summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2015-11-16drm/radeon: unconditionally set sysfs_initializedAlex Deucher
Avoids spew on resume for systems where sysfs may fail even on init. bug: https://bugzilla.kernel.org/show_bug.cgi?id=106851 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
2015-11-16drm/amdgpu: cleanup amdgpu_cs_parser handlingChristian König
No need any more to allocate that structure dynamically, just put it on the stack. This is a start to cleanup some of the scheduler fallouts. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amdgpu: fix leaking the IBs on errorChristian König
Fixing a memory leak when the scheduler is enabled. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amd: add kmem cache for sched fenceChunming Zhou
Change-Id: I45bb8ff10ef05dc3b15e31a77fbcf31117705f11 Signed-off-by: Chunming Zhou <David1.Zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com>
2015-11-16drm/amdgpu: add kmem cache for amdgpu fenceChunming Zhou
Change-Id: I5ad8dd156ccf27a6f18004aa0a215a0925b6e67b Signed-off-by: Chunming Zhou <David1.Zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com>
2015-11-16drm/amdgpu: update fiji_mgcg_cgcg_init tableFlora Cui
Change-Id: If44b8057741c78208f1976f60f31b535c944d0bd Signed-off-by: Flora Cui <Flora.Cui@amd.com> Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
2015-11-16drm/amdgpu: use common fence for amdgpu_vm_fenceChristian König
Just cleanup the function parameters. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amdgpu: use fence_is_later() for vm_flush as well v2Christian König
v2: remove superfluous check Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> (v1) Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
2015-11-16drm/amdgpu: use a timer for fence fallbackChristian König
Less overhead than a work item and also adds proper cleanup handling. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amdgpu: remove fence trace pointsChristian König
Mostly unused and replaced by the common trace points. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/amdgpu: update Fiji's mmPA_SC_RASTER_CONFIG valueFlora Cui
Change-Id: I6d138306a878450e5bf8a77a2f1aacc380a39fe5 Signed-off-by: Flora Cui <Flora.Cui@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/radeon: Only prompt for enabling PAT when we'd allow write-combiningMichel Dänzer
No use bothering users about this for whom we disable write-combining for other reasons anyway. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/radeon: Always disable RADEON_GEM_GTT_UC along with RADEON_GEM_GTT_WCMichel Dänzer
Write-combining is a CPU feature. From the GPU POV, these both simply mean no GPU<->CPU cache coherency. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/radeon: Disable uncacheable CPU mappings of GTT with RV6xxMichel Dänzer
They reportedly cause random GPU hangs. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91268 Reviewed-by: Christian König <christian.koenig@amd.com> Cc: stable@vger.kernel.org Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-11-16drm/i915: get runtime PM reference around GEM set_caching IOCTLImre Deak
After Damien's D3 fix I started to get runtime suspend residency for the first time and that revealed a breakage on the set_caching IOCTL path that accesses the HW but doesn't take an RPM ref. Fix this up. Signed-off-by: Imre Deak <imre.deak@intel.com> Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com> Cc: stable@vger.kernel.org Link: http://patchwork.freedesktop.org/patch/msgid/1446665132-22491-1-git-send-email-imre.deak@intel.com Signed-off-by: Jani Nikula <jani.nikula@intel.com>
2015-11-16drm/i915: Fix GT frequency roundingMika Kuoppala
When we set and later readback a frequency value through sysfs interface, igt/pm_rpm assumes that we get same value back if it matches hw granularity. On bxt we have found out that this is not always the case. Currently frequency - hw ratio - frequency conversions round down, with few exceptions on platforms that have more specific conversions. On bxt the supported range can be for example from 100Mhz to 650Mhz. Midpoint is then calculated by test to be 375 which pm_rps uses to find a closest hw supported frequency. That is 366 (ratio 22), which it then writes back. But as the rounding down kicks in, driver actually sets 350 instead of 366, as 366 is 2/3 below 22 * 50/3. Fix this by rounding to closest instead of rounding down in freq-ratio-freq conversions. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92768 Testcase: igt/pm_rps/basic-api Tested-by: Bob Paauwe <bob.j.paauwe@intel.com> Cc: Bob Paauwe <bob.j.paauwe@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com> Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Bob Paauwe <bob.j.paauwe@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1447435781-23416-1-git-send-email-mika.kuoppala@intel.com Signed-off-by: Jani Nikula <jani.nikula@intel.com>
2015-11-16drm/i915: quirk backlight present on Macbook 4, 1Jani Nikula
Unsurprisingly macbooks have backlights, just the VBT doesn't seem to know it in this case. Reported-and-tested-by: Daniel Nicoletti <dantti12@gmail.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88325 Fixes: c675949ec58c ("drm/i915: do not setup backlight if not available according to VBT") Cc: stable@vger.kernel.org # v3.15+ Reviewed-by: Ander Conselvan de Oliveira <conselvan2@gmail.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1446716999-1796-1-git-send-email-jani.nikula@intel.com
2015-11-16drm/i915: Fix crtc_y assignment in intel_find_initial_plane_obj()Ville Syrjälä
Let's set crtc_y to 0 instead of setting src_y twice. Multiple assignments in one statement is a good way to hide bugs. Please don't do that. Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Fixes: be5651f2d581 ("drm/i915: Update missing properties in find_initial_plane_obj") Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1447434973-12369-1-git-send-email-ville.syrjala@linux.intel.com Signed-off-by: Jani Nikula <jani.nikula@intel.com>
2015-11-16s390: remove SALIPL loaderHeiko Carstens
There is no known user, therefore remove the code. Acked-by: Rob Van Der Heij <robvdheij@nl.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-11-16s390: wire up mlock2 system callHeiko Carstens
Passes mlock2-tests test case in 64 bit and compat mode. Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-11-16s390: remove g5 elf platform supportHeiko Carstens
Remove dead code, since this could only happen on a 31 bit machine where the kernel wouldn't IPL. Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-11-16s390: avoid cache aliasing under z/VM and KVMMartin Schwidefsky
commit 1f6b83e5e4d3 ("s390: avoid z13 cache aliasing") checks for the machine type to optimize address space randomization and zero page allocation to avoid cache aliases. This check might fail under a hypervisor with migration support. z/VMs "Single System Image and Live Guest Relocation" facility will "fake" the machine type of the oldest system in the group. For example in a group of zEC12 and Z13 the guest appears to run on a zEC12 (architecture fencing within the relocation domain) Remove the machine type detection and always use cache aliasing rules that are known to work for all machines. These are the z13 aliasing rules. Suggested-by: Christian Borntraeger <borntraeger@de.ibm.com> Reviewed-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-11-15hwmon : (applesmc) Fix uninitialized variables warningsShuah Khan
Fix the following "maybe used uninitialized" warnings by initializing the variables to keep the compiler quiet. There is no "used uninitialized" in this case. CC [M] drivers/hwmon/applesmc.o drivers/hwmon/applesmc.c: In function ‘applesmc_init_smcreg’: drivers/hwmon/applesmc.c:595:43: warning: ‘right_light_sensor’ may be used uninitialized in this function [-Wmaybe-uninitialized] s->num_light_sensors = left_light_sensor + right_light_sensor; ^ drivers/hwmon/applesmc.c:540:26: note: ‘right_light_sensor’ was declared here bool left_light_sensor, right_light_sensor; ^ drivers/hwmon/applesmc.c:595:43: warning: ‘left_light_sensor’ may be used uninitialized in this function [-Wmaybe-uninitialized] s->num_light_sensors = left_light_sensor + right_light_sensor; ^ drivers/hwmon/applesmc.c:540:7: note: ‘left_light_sensor’ was declared here bool left_light_sensor, right_light_sensor; ^ Signed-off-by: Shuah Khan <shuahkh@osg.samsung.com> Signed-off-by: Guenter Roeck <linux@roeck-us.net>
2015-11-15hwmon: (ina2xx) Fix build issue by selecting REGMAP_I2CLi Yang
Since a0de56c81fcf ("hwmon: (ina2xx) convert driver to using regmap") the driver requires REGMAP_I2C to build. Select it by default in Kconfig. Reported-by: Guo Chunrong <B40290@freescale.com> Cc: Marc Titinger <mtitinger@baylibre.com> Signed-off-by: Li Yang <leoli@freescale.com> Fixes: a0de56c81fcf ("hwmon: (ina2xx) convert driver to using regmap") Signed-off-by: Guenter Roeck <linux@roeck-us.net>
2015-11-15Merge branch 'mv88e6060-fixes'David S. Miller
Neil Armstrong says: ==================== net: dsa: mv88e6060: cleanup and fix setup This patchset introduces some fixes and a registers addressing cleanup for the mv88e6060 DSA driver. The first patch removes the poll_link as mv88e6xxx. The 3 following patches fixes the setup in regards of the datasheet. The 2 last patches introduces a clean header and replaces all magic values. v2: cleanup InitReady patch, add missing Acked-by and fix header copyright notice ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15net: dsa: mv88e6060: replace magic values with register definesNeil Armstrong
To align with the mv88e6xxx code, use the register defines to access all the register addresses and bit fields. Acked-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Neil Armstrong <narmstrong@baylibre.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15net: dsa: mv88e6060: add register defines header fileNeil Armstrong
To align with the mv88e6xxx code, add a similar header file with all the register defines. The file is based on the mv88e6xxx header for coherency. Acked-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Neil Armstrong <narmstrong@baylibre.com> Acked-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15net: dsa: mv88e6060: use the correct bit shift for mac0Neil Armstrong
According to the mv88e6060 datasheet, the first mac byte must be at position 9 instead of 8 since the bit 8 is used to select if the mac address must differ for each port for Pause frames. Use the correct shift and set the same mac address for all port. Acked-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Neil Armstrong <narmstrong@baylibre.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15net: dsa: mv88e6060: use the correct MaxFrameSize bitNeil Armstrong
According to the mv88e6060 datasheet, the MaxFrameSize bit position is 10 instead of 11 which is reserved. Use the bit correctly to setup max frame size to 1536. Acked-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Neil Armstrong <narmstrong@baylibre.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15net: dsa: mv88e6060: use the correct InitReady bitNeil Armstrong
According to the mv88e6060 datasheet, the InitReady bit position is 11 and the polarity is inverted. Use the bit correctly to detect the end of initialization. Acked-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Neil Armstrong <narmstrong@baylibre.com> Acked-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15net: dsa: mv88e6060: remove poll_link callbackNeil Armstrong
As of mv88e6xxx remove the poll_link callback since the link state change polling is now handled by the phylib. Tested on a mv88e6060 B0 device with a TI DM816X SoC. Suggested-by: Andrew Lunn <andrew@lunn.ch> Acked-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Neil Armstrong <narmstrong@baylibre.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15Linux 4.4-rc1v4.4-rc1Linus Torvalds
2015-11-15Merge branch 'mellanox-net-fixes'David S. Miller
Or Gerlitz says: ==================== Mellanox NIC driver update, Nov 12, 2015 Few small mlx5 and mlx4 fixes from the team... done over net commit c5a3788 "Merge branch 'akpm' (patches from Andrew)" Eran's patch needs to go to 4.2 and 4.3 stable kernels. Tariq's patch need to go to 4.3 stable too. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15net/mlx4_core: Avoid returning success in case of an error flowNoa Osherovich
The err variable wasn't set with the correct error value in some cases. Fixes: 47605df95398 ('mlx4: Modify proxy/tunnel QP mechanism [..]') Signed-off-by: Noa Osherovich <noaos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15net/mlx4_core: Fix sleeping while holding spinlock at rem_slave_countersEran Ben Elisha
When cleaning slave's counter resources, we hold a spinlock that protects the slave's counters list. As part of the clean, we call __mlx4_clear_if_stat which calls mlx4_alloc_cmd_mailbox which is a sleepable function. In order to fix this issue, hold the spinlock, and copy all counter indices into a temporary array, and release the spinlock. Afterwards, iterate over this array and free every counter. Repeat this scenario until the original list is empty (a new counter might have been added while releasing the counters from the temporary array). Fixes: b72ca7e96acf ("net/mlx4_core: Reset counters data when freed") Reported-by: Moni Shoua <monis@mellanox.com> Tested-by: Moni Shoua <monis@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15net/mlx5e: Use the right DMA free function on TX pathAchiad Shochat
On xmit path we use skb_frag_dma_map() which is using dma_map_page(), while upon completion we dma-unmap the skb fragments using dma_unmap_single() rather than dma_unmap_page(). To fix this, we now save the dma map type on xmit path and use this info to call the right dma unmap method upon TX completion. Signed-off-by: Achiad Shochat <achiad@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15net/mlx5e: Max mtu comparison fixDoron Tsur
On change mtu the driver compares between hardware queried mtu and software requested mtu. We need to compare between software representation of the queried mtu and the requested mtu. Fixes: facc9699f0fe ('net/mlx5e: Fix HW MTU settings') Signed-off-by: Doron Tsur <doront@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15net/mlx5e: Added self loopback preventionTariq Toukan
Prevent outgoing multicast frames from looping back to the RX queue. By introducing new HW capability self_lb_en_modifiable, which indicates the support to modify self_lb_en bit in modify_tir command. When this capability is set we can prevent TIRs from sending back loopback multicast traffic to their own RQs, by "refreshing TIRs" with modify_tir command, on every time new channels (SQs/RQs) are created at device open. This is needed since TIRs are static and only allocated once on driver load, and the loopback decision is under their responsibility. Fixes issues of the kind: "IPv6: eth2: IPv6 duplicate address fe80::e61d:2dff:fe5c:f2e9 detected!" The issue is seen since the IPv6 solicitations multicast messages are loopedback and the network stack thinks they are coming from another host. Fixes: 5c50368f3831 ("net/mlx5e: Light-weight netdev open/stop") Signed-off-by: Tariq Toukan <tariqt@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15net/mlx5e: Fix inline header size calculationSaeed Mahameed
mlx5e_get_inline_hdr_size didn't take into account the vlan insertion into the inline WQE segment. This could lead to max inline violation in cases where skb_headlen(skb) + VLAN_HLEN >= sq->max_inline. Fixes: 3ea4891db8d0 ("net/mlx5e: Fix LSO vlan insertion") Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Achiad Shochat <achiad@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15ipvs: use skb_to_full_sk() helperEric Dumazet
SYNACK packets might be attached to request sockets. Use skb_to_full_sk() helper to avoid illegal accesses to inet_sk(skb->sk) Fixes: ca6fb0651883 ("tcp: attach SYNACK messages to request sockets instead of listener") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Acked-by: Julian Anastasov <ja@ssi.bg> Acked-by: Simon Horman <horms@verge.net.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15tcp: ensure proper barriers in lockless contextsEric Dumazet
Some functions access TCP sockets without holding a lock and might output non consistent data, depending on compiler and or architecture. tcp_diag_get_info(), tcp_get_info(), tcp_poll(), get_tcp4_sock() ... Introduce sk_state_load() and sk_state_store() to fix the issues, and more clearly document where this lack of locking is happening. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15net: thunder: Fix crash upon shutdown after failed probePavel Fedin
If device probe fails, driver remains bound to the PCI device. However, driver data has been reset to NULL. This causes crash upon dereferencing it in nicvf_remove() Signed-off-by: Pavel Fedin <p.fedin@samsung.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15sctp: translate host order to network order when setting a hmacidlucien
now sctp auth cannot work well when setting a hmacid manually, which is caused by that we didn't use the network order for hmacid, so fix it by adding the transformation in sctp_auth_ep_set_hmacs. even we set hmacid with the network order in userspace, it still can't work, because of this condition in sctp_auth_ep_set_hmacs(): if (id > SCTP_AUTH_HMAC_ID_MAX) return -EOPNOTSUPP; so this wasn't working before and thus it won't break compatibility. Fixes: 65b07e5d0d09 ("[SCTP]: API updates to suport SCTP-AUTH extensions.") Signed-off-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15Merge branch 'packet-fixes'David S. Miller
Daniel Borkmann says: ==================== packet fixes Fixes a couple of issues in packet sockets, i.e. on TX ring side. See individual patches for details. v2 -> v3: - First two patches unchanged, kept Jason's Ack - Reworked 3rd patch and split into 3: - check for dev type as discussed with Willem - infer skb->protocol - fix max len for dgram v1 -> v2: - Added patch 2 as suggested by Dave - Rest is unchanged from previous submission ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15packet: fix tpacket_snd max frame lenDaniel Borkmann
Since it's introduction in commit 69e3c75f4d54 ("net: TX_RING and packet mmap"), TX_RING could be used from SOCK_DGRAM and SOCK_RAW side. When used with SOCK_DGRAM only, the size_max > dev->mtu + reserve check should have reserve as 0, but currently, this is unconditionally set (in it's original form as dev->hard_header_len). I think this is not correct since tpacket_fill_skb() would then take dev->mtu and dev->hard_header_len into account for SOCK_DGRAM, the extra VLAN_HLEN could be possible in both cases. Presumably, the reserve code was copied from packet_snd(), but later on missed the check. Make it similar as we have it in packet_snd(). Fixes: 69e3c75f4d54 ("net: TX_RING and packet mmap") Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Willem de Bruijn <willemb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15packet: infer protocol from ethernet header if unsetDaniel Borkmann
In case no struct sockaddr_ll has been passed to packet socket's sendmsg() when doing a TX_RING flush run, then skb->protocol is set to po->num instead, which is the protocol passed via socket(2)/bind(2). Applications only xmitting can go the path of allocating the socket as socket(PF_PACKET, <mode>, 0) and do a bind(2) on the TX_RING with sll_protocol of 0. That way, register_prot_hook() is neither called on creation nor on bind time, which saves cycles when there's no interest in capturing anyway. That leaves us however with po->num 0 instead and therefore the TX_RING flush run sets skb->protocol to 0 as well. Eric reported that this leads to problems when using tools like trafgen over bonding device. I.e. the bonding's hash function could invoke the kernel's flow dissector, which depends on skb->protocol being properly set. In the current situation, all the traffic is then directed to a single slave. Fix it up by inferring skb->protocol from the Ethernet header when not set and we have ARPHRD_ETHER device type. This is only done in case of SOCK_RAW and where we have a dev->hard_header_len length. In case of ARPHRD_ETHER devices, this is guaranteed to cover ETH_HLEN, and therefore being accessed on the skb after the skb_store_bits(). Reported-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Willem de Bruijn <willemb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15packet: only allow extra vlan len on ethernet devicesDaniel Borkmann
Packet sockets can be used by various net devices and are not really restricted to ARPHRD_ETHER device types. However, when currently checking for the extra 4 bytes that can be transmitted in VLAN case, our assumption is that we generally probe on ARPHRD_ETHER devices. Therefore, before looking into Ethernet header, check the device type first. This also fixes the issue where non-ARPHRD_ETHER devices could have no dev->hard_header_len in TX_RING SOCK_RAW case, and thus the check would test unfilled linear part of the skb (instead of non-linear). Fixes: 57f89bfa2140 ("network: Allow af_packet to transmit +4 bytes for VLAN packets.") Fixes: 52f1454f629f ("packet: allow to transmit +4 byte in TX_RING slot for VLAN case") Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Willem de Bruijn <willemb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15packet: always probe for transport headerDaniel Borkmann
We concluded that the skb_probe_transport_header() should better be called unconditionally. Avoiding the call into the flow dissector has also not really much to do with the direct xmit mode. While it seems that only virtio_net code makes use of GSO from non RX/TX ring packet socket paths, we should probe for a transport header nevertheless before they hit devices. Reference: http://thread.gmane.org/gmane.linux.network/386173/ Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15packet: do skb_probe_transport_header when we actually have dataDaniel Borkmann
In tpacket_fill_skb() commit c1aad275b029 ("packet: set transport header before doing xmit") and later on 40893fd0fd4e ("net: switch to use skb_probe_transport_header()") was probing for a transport header on the skb from a ring buffer slot, but at a time, where the skb has _not even_ been filled with data yet. So that call into the flow dissector is pretty useless. Lets do it after we've set up the skb frags. Fixes: c1aad275b029 ("packet: set transport header before doing xmit") Reported-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-11-15tools/net: Use include/uapi with __EXPORTED_HEADERS__Kamal Mostafa
Use the local uapi headers to keep in sync with "recently" added #define's (e.g. SKF_AD_VLAN_TPID). Refactored CFLAGS, and bpf_asm doesn't need -I. Fixes: 3f356385e8a4 ("filter: bpf_asm: add minimal bpf asm tool") Signed-off-by: Kamal Mostafa <kamal@canonical.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: David S. Miller <davem@davemloft.net>
generated by cgit v1.2.3 (git 2.39.1) at 2025-08-06 01:16:28 +0000