100644 --- a/net/tipc/net.c +++ b/net/tipc/net.c @@ -129,15 +129,6 @@ u32 tipc_net_select_router(u32 addr, u32 ref) return tipc_zone_select_router(tipc_net.zones[tipc_zone(addr)], addr, ref); } -#if 0 -u32 tipc_net_next_node(u32 a) -{ - if (tipc_net.zones[tipc_zone(a)]) - return tipc_zone_next_node(a); - return 0; -} -#endif - void tipc_net_remove_as_router(u32 router) { u32 z_num; @@ -248,6 +239,7 @@ void tipc_net_route_msg(struct sk_buff *buf) /* Handle message for another node */ msg_dbg(msg, "NET>SEND>: "); + skb_trim(buf, msg_size(msg)); tipc_link_send(buf, dnode, msg_link_selector(msg)); } diff --git a/net/tipc/node.c b/net/tipc/node.c index b634942caba..b4d87eb2dc5 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -50,7 +50,8 @@ void node_print(struct print_buf *buf, struct tipc_node *n_ptr, char *str); static void node_lost_contact(struct tipc_node *n_ptr); static void node_established_contact(struct tipc_node *n_ptr); -struct tipc_node *tipc_nodes = NULL; /* sorted list of nodes within cluster */ +/* sorted list of nodes within cluster */ +static struct tipc_node *tipc_nodes = NULL; static DEFINE_SPINLOCK(node_create_lock); @@ -125,16 +126,6 @@ void tipc_node_delete(struct tipc_node *n_ptr) if (!n_ptr) return; -#if 0 - /* Not needed because links are already deleted via tipc_bearer_stop() */ - - u32 l_num; - - for (l_num = 0; l_num < MAX_BEARERS; l_num++) { - link_delete(n_ptr->links[l_num]); - } -#endif - dbg("node %x deleted\n", n_ptr->addr); kfree(n_ptr); } @@ -237,23 +228,22 @@ void tipc_node_link_down(struct tipc_node *n_ptr, struct link *l_ptr) int tipc_node_has_active_links(struct tipc_node *n_ptr) { - return (n_ptr && - ((n_ptr->active_links[0]) || (n_ptr->active_links[1]))); + return n_ptr->active_links[0] != NULL; } int tipc_node_has_redundant_links(struct tipc_node *n_ptr) { - return (n_ptr->working_links > 1); + return n_ptr->working_links > 1; } static int tipc_node_has_active_routes(struct tipc_node *n_ptr) { - return (n_ptr && (n_ptr->last_router >= 0)); + return n_ptr && (n_ptr->last_router >= 0); } int tipc_node_is_up(struct tipc_node *n_ptr) { - return (tipc_node_has_active_links(n_ptr) || tipc_node_has_active_routes(n_ptr)); + return tipc_node_has_active_links(n_ptr) || tipc_node_has_active_routes(n_ptr); } struct tipc_node *tipc_node_attach_link(struct link *l_ptr) @@ -384,6 +374,20 @@ static void node_established_contact(struct tipc_node *n_ptr) tipc_highest_allowed_slave); } +static void node_cleanup_finished(unsigned long node_addr) +{ + struct tipc_node *n_ptr; + + read_lock_bh(&tipc_net_lock); + n_ptr = tipc_node_find(node_addr); + if (n_ptr) { + tipc_node_lock(n_ptr); + n_ptr->cleanup_required = 0; + tipc_node_unlock(n_ptr); + } + read_unlock_bh(&tipc_net_lock); +} + static void node_lost_contact(struct tipc_node *n_ptr) { struct cluster *c_ptr; @@ -458,6 +462,11 @@ static void node_lost_contact(struct tipc_node *n_ptr) tipc_k_signal((Handler)ns->handle_node_down, (unsigned long)ns->usr_handle); } + + /* Prevent re-contact with node until all cleanup is done */ + + n_ptr->cleanup_required = 1; + tipc_k_signal((Handler)node_cleanup_finished, n_ptr->addr); } /** @@ -579,38 +588,6 @@ void tipc_node_remove_router(struct tipc_node *n_ptr, u32 router) node_lost_contact(n_ptr); } -#if 0 -void node_print(struct print_buf *buf, struct tipc_node *n_ptr, char *str) -{ - u32 i; - - tipc_printf(buf, "\n\n%s", str); - for (i = 0; i < MAX_BEARERS; i++) { - if (!n_ptr->links[i]) - continue; - tipc_printf(buf, "Links[%u]: %x, ", i, n_ptr->links[i]); - } - tipc_printf(buf, "Active links: [%x,%x]\n", - n_ptr->active_links[0], n_ptr->active_links[1]); -} -#endif - -u32 tipc_available_nodes(const u32 domain) -{ - struct tipc_node *n_ptr; - u32 cnt = 0; - - read_lock_bh(&tipc_net_lock); - for (n_ptr = tipc_nodes; n_ptr; n_ptr = n_ptr->next) { - if (!tipc_in_scope(domain, n_ptr->addr)) - continue; - if (tipc_node_is_up(n_ptr)) - cnt++; - } - read_unlock_bh(&tipc_net_lock); - return cnt; -} - struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space) { u32 domain; diff --git a/net/tipc/node.h b/net/tipc/node.h index 6f990da5d14..fff331b2d26 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -52,6 +52,7 @@ * @active_links: pointers to active links to node * @links: pointers to all links to node * @working_links: number of working links to node (both active and standby) + * @cleanup_required: non-zero if cleaning up after a prior loss of contact * @link_cnt: number of links to node * @permit_changeover: non-zero if node has redundant links to this system * @routers: bitmap (used for multicluster communication) @@ -78,6 +79,7 @@ struct tipc_node { struct link *links[MAX_BEARERS]; int link_cnt; int working_links; + int cleanup_required; int permit_changeover; u32 routers[512/32]; int last_router; @@ -94,7 +96,6 @@ struct tipc_node { } bclink; }; -extern struct tipc_node *tipc_nodes; extern u32 tipc_own_tag; struct tipc_node *tipc_node_create(u32 addr); diff --git a/net/tipc/port.c b/net/tipc/port.c index 0737680e926..82092eaa153 100644 --- a/net/tipc/port.c +++ b/net/tipc/port.c @@ -293,34 +293,6 @@ int tipc_deleteport(u32 ref) return 0; } -/** - * tipc_get_port() - return port associated with 'ref' - * - * Note: Port is not locked. - */ - -struct tipc_port *tipc_get_port(const u32 ref) -{ - return (struct tipc_port *)tipc_ref_deref(ref); -} - -/** - * tipc_get_handle - return user handle associated to port 'ref' - */ - -void *tipc_get_handle(const u32 ref) -{ - struct port *p_ptr; - void * handle; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return NULL; - handle = p_ptr->publ.usr_handle; - tipc_port_unlock(p_ptr); - return handle; -} - static int port_unreliable(struct port *p_ptr) { return msg_src_droppable(&p_ptr->publ.phdr); @@ -392,7 +364,7 @@ static struct sk_buff *port_build_proto_msg(u32 destport, u32 destnode, struct sk_buff *buf; struct tipc_msg *msg; - buf = buf_acquire(LONG_H_SIZE); + buf = tipc_buf_acquire(LONG_H_SIZE); if (buf) { msg = buf_msg(buf); tipc_msg_init(msg, usr, type, LONG_H_SIZE, destnode); @@ -433,7 +405,7 @@ int tipc_reject_msg(struct sk_buff *buf, u32 err) hdr_sz = MCAST_H_SIZE; else hdr_sz = LONG_H_SIZE; - rbuf = buf_acquire(data_sz + hdr_sz); + rbuf = tipc_buf_acquire(data_sz + hdr_sz); if (rbuf == NULL) { buf_discard(buf); return data_sz; @@ -588,19 +560,10 @@ void tipc_port_recv_proto_msg(struct sk_buff *buf) if (!p_ptr) { err = TIPC_ERR_NO_PORT; } else if (p_ptr->publ.connected) { - if (port_peernode(p_ptr) != msg_orignode(msg)) + if ((port_peernode(p_ptr) != msg_orignode(msg)) || + (port_peerport(p_ptr) != msg_origport(msg))) { err = TIPC_ERR_NO_PORT; - if (port_peerport(p_ptr) != msg_origport(msg)) - err = TIPC_ERR_NO_PORT; - if (!err && msg_routed(msg)) { - u32 seqno = msg_transp_seqno(msg); - u32 myno = ++p_ptr->last_in_seqno; - if (seqno != myno) { - err = TIPC_ERR_NO_PORT; - abort_buf = port_build_self_abort_msg(p_ptr, err); - } - } - if (msg_type(msg) == CONN_ACK) { + } else if (msg_type(msg) == CONN_ACK) { int wakeup = tipc_port_congested(p_ptr) && p_ptr->publ.congested && p_ptr->wakeup; @@ -719,50 +682,6 @@ struct sk_buff *tipc_port_get_ports(void) return buf; } -#if 0 - -#define MAX_PORT_STATS 2000 - -struct sk_buff *port_show_stats(const void *req_tlv_area, int req_tlv_space) -{ - u32 ref; - struct port *p_ptr; - struct sk_buff *buf; - struct tlv_desc *rep_tlv; - struct print_buf pb; - int str_len; - - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_PORT_REF)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - - ref = *(u32 *)TLV_DATA(req_tlv_area); - ref = ntohl(ref); - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return cfg_reply_error_string("port not found"); - - buf = tipc_cfg_reply_alloc(TLV_SPACE(MAX_PORT_STATS)); - if (!buf) { - tipc_port_unlock(p_ptr); - return NULL; - } - rep_tlv = (struct tlv_desc *)buf->data; - - tipc_printbuf_init(&pb, TLV_DATA(rep_tlv), MAX_PORT_STATS); - port_print(p_ptr, &pb, 1); - /* NEED TO FILL IN ADDITIONAL PORT STATISTICS HERE */ - tipc_port_unlock(p_ptr); - str_len = tipc_printbuf_validate(&pb); - - skb_put(buf, TLV_SPACE(str_len)); - TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); - - return buf; -} - -#endif - void tipc_port_reinit(void) { struct port *p_ptr; @@ -1295,50 +1214,13 @@ int tipc_shutdown(u32 ref) return tipc_disconnect(ref); } -int tipc_isconnected(u32 ref, int *isconnected) -{ - struct port *p_ptr; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - *isconnected = p_ptr->publ.connected; - tipc_port_unlock(p_ptr); - return 0; -} - -int tipc_peer(u32 ref, struct tipc_portid *peer) -{ - struct port *p_ptr; - int res; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - if (p_ptr->publ.connected) { - peer->ref = port_peerport(p_ptr); - peer->node = port_peernode(p_ptr); - res = 0; - } else - res = -ENOTCONN; - tipc_port_unlock(p_ptr); - return res; -} - -int tipc_ref_valid(u32 ref) -{ - /* Works irrespective of type */ - return !!tipc_ref_deref(ref); -} - - /* * tipc_port_recv_sections(): Concatenate and deliver sectioned * message for this node. */ -int tipc_port_recv_sections(struct port *sender, unsigned int num_sect, - struct iovec const *msg_sect) +static int tipc_port_recv_sections(struct port *sender, unsigned int num_sect, + struct iovec const *msg_sect) { struct sk_buff *buf; int res; @@ -1389,65 +1271,16 @@ int tipc_send(u32 ref, unsigned int num_sect, struct iovec const *msg_sect) } /** - * tipc_send_buf - send message buffer on connection - */ - -int tipc_send_buf(u32 ref, struct sk_buff *buf, unsigned int dsz) -{ - struct port *p_ptr; - struct tipc_msg *msg; - u32 destnode; - u32 hsz; - u32 sz; - u32 res; - - p_ptr = tipc_port_deref(ref); - if (!p_ptr || !p_ptr->publ.connected) - return -EINVAL; - - msg = &p_ptr->publ.phdr; - hsz = msg_hdr_sz(msg); - sz = hsz + dsz; - msg_set_size(msg, sz); - if (skb_cow(buf, hsz)) - return -ENOMEM; - - skb_push(buf, hsz); - skb_copy_to_linear_data(buf, msg, hsz); - destnode = msg_destnode(msg); - p_ptr->publ.congested = 1; - if (!tipc_port_congested(p_ptr)) { - if (likely(destnode != tipc_own_addr)) - res = tipc_send_buf_fast(buf, destnode); - else { - tipc_port_recv_msg(buf); - res = sz; - } - if (likely(res != -ELINKCONG)) { - port_incr_out_seqno(p_ptr); - p_ptr->sent++; - p_ptr->publ.congested = 0; - return res; - } - } - if (port_unreliable(p_ptr)) { - p_ptr->publ.congested = 0; - return dsz; - } - return -ELINKCONG; -} - -/** * tipc_forward2name - forward message sections to port name */ -int tipc_forward2name(u32 ref, - struct tipc_name const *name, - u32 domain, - u32 num_sect, - struct iovec const *msg_sect, - struct tipc_portid const *orig, - unsigned int importance) +static int tipc_forward2name(u32 ref, + struct tipc_name const *name, + u32 domain, + u32 num_sect, + struct iovec const *msg_sect, + struct tipc_portid const *orig, + unsigned int importance) { struct port *p_ptr; struct tipc_msg *msg; @@ -1473,7 +1306,7 @@ int tipc_forward2name(u32 ref, msg_set_destnode(msg, destnode); msg_set_destport(msg, destport); - if (likely(destport || destnode)) { + if (likely(destport)) { p_ptr->sent++; if (likely(destnode == tipc_own_addr)) return tipc_port_recv_sections(p_ptr, num_sect, msg_sect); @@ -1510,89 +1343,15 @@ int tipc_send2name(u32 ref, } /** - * tipc_forward_buf2name - forward message buffer to port name - */ - -int tipc_forward_buf2name(u32 ref, - struct tipc_name const *name, - u32 domain, - struct sk_buff *buf, - unsigned int dsz, - struct tipc_portid const *orig, - unsigned int importance) -{ - struct port *p_ptr; - struct tipc_msg *msg; - u32 destnode = domain; - u32 destport; - int res; - - p_ptr = (struct port *)tipc_ref_deref(ref); - if (!p_ptr || p_ptr->publ.connected) - return -EINVAL; - - msg = &p_ptr->publ.phdr; - if (importance <= TIPC_CRITICAL_IMPORTANCE) - msg_set_importance(msg, importance); - msg_set_type(msg, TIPC_NAMED_MSG); - msg_set_orignode(msg, orig->node); - msg_set_origport(msg, orig->ref); - msg_set_nametype(msg, name->type); - msg_set_nameinst(msg, name->instance); - msg_set_lookup_scope(msg, tipc_addr_scope(domain)); - msg_set_hdr_sz(msg, LONG_H_SIZE); - msg_set_size(msg, LONG_H_SIZE + dsz); - destport = tipc_nametbl_translate(name->type, name->instance, &destnode); - msg_set_destnode(msg, destnode); - msg_set_destport(msg, destport); - msg_dbg(msg, "forw2name ==> "); - if (skb_cow(buf, LONG_H_SIZE)) - return -ENOMEM; - skb_push(buf, LONG_H_SIZE); - skb_copy_to_linear_data(buf, msg, LONG_H_SIZE); - msg_dbg(buf_msg(buf),"PREP:"); - if (likely(destport || destnode)) { - p_ptr->sent++; - if (destnode == tipc_own_addr) - return tipc_port_recv_msg(buf); - res = tipc_send_buf_fast(buf, destnode); - if (likely(res != -ELINKCONG)) - return res; - if (port_unreliable(p_ptr)) - return dsz; - return -ELINKCONG; - } - return tipc_reject_msg(buf, TIPC_ERR_NO_NAME); -} - -/** - * tipc_send_buf2name - send message buffer to port name - */ - -int tipc_send_buf2name(u32 ref, - struct tipc_name const *dest, - u32 domain, - struct sk_buff *buf, - unsigned int dsz) -{ - struct tipc_portid orig; - - orig.ref = ref; - orig.node = tipc_own_addr; - return tipc_forward_buf2name(ref, dest, domain, buf, dsz, &orig, - TIPC_PORT_IMPORTANCE); -} - -/** * tipc_forward2port - forward message sections to port identity */ -int tipc_forward2port(u32 ref, - struct tipc_portid const *dest, - unsigned int num_sect, - struct iovec const *msg_sect, - struct tipc_portid const *orig, - unsigned int importance) +static int tipc_forward2port(u32 ref, + struct tipc_portid const *dest, + unsigned int num_sect, + struct iovec const *msg_sect, + struct tipc_portid const *orig, + unsigned int importance) { struct port *p_ptr; struct tipc_msg *msg; @@ -1644,12 +1403,12 @@ int tipc_send2port(u32 ref, /** * tipc_forward_buf2port - forward message buffer to port identity */ -int tipc_forward_buf2port(u32 ref, - struct tipc_portid const *dest, - struct sk_buff *buf, - unsigned int dsz, - struct tipc_portid const *orig, - unsigned int importance) +static int tipc_forward_buf2port(u32 ref, + struct tipc_portid const *dest, + struct sk_buff *buf, + unsigned int dsz, + struct tipc_portid const *orig, + unsigned int importance) { struct port *p_ptr; struct tipc_msg *msg; diff --git a/net/tipc/port.h b/net/tipc/port.h index 8d1652aab29..73bbf442b34 100644 --- a/net/tipc/port.h +++ b/net/tipc/port.h @@ -109,8 +109,6 @@ struct port { extern spinlock_t tipc_port_list_lock; struct port_list; -int tipc_port_recv_sections(struct port *p_ptr, u32 num_sect, - struct iovec const *msg_sect); int tipc_port_reject_sections(struct port *p_ptr, struct tipc_msg *hdr, struct iovec const *msg_sect, u32 num_sect, int err); @@ -157,7 +155,7 @@ static inline u32 tipc_peer_node(struct port *p_ptr) static inline int tipc_port_congested(struct port *p_ptr) { - return((p_ptr->sent - p_ptr->acked) >= (TIPC_FLOW_CONTROL_WIN * 2)); + return (p_ptr->sent - p_ptr->acked) >= (TIPC_FLOW_CONTROL_WIN * 2); } /** diff --git a/net/tipc/ref.c b/net/tipc/ref.c index 8dea66500cf..ab8ad32d8c2 100644 --- a/net/tipc/ref.c +++ b/net/tipc/ref.c @@ -282,23 +282,6 @@ void *tipc_ref_lock(u32 ref) return NULL; } -/** - * tipc_ref_unlock - unlock referenced object - */ - -void tipc_ref_unlock(u32 ref) -{ - if (likely(tipc_ref_table.entries)) { - struct reference *entry; - - entry = &tipc_ref_table.entries[ref & - tipc_ref_table.index_mask]; - if (likely((entry->ref == ref) && (entry->object))) - spin_unlock_bh(&entry->lock); - else - err("Attempt to unlock non-existent reference\n"); - } -} /** * tipc_ref_deref - return pointer referenced object (without locking it) diff --git a/net/tipc/ref.h b/net/tipc/ref.h index 7e3798ea93b..5bc8e7ab84d 100644 --- a/net/tipc/ref.h +++ b/net/tipc/ref.h @@ -44,7 +44,6 @@ u32 tipc_ref_acquire(void *object, spinlock_t **lock); void tipc_ref_discard(u32 ref); void *tipc_ref_lock(u32 ref); -void tipc_ref_unlock(u32 ref); void *tipc_ref_deref(u32 ref); #endif diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 66e889ba48f..33217fc3d69 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -64,6 +64,7 @@ struct tipc_sock { struct sock sk; struct tipc_port *p; struct tipc_portid peer_name; + long conn_timeout; }; #define tipc_sk(sk) ((struct tipc_sock *)(sk)) @@ -240,9 +241,9 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol, sock->state = state; sock_init_data(sock, sk); - sk->sk_rcvtimeo = msecs_to_jiffies(CONN_TIMEOUT_DEFAULT); sk->sk_backlog_rcv = backlog_rcv; tipc_sk(sk)->p = tp_ptr; + tipc_sk(sk)->conn_timeout = msecs_to_jiffies(CONN_TIMEOUT_DEFAULT); spin_unlock_bh(tp_ptr->lock); @@ -429,36 +430,55 @@ static int get_name(struct socket *sock, struct sockaddr *uaddr, * to handle any preventable race conditions, so TIPC will do the same ... * * TIPC sets the returned events as follows: - * a) POLLRDNORM and POLLIN are set if the socket's receive queue is non-empty - * or if a connection-oriented socket is does not have an active connection - * (i.e. a read operation will not block). - * b) POLLOUT is set except when a socket's connection has been terminated - * (i.e. a write operation will not block). - * c) POLLHUP is set when a socket's connection has been terminated. - * - * IMPORTANT: The fact that a read or write operation will not block does NOT - * imply that the operation will succeed! + * + * socket state flags set + * ------------ --------- + * unconnected no read flags + * no write flags + * + * connecting POLLIN/POLLRDNORM if ACK/NACK in rx queue + * no write flags + * + * connected POLLIN/POLLRDNORM if data in rx queue + * POLLOUT if port is not congested + * + * disconnecting POLLIN/POLLRDNORM/POLLHUP + * no write flags + * + * listening POLLIN if SYN in rx queue + * no write flags + * + * ready POLLIN/POLLRDNORM if data in rx queue + * [connectionless] POLLOUT (since port cannot be congested) + * + * IMPORTANT: The fact that a read or write operation is indicated does NOT + * imply that the operation will succeed, merely that it should be performed + * and will not block. */ static unsigned int poll(struct file *file, struct socket *sock, poll_table *wait) { struct sock *sk = sock->sk; - u32 mask; + u32 mask = 0; poll_wait(file, sk_sleep(sk), wait); - if (!skb_queue_empty(&sk->sk_receive_queue) || - (sock->state == SS_UNCONNECTED) || - (sock->state == SS_DISCONNECTING)) - mask = (POLLRDNORM | POLLIN); - else - mask = 0; - - if (sock->state == SS_DISCONNECTING) - mask |= POLLHUP; - else - mask |= POLLOUT; + switch ((int)sock->state) { + case SS_READY: + case SS_CONNECTED: + if (!tipc_sk_port(sk)->congested) + mask |= POLLOUT; + /* fall thru' */ + case SS_CONNECTING: + case SS_LISTENING: + if (!skb_queue_empty(&sk->sk_receive_queue)) + mask |= (POLLIN | POLLRDNORM); + break; + case SS_DISCONNECTING: + mask = (POLLIN | POLLRDNORM | POLLHUP); + break; + } return mask; } @@ -1026,9 +1046,8 @@ static int recv_stream(struct kiocb *iocb, struct socket *sock, struct sk_buff *buf; struct tipc_msg *msg; unsigned int sz; - int sz_to_copy; + int sz_to_copy, target, needed; int sz_copied = 0; - int needed; char __user *crs = m->msg_iov->iov_base; unsigned char *buf_crs; u32 err; @@ -1050,6 +1069,8 @@ static int recv_stream(struct kiocb *iocb, struct socket *sock, goto exit; } + target = sock_rcvlowat(sk, flags & MSG_WAITALL, buf_len); + restart: /* Look for a message in receive queue; wait if necessary */ @@ -1138,7 +1159,7 @@ restart: if ((sz_copied < buf_len) && /* didn't get all requested data */ (!skb_queue_empty(&sk->sk_receive_queue) || - (flags & MSG_WAITALL)) && /* and more is ready or required */ + (sz_copied < target)) && /* and more is ready or required */ (!(flags & MSG_PEEK)) && /* and aren't just peeking at data */ (!err)) /* and haven't reached a FIN */ goto restart; @@ -1174,7 +1195,7 @@ static int rx_queue_full(struct tipc_msg *msg, u32 queue_size, u32 base) if (msg_connected(msg)) threshold *= 4; - return (queue_size >= threshold); + return queue_size >= threshold; } /** @@ -1365,6 +1386,7 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen, struct msghdr m = {NULL,}; struct sk_buff *buf; struct tipc_msg *msg; + long timeout; int res; lock_sock(sk); @@ -1379,7 +1401,7 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen, /* For now, TIPC does not support the non-blocking form of connect() */ if (flags & O_NONBLOCK) { - res = -EWOULDBLOCK; + res = -EOPNOTSUPP; goto exit; } @@ -1425,11 +1447,12 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen, /* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */ + timeout = tipc_sk(sk)->conn_timeout; release_sock(sk); res = wait_event_interruptible_timeout(*sk_sleep(sk), (!skb_queue_empty(&sk->sk_receive_queue) || (sock->state != SS_CONNECTING)), - sk->sk_rcvtimeo); + timeout ? timeout : MAX_SCHEDULE_TIMEOUT); lock_sock(sk); if (res > 0) { @@ -1692,7 +1715,7 @@ static int setsockopt(struct socket *sock, res = tipc_set_portunreturnable(tport->ref, value); break; case TIPC_CONN_TIMEOUT: - sk->sk_rcvtimeo = msecs_to_jiffies(value); + tipc_sk(sk)->conn_timeout = msecs_to_jiffies(value); /* no need to set "res", since already 0 at this point */ break; default: @@ -1747,7 +1770,7 @@ static int getsockopt(struct socket *sock, res = tipc_portunreturnable(tport->ref, &value); break; case TIPC_CONN_TIMEOUT: - value = jiffies_to_msecs(sk->sk_rcvtimeo); + value = jiffies_to_msecs(tipc_sk(sk)->conn_timeout); /* no need to set "res", since already 0 at this point */ break; case TIPC_NODE_RECVQ_DEPTH: diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c index ab6eab4c45e..33313961d01 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -76,6 +76,19 @@ struct top_srv { static struct top_srv topsrv = { 0 }; /** + * htohl - convert value to endianness used by destination + * @in: value to convert + * @swap: non-zero if endianness must be reversed + * + * Returns converted value + */ + +static u32 htohl(u32 in, int swap) +{ + return swap ? swab32(in) : in; +} + +/** * subscr_send_event - send a message containing a tipc_event to the subscriber * * Note: Must not hold subscriber's server port lock, since tipc_send() will @@ -94,11 +107,11 @@ static void subscr_send_event(struct subscription *sub, msg_sect.iov_base = (void *)&sub->evt; msg_sect.iov_len = sizeof(struct tipc_event); - sub->evt.event = htonl(event); - sub->evt.found_lower = htonl(found_lower); - sub->evt.found_upper = htonl(found_upper); - sub->evt.port.ref = htonl(port_ref); - sub->evt.port.node = htonl(node); + sub->evt.event = htohl(event, sub->swap); + sub->evt.found_lower = htohl(found_lower, sub->swap); + sub->evt.found_upper = htohl(found_upper, sub->swap); + sub->evt.port.ref = htohl(port_ref, sub->swap); + sub->evt.port.node = htohl(node, sub->swap); tipc_send(sub->server_ref, 1, &msg_sect); } @@ -274,29 +287,16 @@ static void subscr_cancel(struct tipc_subscr *s, { struct subscription *sub; struct subscription *sub_temp; - __u32 type, lower, upper, timeout, filter; int found = 0; /* Find first matching subscription, exit if not found */ - type = ntohl(s->seq.type); - lower = ntohl(s->seq.lower); - upper = ntohl(s->seq.upper); - timeout = ntohl(s->timeout); - filter = ntohl(s->filter) & ~TIPC_SUB_CANCEL; - list_for_each_entry_safe(sub, sub_temp, &subscriber->subscription_list, subscription_list) { - if ((type == sub->seq.type) && - (lower == sub->seq.lower) && - (upper == sub->seq.upper) && - (timeout == sub->timeout) && - (filter == sub->filter) && - !memcmp(s->usr_handle,sub->evt.s.usr_handle, - sizeof(s->usr_handle)) ){ - found = 1; - break; - } + if (!memcmp(s, &sub->evt.s, sizeof(struct tipc_subscr))) { + found = 1; + break; + } } if (!found) return; @@ -310,7 +310,7 @@ static void subscr_cancel(struct tipc_subscr *s, k_term_timer(&sub->timer); spin_lock_bh(subscriber->lock); } - dbg("Cancel: removing sub %u,%u,%u from subscriber %p list\n", + dbg("Cancel: removing sub %u,%u,%u from subscriber %x list\n", sub->seq.type, sub->seq.lower, sub->seq.upper, subscriber); subscr_del(sub); } @@ -325,10 +325,16 @@ static struct subscription *subscr_subscribe(struct tipc_subscr *s, struct subscriber *subscriber) { struct subscription *sub; + int swap; + + /* Determine subscriber's endianness */ + + swap = !(s->filter & (TIPC_SUB_PORTS | TIPC_SUB_SERVICE)); /* Detect & process a subscription cancellation request */ - if (ntohl(s->filter) & TIPC_SUB_CANCEL) { + if (s->filter & htohl(TIPC_SUB_CANCEL, swap)) { + s->filter &= ~htohl(TIPC_SUB_CANCEL, swap); subscr_cancel(s, subscriber); return NULL; } @@ -353,12 +359,13 @@ static struct subscription *subscr_subscribe(struct tipc_subscr *s, /* Initialize subscription object */ - sub->seq.type = ntohl(s->seq.type); - sub->seq.lower = ntohl(s->seq.lower); - sub->seq.upper = ntohl(s->seq.upper); - sub->timeout = ntohl(s->timeout); - sub->filter = ntohl(s->filter); - if ((sub->filter && (sub->filter != TIPC_SUB_PORTS)) || + sub->seq.type = htohl(s->seq.type, swap); + sub->seq.lower = htohl(s->seq.lower, swap); + sub->seq.upper = htohl(s->seq.upper, swap); + sub->timeout = htohl(s->timeout, swap); + sub->filter = htohl(s->filter, swap); + if ((!(sub->filter & TIPC_SUB_PORTS) == + !(sub->filter & TIPC_SUB_SERVICE)) || (sub->seq.lower > sub->seq.upper)) { warn("Subscription rejected, illegal request\n"); kfree(sub); @@ -369,6 +376,7 @@ static struct subscription *subscr_subscribe(struct tipc_subscr *s, INIT_LIST_HEAD(&sub->nameseq_list); list_add(&sub->subscription_list, &subscriber->subscription_list); sub->server_ref = subscriber->port_ref; + sub->swap = swap; memcpy(&sub->evt.s, s, sizeof(struct tipc_subscr)); atomic_inc(&topsrv.subscription_count); if (sub->timeout != TIPC_WAIT_FOREVER) { @@ -598,12 +606,3 @@ void tipc_subscr_stop(void) topsrv.user_ref = 0; } } - - -int tipc_ispublished(struct tipc_name const *name) -{ - u32 domain = 0; - - return(tipc_nametbl_translate(name->type, name->instance,&domain) != 0); -} - diff --git a/net/tipc/subscr.h b/net/tipc/subscr.h index c20f496d95b..45d89bf4d20 100644 --- a/net/tipc/subscr.h +++ b/net/tipc/subscr.h @@ -53,6 +53,7 @@ typedef void (*tipc_subscr_event) (struct subscription *sub, * @nameseq_list: adjacent subscriptions in name sequence's subscription list * @subscription_list: adjacent subscriptions in subscriber's subscription list * @server_ref: object reference of server port associated with subscription + * @swap: indicates if subscriber uses opposite endianness in its messages * @evt: template for events generated by subscription */ @@ -65,6 +66,7 @@ struct subscription { struct list_head nameseq_list; struct list_head subscription_list; u32 server_ref; + int swap; struct tipc_event evt; }; diff --git a/net/tipc/zone.c b/net/tipc/zone.c index 2c01ba2d86b..83f8b5e91fc 100644 --- a/net/tipc/zone.c +++ b/net/tipc/zone.c @@ -160,14 +160,3 @@ u32 tipc_zone_select_router(struct _zone *z_ptr, u32 addr, u32 ref) } return 0; } - - -u32 tipc_zone_next_node(u32 addr) -{ - struct cluster *c_ptr = tipc_cltr_find(addr); - - if (c_ptr) - return tipc_cltr_next_node(c_ptr, addr); - return 0; -} - diff --git a/net/tipc/zone.h b/net/tipc/zone.h index 7bdc3406ba9..bd1c20ce9d0 100644 --- a/net/tipc/zone.h +++ b/net/tipc/zone.h @@ -61,7 +61,6 @@ void tipc_zone_send_external_routes(struct _zone *z_ptr, u32 dest); struct _zone *tipc_zone_create(u32 addr); void tipc_zone_delete(struct _zone *z_ptr); void tipc_zone_attach_cluster(struct _zone *z_ptr, struct cluster *c_ptr); -u32 tipc_zone_next_node(u32 addr); static inline struct _zone *tipc_zone_find(u32 addr) { diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 4414a18c63b..0ebc777a666 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -692,6 +692,7 @@ static int unix_autobind(struct socket *sock) static u32 ordernum = 1; struct unix_address *addr; int err; + unsigned int retries = 0; mutex_lock(&u->readlock); @@ -717,9 +718,17 @@ retry: if (__unix_find_socket_byname(net, addr->name, addr->len, sock->type, addr->hash)) { spin_unlock(&unix_table_lock); - /* Sanity yield. It is unusual case, but yet... */ - if (!(ordernum&0xFF)) - yield(); + /* + * __unix_find_socket_byname() may take long time if many names + * are already in use. + */ + cond_resched(); + /* Give up if all names seems to be in use. */ + if (retries++ == 0xFFFFF) { + err = -ENOSPC; + kfree(addr); + goto out; + } goto retry; } addr->hash ^= sk->sk_type; @@ -1502,6 +1511,8 @@ restart: goto restart; } + if (sock_flag(other, SOCK_RCVTSTAMP)) + __net_timestamp(skb); skb_queue_tail(&other->sk_receive_queue, skb); unix_state_unlock(other); other->sk_data_ready(other, len); @@ -1713,6 +1724,9 @@ static int unix_dgram_recvmsg(struct kiocb *iocb, struct socket *sock, if (err) goto out_free; + if (sock_flag(sk, SOCK_RCVTSTAMP)) + __sock_recv_timestamp(msg, sk, skb); + if (!siocb->scm) { siocb->scm = &tmp_scm; memset(&tmp_scm, 0, sizeof(tmp_scm)); @@ -2024,11 +2038,10 @@ static unsigned int unix_poll(struct file *file, struct socket *sock, poll_table if (sk->sk_shutdown == SHUTDOWN_MASK) mask |= POLLHUP; if (sk->sk_shutdown & RCV_SHUTDOWN) - mask |= POLLRDHUP; + mask |= POLLRDHUP | POLLIN | POLLRDNORM; /* readable? */ - if (!skb_queue_empty(&sk->sk_receive_queue) || - (sk->sk_shutdown & RCV_SHUTDOWN)) + if (!skb_queue_empty(&sk->sk_receive_queue)) mask |= POLLIN | POLLRDNORM; /* Connection-based need to check for termination and startup */ diff --git a/net/wireless/core.c b/net/wireless/core.c index 541e2fff5e9..9c21ebf9780 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -253,11 +253,16 @@ int cfg80211_switch_netns(struct cfg80211_registered_device *rdev, WARN_ON(err); wdev->netdev->features |= NETIF_F_NETNS_LOCAL; } + + return err; } wiphy_net_set(&rdev->wiphy, net); - return err; + err = device_rename(&rdev->wiphy.dev, dev_name(&rdev->wiphy.dev)); + WARN_ON(err); + + return 0; } static void cfg80211_rfkill_poll(struct rfkill *rfkill, void *data) @@ -428,7 +433,7 @@ int wiphy_register(struct wiphy *wiphy) /* sanity check ifmodes */ WARN_ON(!ifmodes); - ifmodes &= ((1 << __NL80211_IFTYPE_AFTER_LAST) - 1) & ~1; + ifmodes &= ((1 << NUM_NL80211_IFTYPES) - 1) & ~1; if (WARN_ON(ifmodes != wiphy->interface_modes)) wiphy->interface_modes = ifmodes; @@ -475,12 +480,10 @@ int wiphy_register(struct wiphy *wiphy) mutex_lock(&cfg80211_mutex); res = device_add(&rdev->wiphy.dev); - if (res) - goto out_unlock; - - res = rfkill_register(rdev->rfkill); - if (res) - goto out_rm_dev; + if (res) { + mutex_unlock(&cfg80211_mutex); + return res; + } /* set up regulatory info */ wiphy_update_regulatory(wiphy, NL80211_REGDOM_SET_BY_CORE); @@ -509,13 +512,18 @@ int wiphy_register(struct wiphy *wiphy) cfg80211_debugfs_rdev_add(rdev); mutex_unlock(&cfg80211_mutex); + /* + * due to a locking dependency this has to be outside of the + * cfg80211_mutex lock + */ + res = rfkill_register(rdev->rfkill); + if (res) + goto out_rm_dev; + return 0; out_rm_dev: device_del(&rdev->wiphy.dev); - -out_unlock: - mutex_unlock(&cfg80211_mutex); return res; } EXPORT_SYMBOL(wiphy_register); @@ -680,8 +688,8 @@ static int cfg80211_netdev_notifier_call(struct notifier_block * nb, INIT_WORK(&wdev->cleanup_work, wdev_cleanup_work); INIT_LIST_HEAD(&wdev->event_list); spin_lock_init(&wdev->event_lock); - INIT_LIST_HEAD(&wdev->action_registrations); - spin_lock_init(&wdev->action_registrations_lock); + INIT_LIST_HEAD(&wdev->mgmt_registrations); + spin_lock_init(&wdev->mgmt_registrations_lock); mutex_lock(&rdev->devlist_mtx); list_add_rcu(&wdev->list, &rdev->netdev_list); @@ -721,6 +729,7 @@ static int cfg80211_netdev_notifier_call(struct notifier_block * nb, dev->ethtool_ops = &cfg80211_ethtool_ops; if ((wdev->iftype == NL80211_IFTYPE_STATION || + wdev->iftype == NL80211_IFTYPE_P2P_CLIENT || wdev->iftype == NL80211_IFTYPE_ADHOC) && !wdev->use_4addr) dev->priv_flags |= IFF_DONT_BRIDGE; break; @@ -729,6 +738,7 @@ static int cfg80211_netdev_notifier_call(struct notifier_block * nb, case NL80211_IFTYPE_ADHOC: cfg80211_leave_ibss(rdev, dev, true); break; + case NL80211_IFTYPE_P2P_CLIENT: case NL80211_IFTYPE_STATION: wdev_lock(wdev); #ifdef CONFIG_CFG80211_WEXT @@ -801,7 +811,7 @@ static int cfg80211_netdev_notifier_call(struct notifier_block * nb, sysfs_remove_link(&dev->dev.kobj, "phy80211"); list_del_rcu(&wdev->list); rdev->devlist_generation++; - cfg80211_mlme_purge_actions(wdev); + cfg80211_mlme_purge_registrations(wdev); #ifdef CONFIG_CFG80211_WEXT kfree(wdev->wext.keys); #endif @@ -907,52 +917,3 @@ static void __exit cfg80211_exit(void) destroy_workqueue(cfg80211_wq); } module_exit(cfg80211_exit); - -static int ___wiphy_printk(const char *level, const struct wiphy *wiphy, - struct va_format *vaf) -{ - if (!wiphy) - return printk("%s(NULL wiphy *): %pV", level, vaf); - - return printk("%s%s: %pV", level, wiphy_name(wiphy), vaf); -} - -int __wiphy_printk(const char *level, const struct wiphy *wiphy, - const char *fmt, ...) -{ - struct va_format vaf; - va_list args; - int r; - - va_start(args, fmt); - - vaf.fmt = fmt; - vaf.va = &args; - - r = ___wiphy_printk(level, wiphy, &vaf); - va_end(args); - - return r; -} -EXPORT_SYMBOL(__wiphy_printk); - -#define define_wiphy_printk_level(func, kern_level) \ -int func(const struct wiphy *wiphy, const char *fmt, ...) \ -{ \ - struct va_format vaf; \ - va_list args; \ - int r; \ - \ - va_start(args, fmt); \ - \ - vaf.fmt = fmt; \ - vaf.va = &args; \ - \ - r = ___wiphy_printk(kern_level, wiphy, &vaf); \ - va_end(args); \ - \ - return r; \ -} \ -EXPORT_SYMBOL(func); - -define_wiphy_printk_level(wiphy_debug, KERN_DEBUG); diff --git a/net/wireless/core.h b/net/wireless/core.h index 63d57ae399c..6583cca0e2e 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -86,7 +86,7 @@ struct cfg80211_registered_device *wiphy_to_dev(struct wiphy *wiphy) static inline bool wiphy_idx_valid(int wiphy_idx) { - return (wiphy_idx >= 0); + return wiphy_idx >= 0; } @@ -95,7 +95,10 @@ extern struct mutex cfg80211_mutex; extern struct list_head cfg80211_rdev_list; extern int cfg80211_rdev_list_generation; -#define assert_cfg80211_lock() WARN_ON(!mutex_is_locked(&cfg80211_mutex)) +static inline void assert_cfg80211_lock(void) +{ + lockdep_assert_held(&cfg80211_mutex); +} /* * You can use this to mark a wiphy_idx as not having an associated wiphy. @@ -202,8 +205,8 @@ static inline void wdev_unlock(struct wireless_dev *wdev) mutex_unlock(&wdev->mtx); } -#define ASSERT_RDEV_LOCK(rdev) WARN_ON(!mutex_is_locked(&(rdev)->mtx)); -#define ASSERT_WDEV_LOCK(wdev) WARN_ON(!mutex_is_locked(&(wdev)->mtx)); +#define ASSERT_RDEV_LOCK(rdev) lockdep_assert_held(&(rdev)->mtx) +#define ASSERT_WDEV_LOCK(wdev) lockdep_assert_held(&(wdev)->mtx) enum cfg80211_event_type { EVENT_CONNECT_RESULT, @@ -331,16 +334,17 @@ void __cfg80211_connect_result(struct net_device *dev, const u8 *bssid, const u8 *resp_ie, size_t resp_ie_len, u16 status, bool wextev, struct cfg80211_bss *bss); -int cfg80211_mlme_register_action(struct wireless_dev *wdev, u32 snd_pid, - const u8 *match_data, int match_len); -void cfg80211_mlme_unregister_actions(struct wireless_dev *wdev, u32 nlpid); -void cfg80211_mlme_purge_actions(struct wireless_dev *wdev); -int cfg80211_mlme_action(struct cfg80211_registered_device *rdev, - struct net_device *dev, - struct ieee80211_channel *chan, - enum nl80211_channel_type channel_type, - bool channel_type_valid, - const u8 *buf, size_t len, u64 *cookie); +int cfg80211_mlme_register_mgmt(struct wireless_dev *wdev, u32 snd_pid, + u16 frame_type, const u8 *match_data, + int match_len); +void cfg80211_mlme_unregister_socket(struct wireless_dev *wdev, u32 nlpid); +void cfg80211_mlme_purge_registrations(struct wireless_dev *wdev); +int cfg80211_mlme_mgmt_tx(struct cfg80211_registered_device *rdev, + struct net_device *dev, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + bool channel_type_valid, + const u8 *buf, size_t len, u64 *cookie); /* SME */ int __cfg80211_connect(struct cfg80211_registered_device *rdev, @@ -371,7 +375,7 @@ bool cfg80211_sme_failed_reassoc(struct wireless_dev *wdev); /* internal helpers */ int cfg80211_validate_key_settings(struct cfg80211_registered_device *rdev, struct key_params *params, int key_idx, - const u8 *mac_addr); + bool pairwise, const u8 *mac_addr); void __cfg80211_disconnected(struct net_device *dev, const u8 *ie, size_t ie_len, u16 reason, bool from_ap); void cfg80211_sme_scan_done(struct net_device *dev); diff --git a/net/wireless/debugfs.c b/net/wireless/debugfs.c index a4991a3efec..39765bcfb47 100644 --- a/net/wireless/debugfs.c +++ b/net/wireless/debugfs.c @@ -34,6 +34,7 @@ static ssize_t name## _read(struct file *file, char __user *userbuf, \ static const struct file_operations name## _ops = { \ .read = name## _read, \ .open = cfg80211_open_file_generic, \ + .llseek = generic_file_llseek, \ }; DEBUGFS_READONLY_FILE(rts_threshold, 20, "%d", @@ -102,6 +103,7 @@ static ssize_t ht40allow_map_read(struct file *file, static const struct file_operations ht40allow_map_ops = { .read = ht40allow_map_read, .open = cfg80211_open_file_generic, + .llseek = default_llseek, }; #define DEBUGFS_ADD(name) \ diff --git a/net/wireless/ibss.c b/net/wireless/ibss.c index 27a8ce9343c..f33fbb79437 100644 --- a/net/wireless/ibss.c +++ b/net/wireless/ibss.c @@ -88,6 +88,25 @@ int __cfg80211_join_ibss(struct cfg80211_registered_device *rdev, if (wdev->ssid_len) return -EALREADY; + if (!params->basic_rates) { + /* + * If no rates were explicitly configured, + * use the mandatory rate set for 11b or + * 11a for maximum compatibility. + */ + struct ieee80211_supported_band *sband = + rdev->wiphy.bands[params->channel->band]; + int j; + u32 flag = params->channel->band == IEEE80211_BAND_5GHZ ? + IEEE80211_RATE_MANDATORY_A : + IEEE80211_RATE_MANDATORY_B; + + for (j = 0; j < sband->n_bitrates; j++) { + if (sband->bitrates[j].flags & flag) + params->basic_rates |= BIT(j); + } + } + if (WARN_ON(wdev->connect_keys)) kfree(wdev->connect_keys); wdev->connect_keys = connkeys; @@ -141,7 +160,7 @@ static void __cfg80211_clear_ibss(struct net_device *dev, bool nowext) */ if (rdev->ops->del_key) for (i = 0; i < 6; i++) - rdev->ops->del_key(wdev->wiphy, dev, i, NULL); + rdev->ops->del_key(wdev->wiphy, dev, i, false, NULL); if (wdev->current_bss) { cfg80211_unhold_bss(wdev->current_bss); diff --git a/net/wireless/mlme.c b/net/wireless/mlme.c index e74a1a2119d..26838d903b9 100644 --- a/net/wireless/mlme.c +++ b/net/wireless/mlme.c @@ -149,7 +149,7 @@ void __cfg80211_send_deauth(struct net_device *dev, struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *)buf; const u8 *bssid = mgmt->bssid; int i; - bool found = false; + bool found = false, was_current = false; ASSERT_WDEV_LOCK(wdev); @@ -159,6 +159,7 @@ void __cfg80211_send_deauth(struct net_device *dev, cfg80211_put_bss(&wdev->current_bss->pub); wdev->current_bss = NULL; found = true; + was_current = true; } else for (i = 0; i < MAX_AUTH_BSSES; i++) { if (wdev->auth_bsses[i] && memcmp(wdev->auth_bsses[i]->pub.bssid, bssid, ETH_ALEN) == 0) { @@ -183,7 +184,7 @@ void __cfg80211_send_deauth(struct net_device *dev, nl80211_send_deauth(rdev, dev, buf, len, GFP_KERNEL); - if (wdev->sme_state == CFG80211_SME_CONNECTED) { + if (wdev->sme_state == CFG80211_SME_CONNECTED && was_current) { u16 reason_code; bool from_ap; @@ -747,31 +748,53 @@ void cfg80211_new_sta(struct net_device *dev, const u8 *mac_addr, } EXPORT_SYMBOL(cfg80211_new_sta); -struct cfg80211_action_registration { +struct cfg80211_mgmt_registration { struct list_head list; u32 nlpid; int match_len; + __le16 frame_type; + u8 match[]; }; -int cfg80211_mlme_register_action(struct wireless_dev *wdev, u32 snd_pid, - const u8 *match_data, int match_len) +int cfg80211_mlme_register_mgmt(struct wireless_dev *wdev, u32 snd_pid, + u16 frame_type, const u8 *match_data, + int match_len) { - struct cfg80211_action_registration *reg, *nreg; + struct wiphy *wiphy = wdev->wiphy; + struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy); + struct cfg80211_mgmt_registration *reg, *nreg; int err = 0; + u16 mgmt_type; + + if (!wdev->wiphy->mgmt_stypes) + return -EOPNOTSUPP; + + if ((frame_type & IEEE80211_FCTL_FTYPE) != IEEE80211_FTYPE_MGMT) + return -EINVAL; + + if (frame_type & ~(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) + return -EINVAL; + + mgmt_type = (frame_type & IEEE80211_FCTL_STYPE) >> 4; + if (!(wdev->wiphy->mgmt_stypes[wdev->iftype].rx & BIT(mgmt_type))) + return -EINVAL; nreg = kzalloc(sizeof(*reg) + match_len, GFP_KERNEL); if (!nreg) return -ENOMEM; - spin_lock_bh(&wdev->action_registrations_lock); + spin_lock_bh(&wdev->mgmt_registrations_lock); - list_for_each_entry(reg, &wdev->action_registrations, list) { + list_for_each_entry(reg, &wdev->mgmt_registrations, list) { int mlen = min(match_len, reg->match_len); + if (frame_type != le16_to_cpu(reg->frame_type)) + continue; + if (memcmp(reg->match, match_data, mlen) == 0) { err = -EALREADY; break; @@ -786,134 +809,212 @@ int cfg80211_mlme_register_action(struct wireless_dev *wdev, u32 snd_pid, memcpy(nreg->match, match_data, match_len); nreg->match_len = match_len; nreg->nlpid = snd_pid; - list_add(&nreg->list, &wdev->action_registrations); + nreg->frame_type = cpu_to_le16(frame_type); + list_add(&nreg->list, &wdev->mgmt_registrations); + + if (rdev->ops->mgmt_frame_register) + rdev->ops->mgmt_frame_register(wiphy, wdev->netdev, + frame_type, true); out: - spin_unlock_bh(&wdev->action_registrations_lock); + spin_unlock_bh(&wdev->mgmt_registrations_lock); + return err; } -void cfg80211_mlme_unregister_actions(struct wireless_dev *wdev, u32 nlpid) +void cfg80211_mlme_unregister_socket(struct wireless_dev *wdev, u32 nlpid) { - struct cfg80211_action_registration *reg, *tmp; + struct wiphy *wiphy = wdev->wiphy; + struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy); + struct cfg80211_mgmt_registration *reg, *tmp; - spin_lock_bh(&wdev->action_registrations_lock); + spin_lock_bh(&wdev->mgmt_registrations_lock); - list_for_each_entry_safe(reg, tmp, &wdev->action_registrations, list) { - if (reg->nlpid == nlpid) { - list_del(®->list); - kfree(reg); + list_for_each_entry_safe(reg, tmp, &wdev->mgmt_registrations, list) { + if (reg->nlpid != nlpid) + continue; + + if (rdev->ops->mgmt_frame_register) { + u16 frame_type = le16_to_cpu(reg->frame_type); + + rdev->ops->mgmt_frame_register(wiphy, wdev->netdev, + frame_type, false); } + + list_del(®->list); + kfree(reg); } - spin_unlock_bh(&wdev->action_registrations_lock); + spin_unlock_bh(&wdev->mgmt_registrations_lock); } -void cfg80211_mlme_purge_actions(struct wireless_dev *wdev) +void cfg80211_mlme_purge_registrations(struct wireless_dev *wdev) { - struct cfg80211_action_registration *reg, *tmp; + struct cfg80211_mgmt_registration *reg, *tmp; - spin_lock_bh(&wdev->action_registrations_lock); + spin_lock_bh(&wdev->mgmt_registrations_lock); - list_for_each_entry_safe(reg, tmp, &wdev->action_registrations, list) { + list_for_each_entry_safe(reg, tmp, &wdev->mgmt_registrations, list) { list_del(®->list); kfree(reg); } - spin_unlock_bh(&wdev->action_registrations_lock); + spin_unlock_bh(&wdev->mgmt_registrations_lock); } -int cfg80211_mlme_action(struct cfg80211_registered_device *rdev, - struct net_device *dev, - struct ieee80211_channel *chan, - enum nl80211_channel_type channel_type, - bool channel_type_valid, - const u8 *buf, size_t len, u64 *cookie) +int cfg80211_mlme_mgmt_tx(struct cfg80211_registered_device *rdev, + struct net_device *dev, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + bool channel_type_valid, + const u8 *buf, size_t len, u64 *cookie) { struct wireless_dev *wdev = dev->ieee80211_ptr; const struct ieee80211_mgmt *mgmt; + u16 stype; - if (rdev->ops->action == NULL) + if (!wdev->wiphy->mgmt_stypes) return -EOPNOTSUPP; + + if (!rdev->ops->mgmt_tx) + return -EOPNOTSUPP; + if (len < 24 + 1) return -EINVAL; mgmt = (const struct ieee80211_mgmt *) buf; - if (!ieee80211_is_action(mgmt->frame_control)) + + if (!ieee80211_is_mgmt(mgmt->frame_control)) return -EINVAL; - if (mgmt->u.action.category != WLAN_CATEGORY_PUBLIC) { - /* Verify that we are associated with the destination AP */ - if (!wdev->current_bss || - memcmp(wdev->current_bss->pub.bssid, mgmt->bssid, - ETH_ALEN) != 0 || - (wdev->iftype == NL80211_IFTYPE_STATION && - memcmp(wdev->current_bss->pub.bssid, mgmt->da, - ETH_ALEN) != 0)) - return -ENOTCONN; + + stype = le16_to_cpu(mgmt->frame_control) & IEEE80211_FCTL_STYPE; + if (!(wdev->wiphy->mgmt_stypes[wdev->iftype].tx & BIT(stype >> 4))) + return -EINVAL; + + if (ieee80211_is_action(mgmt->frame_control) && + mgmt->u.action.category != WLAN_CATEGORY_PUBLIC) { + int err = 0; + + wdev_lock(wdev); + + switch (wdev->iftype) { + case NL80211_IFTYPE_ADHOC: + case NL80211_IFTYPE_STATION: + case NL80211_IFTYPE_P2P_CLIENT: + if (!wdev->current_bss) { + err = -ENOTCONN; + break; + } + + if (memcmp(wdev->current_bss->pub.bssid, + mgmt->bssid, ETH_ALEN)) { + err = -ENOTCONN; + break; + } + + /* + * check for IBSS DA must be done by driver as + * cfg80211 doesn't track the stations + */ + if (wdev->iftype == NL80211_IFTYPE_ADHOC) + break; + + /* for station, check that DA is the AP */ + if (memcmp(wdev->current_bss->pub.bssid, + mgmt->da, ETH_ALEN)) { + err = -ENOTCONN; + break; + } + break; + case NL80211_IFTYPE_AP: + case NL80211_IFTYPE_P2P_GO: + case NL80211_IFTYPE_AP_VLAN: + if (memcmp(mgmt->bssid, dev->dev_addr, ETH_ALEN)) + err = -EINVAL; + break; + default: + err = -EOPNOTSUPP; + break; + } + wdev_unlock(wdev); + + if (err) + return err; } if (memcmp(mgmt->sa, dev->dev_addr, ETH_ALEN) != 0) return -EINVAL; /* Transmit the Action frame as requested by user space */ - return rdev->ops->action(&rdev->wiphy, dev, chan, channel_type, - channel_type_valid, buf, len, cookie); + return rdev->ops->mgmt_tx(&rdev->wiphy, dev, chan, channel_type, + channel_type_valid, buf, len, cookie); } -bool cfg80211_rx_action(struct net_device *dev, int freq, const u8 *buf, - size_t len, gfp_t gfp) +bool cfg80211_rx_mgmt(struct net_device *dev, int freq, const u8 *buf, + size_t len, gfp_t gfp) { struct wireless_dev *wdev = dev->ieee80211_ptr; struct wiphy *wiphy = wdev->wiphy; struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy); - struct cfg80211_action_registration *reg; - const u8 *action_data; - int action_data_len; + struct cfg80211_mgmt_registration *reg; + const struct ieee80211_txrx_stypes *stypes = + &wiphy->mgmt_stypes[wdev->iftype]; + struct ieee80211_mgmt *mgmt = (void *)buf; + const u8 *data; + int data_len; bool result = false; + __le16 ftype = mgmt->frame_control & + cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE); + u16 stype; - /* frame length - min size excluding category */ - action_data_len = len - (IEEE80211_MIN_ACTION_SIZE - 1); + stype = (le16_to_cpu(mgmt->frame_control) & IEEE80211_FCTL_STYPE) >> 4; - /* action data starts with category */ - action_data = buf + IEEE80211_MIN_ACTION_SIZE - 1; + if (!(stypes->rx & BIT(stype))) + return false; - spin_lock_bh(&wdev->action_registrations_lock); + data = buf + ieee80211_hdrlen(mgmt->frame_control); + data_len = len - ieee80211_hdrlen(mgmt->frame_control); + + spin_lock_bh(&wdev->mgmt_registrations_lock); + + list_for_each_entry(reg, &wdev->mgmt_registrations, list) { + if (reg->frame_type != ftype) + continue; - list_for_each_entry(reg, &wdev->action_registrations, list) { - if (reg->match_len > action_data_len) + if (reg->match_len > data_len) continue; - if (memcmp(reg->match, action_data, reg->match_len)) + if (memcmp(reg->match, data, reg->match_len)) continue; /* found match! */ /* Indicate the received Action frame to user space */ - if (nl80211_send_action(rdev, dev, reg->nlpid, freq, - buf, len, gfp)) + if (nl80211_send_mgmt(rdev, dev, reg->nlpid, freq, + buf, len, gfp)) continue; result = true; break; } - spin_unlock_bh(&wdev->action_registrations_lock); + spin_unlock_bh(&wdev->mgmt_registrations_lock); return result; } -EXPORT_SYMBOL(cfg80211_rx_action); +EXPORT_SYMBOL(cfg80211_rx_mgmt); -void cfg80211_action_tx_status(struct net_device *dev, u64 cookie, - const u8 *buf, size_t len, bool ack, gfp_t gfp) +void cfg80211_mgmt_tx_status(struct net_device *dev, u64 cookie, + const u8 *buf, size_t len, bool ack, gfp_t gfp) { struct wireless_dev *wdev = dev->ieee80211_ptr; struct wiphy *wiphy = wdev->wiphy; struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy); /* Indicate TX status of the Action frame to user space */ - nl80211_send_action_tx_status(rdev, dev, cookie, buf, len, ack, gfp); + nl80211_send_mgmt_tx_status(rdev, dev, cookie, buf, len, ack, gfp); } -EXPORT_SYMBOL(cfg80211_action_tx_status); +EXPORT_SYMBOL(cfg80211_mgmt_tx_status); void cfg80211_cqm_rssi_notify(struct net_device *dev, enum nl80211_cqm_rssi_threshold_event rssi_event, diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 37902a54e9c..c506241f863 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -23,6 +23,11 @@ #include "nl80211.h" #include "reg.h" +static int nl80211_pre_doit(struct genl_ops *ops, struct sk_buff *skb, + struct genl_info *info); +static void nl80211_post_doit(struct genl_ops *ops, struct sk_buff *skb, + struct genl_info *info); + /* the netlink family */ static struct genl_family nl80211_fam = { .id = GENL_ID_GENERATE, /* don't bother with a hardcoded ID */ @@ -31,6 +36,8 @@ static struct genl_family nl80211_fam = { .version = 1, /* no particular meaning now */ .maxattr = NL80211_ATTR_MAX, .netnsok = true, + .pre_doit = nl80211_pre_doit, + .post_doit = nl80211_post_doit, }; /* internal helper: get rdev and dev */ @@ -86,6 +93,7 @@ static const struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] = { [NL80211_ATTR_KEY_CIPHER] = { .type = NLA_U32 }, [NL80211_ATTR_KEY_DEFAULT] = { .type = NLA_FLAG }, [NL80211_ATTR_KEY_SEQ] = { .type = NLA_BINARY, .len = 8 }, + [NL80211_ATTR_KEY_TYPE] = { .type = NLA_U32 }, [NL80211_ATTR_BEACON_INTERVAL] = { .type = NLA_U32 }, [NL80211_ATTR_DTIM_PERIOD] = { .type = NLA_U32 }, @@ -136,6 +144,8 @@ static const struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] = { .len = sizeof(struct nl80211_sta_flag_update), }, [NL80211_ATTR_CONTROL_PORT] = { .type = NLA_FLAG }, + [NL80211_ATTR_CONTROL_PORT_ETHERTYPE] = { .type = NLA_U16 }, + [NL80211_ATTR_CONTROL_PORT_NO_ENCRYPT] = { .type = NLA_FLAG }, [NL80211_ATTR_PRIVACY] = { .type = NLA_FLAG }, [NL80211_ATTR_CIPHER_SUITE_GROUP] = { .type = NLA_U32 }, [NL80211_ATTR_WPA_VERSIONS] = { .type = NLA_U32 }, @@ -156,9 +166,10 @@ static const struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] = { [NL80211_ATTR_WIPHY_TX_POWER_SETTING] = { .type = NLA_U32 }, [NL80211_ATTR_WIPHY_TX_POWER_LEVEL] = { .type = NLA_U32 }, + [NL80211_ATTR_FRAME_TYPE] = { .type = NLA_U16 }, }; -/* policy for the attributes */ +/* policy for the key attributes */ static const struct nla_policy nl80211_key_policy[NL80211_KEY_MAX + 1] = { [NL80211_KEY_DATA] = { .type = NLA_BINARY, .len = WLAN_MAX_KEY_LEN }, [NL80211_KEY_IDX] = { .type = NLA_U8 }, @@ -166,6 +177,7 @@ static const struct nla_policy nl80211_key_policy[NL80211_KEY_MAX + 1] = { [NL80211_KEY_SEQ] = { .type = NLA_BINARY, .len = 8 }, [NL80211_KEY_DEFAULT] = { .type = NLA_FLAG }, [NL80211_KEY_DEFAULT_MGMT] = { .type = NLA_FLAG }, + [NL80211_KEY_TYPE] = { .type = NLA_U32 }, }; /* ifidx get helper */ @@ -188,6 +200,47 @@ static int nl80211_get_ifidx(struct netlink_callback *cb) return res; } +static int nl80211_prepare_netdev_dump(struct sk_buff *skb, + struct netlink_callback *cb, + struct cfg80211_registered_device **rdev, + struct net_device **dev) +{ + int ifidx = cb->args[0]; + int err; + + if (!ifidx) + ifidx = nl80211_get_ifidx(cb); + if (ifidx < 0) + return ifidx; + + cb->args[0] = ifidx; + + rtnl_lock(); + + *dev = __dev_get_by_index(sock_net(skb->sk), ifidx); + if (!*dev) { + err = -ENODEV; + goto out_rtnl; + } + + *rdev = cfg80211_get_dev_from_ifindex(sock_net(skb->sk), ifidx); + if (IS_ERR(dev)) { + err = PTR_ERR(dev); + goto out_rtnl; + } + + return 0; + out_rtnl: + rtnl_unlock(); + return err; +} + +static void nl80211_finish_netdev_dump(struct cfg80211_registered_device *rdev) +{ + cfg80211_unlock_rdev(rdev); + rtnl_unlock(); +} + /* IE validation */ static bool is_valid_ie_attr(const struct nlattr *attr) { @@ -255,6 +308,7 @@ static int nl80211_msg_put_channel(struct sk_buff *msg, struct key_parse { struct key_params p; int idx; + int type; bool def, defmgmt; }; @@ -285,6 +339,12 @@ static int nl80211_parse_key_new(struct nlattr *key, struct key_parse *k) if (tb[NL80211_KEY_CIPHER]) k->p.cipher = nla_get_u32(tb[NL80211_KEY_CIPHER]); + if (tb[NL80211_KEY_TYPE]) { + k->type = nla_get_u32(tb[NL80211_KEY_TYPE]); + if (k->type < 0 || k->type >= NUM_NL80211_KEYTYPES) + return -EINVAL; + } + return 0; } @@ -309,6 +369,12 @@ static int nl80211_parse_key_old(struct genl_info *info, struct key_parse *k) k->def = !!info->attrs[NL80211_ATTR_KEY_DEFAULT]; k->defmgmt = !!info->attrs[NL80211_ATTR_KEY_DEFAULT_MGMT]; + if (info->attrs[NL80211_ATTR_KEY_TYPE]) { + k->type = nla_get_u32(info->attrs[NL80211_ATTR_KEY_TYPE]); + if (k->type < 0 || k->type >= NUM_NL80211_KEYTYPES) + return -EINVAL; + } + return 0; } @@ -318,6 +384,7 @@ static int nl80211_parse_key(struct genl_info *info, struct key_parse *k) memset(k, 0, sizeof(*k)); k->idx = -1; + k->type = -1; if (info->attrs[NL80211_ATTR_KEY]) err = nl80211_parse_key_new(info->attrs[NL80211_ATTR_KEY], k); @@ -382,7 +449,7 @@ nl80211_parse_connkeys(struct cfg80211_registered_device *rdev, } else if (parse.defmgmt) goto error; err = cfg80211_validate_key_settings(rdev, &parse.p, - parse.idx, NULL); + parse.idx, false, NULL); if (err) goto error; result->params[parse.idx].cipher = parse.p.cipher; @@ -401,18 +468,17 @@ static int nl80211_key_allowed(struct wireless_dev *wdev) { ASSERT_WDEV_LOCK(wdev); - if (!netif_running(wdev->netdev)) - return -ENETDOWN; - switch (wdev->iftype) { case NL80211_IFTYPE_AP: case NL80211_IFTYPE_AP_VLAN: + case NL80211_IFTYPE_P2P_GO: break; case NL80211_IFTYPE_ADHOC: if (!wdev->current_bss) return -ENOLINK; break; case NL80211_IFTYPE_STATION: + case NL80211_IFTYPE_P2P_CLIENT: if (wdev->sme_state != CFG80211_SME_CONNECTED) return -ENOLINK; break; @@ -437,6 +503,8 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags, struct ieee80211_rate *rate; int i; u16 ifmodes = dev->wiphy.interface_modes; + const struct ieee80211_txrx_stypes *mgmt_stypes = + dev->wiphy.mgmt_stypes; hdr = nl80211hdr_put(msg, pid, seq, flags, NL80211_CMD_NEW_WIPHY); if (!hdr) @@ -464,6 +532,9 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags, NLA_PUT_U16(msg, NL80211_ATTR_MAX_SCAN_IE_LEN, dev->wiphy.max_scan_ie_len); + if (dev->wiphy.flags & WIPHY_FLAG_IBSS_RSN) + NLA_PUT_FLAG(msg, NL80211_ATTR_SUPPORT_IBSS_RSN); + NLA_PUT(msg, NL80211_ATTR_CIPHER_SUITES, sizeof(u32) * dev->wiphy.n_cipher_suites, dev->wiphy.cipher_suites); @@ -471,6 +542,9 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags, NLA_PUT_U8(msg, NL80211_ATTR_MAX_NUM_PMKIDS, dev->wiphy.max_num_pmkids); + if (dev->wiphy.flags & WIPHY_FLAG_CONTROL_PORT_PROTOCOL) + NLA_PUT_FLAG(msg, NL80211_ATTR_CONTROL_PORT_ETHERTYPE); + nl_modes = nla_nest_start(msg, NL80211_ATTR_SUPPORTED_IFTYPES); if (!nl_modes) goto nla_put_failure; @@ -587,12 +661,13 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags, CMD(flush_pmksa, FLUSH_PMKSA); CMD(remain_on_channel, REMAIN_ON_CHANNEL); CMD(set_bitrate_mask, SET_TX_BITRATE_MASK); - CMD(action, ACTION); + CMD(mgmt_tx, FRAME); if (dev->wiphy.flags & WIPHY_FLAG_NETNS_OK) { i++; NLA_PUT_U32(msg, i, NL80211_CMD_SET_WIPHY_NETNS); } CMD(set_channel, SET_CHANNEL); + CMD(set_wds_peer, SET_WDS_PEER); #undef CMD @@ -608,6 +683,55 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags, nla_nest_end(msg, nl_cmds); + if (mgmt_stypes) { + u16 stypes; + struct nlattr *nl_ftypes, *nl_ifs; + enum nl80211_iftype ift; + + nl_ifs = nla_nest_start(msg, NL80211_ATTR_TX_FRAME_TYPES); + if (!nl_ifs) + goto nla_put_failure; + + for (ift = 0; ift < NUM_NL80211_IFTYPES; ift++) { + nl_ftypes = nla_nest_start(msg, ift); + if (!nl_ftypes) + goto nla_put_failure; + i = 0; + stypes = mgmt_stypes[ift].tx; + while (stypes) { + if (stypes & 1) + NLA_PUT_U16(msg, NL80211_ATTR_FRAME_TYPE, + (i << 4) | IEEE80211_FTYPE_MGMT); + stypes >>= 1; + i++; + } + nla_nest_end(msg, nl_ftypes); + } + + nla_nest_end(msg, nl_ifs); + + nl_ifs = nla_nest_start(msg, NL80211_ATTR_RX_FRAME_TYPES); + if (!nl_ifs) + goto nla_put_failure; + + for (ift = 0; ift < NUM_NL80211_IFTYPES; ift++) { + nl_ftypes = nla_nest_start(msg, ift); + if (!nl_ftypes) + goto nla_put_failure; + i = 0; + stypes = mgmt_stypes[ift].rx; + while (stypes) { + if (stypes & 1) + NLA_PUT_U16(msg, NL80211_ATTR_FRAME_TYPE, + (i << 4) | IEEE80211_FTYPE_MGMT); + stypes >>= 1; + i++; + } + nla_nest_end(msg, nl_ftypes); + } + nla_nest_end(msg, nl_ifs); + } + return genlmsg_end(msg, hdr); nla_put_failure: @@ -644,28 +768,18 @@ static int nl80211_dump_wiphy(struct sk_buff *skb, struct netlink_callback *cb) static int nl80211_get_wiphy(struct sk_buff *skb, struct genl_info *info) { struct sk_buff *msg; - struct cfg80211_registered_device *dev; - - dev = cfg80211_get_dev_from_info(info); - if (IS_ERR(dev)) - return PTR_ERR(dev); + struct cfg80211_registered_device *dev = info->user_ptr[0]; msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (!msg) - goto out_err; - - if (nl80211_send_wiphy(msg, info->snd_pid, info->snd_seq, 0, dev) < 0) - goto out_free; + return -ENOMEM; - cfg80211_unlock_rdev(dev); + if (nl80211_send_wiphy(msg, info->snd_pid, info->snd_seq, 0, dev) < 0) { + nlmsg_free(msg); + return -ENOBUFS; + } return genlmsg_reply(msg, info); - - out_free: - nlmsg_free(msg); - out_err: - cfg80211_unlock_rdev(dev); - return -ENOBUFS; } static const struct nla_policy txq_params_policy[NL80211_TXQ_ATTR_MAX + 1] = { @@ -709,7 +823,8 @@ static bool nl80211_can_set_dev_channel(struct wireless_dev *wdev) wdev->iftype == NL80211_IFTYPE_AP || wdev->iftype == NL80211_IFTYPE_WDS || wdev->iftype == NL80211_IFTYPE_MESH_POINT || - wdev->iftype == NL80211_IFTYPE_MONITOR; + wdev->iftype == NL80211_IFTYPE_MONITOR || + wdev->iftype == NL80211_IFTYPE_P2P_GO; } static int __nl80211_set_channel(struct cfg80211_registered_device *rdev, @@ -753,38 +868,48 @@ static int __nl80211_set_channel(struct cfg80211_registered_device *rdev, static int nl80211_set_channel(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *netdev; - int result; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *netdev = info->user_ptr[1]; - rtnl_lock(); + return __nl80211_set_channel(rdev, netdev->ieee80211_ptr, info); +} - result = get_rdev_dev_by_info_ifindex(info, &rdev, &netdev); - if (result) - goto unlock; +static int nl80211_set_wds_peer(struct sk_buff *skb, struct genl_info *info) +{ + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; + struct wireless_dev *wdev = dev->ieee80211_ptr; + const u8 *bssid; - result = __nl80211_set_channel(rdev, netdev->ieee80211_ptr, info); + if (!info->attrs[NL80211_ATTR_MAC]) + return -EINVAL; - unlock: - rtnl_unlock(); + if (netif_running(dev)) + return -EBUSY; - return result; + if (!rdev->ops->set_wds_peer) + return -EOPNOTSUPP; + + if (wdev->iftype != NL80211_IFTYPE_WDS) + return -EOPNOTSUPP; + + bssid = nla_data(info->attrs[NL80211_ATTR_MAC]); + return rdev->ops->set_wds_peer(wdev->wiphy, dev, bssid); } + static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) { struct cfg80211_registered_device *rdev; struct net_device *netdev = NULL; struct wireless_dev *wdev; - int result, rem_txq_params = 0; + int result = 0, rem_txq_params = 0; struct nlattr *nl_txq_params; u32 changed; u8 retry_short = 0, retry_long = 0; u32 frag_threshold = 0, rts_threshold = 0; u8 coverage_class = 0; - rtnl_lock(); - /* * Try to find the wiphy and netdev. Normally this * function shouldn't need the netdev, but this is @@ -811,8 +936,7 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) rdev = __cfg80211_rdev_from_info(info); if (IS_ERR(rdev)) { mutex_unlock(&cfg80211_mutex); - result = PTR_ERR(rdev); - goto unlock; + return PTR_ERR(rdev); } wdev = NULL; netdev = NULL; @@ -994,8 +1118,6 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) mutex_unlock(&rdev->mtx); if (netdev) dev_put(netdev); - unlock: - rtnl_unlock(); return result; } @@ -1075,33 +1197,20 @@ static int nl80211_dump_interface(struct sk_buff *skb, struct netlink_callback * static int nl80211_get_interface(struct sk_buff *skb, struct genl_info *info) { struct sk_buff *msg; - struct cfg80211_registered_device *dev; - struct net_device *netdev; - int err; - - err = get_rdev_dev_by_info_ifindex(info, &dev, &netdev); - if (err) - return err; + struct cfg80211_registered_device *dev = info->user_ptr[0]; + struct net_device *netdev = info->user_ptr[1]; msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (!msg) - goto out_err; + return -ENOMEM; if (nl80211_send_iface(msg, info->snd_pid, info->snd_seq, 0, - dev, netdev) < 0) - goto out_free; - - dev_put(netdev); - cfg80211_unlock_rdev(dev); + dev, netdev) < 0) { + nlmsg_free(msg); + return -ENOBUFS; + } return genlmsg_reply(msg, info); - - out_free: - nlmsg_free(msg); - out_err: - dev_put(netdev); - cfg80211_unlock_rdev(dev); - return -ENOBUFS; } static const struct nla_policy mntr_flags_policy[NL80211_MNTR_FLAG_MAX + 1] = { @@ -1161,39 +1270,29 @@ static int nl80211_valid_4addr(struct cfg80211_registered_device *rdev, static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; struct vif_params params; int err; enum nl80211_iftype otype, ntype; - struct net_device *dev; + struct net_device *dev = info->user_ptr[1]; u32 _flags, *flags = NULL; bool change = false; memset(¶ms, 0, sizeof(params)); - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - otype = ntype = dev->ieee80211_ptr->iftype; if (info->attrs[NL80211_ATTR_IFTYPE]) { ntype = nla_get_u32(info->attrs[NL80211_ATTR_IFTYPE]); if (otype != ntype) change = true; - if (ntype > NL80211_IFTYPE_MAX) { - err = -EINVAL; - goto unlock; - } + if (ntype > NL80211_IFTYPE_MAX) + return -EINVAL; } if (info->attrs[NL80211_ATTR_MESH_ID]) { - if (ntype != NL80211_IFTYPE_MESH_POINT) { - err = -EINVAL; - goto unlock; - } + if (ntype != NL80211_IFTYPE_MESH_POINT) + return -EINVAL; params.mesh_id = nla_data(info->attrs[NL80211_ATTR_MESH_ID]); params.mesh_id_len = nla_len(info->attrs[NL80211_ATTR_MESH_ID]); change = true; @@ -1204,20 +1303,18 @@ static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info) change = true; err = nl80211_valid_4addr(rdev, dev, params.use_4addr, ntype); if (err) - goto unlock; + return err; } else { params.use_4addr = -1; } if (info->attrs[NL80211_ATTR_MNTR_FLAGS]) { - if (ntype != NL80211_IFTYPE_MONITOR) { - err = -EINVAL; - goto unlock; - } + if (ntype != NL80211_IFTYPE_MONITOR) + return -EINVAL; err = parse_monitor_flags(info->attrs[NL80211_ATTR_MNTR_FLAGS], &_flags); if (err) - goto unlock; + return err; flags = &_flags; change = true; @@ -1231,17 +1328,12 @@ static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info) if (!err && params.use_4addr != -1) dev->ieee80211_ptr->use_4addr = params.use_4addr; - unlock: - dev_put(dev); - cfg80211_unlock_rdev(rdev); - unlock_rtnl: - rtnl_unlock(); return err; } static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; struct vif_params params; int err; enum nl80211_iftype type = NL80211_IFTYPE_UNSPECIFIED; @@ -1258,19 +1350,9 @@ static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info) return -EINVAL; } - rtnl_lock(); - - rdev = cfg80211_get_dev_from_info(info); - if (IS_ERR(rdev)) { - err = PTR_ERR(rdev); - goto unlock_rtnl; - } - if (!rdev->ops->add_virtual_intf || - !(rdev->wiphy.interface_modes & (1 << type))) { - err = -EOPNOTSUPP; - goto unlock; - } + !(rdev->wiphy.interface_modes & (1 << type))) + return -EOPNOTSUPP; if (type == NL80211_IFTYPE_MESH_POINT && info->attrs[NL80211_ATTR_MESH_ID]) { @@ -1282,7 +1364,7 @@ static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info) params.use_4addr = !!nla_get_u8(info->attrs[NL80211_ATTR_4ADDR]); err = nl80211_valid_4addr(rdev, NULL, params.use_4addr, type); if (err) - goto unlock; + return err; } err = parse_monitor_flags(type == NL80211_IFTYPE_MONITOR ? @@ -1292,38 +1374,18 @@ static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info) nla_data(info->attrs[NL80211_ATTR_IFNAME]), type, err ? NULL : &flags, ¶ms); - unlock: - cfg80211_unlock_rdev(rdev); - unlock_rtnl: - rtnl_unlock(); return err; } static int nl80211_del_interface(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - int err; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (!rdev->ops->del_virtual_intf) { - err = -EOPNOTSUPP; - goto out; - } - - err = rdev->ops->del_virtual_intf(&rdev->wiphy, dev); + if (!rdev->ops->del_virtual_intf) + return -EOPNOTSUPP; - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - unlock_rtnl: - rtnl_unlock(); - return err; + return rdev->ops->del_virtual_intf(&rdev->wiphy, dev); } struct get_key_cookie { @@ -1376,11 +1438,12 @@ static void get_key_callback(void *c, struct key_params *params) static int nl80211_get_key(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; int err; - struct net_device *dev; + struct net_device *dev = info->user_ptr[1]; u8 key_idx = 0; - u8 *mac_addr = NULL; + const u8 *mac_addr = NULL; + bool pairwise; struct get_key_cookie cookie = { .error = 0, }; @@ -1396,30 +1459,28 @@ static int nl80211_get_key(struct sk_buff *skb, struct genl_info *info) if (info->attrs[NL80211_ATTR_MAC]) mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]); - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (!rdev->ops->get_key) { - err = -EOPNOTSUPP; - goto out; + pairwise = !!mac_addr; + if (info->attrs[NL80211_ATTR_KEY_TYPE]) { + u32 kt = nla_get_u32(info->attrs[NL80211_ATTR_KEY_TYPE]); + if (kt >= NUM_NL80211_KEYTYPES) + return -EINVAL; + if (kt != NL80211_KEYTYPE_GROUP && + kt != NL80211_KEYTYPE_PAIRWISE) + return -EINVAL; + pairwise = kt == NL80211_KEYTYPE_PAIRWISE; } + if (!rdev->ops->get_key) + return -EOPNOTSUPP; + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (!msg) { - err = -ENOMEM; - goto out; - } + if (!msg) + return -ENOMEM; hdr = nl80211hdr_put(msg, info->snd_pid, info->snd_seq, 0, NL80211_CMD_NEW_KEY); - - if (IS_ERR(hdr)) { - err = PTR_ERR(hdr); - goto free_msg; - } + if (IS_ERR(hdr)) + return PTR_ERR(hdr); cookie.msg = msg; cookie.idx = key_idx; @@ -1429,8 +1490,12 @@ static int nl80211_get_key(struct sk_buff *skb, struct genl_info *info) if (mac_addr) NLA_PUT(msg, NL80211_ATTR_MAC, ETH_ALEN, mac_addr); - err = rdev->ops->get_key(&rdev->wiphy, dev, key_idx, mac_addr, - &cookie, get_key_callback); + if (pairwise && mac_addr && + !(rdev->wiphy.flags & WIPHY_FLAG_IBSS_RSN)) + return -ENOENT; + + err = rdev->ops->get_key(&rdev->wiphy, dev, key_idx, pairwise, + mac_addr, &cookie, get_key_callback); if (err) goto free_msg; @@ -1439,28 +1504,21 @@ static int nl80211_get_key(struct sk_buff *skb, struct genl_info *info) goto nla_put_failure; genlmsg_end(msg, hdr); - err = genlmsg_reply(msg, info); - goto out; + return genlmsg_reply(msg, info); nla_put_failure: err = -ENOBUFS; free_msg: nlmsg_free(msg); - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - unlock_rtnl: - rtnl_unlock(); - return err; } static int nl80211_set_key(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; struct key_parse key; int err; - struct net_device *dev; + struct net_device *dev = info->user_ptr[1]; int (*func)(struct wiphy *wiphy, struct net_device *netdev, u8 key_index); @@ -1475,21 +1533,13 @@ static int nl80211_set_key(struct sk_buff *skb, struct genl_info *info) if (!key.def && !key.defmgmt) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - if (key.def) func = rdev->ops->set_default_key; else func = rdev->ops->set_default_mgmt_key; - if (!func) { - err = -EOPNOTSUPP; - goto out; - } + if (!func) + return -EOPNOTSUPP; wdev_lock(dev->ieee80211_ptr); err = nl80211_key_allowed(dev->ieee80211_ptr); @@ -1506,23 +1556,16 @@ static int nl80211_set_key(struct sk_buff *skb, struct genl_info *info) #endif wdev_unlock(dev->ieee80211_ptr); - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - - unlock_rtnl: - rtnl_unlock(); - return err; } static int nl80211_new_key(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; int err; - struct net_device *dev; + struct net_device *dev = info->user_ptr[1]; struct key_parse key; - u8 *mac_addr = NULL; + const u8 *mac_addr = NULL; err = nl80211_parse_key(info, &key); if (err) @@ -1534,43 +1577,42 @@ static int nl80211_new_key(struct sk_buff *skb, struct genl_info *info) if (info->attrs[NL80211_ATTR_MAC]) mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]); - rtnl_lock(); + if (key.type == -1) { + if (mac_addr) + key.type = NL80211_KEYTYPE_PAIRWISE; + else + key.type = NL80211_KEYTYPE_GROUP; + } - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; + /* for now */ + if (key.type != NL80211_KEYTYPE_PAIRWISE && + key.type != NL80211_KEYTYPE_GROUP) + return -EINVAL; - if (!rdev->ops->add_key) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->add_key) + return -EOPNOTSUPP; - if (cfg80211_validate_key_settings(rdev, &key.p, key.idx, mac_addr)) { - err = -EINVAL; - goto out; - } + if (cfg80211_validate_key_settings(rdev, &key.p, key.idx, + key.type == NL80211_KEYTYPE_PAIRWISE, + mac_addr)) + return -EINVAL; wdev_lock(dev->ieee80211_ptr); err = nl80211_key_allowed(dev->ieee80211_ptr); if (!err) err = rdev->ops->add_key(&rdev->wiphy, dev, key.idx, + key.type == NL80211_KEYTYPE_PAIRWISE, mac_addr, &key.p); wdev_unlock(dev->ieee80211_ptr); - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - unlock_rtnl: - rtnl_unlock(); - return err; } static int nl80211_del_key(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; int err; - struct net_device *dev; + struct net_device *dev = info->user_ptr[1]; u8 *mac_addr = NULL; struct key_parse key; @@ -1581,21 +1623,32 @@ static int nl80211_del_key(struct sk_buff *skb, struct genl_info *info) if (info->attrs[NL80211_ATTR_MAC]) mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]); - rtnl_lock(); + if (key.type == -1) { + if (mac_addr) + key.type = NL80211_KEYTYPE_PAIRWISE; + else + key.type = NL80211_KEYTYPE_GROUP; + } - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; + /* for now */ + if (key.type != NL80211_KEYTYPE_PAIRWISE && + key.type != NL80211_KEYTYPE_GROUP) + return -EINVAL; - if (!rdev->ops->del_key) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->del_key) + return -EOPNOTSUPP; wdev_lock(dev->ieee80211_ptr); err = nl80211_key_allowed(dev->ieee80211_ptr); + + if (key.type == NL80211_KEYTYPE_PAIRWISE && mac_addr && + !(rdev->wiphy.flags & WIPHY_FLAG_IBSS_RSN)) + err = -ENOENT; + if (!err) - err = rdev->ops->del_key(&rdev->wiphy, dev, key.idx, mac_addr); + err = rdev->ops->del_key(&rdev->wiphy, dev, key.idx, + key.type == NL80211_KEYTYPE_PAIRWISE, + mac_addr); #ifdef CONFIG_CFG80211_WEXT if (!err) { @@ -1607,13 +1660,6 @@ static int nl80211_del_key(struct sk_buff *skb, struct genl_info *info) #endif wdev_unlock(dev->ieee80211_ptr); - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - - unlock_rtnl: - rtnl_unlock(); - return err; } @@ -1621,35 +1667,25 @@ static int nl80211_addset_beacon(struct sk_buff *skb, struct genl_info *info) { int (*call)(struct wiphy *wiphy, struct net_device *dev, struct beacon_parameters *info); - struct cfg80211_registered_device *rdev; - int err; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; struct beacon_parameters params; int haveinfo = 0; if (!is_valid_ie_attr(info->attrs[NL80211_ATTR_BEACON_TAIL])) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP) { - err = -EOPNOTSUPP; - goto out; - } + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) + return -EOPNOTSUPP; switch (info->genlhdr->cmd) { case NL80211_CMD_NEW_BEACON: /* these are required for NEW_BEACON */ if (!info->attrs[NL80211_ATTR_BEACON_INTERVAL] || !info->attrs[NL80211_ATTR_DTIM_PERIOD] || - !info->attrs[NL80211_ATTR_BEACON_HEAD]) { - err = -EINVAL; - goto out; - } + !info->attrs[NL80211_ATTR_BEACON_HEAD]) + return -EINVAL; call = rdev->ops->add_beacon; break; @@ -1658,14 +1694,11 @@ static int nl80211_addset_beacon(struct sk_buff *skb, struct genl_info *info) break; default: WARN_ON(1); - err = -EOPNOTSUPP; - goto out; + return -EOPNOTSUPP; } - if (!call) { - err = -EOPNOTSUPP; - goto out; - } + if (!call) + return -EOPNOTSUPP; memset(¶ms, 0, sizeof(params)); @@ -1695,52 +1728,25 @@ static int nl80211_addset_beacon(struct sk_buff *skb, struct genl_info *info) haveinfo = 1; } - if (!haveinfo) { - err = -EINVAL; - goto out; - } - - err = call(&rdev->wiphy, dev, ¶ms); - - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - unlock_rtnl: - rtnl_unlock(); + if (!haveinfo) + return -EINVAL; - return err; + return call(&rdev->wiphy, dev, ¶ms); } static int nl80211_del_beacon(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - int err; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (!rdev->ops->del_beacon) { - err = -EOPNOTSUPP; - goto out; - } - - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP) { - err = -EOPNOTSUPP; - goto out; - } - err = rdev->ops->del_beacon(&rdev->wiphy, dev); + if (!rdev->ops->del_beacon) + return -EOPNOTSUPP; - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - unlock_rtnl: - rtnl_unlock(); + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) + return -EOPNOTSUPP; - return err; + return rdev->ops->del_beacon(&rdev->wiphy, dev); } static const struct nla_policy sta_flags_policy[NL80211_STA_FLAG_MAX + 1] = { @@ -1861,6 +1867,12 @@ static int nl80211_send_station(struct sk_buff *msg, u32 pid, u32 seq, if (sinfo->filled & STATION_INFO_TX_PACKETS) NLA_PUT_U32(msg, NL80211_STA_INFO_TX_PACKETS, sinfo->tx_packets); + if (sinfo->filled & STATION_INFO_TX_RETRIES) + NLA_PUT_U32(msg, NL80211_STA_INFO_TX_RETRIES, + sinfo->tx_retries); + if (sinfo->filled & STATION_INFO_TX_FAILED) + NLA_PUT_U32(msg, NL80211_STA_INFO_TX_FAILED, + sinfo->tx_failed); nla_nest_end(msg, sinfoattr); return genlmsg_end(msg, hdr); @@ -1877,28 +1889,12 @@ static int nl80211_dump_station(struct sk_buff *skb, struct cfg80211_registered_device *dev; struct net_device *netdev; u8 mac_addr[ETH_ALEN]; - int ifidx = cb->args[0]; int sta_idx = cb->args[1]; int err; - if (!ifidx) - ifidx = nl80211_get_ifidx(cb); - if (ifidx < 0) - return ifidx; - - rtnl_lock(); - - netdev = __dev_get_by_index(sock_net(skb->sk), ifidx); - if (!netdev) { - err = -ENODEV; - goto out_rtnl; - } - - dev = cfg80211_get_dev_from_ifindex(sock_net(skb->sk), ifidx); - if (IS_ERR(dev)) { - err = PTR_ERR(dev); - goto out_rtnl; - } + err = nl80211_prepare_netdev_dump(skb, cb, &dev, &netdev); + if (err) + return err; if (!dev->ops->dump_station) { err = -EOPNOTSUPP; @@ -1928,21 +1924,19 @@ static int nl80211_dump_station(struct sk_buff *skb, cb->args[1] = sta_idx; err = skb->len; out_err: - cfg80211_unlock_rdev(dev); - out_rtnl: - rtnl_unlock(); + nl80211_finish_netdev_dump(dev); return err; } static int nl80211_get_station(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - int err; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; struct station_info sinfo; struct sk_buff *msg; u8 *mac_addr = NULL; + int err; memset(&sinfo, 0, sizeof(sinfo)); @@ -1951,41 +1945,24 @@ static int nl80211_get_station(struct sk_buff *skb, struct genl_info *info) mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]); - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - - if (!rdev->ops->get_station) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->get_station) + return -EOPNOTSUPP; err = rdev->ops->get_station(&rdev->wiphy, dev, mac_addr, &sinfo); if (err) - goto out; + return err; msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (!msg) - goto out; + return -ENOMEM; if (nl80211_send_station(msg, info->snd_pid, info->snd_seq, 0, - dev, mac_addr, &sinfo) < 0) - goto out_free; - - err = genlmsg_reply(msg, info); - goto out; - - out_free: - nlmsg_free(msg); - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); + dev, mac_addr, &sinfo) < 0) { + nlmsg_free(msg); + return -ENOBUFS; + } - return err; + return genlmsg_reply(msg, info); } /* @@ -2015,9 +1992,9 @@ static int get_vlan(struct genl_info *info, static int nl80211_set_station(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; int err; - struct net_device *dev; + struct net_device *dev = info->user_ptr[1]; struct station_parameters params; u8 *mac_addr = NULL; @@ -2055,12 +2032,6 @@ static int nl80211_set_station(struct sk_buff *skb, struct genl_info *info) params.plink_action = nla_get_u8(info->attrs[NL80211_ATTR_STA_PLINK_ACTION]); - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - err = get_vlan(info, rdev, ¶ms.vlan); if (err) goto out; @@ -2071,10 +2042,12 @@ static int nl80211_set_station(struct sk_buff *skb, struct genl_info *info) switch (dev->ieee80211_ptr->iftype) { case NL80211_IFTYPE_AP: case NL80211_IFTYPE_AP_VLAN: + case NL80211_IFTYPE_P2P_GO: /* disallow mesh-specific things */ if (params.plink_action) err = -EINVAL; break; + case NL80211_IFTYPE_P2P_CLIENT: case NL80211_IFTYPE_STATION: /* disallow everything but AUTHORIZED flag */ if (params.plink_action) @@ -2120,19 +2093,15 @@ static int nl80211_set_station(struct sk_buff *skb, struct genl_info *info) out: if (params.vlan) dev_put(params.vlan); - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); return err; } static int nl80211_new_station(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; int err; - struct net_device *dev; + struct net_device *dev = info->user_ptr[1]; struct station_parameters params; u8 *mac_addr = NULL; @@ -2169,17 +2138,10 @@ static int nl80211_new_station(struct sk_buff *skb, struct genl_info *info) if (parse_station_flags(info, ¶ms)) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP && - dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN) { - err = -EINVAL; - goto out; - } + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) + return -EINVAL; err = get_vlan(info, rdev, ¶ms.vlan); if (err) @@ -2193,61 +2155,33 @@ static int nl80211_new_station(struct sk_buff *skb, struct genl_info *info) goto out; } - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } - err = rdev->ops->add_station(&rdev->wiphy, dev, mac_addr, ¶ms); out: if (params.vlan) dev_put(params.vlan); - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); - return err; } static int nl80211_del_station(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - int err; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; u8 *mac_addr = NULL; if (info->attrs[NL80211_ATTR_MAC]) mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]); - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP && dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN && - dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT) { - err = -EINVAL; - goto out; - } - - if (!rdev->ops->del_station) { - err = -EOPNOTSUPP; - goto out; - } - - err = rdev->ops->del_station(&rdev->wiphy, dev, mac_addr); + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) + return -EINVAL; - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); + if (!rdev->ops->del_station) + return -EOPNOTSUPP; - return err; + return rdev->ops->del_station(&rdev->wiphy, dev, mac_addr); } static int nl80211_send_mpath(struct sk_buff *msg, u32 pid, u32 seq, @@ -2310,28 +2244,12 @@ static int nl80211_dump_mpath(struct sk_buff *skb, struct net_device *netdev; u8 dst[ETH_ALEN]; u8 next_hop[ETH_ALEN]; - int ifidx = cb->args[0]; int path_idx = cb->args[1]; int err; - if (!ifidx) - ifidx = nl80211_get_ifidx(cb); - if (ifidx < 0) - return ifidx; - - rtnl_lock(); - - netdev = __dev_get_by_index(sock_net(skb->sk), ifidx); - if (!netdev) { - err = -ENODEV; - goto out_rtnl; - } - - dev = cfg80211_get_dev_from_ifindex(sock_net(skb->sk), ifidx); - if (IS_ERR(dev)) { - err = PTR_ERR(dev); - goto out_rtnl; - } + err = nl80211_prepare_netdev_dump(skb, cb, &dev, &netdev); + if (err) + return err; if (!dev->ops->dump_mpath) { err = -EOPNOTSUPP; @@ -2365,18 +2283,15 @@ static int nl80211_dump_mpath(struct sk_buff *skb, cb->args[1] = path_idx; err = skb->len; out_err: - cfg80211_unlock_rdev(dev); - out_rtnl: - rtnl_unlock(); - + nl80211_finish_netdev_dump(dev); return err; } static int nl80211_get_mpath(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; int err; - struct net_device *dev; + struct net_device *dev = info->user_ptr[1]; struct mpath_info pinfo; struct sk_buff *msg; u8 *dst = NULL; @@ -2389,53 +2304,33 @@ static int nl80211_get_mpath(struct sk_buff *skb, struct genl_info *info) dst = nla_data(info->attrs[NL80211_ATTR_MAC]); - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - - if (!rdev->ops->get_mpath) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->get_mpath) + return -EOPNOTSUPP; - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT) { - err = -EOPNOTSUPP; - goto out; - } + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT) + return -EOPNOTSUPP; err = rdev->ops->get_mpath(&rdev->wiphy, dev, dst, next_hop, &pinfo); if (err) - goto out; + return err; msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (!msg) - goto out; + return -ENOMEM; if (nl80211_send_mpath(msg, info->snd_pid, info->snd_seq, 0, - dev, dst, next_hop, &pinfo) < 0) - goto out_free; - - err = genlmsg_reply(msg, info); - goto out; - - out_free: - nlmsg_free(msg); - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); + dev, dst, next_hop, &pinfo) < 0) { + nlmsg_free(msg); + return -ENOBUFS; + } - return err; + return genlmsg_reply(msg, info); } static int nl80211_set_mpath(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - int err; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; u8 *dst = NULL; u8 *next_hop = NULL; @@ -2448,42 +2343,19 @@ static int nl80211_set_mpath(struct sk_buff *skb, struct genl_info *info) dst = nla_data(info->attrs[NL80211_ATTR_MAC]); next_hop = nla_data(info->attrs[NL80211_ATTR_MPATH_NEXT_HOP]); - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - - if (!rdev->ops->change_mpath) { - err = -EOPNOTSUPP; - goto out; - } - - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT) { - err = -EOPNOTSUPP; - goto out; - } - - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } - - err = rdev->ops->change_mpath(&rdev->wiphy, dev, dst, next_hop); + if (!rdev->ops->change_mpath) + return -EOPNOTSUPP; - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT) + return -EOPNOTSUPP; - return err; + return rdev->ops->change_mpath(&rdev->wiphy, dev, dst, next_hop); } + static int nl80211_new_mpath(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - int err; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; u8 *dst = NULL; u8 *next_hop = NULL; @@ -2496,75 +2368,34 @@ static int nl80211_new_mpath(struct sk_buff *skb, struct genl_info *info) dst = nla_data(info->attrs[NL80211_ATTR_MAC]); next_hop = nla_data(info->attrs[NL80211_ATTR_MPATH_NEXT_HOP]); - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - - if (!rdev->ops->add_mpath) { - err = -EOPNOTSUPP; - goto out; - } - - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT) { - err = -EOPNOTSUPP; - goto out; - } - - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } - - err = rdev->ops->add_mpath(&rdev->wiphy, dev, dst, next_hop); + if (!rdev->ops->add_mpath) + return -EOPNOTSUPP; - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT) + return -EOPNOTSUPP; - return err; + return rdev->ops->add_mpath(&rdev->wiphy, dev, dst, next_hop); } static int nl80211_del_mpath(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - int err; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; u8 *dst = NULL; if (info->attrs[NL80211_ATTR_MAC]) dst = nla_data(info->attrs[NL80211_ATTR_MAC]); - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - - if (!rdev->ops->del_mpath) { - err = -EOPNOTSUPP; - goto out; - } - - err = rdev->ops->del_mpath(&rdev->wiphy, dev, dst); - - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); + if (!rdev->ops->del_mpath) + return -EOPNOTSUPP; - return err; + return rdev->ops->del_mpath(&rdev->wiphy, dev, dst); } static int nl80211_set_bss(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - int err; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; struct bss_parameters params; memset(¶ms, 0, sizeof(params)); @@ -2592,31 +2423,14 @@ static int nl80211_set_bss(struct sk_buff *skb, struct genl_info *info) if (info->attrs[NL80211_ATTR_AP_ISOLATE]) params.ap_isolate = !!nla_get_u8(info->attrs[NL80211_ATTR_AP_ISOLATE]); - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - - if (!rdev->ops->change_bss) { - err = -EOPNOTSUPP; - goto out; - } - - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP) { - err = -EOPNOTSUPP; - goto out; - } - - err = rdev->ops->change_bss(&rdev->wiphy, dev, ¶ms); + if (!rdev->ops->change_bss) + return -EOPNOTSUPP; - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) + return -EOPNOTSUPP; - return err; + return rdev->ops->change_bss(&rdev->wiphy, dev, ¶ms); } static const struct nla_policy reg_rule_policy[NL80211_REG_RULE_ATTR_MAX + 1] = { @@ -2695,37 +2509,26 @@ static int nl80211_req_set_reg(struct sk_buff *skb, struct genl_info *info) static int nl80211_get_mesh_params(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; struct mesh_config cur_params; int err; - struct net_device *dev; + struct net_device *dev = info->user_ptr[1]; void *hdr; struct nlattr *pinfoattr; struct sk_buff *msg; - rtnl_lock(); - - /* Look up our device */ - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - - if (!rdev->ops->get_mesh_params) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->get_mesh_params) + return -EOPNOTSUPP; /* Get the mesh params */ err = rdev->ops->get_mesh_params(&rdev->wiphy, dev, &cur_params); if (err) - goto out; + return err; /* Draw up a netlink message to send back */ msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (!msg) { - err = -ENOBUFS; - goto out; - } + if (!msg) + return -ENOMEM; hdr = nl80211hdr_put(msg, info->snd_pid, info->snd_seq, 0, NL80211_CMD_GET_MESH_PARAMS); if (!hdr) @@ -2764,21 +2567,12 @@ static int nl80211_get_mesh_params(struct sk_buff *skb, cur_params.dot11MeshHWMPRootMode); nla_nest_end(msg, pinfoattr); genlmsg_end(msg, hdr); - err = genlmsg_reply(msg, info); - goto out; + return genlmsg_reply(msg, info); nla_put_failure: genlmsg_cancel(msg, hdr); nlmsg_free(msg); - err = -EMSGSIZE; - out: - /* Cleanup */ - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); - - return err; + return -ENOBUFS; } #define FILL_IN_MESH_PARAM_IF_SET(table, cfg, param, mask, attr_num, nla_fn) \ @@ -2808,10 +2602,9 @@ static const struct nla_policy nl80211_meshconf_params_policy[NL80211_MESHCONF_A static int nl80211_set_mesh_params(struct sk_buff *skb, struct genl_info *info) { - int err; u32 mask; - struct cfg80211_registered_device *rdev; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; struct mesh_config cfg; struct nlattr *tb[NL80211_MESHCONF_ATTR_MAX + 1]; struct nlattr *parent_attr; @@ -2823,16 +2616,8 @@ static int nl80211_set_mesh_params(struct sk_buff *skb, struct genl_info *info) parent_attr, nl80211_meshconf_params_policy)) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - - if (!rdev->ops->set_mesh_params) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->set_mesh_params) + return -EOPNOTSUPP; /* This makes sure that there aren't more than 32 mesh config * parameters (otherwise our bitfield scheme would not work.) */ @@ -2878,16 +2663,7 @@ static int nl80211_set_mesh_params(struct sk_buff *skb, struct genl_info *info) nla_get_u8); /* Apply changes */ - err = rdev->ops->set_mesh_params(&rdev->wiphy, dev, &cfg, mask); - - out: - /* cleanup */ - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); - - return err; + return rdev->ops->set_mesh_params(&rdev->wiphy, dev, &cfg, mask); } #undef FILL_IN_MESH_PARAM_IF_SET @@ -3070,8 +2846,8 @@ static int validate_scan_freqs(struct nlattr *freqs) static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; struct cfg80211_scan_request *request; struct cfg80211_ssid *ssid; struct ieee80211_channel *channel; @@ -3084,36 +2860,19 @@ static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info) if (!is_valid_ie_attr(info->attrs[NL80211_ATTR_IE])) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - wiphy = &rdev->wiphy; - if (!rdev->ops->scan) { - err = -EOPNOTSUPP; - goto out; - } - - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } + if (!rdev->ops->scan) + return -EOPNOTSUPP; - if (rdev->scan_req) { - err = -EBUSY; - goto out; - } + if (rdev->scan_req) + return -EBUSY; if (info->attrs[NL80211_ATTR_SCAN_FREQUENCIES]) { n_channels = validate_scan_freqs( info->attrs[NL80211_ATTR_SCAN_FREQUENCIES]); - if (!n_channels) { - err = -EINVAL; - goto out; - } + if (!n_channels) + return -EINVAL; } else { n_channels = 0; @@ -3126,29 +2885,23 @@ static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info) nla_for_each_nested(attr, info->attrs[NL80211_ATTR_SCAN_SSIDS], tmp) n_ssids++; - if (n_ssids > wiphy->max_scan_ssids) { - err = -EINVAL; - goto out; - } + if (n_ssids > wiphy->max_scan_ssids) + return -EINVAL; if (info->attrs[NL80211_ATTR_IE]) ie_len = nla_len(info->attrs[NL80211_ATTR_IE]); else ie_len = 0; - if (ie_len > wiphy->max_scan_ie_len) { - err = -EINVAL; - goto out; - } + if (ie_len > wiphy->max_scan_ie_len) + return -EINVAL; request = kzalloc(sizeof(*request) + sizeof(*ssid) * n_ssids + sizeof(channel) * n_channels + ie_len, GFP_KERNEL); - if (!request) { - err = -ENOMEM; - goto out; - } + if (!request) + return -ENOMEM; if (n_ssids) request->ssids = (void *)&request->channels[n_channels]; @@ -3236,18 +2989,11 @@ static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info) if (!err) { nl80211_send_scan_start(rdev, dev); dev_hold(dev); - } - + } else { out_free: - if (err) { rdev->scan_req = NULL; kfree(request); } - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); return err; } @@ -3306,6 +3052,7 @@ static int nl80211_send_bss(struct sk_buff *msg, u32 pid, u32 seq, int flags, } switch (wdev->iftype) { + case NL80211_IFTYPE_P2P_CLIENT: case NL80211_IFTYPE_STATION: if (intbss == wdev->current_bss) NLA_PUT_U32(msg, NL80211_BSS_STATUS, @@ -3343,25 +3090,12 @@ static int nl80211_dump_scan(struct sk_buff *skb, struct net_device *dev; struct cfg80211_internal_bss *scan; struct wireless_dev *wdev; - int ifidx = cb->args[0]; int start = cb->args[1], idx = 0; int err; - if (!ifidx) - ifidx = nl80211_get_ifidx(cb); - if (ifidx < 0) - return ifidx; - cb->args[0] = ifidx; - - dev = dev_get_by_index(sock_net(skb->sk), ifidx); - if (!dev) - return -ENODEV; - - rdev = cfg80211_get_dev_from_ifindex(sock_net(skb->sk), ifidx); - if (IS_ERR(rdev)) { - err = PTR_ERR(rdev); - goto out_put_netdev; - } + err = nl80211_prepare_netdev_dump(skb, cb, &rdev, &dev); + if (err) + return err; wdev = dev->ieee80211_ptr; @@ -3377,21 +3111,17 @@ static int nl80211_dump_scan(struct sk_buff *skb, cb->nlh->nlmsg_seq, NLM_F_MULTI, rdev, wdev, scan) < 0) { idx--; - goto out; + break; } } - out: spin_unlock_bh(&rdev->bss_lock); wdev_unlock(wdev); cb->args[1] = idx; - err = skb->len; - cfg80211_unlock_rdev(rdev); - out_put_netdev: - dev_put(dev); + nl80211_finish_netdev_dump(rdev); - return err; + return skb->len; } static int nl80211_send_survey(struct sk_buff *msg, u32 pid, u32 seq, @@ -3421,6 +3151,23 @@ static int nl80211_send_survey(struct sk_buff *msg, u32 pid, u32 seq, if (survey->filled & SURVEY_INFO_NOISE_DBM) NLA_PUT_U8(msg, NL80211_SURVEY_INFO_NOISE, survey->noise); + if (survey->filled & SURVEY_INFO_IN_USE) + NLA_PUT_FLAG(msg, NL80211_SURVEY_INFO_IN_USE); + if (survey->filled & SURVEY_INFO_CHANNEL_TIME) + NLA_PUT_U64(msg, NL80211_SURVEY_INFO_CHANNEL_TIME, + survey->channel_time); + if (survey->filled & SURVEY_INFO_CHANNEL_TIME_BUSY) + NLA_PUT_U64(msg, NL80211_SURVEY_INFO_CHANNEL_TIME_BUSY, + survey->channel_time_busy); + if (survey->filled & SURVEY_INFO_CHANNEL_TIME_EXT_BUSY) + NLA_PUT_U64(msg, NL80211_SURVEY_INFO_CHANNEL_TIME_EXT_BUSY, + survey->channel_time_ext_busy); + if (survey->filled & SURVEY_INFO_CHANNEL_TIME_RX) + NLA_PUT_U64(msg, NL80211_SURVEY_INFO_CHANNEL_TIME_RX, + survey->channel_time_rx); + if (survey->filled & SURVEY_INFO_CHANNEL_TIME_TX) + NLA_PUT_U64(msg, NL80211_SURVEY_INFO_CHANNEL_TIME_TX, + survey->channel_time_tx); nla_nest_end(msg, infoattr); @@ -3437,29 +3184,12 @@ static int nl80211_dump_survey(struct sk_buff *skb, struct survey_info survey; struct cfg80211_registered_device *dev; struct net_device *netdev; - int ifidx = cb->args[0]; int survey_idx = cb->args[1]; int res; - if (!ifidx) - ifidx = nl80211_get_ifidx(cb); - if (ifidx < 0) - return ifidx; - cb->args[0] = ifidx; - - rtnl_lock(); - - netdev = __dev_get_by_index(sock_net(skb->sk), ifidx); - if (!netdev) { - res = -ENODEV; - goto out_rtnl; - } - - dev = cfg80211_get_dev_from_ifindex(sock_net(skb->sk), ifidx); - if (IS_ERR(dev)) { - res = PTR_ERR(dev); - goto out_rtnl; - } + res = nl80211_prepare_netdev_dump(skb, cb, &dev, &netdev); + if (res) + return res; if (!dev->ops->dump_survey) { res = -EOPNOTSUPP; @@ -3487,10 +3217,7 @@ static int nl80211_dump_survey(struct sk_buff *skb, cb->args[1] = survey_idx; res = skb->len; out_err: - cfg80211_unlock_rdev(dev); - out_rtnl: - rtnl_unlock(); - + nl80211_finish_netdev_dump(dev); return res; } @@ -3523,8 +3250,8 @@ static bool nl80211_valid_cipher_suite(u32 cipher) static int nl80211_authenticate(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; struct ieee80211_channel *chan; const u8 *bssid, *ssid, *ie = NULL; int err, ssid_len, ie_len = 0; @@ -3552,6 +3279,8 @@ static int nl80211_authenticate(struct sk_buff *skb, struct genl_info *info) return err; if (key.idx >= 0) { + if (key.type != -1 && key.type != NL80211_KEYTYPE_GROUP) + return -EINVAL; if (!key.p.key || !key.p.key_len) return -EINVAL; if ((key.p.cipher != WLAN_CIPHER_SUITE_WEP40 || @@ -3566,34 +3295,31 @@ static int nl80211_authenticate(struct sk_buff *skb, struct genl_info *info) key.p.key = NULL; } - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (!rdev->ops->auth) { - err = -EOPNOTSUPP; - goto out; + if (key.idx >= 0) { + int i; + bool ok = false; + for (i = 0; i < rdev->wiphy.n_cipher_suites; i++) { + if (key.p.cipher == rdev->wiphy.cipher_suites[i]) { + ok = true; + break; + } + } + if (!ok) + return -EINVAL; } - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->auth) + return -EOPNOTSUPP; - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) + return -EOPNOTSUPP; bssid = nla_data(info->attrs[NL80211_ATTR_MAC]); chan = ieee80211_get_channel(&rdev->wiphy, nla_get_u32(info->attrs[NL80211_ATTR_WIPHY_FREQ])); - if (!chan || (chan->flags & IEEE80211_CHAN_DISABLED)) { - err = -EINVAL; - goto out; - } + if (!chan || (chan->flags & IEEE80211_CHAN_DISABLED)) + return -EINVAL; ssid = nla_data(info->attrs[NL80211_ATTR_SSID]); ssid_len = nla_len(info->attrs[NL80211_ATTR_SSID]); @@ -3604,27 +3330,19 @@ static int nl80211_authenticate(struct sk_buff *skb, struct genl_info *info) } auth_type = nla_get_u32(info->attrs[NL80211_ATTR_AUTH_TYPE]); - if (!nl80211_valid_auth_type(auth_type)) { - err = -EINVAL; - goto out; - } + if (!nl80211_valid_auth_type(auth_type)) + return -EINVAL; local_state_change = !!info->attrs[NL80211_ATTR_LOCAL_STATE_CHANGE]; - err = cfg80211_mlme_auth(rdev, dev, chan, auth_type, bssid, - ssid, ssid_len, ie, ie_len, - key.p.key, key.p.key_len, key.idx, - local_state_change); - -out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); -unlock_rtnl: - rtnl_unlock(); - return err; + return cfg80211_mlme_auth(rdev, dev, chan, auth_type, bssid, + ssid, ssid_len, ie, ie_len, + key.p.key, key.p.key_len, key.idx, + local_state_change); } -static int nl80211_crypto_settings(struct genl_info *info, +static int nl80211_crypto_settings(struct cfg80211_registered_device *rdev, + struct genl_info *info, struct cfg80211_crypto_settings *settings, int cipher_limit) { @@ -3632,6 +3350,19 @@ static int nl80211_crypto_settings(struct genl_info *info, settings->control_port = info->attrs[NL80211_ATTR_CONTROL_PORT]; + if (info->attrs[NL80211_ATTR_CONTROL_PORT_ETHERTYPE]) { + u16 proto; + proto = nla_get_u16( + info->attrs[NL80211_ATTR_CONTROL_PORT_ETHERTYPE]); + settings->control_port_ethertype = cpu_to_be16(proto); + if (!(rdev->wiphy.flags & WIPHY_FLAG_CONTROL_PORT_PROTOCOL) && + proto != ETH_P_PAE) + return -EINVAL; + if (info->attrs[NL80211_ATTR_CONTROL_PORT_NO_ENCRYPT]) + settings->control_port_no_encrypt = true; + } else + settings->control_port_ethertype = cpu_to_be16(ETH_P_PAE); + if (info->attrs[NL80211_ATTR_CIPHER_SUITES_PAIRWISE]) { void *data; int len, i; @@ -3691,8 +3422,8 @@ static int nl80211_crypto_settings(struct genl_info *info, static int nl80211_associate(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; struct cfg80211_crypto_settings crypto; struct ieee80211_channel *chan; const u8 *bssid, *ssid, *ie = NULL, *prev_bssid = NULL; @@ -3707,35 +3438,19 @@ static int nl80211_associate(struct sk_buff *skb, struct genl_info *info) !info->attrs[NL80211_ATTR_WIPHY_FREQ]) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (!rdev->ops->assoc) { - err = -EOPNOTSUPP; - goto out; - } - - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->assoc) + return -EOPNOTSUPP; - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) + return -EOPNOTSUPP; bssid = nla_data(info->attrs[NL80211_ATTR_MAC]); chan = ieee80211_get_channel(&rdev->wiphy, nla_get_u32(info->attrs[NL80211_ATTR_WIPHY_FREQ])); - if (!chan || (chan->flags & IEEE80211_CHAN_DISABLED)) { - err = -EINVAL; - goto out; - } + if (!chan || (chan->flags & IEEE80211_CHAN_DISABLED)) + return -EINVAL; ssid = nla_data(info->attrs[NL80211_ATTR_SSID]); ssid_len = nla_len(info->attrs[NL80211_ATTR_SSID]); @@ -3750,35 +3465,28 @@ static int nl80211_associate(struct sk_buff *skb, struct genl_info *info) nla_get_u32(info->attrs[NL80211_ATTR_USE_MFP]); if (mfp == NL80211_MFP_REQUIRED) use_mfp = true; - else if (mfp != NL80211_MFP_NO) { - err = -EINVAL; - goto out; - } + else if (mfp != NL80211_MFP_NO) + return -EINVAL; } if (info->attrs[NL80211_ATTR_PREV_BSSID]) prev_bssid = nla_data(info->attrs[NL80211_ATTR_PREV_BSSID]); - err = nl80211_crypto_settings(info, &crypto, 1); + err = nl80211_crypto_settings(rdev, info, &crypto, 1); if (!err) err = cfg80211_mlme_assoc(rdev, dev, chan, bssid, prev_bssid, ssid, ssid_len, ie, ie_len, use_mfp, &crypto); -out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); -unlock_rtnl: - rtnl_unlock(); return err; } static int nl80211_deauthenticate(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; const u8 *ie = NULL, *bssid; - int err, ie_len = 0; + int ie_len = 0; u16 reason_code; bool local_state_change; @@ -3791,34 +3499,19 @@ static int nl80211_deauthenticate(struct sk_buff *skb, struct genl_info *info) if (!info->attrs[NL80211_ATTR_REASON_CODE]) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (!rdev->ops->deauth) { - err = -EOPNOTSUPP; - goto out; - } - - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->deauth) + return -EOPNOTSUPP; - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) + return -EOPNOTSUPP; bssid = nla_data(info->attrs[NL80211_ATTR_MAC]); reason_code = nla_get_u16(info->attrs[NL80211_ATTR_REASON_CODE]); if (reason_code == 0) { /* Reason Code 0 is reserved */ - err = -EINVAL; - goto out; + return -EINVAL; } if (info->attrs[NL80211_ATTR_IE]) { @@ -3828,23 +3521,16 @@ static int nl80211_deauthenticate(struct sk_buff *skb, struct genl_info *info) local_state_change = !!info->attrs[NL80211_ATTR_LOCAL_STATE_CHANGE]; - err = cfg80211_mlme_deauth(rdev, dev, bssid, ie, ie_len, reason_code, - local_state_change); - -out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); -unlock_rtnl: - rtnl_unlock(); - return err; + return cfg80211_mlme_deauth(rdev, dev, bssid, ie, ie_len, reason_code, + local_state_change); } static int nl80211_disassociate(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; const u8 *ie = NULL, *bssid; - int err, ie_len = 0; + int ie_len = 0; u16 reason_code; bool local_state_change; @@ -3857,34 +3543,19 @@ static int nl80211_disassociate(struct sk_buff *skb, struct genl_info *info) if (!info->attrs[NL80211_ATTR_REASON_CODE]) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (!rdev->ops->disassoc) { - err = -EOPNOTSUPP; - goto out; - } - - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->disassoc) + return -EOPNOTSUPP; - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) + return -EOPNOTSUPP; bssid = nla_data(info->attrs[NL80211_ATTR_MAC]); reason_code = nla_get_u16(info->attrs[NL80211_ATTR_REASON_CODE]); if (reason_code == 0) { /* Reason Code 0 is reserved */ - err = -EINVAL; - goto out; + return -EINVAL; } if (info->attrs[NL80211_ATTR_IE]) { @@ -3894,21 +3565,14 @@ static int nl80211_disassociate(struct sk_buff *skb, struct genl_info *info) local_state_change = !!info->attrs[NL80211_ATTR_LOCAL_STATE_CHANGE]; - err = cfg80211_mlme_disassoc(rdev, dev, bssid, ie, ie_len, reason_code, - local_state_change); - -out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); -unlock_rtnl: - rtnl_unlock(); - return err; + return cfg80211_mlme_disassoc(rdev, dev, bssid, ie, ie_len, reason_code, + local_state_change); } static int nl80211_join_ibss(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; struct cfg80211_ibss_params ibss; struct wiphy *wiphy; struct cfg80211_cached_keys *connkeys = NULL; @@ -3933,26 +3597,11 @@ static int nl80211_join_ibss(struct sk_buff *skb, struct genl_info *info) return -EINVAL; } - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (!rdev->ops->join_ibss) { - err = -EOPNOTSUPP; - goto out; - } - - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->join_ibss) + return -EOPNOTSUPP; - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC) + return -EOPNOTSUPP; wiphy = &rdev->wiphy; @@ -3970,24 +3619,12 @@ static int nl80211_join_ibss(struct sk_buff *skb, struct genl_info *info) nla_get_u32(info->attrs[NL80211_ATTR_WIPHY_FREQ])); if (!ibss.channel || ibss.channel->flags & IEEE80211_CHAN_NO_IBSS || - ibss.channel->flags & IEEE80211_CHAN_DISABLED) { - err = -EINVAL; - goto out; - } + ibss.channel->flags & IEEE80211_CHAN_DISABLED) + return -EINVAL; ibss.channel_fixed = !!info->attrs[NL80211_ATTR_FREQ_FIXED]; ibss.privacy = !!info->attrs[NL80211_ATTR_PRIVACY]; - if (ibss.privacy && info->attrs[NL80211_ATTR_KEYS]) { - connkeys = nl80211_parse_connkeys(rdev, - info->attrs[NL80211_ATTR_KEYS]); - if (IS_ERR(connkeys)) { - err = PTR_ERR(connkeys); - connkeys = NULL; - goto out; - } - } - if (info->attrs[NL80211_ATTR_BSS_BASIC_RATES]) { u8 *rates = nla_data(info->attrs[NL80211_ATTR_BSS_BASIC_RATES]); @@ -3997,10 +3634,8 @@ static int nl80211_join_ibss(struct sk_buff *skb, struct genl_info *info) wiphy->bands[ibss.channel->band]; int i, j; - if (n_rates == 0) { - err = -EINVAL; - goto out; - } + if (n_rates == 0) + return -EINVAL; for (i = 0; i < n_rates; i++) { int rate = (rates[i] & 0x7f) * 5; @@ -4013,77 +3648,36 @@ static int nl80211_join_ibss(struct sk_buff *skb, struct genl_info *info) break; } } - if (!found) { - err = -EINVAL; - goto out; - } - } - } else { - /* - * If no rates were explicitly configured, - * use the mandatory rate set for 11b or - * 11a for maximum compatibility. - */ - struct ieee80211_supported_band *sband = - wiphy->bands[ibss.channel->band]; - int j; - u32 flag = ibss.channel->band == IEEE80211_BAND_5GHZ ? - IEEE80211_RATE_MANDATORY_A : - IEEE80211_RATE_MANDATORY_B; - - for (j = 0; j < sband->n_bitrates; j++) { - if (sband->bitrates[j].flags & flag) - ibss.basic_rates |= BIT(j); + if (!found) + return -EINVAL; } } - err = cfg80211_join_ibss(rdev, dev, &ibss, connkeys); + if (ibss.privacy && info->attrs[NL80211_ATTR_KEYS]) { + connkeys = nl80211_parse_connkeys(rdev, + info->attrs[NL80211_ATTR_KEYS]); + if (IS_ERR(connkeys)) + return PTR_ERR(connkeys); + } -out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); -unlock_rtnl: + err = cfg80211_join_ibss(rdev, dev, &ibss, connkeys); if (err) kfree(connkeys); - rtnl_unlock(); return err; } static int nl80211_leave_ibss(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *dev; - int err; - - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (!rdev->ops->leave_ibss) { - err = -EOPNOTSUPP; - goto out; - } + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC) { - err = -EOPNOTSUPP; - goto out; - } - - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } + if (!rdev->ops->leave_ibss) + return -EOPNOTSUPP; - err = cfg80211_leave_ibss(rdev, dev, false); + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC) + return -EOPNOTSUPP; -out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); -unlock_rtnl: - rtnl_unlock(); - return err; + return cfg80211_leave_ibss(rdev, dev, false); } #ifdef CONFIG_NL80211_TESTMODE @@ -4093,20 +3687,12 @@ static struct genl_multicast_group nl80211_testmode_mcgrp = { static int nl80211_testmode_do(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; int err; if (!info->attrs[NL80211_ATTR_TESTDATA]) return -EINVAL; - rtnl_lock(); - - rdev = cfg80211_get_dev_from_info(info); - if (IS_ERR(rdev)) { - err = PTR_ERR(rdev); - goto unlock_rtnl; - } - err = -EOPNOTSUPP; if (rdev->ops->testmode_cmd) { rdev->testmode_info = info; @@ -4116,10 +3702,6 @@ static int nl80211_testmode_do(struct sk_buff *skb, struct genl_info *info) rdev->testmode_info = NULL; } - cfg80211_unlock_rdev(rdev); - - unlock_rtnl: - rtnl_unlock(); return err; } @@ -4210,8 +3792,8 @@ EXPORT_SYMBOL(cfg80211_testmode_event); static int nl80211_connect(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; struct cfg80211_connect_params connect; struct wiphy *wiphy; struct cfg80211_cached_keys *connkeys = NULL; @@ -4236,25 +3818,14 @@ static int nl80211_connect(struct sk_buff *skb, struct genl_info *info) connect.privacy = info->attrs[NL80211_ATTR_PRIVACY]; - err = nl80211_crypto_settings(info, &connect.crypto, + err = nl80211_crypto_settings(rdev, info, &connect.crypto, NL80211_MAX_NR_CIPHER_SUITES); if (err) return err; - rtnl_lock(); - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) { - err = -EOPNOTSUPP; - goto out; - } - - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) + return -EOPNOTSUPP; wiphy = &rdev->wiphy; @@ -4273,39 +3844,27 @@ static int nl80211_connect(struct sk_buff *skb, struct genl_info *info) ieee80211_get_channel(wiphy, nla_get_u32(info->attrs[NL80211_ATTR_WIPHY_FREQ])); if (!connect.channel || - connect.channel->flags & IEEE80211_CHAN_DISABLED) { - err = -EINVAL; - goto out; - } + connect.channel->flags & IEEE80211_CHAN_DISABLED) + return -EINVAL; } if (connect.privacy && info->attrs[NL80211_ATTR_KEYS]) { connkeys = nl80211_parse_connkeys(rdev, info->attrs[NL80211_ATTR_KEYS]); - if (IS_ERR(connkeys)) { - err = PTR_ERR(connkeys); - connkeys = NULL; - goto out; - } + if (IS_ERR(connkeys)) + return PTR_ERR(connkeys); } err = cfg80211_connect(rdev, dev, &connect, connkeys); - -out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); -unlock_rtnl: if (err) kfree(connkeys); - rtnl_unlock(); return err; } static int nl80211_disconnect(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *dev; - int err; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; u16 reason; if (!info->attrs[NL80211_ATTR_REASON_CODE]) @@ -4316,35 +3875,16 @@ static int nl80211_disconnect(struct sk_buff *skb, struct genl_info *info) if (reason == 0) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) { - err = -EOPNOTSUPP; - goto out; - } - - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } - - err = cfg80211_disconnect(rdev, dev, reason, true); + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) + return -EOPNOTSUPP; -out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); -unlock_rtnl: - rtnl_unlock(); - return err; + return cfg80211_disconnect(rdev, dev, reason, true); } static int nl80211_wiphy_netns(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; struct net *net; int err; u32 pid; @@ -4354,43 +3894,26 @@ static int nl80211_wiphy_netns(struct sk_buff *skb, struct genl_info *info) pid = nla_get_u32(info->attrs[NL80211_ATTR_PID]); - rtnl_lock(); - - rdev = cfg80211_get_dev_from_info(info); - if (IS_ERR(rdev)) { - err = PTR_ERR(rdev); - goto out_rtnl; - } - net = get_net_ns_by_pid(pid); - if (IS_ERR(net)) { - err = PTR_ERR(net); - goto out; - } + if (IS_ERR(net)) + return PTR_ERR(net); err = 0; /* check if anything to do */ - if (net_eq(wiphy_net(&rdev->wiphy), net)) - goto out_put_net; + if (!net_eq(wiphy_net(&rdev->wiphy), net)) + err = cfg80211_switch_netns(rdev, net); - err = cfg80211_switch_netns(rdev, net); - out_put_net: put_net(net); - out: - cfg80211_unlock_rdev(rdev); - out_rtnl: - rtnl_unlock(); return err; } static int nl80211_setdel_pmksa(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; int (*rdev_ops)(struct wiphy *wiphy, struct net_device *dev, struct cfg80211_pmksa *pmksa) = NULL; - int err; - struct net_device *dev; + struct net_device *dev = info->user_ptr[1]; struct cfg80211_pmksa pmksa; memset(&pmksa, 0, sizeof(struct cfg80211_pmksa)); @@ -4401,19 +3924,12 @@ static int nl80211_setdel_pmksa(struct sk_buff *skb, struct genl_info *info) if (!info->attrs[NL80211_ATTR_PMKID]) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - pmksa.pmkid = nla_data(info->attrs[NL80211_ATTR_PMKID]); pmksa.bssid = nla_data(info->attrs[NL80211_ATTR_MAC]); - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) { - err = -EOPNOTSUPP; - goto out; - } + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) + return -EOPNOTSUPP; switch (info->genlhdr->cmd) { case NL80211_CMD_SET_PMKSA: @@ -4427,61 +3943,32 @@ static int nl80211_setdel_pmksa(struct sk_buff *skb, struct genl_info *info) break; } - if (!rdev_ops) { - err = -EOPNOTSUPP; - goto out; - } - - err = rdev_ops(&rdev->wiphy, dev, &pmksa); - - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); + if (!rdev_ops) + return -EOPNOTSUPP; - return err; + return rdev_ops(&rdev->wiphy, dev, &pmksa); } static int nl80211_flush_pmksa(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - int err; - struct net_device *dev; - - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto out_rtnl; - - if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) { - err = -EOPNOTSUPP; - goto out; - } - - if (!rdev->ops->flush_pmksa) { - err = -EOPNOTSUPP; - goto out; - } - - err = rdev->ops->flush_pmksa(&rdev->wiphy, dev); + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - out_rtnl: - rtnl_unlock(); + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) + return -EOPNOTSUPP; - return err; + if (!rdev->ops->flush_pmksa) + return -EOPNOTSUPP; + return rdev->ops->flush_pmksa(&rdev->wiphy, dev); } static int nl80211_remain_on_channel(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; struct ieee80211_channel *chan; struct sk_buff *msg; void *hdr; @@ -4503,21 +3990,8 @@ static int nl80211_remain_on_channel(struct sk_buff *skb, if (!duration || !msecs_to_jiffies(duration) || duration > 5000) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (!rdev->ops->remain_on_channel) { - err = -EOPNOTSUPP; - goto out; - } - - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } + if (!rdev->ops->remain_on_channel) + return -EOPNOTSUPP; if (info->attrs[NL80211_ATTR_WIPHY_CHANNEL_TYPE]) { channel_type = nla_get_u32( @@ -4525,24 +3999,18 @@ static int nl80211_remain_on_channel(struct sk_buff *skb, if (channel_type != NL80211_CHAN_NO_HT && channel_type != NL80211_CHAN_HT20 && channel_type != NL80211_CHAN_HT40PLUS && - channel_type != NL80211_CHAN_HT40MINUS) { - err = -EINVAL; - goto out; - } + channel_type != NL80211_CHAN_HT40MINUS) + return -EINVAL; } freq = nla_get_u32(info->attrs[NL80211_ATTR_WIPHY_FREQ]); chan = rdev_freq_to_chan(rdev, freq, channel_type); - if (chan == NULL) { - err = -EINVAL; - goto out; - } + if (chan == NULL) + return -EINVAL; msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (!msg) { - err = -ENOMEM; - goto out; - } + if (!msg) + return -ENOMEM; hdr = nl80211hdr_put(msg, info->snd_pid, info->snd_seq, 0, NL80211_CMD_REMAIN_ON_CHANNEL); @@ -4561,58 +4029,32 @@ static int nl80211_remain_on_channel(struct sk_buff *skb, NLA_PUT_U64(msg, NL80211_ATTR_COOKIE, cookie); genlmsg_end(msg, hdr); - err = genlmsg_reply(msg, info); - goto out; + + return genlmsg_reply(msg, info); nla_put_failure: err = -ENOBUFS; free_msg: nlmsg_free(msg); - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - unlock_rtnl: - rtnl_unlock(); return err; } static int nl80211_cancel_remain_on_channel(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; u64 cookie; - int err; if (!info->attrs[NL80211_ATTR_COOKIE]) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (!rdev->ops->cancel_remain_on_channel) { - err = -EOPNOTSUPP; - goto out; - } - - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } + if (!rdev->ops->cancel_remain_on_channel) + return -EOPNOTSUPP; cookie = nla_get_u64(info->attrs[NL80211_ATTR_COOKIE]); - err = rdev->ops->cancel_remain_on_channel(&rdev->wiphy, dev, cookie); - - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - unlock_rtnl: - rtnl_unlock(); - return err; + return rdev->ops->cancel_remain_on_channel(&rdev->wiphy, dev, cookie); } static u32 rateset_to_mask(struct ieee80211_supported_band *sband, @@ -4648,26 +4090,18 @@ static int nl80211_set_tx_bitrate_mask(struct sk_buff *skb, struct genl_info *info) { struct nlattr *tb[NL80211_TXRATE_MAX + 1]; - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; struct cfg80211_bitrate_mask mask; - int err, rem, i; - struct net_device *dev; + int rem, i; + struct net_device *dev = info->user_ptr[1]; struct nlattr *tx_rates; struct ieee80211_supported_band *sband; if (info->attrs[NL80211_ATTR_TX_RATES] == NULL) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (!rdev->ops->set_bitrate_mask) { - err = -EOPNOTSUPP; - goto unlock; - } + if (!rdev->ops->set_bitrate_mask) + return -EOPNOTSUPP; memset(&mask, 0, sizeof(mask)); /* Default to all rates enabled */ @@ -4684,15 +4118,11 @@ static int nl80211_set_tx_bitrate_mask(struct sk_buff *skb, nla_for_each_nested(tx_rates, info->attrs[NL80211_ATTR_TX_RATES], rem) { enum ieee80211_band band = nla_type(tx_rates); - if (band < 0 || band >= IEEE80211_NUM_BANDS) { - err = -EINVAL; - goto unlock; - } + if (band < 0 || band >= IEEE80211_NUM_BANDS) + return -EINVAL; sband = rdev->wiphy.bands[band]; - if (sband == NULL) { - err = -EINVAL; - goto unlock; - } + if (sband == NULL) + return -EINVAL; nla_parse(tb, NL80211_TXRATE_MAX, nla_data(tx_rates), nla_len(tx_rates), nl80211_txattr_policy); if (tb[NL80211_TXRATE_LEGACY]) { @@ -4700,68 +4130,48 @@ static int nl80211_set_tx_bitrate_mask(struct sk_buff *skb, sband, nla_data(tb[NL80211_TXRATE_LEGACY]), nla_len(tb[NL80211_TXRATE_LEGACY])); - if (mask.control[band].legacy == 0) { - err = -EINVAL; - goto unlock; - } + if (mask.control[band].legacy == 0) + return -EINVAL; } } - err = rdev->ops->set_bitrate_mask(&rdev->wiphy, dev, NULL, &mask); - - unlock: - dev_put(dev); - cfg80211_unlock_rdev(rdev); - unlock_rtnl: - rtnl_unlock(); - return err; + return rdev->ops->set_bitrate_mask(&rdev->wiphy, dev, NULL, &mask); } -static int nl80211_register_action(struct sk_buff *skb, struct genl_info *info) +static int nl80211_register_mgmt(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *dev; - int err; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; + u16 frame_type = IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ACTION; if (!info->attrs[NL80211_ATTR_FRAME_MATCH]) return -EINVAL; - if (nla_len(info->attrs[NL80211_ATTR_FRAME_MATCH]) < 1) - return -EINVAL; - - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; + if (info->attrs[NL80211_ATTR_FRAME_TYPE]) + frame_type = nla_get_u16(info->attrs[NL80211_ATTR_FRAME_TYPE]); if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION && - dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC) { - err = -EOPNOTSUPP; - goto out; - } + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) + return -EOPNOTSUPP; /* not much point in registering if we can't reply */ - if (!rdev->ops->action) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->mgmt_tx) + return -EOPNOTSUPP; - err = cfg80211_mlme_register_action(dev->ieee80211_ptr, info->snd_pid, + return cfg80211_mlme_register_mgmt(dev->ieee80211_ptr, info->snd_pid, + frame_type, nla_data(info->attrs[NL80211_ATTR_FRAME_MATCH]), nla_len(info->attrs[NL80211_ATTR_FRAME_MATCH])); - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - unlock_rtnl: - rtnl_unlock(); - return err; } -static int nl80211_action(struct sk_buff *skb, struct genl_info *info) +static int nl80211_tx_mgmt(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; - struct net_device *dev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct net_device *dev = info->user_ptr[1]; struct ieee80211_channel *chan; enum nl80211_channel_type channel_type = NL80211_CHAN_NO_HT; bool channel_type_valid = false; @@ -4775,27 +4185,16 @@ static int nl80211_action(struct sk_buff *skb, struct genl_info *info) !info->attrs[NL80211_ATTR_WIPHY_FREQ]) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - - if (!rdev->ops->action) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->mgmt_tx) + return -EOPNOTSUPP; if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION && - dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC) { - err = -EOPNOTSUPP; - goto out; - } - - if (!netif_running(dev)) { - err = -ENETDOWN; - goto out; - } + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN && + dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) + return -EOPNOTSUPP; if (info->attrs[NL80211_ATTR_WIPHY_CHANNEL_TYPE]) { channel_type = nla_get_u32( @@ -4803,147 +4202,104 @@ static int nl80211_action(struct sk_buff *skb, struct genl_info *info) if (channel_type != NL80211_CHAN_NO_HT && channel_type != NL80211_CHAN_HT20 && channel_type != NL80211_CHAN_HT40PLUS && - channel_type != NL80211_CHAN_HT40MINUS) { - err = -EINVAL; - goto out; - } + channel_type != NL80211_CHAN_HT40MINUS) + return -EINVAL; channel_type_valid = true; } freq = nla_get_u32(info->attrs[NL80211_ATTR_WIPHY_FREQ]); chan = rdev_freq_to_chan(rdev, freq, channel_type); - if (chan == NULL) { - err = -EINVAL; - goto out; - } + if (chan == NULL) + return -EINVAL; msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (!msg) { - err = -ENOMEM; - goto out; - } + if (!msg) + return -ENOMEM; hdr = nl80211hdr_put(msg, info->snd_pid, info->snd_seq, 0, - NL80211_CMD_ACTION); + NL80211_CMD_FRAME); if (IS_ERR(hdr)) { err = PTR_ERR(hdr); goto free_msg; } - err = cfg80211_mlme_action(rdev, dev, chan, channel_type, - channel_type_valid, - nla_data(info->attrs[NL80211_ATTR_FRAME]), - nla_len(info->attrs[NL80211_ATTR_FRAME]), - &cookie); + err = cfg80211_mlme_mgmt_tx(rdev, dev, chan, channel_type, + channel_type_valid, + nla_data(info->attrs[NL80211_ATTR_FRAME]), + nla_len(info->attrs[NL80211_ATTR_FRAME]), + &cookie); if (err) goto free_msg; NLA_PUT_U64(msg, NL80211_ATTR_COOKIE, cookie); genlmsg_end(msg, hdr); - err = genlmsg_reply(msg, info); - goto out; + return genlmsg_reply(msg, info); nla_put_failure: err = -ENOBUFS; free_msg: nlmsg_free(msg); - out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); -unlock_rtnl: - rtnl_unlock(); return err; } static int nl80211_set_power_save(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; struct wireless_dev *wdev; - struct net_device *dev; + struct net_device *dev = info->user_ptr[1]; u8 ps_state; bool state; int err; - if (!info->attrs[NL80211_ATTR_PS_STATE]) { - err = -EINVAL; - goto out; - } + if (!info->attrs[NL80211_ATTR_PS_STATE]) + return -EINVAL; ps_state = nla_get_u32(info->attrs[NL80211_ATTR_PS_STATE]); - if (ps_state != NL80211_PS_DISABLED && ps_state != NL80211_PS_ENABLED) { - err = -EINVAL; - goto out; - } - - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rdev; + if (ps_state != NL80211_PS_DISABLED && ps_state != NL80211_PS_ENABLED) + return -EINVAL; wdev = dev->ieee80211_ptr; - if (!rdev->ops->set_power_mgmt) { - err = -EOPNOTSUPP; - goto unlock_rdev; - } + if (!rdev->ops->set_power_mgmt) + return -EOPNOTSUPP; state = (ps_state == NL80211_PS_ENABLED) ? true : false; if (state == wdev->ps) - goto unlock_rdev; - - wdev->ps = state; - - if (rdev->ops->set_power_mgmt(wdev->wiphy, dev, wdev->ps, - wdev->ps_timeout)) - /* assume this means it's off */ - wdev->ps = false; - -unlock_rdev: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - rtnl_unlock(); + return 0; -out: + err = rdev->ops->set_power_mgmt(wdev->wiphy, dev, state, + wdev->ps_timeout); + if (!err) + wdev->ps = state; return err; } static int nl80211_get_power_save(struct sk_buff *skb, struct genl_info *info) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; enum nl80211_ps_state ps_state; struct wireless_dev *wdev; - struct net_device *dev; + struct net_device *dev = info->user_ptr[1]; struct sk_buff *msg; void *hdr; int err; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rtnl; - wdev = dev->ieee80211_ptr; - if (!rdev->ops->set_power_mgmt) { - err = -EOPNOTSUPP; - goto out; - } + if (!rdev->ops->set_power_mgmt) + return -EOPNOTSUPP; msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (!msg) { - err = -ENOMEM; - goto out; - } + if (!msg) + return -ENOMEM; hdr = nl80211hdr_put(msg, info->snd_pid, info->snd_seq, 0, NL80211_CMD_GET_POWER_SAVE); if (!hdr) { - err = -ENOMEM; + err = -ENOBUFS; goto free_msg; } @@ -4955,22 +4311,12 @@ static int nl80211_get_power_save(struct sk_buff *skb, struct genl_info *info) NLA_PUT_U32(msg, NL80211_ATTR_PS_STATE, ps_state); genlmsg_end(msg, hdr); - err = genlmsg_reply(msg, info); - goto out; + return genlmsg_reply(msg, info); -nla_put_failure: + nla_put_failure: err = -ENOBUFS; - -free_msg: + free_msg: nlmsg_free(msg); - -out: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - -unlock_rtnl: - rtnl_unlock(); - return err; } @@ -4984,41 +4330,24 @@ nl80211_attr_cqm_policy[NL80211_ATTR_CQM_MAX + 1] __read_mostly = { static int nl80211_set_cqm_rssi(struct genl_info *info, s32 threshold, u32 hysteresis) { - struct cfg80211_registered_device *rdev; + struct cfg80211_registered_device *rdev = info->user_ptr[0]; struct wireless_dev *wdev; - struct net_device *dev; - int err; + struct net_device *dev = info->user_ptr[1]; if (threshold > 0) return -EINVAL; - rtnl_lock(); - - err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); - if (err) - goto unlock_rdev; - wdev = dev->ieee80211_ptr; - if (!rdev->ops->set_cqm_rssi_config) { - err = -EOPNOTSUPP; - goto unlock_rdev; - } - - if (wdev->iftype != NL80211_IFTYPE_STATION) { - err = -EOPNOTSUPP; - goto unlock_rdev; - } - - err = rdev->ops->set_cqm_rssi_config(wdev->wiphy, dev, - threshold, hysteresis); + if (!rdev->ops->set_cqm_rssi_config) + return -EOPNOTSUPP; -unlock_rdev: - cfg80211_unlock_rdev(rdev); - dev_put(dev); - rtnl_unlock(); + if (wdev->iftype != NL80211_IFTYPE_STATION && + wdev->iftype != NL80211_IFTYPE_P2P_CLIENT) + return -EOPNOTSUPP; - return err; + return rdev->ops->set_cqm_rssi_config(wdev->wiphy, dev, + threshold, hysteresis); } static int nl80211_set_cqm(struct sk_buff *skb, struct genl_info *info) @@ -5052,6 +4381,65 @@ out: return err; } +#define NL80211_FLAG_NEED_WIPHY 0x01 +#define NL80211_FLAG_NEED_NETDEV 0x02 +#define NL80211_FLAG_NEED_RTNL 0x04 +#define NL80211_FLAG_CHECK_NETDEV_UP 0x08 +#define NL80211_FLAG_NEED_NETDEV_UP (NL80211_FLAG_NEED_NETDEV |\ + NL80211_FLAG_CHECK_NETDEV_UP) + +static int nl80211_pre_doit(struct genl_ops *ops, struct sk_buff *skb, + struct genl_info *info) +{ + struct cfg80211_registered_device *rdev; + struct net_device *dev; + int err; + bool rtnl = ops->internal_flags & NL80211_FLAG_NEED_RTNL; + + if (rtnl) + rtnl_lock(); + + if (ops->internal_flags & NL80211_FLAG_NEED_WIPHY) { + rdev = cfg80211_get_dev_from_info(info); + if (IS_ERR(rdev)) { + if (rtnl) + rtnl_unlock(); + return PTR_ERR(rdev); + } + info->user_ptr[0] = rdev; + } else if (ops->internal_flags & NL80211_FLAG_NEED_NETDEV) { + err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); + if (err) { + if (rtnl) + rtnl_unlock(); + return err; + } + if (ops->internal_flags & NL80211_FLAG_CHECK_NETDEV_UP && + !netif_running(dev)) { + cfg80211_unlock_rdev(rdev); + dev_put(dev); + if (rtnl) + rtnl_unlock(); + return -ENETDOWN; + } + info->user_ptr[0] = rdev; + info->user_ptr[1] = dev; + } + + return 0; +} + +static void nl80211_post_doit(struct genl_ops *ops, struct sk_buff *skb, + struct genl_info *info) +{ + if (info->user_ptr[0]) + cfg80211_unlock_rdev(info->user_ptr[0]); + if (info->user_ptr[1]) + dev_put(info->user_ptr[1]); + if (ops->internal_flags & NL80211_FLAG_NEED_RTNL) + rtnl_unlock(); +} + static struct genl_ops nl80211_ops[] = { { .cmd = NL80211_CMD_GET_WIPHY, @@ -5059,12 +4447,14 @@ static struct genl_ops nl80211_ops[] = { .dumpit = nl80211_dump_wiphy, .policy = nl80211_policy, /* can be retrieved by unprivileged users */ + .internal_flags = NL80211_FLAG_NEED_WIPHY, }, { .cmd = NL80211_CMD_SET_WIPHY, .doit = nl80211_set_wiphy, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_GET_INTERFACE, @@ -5072,90 +4462,119 @@ static struct genl_ops nl80211_ops[] = { .dumpit = nl80211_dump_interface, .policy = nl80211_policy, /* can be retrieved by unprivileged users */ + .internal_flags = NL80211_FLAG_NEED_NETDEV, }, { .cmd = NL80211_CMD_SET_INTERFACE, .doit = nl80211_set_interface, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_NEW_INTERFACE, .doit = nl80211_new_interface, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_WIPHY | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_DEL_INTERFACE, .doit = nl80211_del_interface, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_GET_KEY, .doit = nl80211_get_key, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_SET_KEY, .doit = nl80211_set_key, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_NEW_KEY, .doit = nl80211_new_key, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_DEL_KEY, .doit = nl80211_del_key, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_SET_BEACON, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, .doit = nl80211_addset_beacon, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_NEW_BEACON, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, .doit = nl80211_addset_beacon, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_DEL_BEACON, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, .doit = nl80211_del_beacon, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_GET_STATION, .doit = nl80211_get_station, .dumpit = nl80211_dump_station, .policy = nl80211_policy, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_SET_STATION, .doit = nl80211_set_station, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_NEW_STATION, .doit = nl80211_new_station, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_DEL_STATION, .doit = nl80211_del_station, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_GET_MPATH, @@ -5163,30 +4582,40 @@ static struct genl_ops nl80211_ops[] = { .dumpit = nl80211_dump_mpath, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_SET_MPATH, .doit = nl80211_set_mpath, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_NEW_MPATH, .doit = nl80211_new_mpath, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_DEL_MPATH, .doit = nl80211_del_mpath, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_SET_BSS, .doit = nl80211_set_bss, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_GET_REG, @@ -5211,18 +4640,24 @@ static struct genl_ops nl80211_ops[] = { .doit = nl80211_get_mesh_params, .policy = nl80211_policy, /* can be retrieved by unprivileged users */ + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_SET_MESH_PARAMS, .doit = nl80211_set_mesh_params, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_TRIGGER_SCAN, .doit = nl80211_trigger_scan, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_GET_SCAN, @@ -5234,36 +4669,48 @@ static struct genl_ops nl80211_ops[] = { .doit = nl80211_authenticate, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_ASSOCIATE, .doit = nl80211_associate, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_DEAUTHENTICATE, .doit = nl80211_deauthenticate, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_DISASSOCIATE, .doit = nl80211_disassociate, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_JOIN_IBSS, .doit = nl80211_join_ibss, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_LEAVE_IBSS, .doit = nl80211_leave_ibss, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, #ifdef CONFIG_NL80211_TESTMODE { @@ -5271,6 +4718,8 @@ static struct genl_ops nl80211_ops[] = { .doit = nl80211_testmode_do, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_WIPHY | + NL80211_FLAG_NEED_RTNL, }, #endif { @@ -5278,18 +4727,24 @@ static struct genl_ops nl80211_ops[] = { .doit = nl80211_connect, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_DISCONNECT, .doit = nl80211_disconnect, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_SET_WIPHY_NETNS, .doit = nl80211_wiphy_netns, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_WIPHY | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_GET_SURVEY, @@ -5301,72 +4756,104 @@ static struct genl_ops nl80211_ops[] = { .doit = nl80211_setdel_pmksa, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_DEL_PMKSA, .doit = nl80211_setdel_pmksa, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_FLUSH_PMKSA, .doit = nl80211_flush_pmksa, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_REMAIN_ON_CHANNEL, .doit = nl80211_remain_on_channel, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_CANCEL_REMAIN_ON_CHANNEL, .doit = nl80211_cancel_remain_on_channel, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_SET_TX_BITRATE_MASK, .doit = nl80211_set_tx_bitrate_mask, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { - .cmd = NL80211_CMD_REGISTER_ACTION, - .doit = nl80211_register_action, + .cmd = NL80211_CMD_REGISTER_FRAME, + .doit = nl80211_register_mgmt, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { - .cmd = NL80211_CMD_ACTION, - .doit = nl80211_action, + .cmd = NL80211_CMD_FRAME, + .doit = nl80211_tx_mgmt, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_SET_POWER_SAVE, .doit = nl80211_set_power_save, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_GET_POWER_SAVE, .doit = nl80211_get_power_save, .policy = nl80211_policy, /* can be retrieved by unprivileged users */ + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_SET_CQM, .doit = nl80211_set_cqm, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_SET_CHANNEL, .doit = nl80211_set_channel, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, + }, + { + .cmd = NL80211_CMD_SET_WDS_PEER, + .doit = nl80211_set_wds_peer, + .policy = nl80211_policy, + .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_NETDEV | + NL80211_FLAG_NEED_RTNL, }, }; @@ -6040,9 +5527,9 @@ void nl80211_send_sta_event(struct cfg80211_registered_device *rdev, nl80211_mlme_mcgrp.id, gfp); } -int nl80211_send_action(struct cfg80211_registered_device *rdev, - struct net_device *netdev, u32 nlpid, - int freq, const u8 *buf, size_t len, gfp_t gfp) +int nl80211_send_mgmt(struct cfg80211_registered_device *rdev, + struct net_device *netdev, u32 nlpid, + int freq, const u8 *buf, size_t len, gfp_t gfp) { struct sk_buff *msg; void *hdr; @@ -6052,7 +5539,7 @@ int nl80211_send_action(struct cfg80211_registered_device *rdev, if (!msg) return -ENOMEM; - hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_ACTION); + hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_FRAME); if (!hdr) { nlmsg_free(msg); return -ENOMEM; @@ -6080,10 +5567,10 @@ int nl80211_send_action(struct cfg80211_registered_device *rdev, return -ENOBUFS; } -void nl80211_send_action_tx_status(struct cfg80211_registered_device *rdev, - struct net_device *netdev, u64 cookie, - const u8 *buf, size_t len, bool ack, - gfp_t gfp) +void nl80211_send_mgmt_tx_status(struct cfg80211_registered_device *rdev, + struct net_device *netdev, u64 cookie, + const u8 *buf, size_t len, bool ack, + gfp_t gfp) { struct sk_buff *msg; void *hdr; @@ -6092,7 +5579,7 @@ void nl80211_send_action_tx_status(struct cfg80211_registered_device *rdev, if (!msg) return; - hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_ACTION_TX_STATUS); + hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_FRAME_TX_STATUS); if (!hdr) { nlmsg_free(msg); return; @@ -6179,7 +5666,7 @@ static int nl80211_netlink_notify(struct notifier_block * nb, list_for_each_entry_rcu(rdev, &cfg80211_rdev_list, list) list_for_each_entry_rcu(wdev, &rdev->netdev_list, list) - cfg80211_mlme_unregister_actions(wdev, notify->pid); + cfg80211_mlme_unregister_socket(wdev, notify->pid); rcu_read_unlock(); diff --git a/net/wireless/nl80211.h b/net/wireless/nl80211.h index 2ad7fbc7d9f..30d2f939150 100644 --- a/net/wireless/nl80211.h +++ b/net/wireless/nl80211.h @@ -74,13 +74,13 @@ void nl80211_send_sta_event(struct cfg80211_registered_device *rdev, struct net_device *dev, const u8 *mac_addr, struct station_info *sinfo, gfp_t gfp); -int nl80211_send_action(struct cfg80211_registered_device *rdev, - struct net_device *netdev, u32 nlpid, int freq, - const u8 *buf, size_t len, gfp_t gfp); -void nl80211_send_action_tx_status(struct cfg80211_registered_device *rdev, - struct net_device *netdev, u64 cookie, - const u8 *buf, size_t len, bool ack, - gfp_t gfp); +int nl80211_send_mgmt(struct cfg80211_registered_device *rdev, + struct net_device *netdev, u32 nlpid, int freq, + const u8 *buf, size_t len, gfp_t gfp); +void nl80211_send_mgmt_tx_status(struct cfg80211_registered_device *rdev, + struct net_device *netdev, u64 cookie, + const u8 *buf, size_t len, bool ack, + gfp_t gfp); void nl80211_send_cqm_rssi_notify(struct cfg80211_registered_device *rdev, diff --git a/net/wireless/radiotap.c b/net/wireless/radiotap.c index 1332c445d1c..dbe35e138e9 100644 --- a/net/wireless/radiotap.c +++ b/net/wireless/radiotap.c @@ -14,6 +14,7 @@ * See COPYING for more details. */ +#include <linux/kernel.h> #include <net/cfg80211.h> #include <net/ieee80211_radiotap.h> #include <asm/unaligned.h> @@ -45,7 +46,7 @@ static const struct radiotap_align_size rtap_namespace_sizes[] = { }; static const struct ieee80211_radiotap_namespace radiotap_ns = { - .n_bits = sizeof(rtap_namespace_sizes) / sizeof(rtap_namespace_sizes[0]), + .n_bits = ARRAY_SIZE(rtap_namespace_sizes), .align_size = rtap_namespace_sizes, }; @@ -200,7 +201,7 @@ int ieee80211_radiotap_iterator_next( { while (1) { int hit = 0; - int pad, align, size, subns, vnslen; + int pad, align, size, subns; uint32_t oui; /* if no more EXT bits, that's it */ @@ -260,6 +261,27 @@ int ieee80211_radiotap_iterator_next( if (pad) iterator->_arg += align - pad; + if (iterator->_arg_index % 32 == IEEE80211_RADIOTAP_VENDOR_NAMESPACE) { + int vnslen; + + if ((unsigned long)iterator->_arg + size - + (unsigned long)iterator->_rtheader > + (unsigned long)iterator->_max_length) + return -EINVAL; + + oui = (*iterator->_arg << 16) | + (*(iterator->_arg + 1) << 8) | + *(iterator->_arg + 2); + subns = *(iterator->_arg + 3); + + find_ns(iterator, oui, subns); + + vnslen = get_unaligned_le16(iterator->_arg + 4); + iterator->_next_ns_data = iterator->_arg + size + vnslen; + if (!iterator->current_namespace) + size += vnslen; + } + /* * this is what we will return to user, but we need to * move on first so next call has something fresh to test @@ -286,40 +308,25 @@ int ieee80211_radiotap_iterator_next( /* these special ones are valid in each bitmap word */ switch (iterator->_arg_index % 32) { case IEEE80211_RADIOTAP_VENDOR_NAMESPACE: - iterator->_bitmap_shifter >>= 1; - iterator->_arg_index++; - iterator->_reset_on_ext = 1; - vnslen = get_unaligned_le16(iterator->this_arg + 4); - iterator->_next_ns_data = iterator->_arg + vnslen; - oui = (*iterator->this_arg << 16) | - (*(iterator->this_arg + 1) << 8) | - *(iterator->this_arg + 2); - subns = *(iterator->this_arg + 3); - - find_ns(iterator, oui, subns); - iterator->is_radiotap_ns = 0; - /* allow parsers to show this information */ + /* + * If parser didn't register this vendor + * namespace with us, allow it to show it + * as 'raw. Do do that, set argument index + * to vendor namespace. + */ iterator->this_arg_index = IEEE80211_RADIOTAP_VENDOR_NAMESPACE; - iterator->this_arg_size += vnslen; - if ((unsigned long)iterator->this_arg + - iterator->this_arg_size - - (unsigned long)iterator->_rtheader > - (unsigned long)(unsigned long)iterator->_max_length) - return -EINVAL; - hit = 1; - break; + if (!iterator->current_namespace) + hit = 1; + goto next_entry; case IEEE80211_RADIOTAP_RADIOTAP_NAMESPACE: - iterator->_bitmap_shifter >>= 1; - iterator->_arg_index++; - iterator->_reset_on_ext = 1; iterator->current_namespace = &radiotap_ns; iterator->is_radiotap_ns = 1; - break; + goto next_entry; case IEEE80211_RADIOTAP_EXT: /* * bit 31 was set, there is more diff --git a/net/wireless/reg.c b/net/wireless/reg.c index f180db0de66..d14bbf960c1 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -36,6 +36,7 @@ #include <linux/slab.h> #include <linux/list.h> #include <linux/random.h> +#include <linux/ctype.h> #include <linux/nl80211.h> #include <linux/platform_device.h> #include <net/cfg80211.h> @@ -73,7 +74,11 @@ const struct ieee80211_regdomain *cfg80211_regdomain; * - last_request */ static DEFINE_MUTEX(reg_mutex); -#define assert_reg_lock() WARN_ON(!mutex_is_locked(®_mutex)) + +static inline void assert_reg_lock(void) +{ + lockdep_assert_held(®_mutex); +} /* Used to queue up regulatory hints */ static LIST_HEAD(reg_requests_list); @@ -181,14 +186,6 @@ static bool is_alpha2_set(const char *alpha2) return false; } -static bool is_alpha_upper(char letter) -{ - /* ASCII A - Z */ - if (letter >= 65 && letter <= 90) - return true; - return false; -} - static bool is_unknown_alpha2(const char *alpha2) { if (!alpha2) @@ -220,7 +217,7 @@ static bool is_an_alpha2(const char *alpha2) { if (!alpha2) return false; - if (is_alpha_upper(alpha2[0]) && is_alpha_upper(alpha2[1])) + if (isalpha(alpha2[0]) && isalpha(alpha2[1])) return true; return false; } @@ -1399,6 +1396,11 @@ static DECLARE_WORK(reg_work, reg_todo); static void queue_regulatory_request(struct regulatory_request *request) { + if (isalpha(request->alpha2[0])) + request->alpha2[0] = toupper(request->alpha2[0]); + if (isalpha(request->alpha2[1])) + request->alpha2[1] = toupper(request->alpha2[1]); + spin_lock(®_requests_lock); list_add_tail(&request->list, ®_requests_list); spin_unlock(®_requests_lock); diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 5ca8c718014..503ebb86ba1 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -650,14 +650,14 @@ void cfg80211_unlink_bss(struct wiphy *wiphy, struct cfg80211_bss *pub) bss = container_of(pub, struct cfg80211_internal_bss, pub); spin_lock_bh(&dev->bss_lock); + if (!list_empty(&bss->list)) { + list_del_init(&bss->list); + dev->bss_generation++; + rb_erase(&bss->rbn, &dev->bss_tree); - list_del(&bss->list); - dev->bss_generation++; - rb_erase(&bss->rbn, &dev->bss_tree); - + kref_put(&bss->ref, bss_release); + } spin_unlock_bh(&dev->bss_lock); - - kref_put(&bss->ref, bss_release); } EXPORT_SYMBOL(cfg80211_unlink_bss); diff --git a/net/wireless/sme.c b/net/wireless/sme.c index a8c2d6b877a..e17b0bee6bd 100644 --- a/net/wireless/sme.c +++ b/net/wireless/sme.c @@ -411,7 +411,8 @@ void __cfg80211_connect_result(struct net_device *dev, const u8 *bssid, ASSERT_WDEV_LOCK(wdev); - if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION)) + if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION && + wdev->iftype != NL80211_IFTYPE_P2P_CLIENT)) return; if (wdev->sme_state != CFG80211_SME_CONNECTING) @@ -548,7 +549,8 @@ void __cfg80211_roamed(struct wireless_dev *wdev, const u8 *bssid, ASSERT_WDEV_LOCK(wdev); - if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION)) + if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION && + wdev->iftype != NL80211_IFTYPE_P2P_CLIENT)) return; if (wdev->sme_state != CFG80211_SME_CONNECTED) @@ -644,7 +646,8 @@ void __cfg80211_disconnected(struct net_device *dev, const u8 *ie, ASSERT_WDEV_LOCK(wdev); - if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION)) + if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION && + wdev->iftype != NL80211_IFTYPE_P2P_CLIENT)) return; if (wdev->sme_state != CFG80211_SME_CONNECTED) @@ -695,7 +698,7 @@ void __cfg80211_disconnected(struct net_device *dev, const u8 *ie, */ if (rdev->ops->del_key) for (i = 0; i < 6; i++) - rdev->ops->del_key(wdev->wiphy, dev, i, NULL); + rdev->ops->del_key(wdev->wiphy, dev, i, false, NULL); #ifdef CONFIG_CFG80211_WEXT memset(&wrqu, 0, sizeof(wrqu)); diff --git a/net/wireless/sysfs.c b/net/wireless/sysfs.c index 9f2cef3e0ca..4294fa22bb2 100644 --- a/net/wireless/sysfs.c +++ b/net/wireless/sysfs.c @@ -35,6 +35,14 @@ SHOW_FMT(index, "%d", wiphy_idx); SHOW_FMT(macaddress, "%pM", wiphy.perm_addr); SHOW_FMT(address_mask, "%pM", wiphy.addr_mask); +static ssize_t name_show(struct device *dev, + struct device_attribute *attr, + char *buf) { + struct wiphy *wiphy = &dev_to_rdev(dev)->wiphy; + return sprintf(buf, "%s\n", dev_name(&wiphy->dev)); +} + + static ssize_t addresses_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -57,6 +65,7 @@ static struct device_attribute ieee80211_dev_attrs[] = { __ATTR_RO(macaddress), __ATTR_RO(address_mask), __ATTR_RO(addresses), + __ATTR_RO(name), {} }; @@ -110,6 +119,13 @@ static int wiphy_resume(struct device *dev) return ret; } +static const void *wiphy_namespace(struct device *d) +{ + struct wiphy *wiphy = container_of(d, struct wiphy, dev); + + return wiphy_net(wiphy); +} + struct class ieee80211_class = { .name = "ieee80211", .owner = THIS_MODULE, @@ -120,6 +136,8 @@ struct class ieee80211_class = { #endif .suspend = wiphy_suspend, .resume = wiphy_resume, + .ns_type = &net_ns_type_operations, + .namespace = wiphy_namespace, }; int wiphy_sysfs_init(void) diff --git a/net/wireless/util.c b/net/wireless/util.c index 0c8a1e8b769..76120aeda57 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -144,19 +144,25 @@ void ieee80211_set_bitrate_flags(struct wiphy *wiphy) int cfg80211_validate_key_settings(struct cfg80211_registered_device *rdev, struct key_params *params, int key_idx, - const u8 *mac_addr) + bool pairwise, const u8 *mac_addr) { int i; if (key_idx > 5) return -EINVAL; + if (!pairwise && mac_addr && !(rdev->wiphy.flags & WIPHY_FLAG_IBSS_RSN)) + return -EINVAL; + + if (pairwise && !mac_addr) + return -EINVAL; + /* * Disallow pairwise keys with non-zero index unless it's WEP * (because current deployments use pairwise WEP keys with * non-zero indizes but 802.11i clearly specifies to use zero) */ - if (mac_addr && key_idx && + if (pairwise && key_idx && params->cipher != WLAN_CIPHER_SUITE_WEP40 && params->cipher != WLAN_CIPHER_SUITE_WEP104) return -EINVAL; @@ -183,7 +189,14 @@ int cfg80211_validate_key_settings(struct cfg80211_registered_device *rdev, return -EINVAL; break; default: - return -EINVAL; + /* + * We don't know anything about this algorithm, + * allow using it -- but the driver must check + * all parameters! We still check below whether + * or not the driver supports this algorithm, + * of course. + */ + break; } if (params->seq) { @@ -221,7 +234,7 @@ const unsigned char bridge_tunnel_header[] __aligned(2) = { 0xaa, 0xaa, 0x03, 0x00, 0x00, 0xf8 }; EXPORT_SYMBOL(bridge_tunnel_header); -unsigned int ieee80211_hdrlen(__le16 fc) +unsigned int __attribute_const__ ieee80211_hdrlen(__le16 fc) { unsigned int hdrlen = 24; @@ -319,7 +332,8 @@ int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr, cpu_to_le16(IEEE80211_FCTL_TODS | IEEE80211_FCTL_FROMDS)) { case cpu_to_le16(IEEE80211_FCTL_TODS): if (unlikely(iftype != NL80211_IFTYPE_AP && - iftype != NL80211_IFTYPE_AP_VLAN)) + iftype != NL80211_IFTYPE_AP_VLAN && + iftype != NL80211_IFTYPE_P2P_GO)) return -1; break; case cpu_to_le16(IEEE80211_FCTL_TODS | IEEE80211_FCTL_FROMDS): @@ -347,7 +361,8 @@ int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr, break; case cpu_to_le16(IEEE80211_FCTL_FROMDS): if ((iftype != NL80211_IFTYPE_STATION && - iftype != NL80211_IFTYPE_MESH_POINT) || + iftype != NL80211_IFTYPE_P2P_CLIENT && + iftype != NL80211_IFTYPE_MESH_POINT) || (is_multicast_ether_addr(dst) && !compare_ether_addr(src, addr))) return -1; @@ -424,6 +439,7 @@ int ieee80211_data_from_8023(struct sk_buff *skb, const u8 *addr, switch (iftype) { case NL80211_IFTYPE_AP: case NL80211_IFTYPE_AP_VLAN: + case NL80211_IFTYPE_P2P_GO: fc |= cpu_to_le16(IEEE80211_FCTL_FROMDS); /* DA BSSID SA */ memcpy(hdr.addr1, skb->data, ETH_ALEN); @@ -432,6 +448,7 @@ int ieee80211_data_from_8023(struct sk_buff *skb, const u8 *addr, hdrlen = 24; break; case NL80211_IFTYPE_STATION: + case NL80211_IFTYPE_P2P_CLIENT: fc |= cpu_to_le16(IEEE80211_FCTL_TODS); /* BSSID SA DA */ memcpy(hdr.addr1, bssid, ETH_ALEN); @@ -666,7 +683,7 @@ void cfg80211_upload_connect_keys(struct wireless_dev *wdev) for (i = 0; i < 6; i++) { if (!wdev->connect_keys->params[i].cipher) continue; - if (rdev->ops->add_key(wdev->wiphy, dev, i, NULL, + if (rdev->ops->add_key(wdev->wiphy, dev, i, false, NULL, &wdev->connect_keys->params[i])) { printk(KERN_ERR "%s: failed to set key %d\n", dev->name, i); @@ -771,7 +788,9 @@ int cfg80211_change_iface(struct cfg80211_registered_device *rdev, /* if it's part of a bridge, reject changing type to station/ibss */ if ((dev->priv_flags & IFF_BRIDGE_PORT) && - (ntype == NL80211_IFTYPE_ADHOC || ntype == NL80211_IFTYPE_STATION)) + (ntype == NL80211_IFTYPE_ADHOC || + ntype == NL80211_IFTYPE_STATION || + ntype == NL80211_IFTYPE_P2P_CLIENT)) return -EBUSY; if (ntype != otype) { @@ -782,6 +801,7 @@ int cfg80211_change_iface(struct cfg80211_registered_device *rdev, cfg80211_leave_ibss(rdev, dev, false); break; case NL80211_IFTYPE_STATION: + case NL80211_IFTYPE_P2P_CLIENT: cfg80211_disconnect(rdev, dev, WLAN_REASON_DEAUTH_LEAVING, true); break; @@ -810,9 +830,11 @@ int cfg80211_change_iface(struct cfg80211_registered_device *rdev, if (dev->ieee80211_ptr->use_4addr) break; /* fall through */ + case NL80211_IFTYPE_P2P_CLIENT: case NL80211_IFTYPE_ADHOC: dev->priv_flags |= IFF_DONT_BRIDGE; break; + case NL80211_IFTYPE_P2P_GO: case NL80211_IFTYPE_AP: case NL80211_IFTYPE_AP_VLAN: case NL80211_IFTYPE_WDS: @@ -823,7 +845,7 @@ int cfg80211_change_iface(struct cfg80211_registered_device *rdev, /* monitor can't bridge anyway */ break; case NL80211_IFTYPE_UNSPECIFIED: - case __NL80211_IFTYPE_AFTER_LAST: + case NUM_NL80211_IFTYPES: /* not happening */ break; } diff --git a/net/wireless/wext-compat.c b/net/wireless/wext-compat.c index bb5e0a5ecfa..12222ee6ebf 100644 --- a/net/wireless/wext-compat.c +++ b/net/wireless/wext-compat.c @@ -432,14 +432,17 @@ int cfg80211_wext_giwretry(struct net_device *dev, EXPORT_SYMBOL_GPL(cfg80211_wext_giwretry); static int __cfg80211_set_encryption(struct cfg80211_registered_device *rdev, - struct net_device *dev, const u8 *addr, - bool remove, bool tx_key, int idx, - struct key_params *params) + struct net_device *dev, bool pairwise, + const u8 *addr, bool remove, bool tx_key, + int idx, struct key_params *params) { struct wireless_dev *wdev = dev->ieee80211_ptr; int err, i; bool rejoin = false; + if (pairwise && !addr) + return -EINVAL; + if (!wdev->wext.keys) { wdev->wext.keys = kzalloc(sizeof(*wdev->wext.keys), GFP_KERNEL); @@ -478,7 +481,13 @@ static int __cfg80211_set_encryption(struct cfg80211_registered_device *rdev, __cfg80211_leave_ibss(rdev, wdev->netdev, true); rejoin = true; } - err = rdev->ops->del_key(&rdev->wiphy, dev, idx, addr); + + if (!pairwise && addr && + !(rdev->wiphy.flags & WIPHY_FLAG_IBSS_RSN)) + err = -ENOENT; + else + err = rdev->ops->del_key(&rdev->wiphy, dev, idx, + pairwise, addr); } wdev->wext.connect.privacy = false; /* @@ -507,12 +516,13 @@ static int __cfg80211_set_encryption(struct cfg80211_registered_device *rdev, if (addr) tx_key = false; - if (cfg80211_validate_key_settings(rdev, params, idx, addr)) + if (cfg80211_validate_key_settings(rdev, params, idx, pairwise, addr)) return -EINVAL; err = 0; if (wdev->current_bss) - err = rdev->ops->add_key(&rdev->wiphy, dev, idx, addr, params); + err = rdev->ops->add_key(&rdev->wiphy, dev, idx, + pairwise, addr, params); if (err) return err; @@ -563,17 +573,17 @@ static int __cfg80211_set_encryption(struct cfg80211_registered_device *rdev, } static int cfg80211_set_encryption(struct cfg80211_registered_device *rdev, - struct net_device *dev, const u8 *addr, - bool remove, bool tx_key, int idx, - struct key_params *params) + struct net_device *dev, bool pairwise, + const u8 *addr, bool remove, bool tx_key, + int idx, struct key_params *params) { int err; /* devlist mutex needed for possible IBSS re-join */ mutex_lock(&rdev->devlist_mtx); wdev_lock(dev->ieee80211_ptr); - err = __cfg80211_set_encryption(rdev, dev, addr, remove, - tx_key, idx, params); + err = __cfg80211_set_encryption(rdev, dev, pairwise, addr, + remove, tx_key, idx, params); wdev_unlock(dev->ieee80211_ptr); mutex_unlock(&rdev->devlist_mtx); @@ -635,7 +645,7 @@ int cfg80211_wext_siwencode(struct net_device *dev, else if (!remove) return -EINVAL; - return cfg80211_set_encryption(rdev, dev, NULL, remove, + return cfg80211_set_encryption(rdev, dev, false, NULL, remove, wdev->wext.default_key == -1, idx, ¶ms); } @@ -725,7 +735,9 @@ int cfg80211_wext_siwencodeext(struct net_device *dev, } return cfg80211_set_encryption( - rdev, dev, addr, remove, + rdev, dev, + !(ext->ext_flags & IW_ENCODE_EXT_GROUP_KEY), + addr, remove, ext->ext_flags & IW_ENCODE_EXT_SET_TX_KEY, idx, ¶ms); } @@ -1354,6 +1366,10 @@ struct iw_statistics *cfg80211_wireless_stats(struct net_device *dev) } wstats.qual.updated |= IW_QUAL_NOISE_INVALID; + if (sinfo.filled & STATION_INFO_RX_DROP_MISC) + wstats.discard.misc = sinfo.rx_dropped_misc; + if (sinfo.filled & STATION_INFO_TX_FAILED) + wstats.discard.retries = sinfo.tx_failed; return &wstats; } @@ -1420,6 +1436,9 @@ int cfg80211_wext_giwessid(struct net_device *dev, { struct wireless_dev *wdev = dev->ieee80211_ptr; + data->flags = 0; + data->length = 0; + switch (wdev->iftype) { case NL80211_IFTYPE_ADHOC: return cfg80211_ibss_wext_giwessid(dev, info, data, ssid); diff --git a/net/wireless/wext-core.c b/net/wireless/wext-core.c index 0ef17bc42ba..dc675a3daa3 100644 --- a/net/wireless/wext-core.c +++ b/net/wireless/wext-core.c @@ -611,7 +611,7 @@ struct iw_statistics *get_wireless_stats(struct net_device *dev) #endif #ifdef CONFIG_CFG80211_WEXT - if (dev->ieee80211_ptr && dev->ieee80211_ptr && + if (dev->ieee80211_ptr && dev->ieee80211_ptr->wiphy && dev->ieee80211_ptr->wiphy->wext && dev->ieee80211_ptr->wiphy->wext->get_wireless_stats) @@ -782,6 +782,22 @@ static int ioctl_standard_iw_point(struct iw_point *iwp, unsigned int cmd, } } + if (IW_IS_GET(cmd) && !(descr->flags & IW_DESCR_FLAG_NOMAX)) { + /* + * If this is a GET, but not NOMAX, it means that the extra + * data is not bounded by userspace, but by max_tokens. Thus + * set the length to max_tokens. This matches the extra data + * allocation. + * The driver should fill it with the number of tokens it + * provided, and it may check iwp->length rather than having + * knowledge of max_tokens. If the driver doesn't change the + * iwp->length, this ioctl just copies back max_token tokens + * filled with zeroes. Hopefully the driver isn't claiming + * them to be valid data. + */ + iwp->length = descr->max_tokens; + } + err = handler(dev, info, (union iwreq_data *) iwp, extra); iwp->length += essid_compat; diff --git a/net/wireless/wext-priv.c b/net/wireless/wext-priv.c index 3feb28e41c5..674d426a9d2 100644 --- a/net/wireless/wext-priv.c +++ b/net/wireless/wext-priv.c @@ -152,7 +152,7 @@ static int ioctl_private_iw_point(struct iw_point *iwp, unsigned int cmd, } else if (!iwp->pointer) return -EFAULT; - extra = kmalloc(extra_size, GFP_KERNEL); + extra = kzalloc(extra_size, GFP_KERNEL); if (!extra) return -ENOMEM; diff --git a/net/wireless/wext-sme.c b/net/wireless/wext-sme.c index 9818198add8..6fffe62d7c2 100644 --- a/net/wireless/wext-sme.c +++ b/net/wireless/wext-sme.c @@ -197,6 +197,8 @@ int cfg80211_mgd_wext_siwessid(struct net_device *dev, wdev->wext.connect.ssid_len = len; wdev->wext.connect.crypto.control_port = false; + wdev->wext.connect.crypto.control_port_ethertype = + cpu_to_be16(ETH_P_PAE); err = cfg80211_mgd_wext_connect(rdev, wdev); out: diff --git a/net/x25/Kconfig b/net/x25/Kconfig index e6759c9660b..2196e55e4f6 100644 --- a/net/x25/Kconfig +++ b/net/x25/Kconfig @@ -5,6 +5,7 @@ config X25 tristate "CCITT X.25 Packet Layer (EXPERIMENTAL)" depends on EXPERIMENTAL + depends on BKL # should be fixable ---help--- X.25 is a set of standardized network protocols, similar in scope to frame relay; the one physical line from your box to the X.25 network diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c index 5e86d4e97dc..f7af98dff40 100644 --- a/net/x25/af_x25.c +++ b/net/x25/af_x25.c @@ -507,14 +507,14 @@ static int x25_listen(struct socket *sock, int backlog) struct sock *sk = sock->sk; int rc = -EOPNOTSUPP; - lock_kernel(); + lock_sock(sk); if (sk->sk_state != TCP_LISTEN) { memset(&x25_sk(sk)->dest_addr, 0, X25_ADDR_LEN); sk->sk_max_ack_backlog = backlog; sk->sk_state = TCP_LISTEN; rc = 0; } - unlock_kernel(); + release_sock(sk); return rc; } @@ -688,7 +688,6 @@ static int x25_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) struct sockaddr_x25 *addr = (struct sockaddr_x25 *)uaddr; int len, i, rc = 0; - lock_kernel(); if (!sock_flag(sk, SOCK_ZAPPED) || addr_len != sizeof(struct sockaddr_x25) || addr->sx25_family != AF_X25) { @@ -704,12 +703,13 @@ static int x25_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) } } + lock_sock(sk); x25_sk(sk)->source_addr = addr->sx25_addr; x25_insert_socket(sk); sock_reset_flag(sk, SOCK_ZAPPED); + release_sock(sk); SOCK_DEBUG(sk, "x25_bind: socket is bound\n"); out: - unlock_kernel(); return rc; } @@ -751,7 +751,6 @@ static int x25_connect(struct socket *sock, struct sockaddr *uaddr, struct x25_route *rt; int rc = 0; - lock_kernel(); lock_sock(sk); if (sk->sk_state == TCP_ESTABLISHED && sock->state == SS_CONNECTING) { sock->state = SS_CONNECTED; @@ -829,7 +828,6 @@ out_put_route: x25_route_put(rt); out: release_sock(sk); - unlock_kernel(); return rc; } @@ -869,8 +867,7 @@ static int x25_accept(struct socket *sock, struct socket *newsock, int flags) struct sk_buff *skb; int rc = -EINVAL; - lock_kernel(); - if (!sk || sk->sk_state != TCP_LISTEN) + if (!sk) goto out; rc = -EOPNOTSUPP; @@ -878,6 +875,10 @@ static int x25_accept(struct socket *sock, struct socket *newsock, int flags) goto out; lock_sock(sk); + rc = -EINVAL; + if (sk->sk_state != TCP_LISTEN) + goto out2; + rc = x25_wait_for_data(sk, sk->sk_rcvtimeo); if (rc) goto out2; @@ -897,7 +898,6 @@ static int x25_accept(struct socket *sock, struct socket *newsock, int flags) out2: release_sock(sk); out: - unlock_kernel(); return rc; } @@ -909,7 +909,6 @@ static int x25_getname(struct socket *sock, struct sockaddr *uaddr, struct x25_sock *x25 = x25_sk(sk); int rc = 0; - lock_kernel(); if (peer) { if (sk->sk_state != TCP_ESTABLISHED) { rc = -ENOTCONN; @@ -923,19 +922,6 @@ static int x25_getname(struct socket *sock, struct sockaddr *uaddr, *uaddr_len = sizeof(*sx25); out: - unlock_kernel(); - return rc; -} - -static unsigned int x25_datagram_poll(struct file *file, struct socket *sock, - poll_table *wait) -{ - int rc; - - lock_kernel(); - rc = datagram_poll(file, sock, wait); - unlock_kernel(); - return rc; } @@ -1746,7 +1732,7 @@ static const struct proto_ops x25_proto_ops = { .socketpair = sock_no_socketpair, .accept = x25_accept, .getname = x25_getname, - .poll = x25_datagram_poll, + .poll = datagram_poll, .ioctl = x25_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = compat_x25_ioctl, diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c index a3cca0a9434..64f2ae1fdc1 100644 --- a/net/xfrm/xfrm_output.c +++ b/net/xfrm/xfrm_output.c @@ -101,7 +101,7 @@ resume: err = -EHOSTUNREACH; goto error_nolock; } - skb_dst_set_noref(skb, dst); + skb_dst_set(skb, dst_clone(dst)); x = dst->xfrm; } while (x && !(x->outer_mode->flags & XFRM_MODE_FLAG_TUNNEL)); diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 2b3ed7ad493..044e7789851 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -50,6 +50,9 @@ static struct xfrm_policy_afinfo *xfrm_policy_get_afinfo(unsigned short family); static void xfrm_policy_put_afinfo(struct xfrm_policy_afinfo *afinfo); static void xfrm_init_pmtu(struct dst_entry *dst); static int stale_bundle(struct dst_entry *dst); +static int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *xdst, + struct flowi *fl, int family, int strict); + static struct xfrm_policy *__xfrm_policy_unlink(struct xfrm_policy *pol, int dir); @@ -1175,9 +1178,8 @@ xfrm_tmpl_resolve_one(struct xfrm_policy *policy, struct flowi *fl, tmpl->mode == XFRM_MODE_BEET) { remote = &tmpl->id.daddr; local = &tmpl->saddr; - family = tmpl->encap_family; - if (xfrm_addr_any(local, family)) { - error = xfrm_get_saddr(net, &tmp, remote, family); + if (xfrm_addr_any(local, tmpl->encap_family)) { + error = xfrm_get_saddr(net, &tmp, remote, tmpl->encap_family); if (error) goto fail; local = &tmp; @@ -2277,7 +2279,7 @@ static void xfrm_init_pmtu(struct dst_entry *dst) * still valid. */ -int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *first, +static int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *first, struct flowi *fl, int family, int strict) { struct dst_entry *dst = &first->u.dst; @@ -2359,8 +2361,6 @@ int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *first, return 1; } -EXPORT_SYMBOL(xfrm_bundle_ok); - int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo) { struct net *net; diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 5208b12fbfb..eb96ce52f17 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -656,15 +656,23 @@ void xfrm_sad_getinfo(struct net *net, struct xfrmk_sadinfo *si) EXPORT_SYMBOL(xfrm_sad_getinfo); static int -xfrm_init_tempsel(struct xfrm_state *x, struct flowi *fl, - struct xfrm_tmpl *tmpl, - xfrm_address_t *daddr, xfrm_address_t *saddr, - unsigned short family) +xfrm_init_tempstate(struct xfrm_state *x, struct flowi *fl, + struct xfrm_tmpl *tmpl, + xfrm_address_t *daddr, xfrm_address_t *saddr, + unsigned short family) { struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family); if (!afinfo) return -1; - afinfo->init_tempsel(x, fl, tmpl, daddr, saddr); + afinfo->init_tempsel(&x->sel, fl); + + if (family != tmpl->encap_family) { + xfrm_state_put_afinfo(afinfo); + afinfo = xfrm_state_get_afinfo(tmpl->encap_family); + if (!afinfo) + return -1; + } + afinfo->init_temprop(x, tmpl, daddr, saddr); xfrm_state_put_afinfo(afinfo); return 0; } @@ -790,37 +798,38 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, int error = 0; struct xfrm_state *best = NULL; u32 mark = pol->mark.v & pol->mark.m; + unsigned short encap_family = tmpl->encap_family; to_put = NULL; spin_lock_bh(&xfrm_state_lock); - h = xfrm_dst_hash(net, daddr, saddr, tmpl->reqid, family); + h = xfrm_dst_hash(net, daddr, saddr, tmpl->reqid, encap_family); hlist_for_each_entry(x, entry, net->xfrm.state_bydst+h, bydst) { - if (x->props.family == family && + if (x->props.family == encap_family && x->props.reqid == tmpl->reqid && (mark & x->mark.m) == x->mark.v && !(x->props.flags & XFRM_STATE_WILDRECV) && - xfrm_state_addr_check(x, daddr, saddr, family) && + xfrm_state_addr_check(x, daddr, saddr, encap_family) && tmpl->mode == x->props.mode && tmpl->id.proto == x->id.proto && (tmpl->id.spi == x->id.spi || !tmpl->id.spi)) - xfrm_state_look_at(pol, x, fl, family, daddr, saddr, + xfrm_state_look_at(pol, x, fl, encap_family, daddr, saddr, &best, &acquire_in_progress, &error); } if (best) goto found; - h_wildcard = xfrm_dst_hash(net, daddr, &saddr_wildcard, tmpl->reqid, family); + h_wildcard = xfrm_dst_hash(net, daddr, &saddr_wildcard, tmpl->reqid, encap_family); hlist_for_each_entry(x, entry, net->xfrm.state_bydst+h_wildcard, bydst) { - if (x->props.family == family && + if (x->props.family == encap_family && x->props.reqid == tmpl->reqid && (mark & x->mark.m) == x->mark.v && !(x->props.flags & XFRM_STATE_WILDRECV) && - xfrm_state_addr_check(x, daddr, saddr, family) && + xfrm_state_addr_check(x, daddr, saddr, encap_family) && tmpl->mode == x->props.mode && tmpl->id.proto == x->id.proto && (tmpl->id.spi == x->id.spi || !tmpl->id.spi)) - xfrm_state_look_at(pol, x, fl, family, daddr, saddr, + xfrm_state_look_at(pol, x, fl, encap_family, daddr, saddr, &best, &acquire_in_progress, &error); } @@ -829,7 +838,7 @@ found: if (!x && !error && !acquire_in_progress) { if (tmpl->id.spi && (x0 = __xfrm_state_lookup(net, mark, daddr, tmpl->id.spi, - tmpl->id.proto, family)) != NULL) { + tmpl->id.proto, encap_family)) != NULL) { to_put = x0; error = -EEXIST; goto out; @@ -839,9 +848,9 @@ found: error = -ENOMEM; goto out; } - /* Initialize temporary selector matching only + /* Initialize temporary state matching only * to current session. */ - xfrm_init_tempsel(x, fl, tmpl, daddr, saddr, family); + xfrm_init_tempstate(x, fl, tmpl, daddr, saddr, family); memcpy(&x->mark, &pol->mark, sizeof(x->mark)); error = security_xfrm_state_alloc_acquire(x, pol->security, fl->secid); @@ -856,10 +865,10 @@ found: x->km.state = XFRM_STATE_ACQ; list_add(&x->km.all, &net->xfrm.state_all); hlist_add_head(&x->bydst, net->xfrm.state_bydst+h); - h = xfrm_src_hash(net, daddr, saddr, family); + h = xfrm_src_hash(net, daddr, saddr, encap_family); hlist_add_head(&x->bysrc, net->xfrm.state_bysrc+h); if (x->id.spi) { - h = xfrm_spi_hash(net, &x->id.daddr, x->id.spi, x->id.proto, family); + h = xfrm_spi_hash(net, &x->id.daddr, x->id.spi, x->id.proto, encap_family); hlist_add_head(&x->byspi, net->xfrm.state_byspi+h); } x->lft.hard_add_expires_seconds = net->xfrm.sysctl_acq_expires; diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index ba59983aaff..8bae6b22c84 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1801,7 +1801,7 @@ static int xfrm_add_sa_expire(struct sk_buff *skb, struct nlmsghdr *nlh, struct xfrm_user_expire *ue = nlmsg_data(nlh); struct xfrm_usersa_info *p = &ue->state; struct xfrm_mark m; - u32 mark = xfrm_mark_get(attrs, &m);; + u32 mark = xfrm_mark_get(attrs, &m); x = xfrm_state_lookup(net, mark, &p->id.daddr, p->id.spi, p->id.proto, p->family); @@ -2504,7 +2504,7 @@ static struct xfrm_policy *xfrm_compile_policy(struct sock *sk, int opt, if (p->dir > XFRM_POLICY_OUT) return NULL; - xp = xfrm_policy_alloc(net, GFP_KERNEL); + xp = xfrm_policy_alloc(net, GFP_ATOMIC); if (xp == NULL) { *dir = -ENOBUFS; return NULL; |