| Age | Commit message (Collapse) | Author |
|---|
|
clang warns:
net/netfilter/nft_bitwise.c:138:50: error: size argument in 'memcmp'
call is a comparison [-Werror,-Wmemsize-comparison]
if (memcmp(&priv->xor, &zero, sizeof(priv->xor) ||
~~~~~~~~~~~~~~~~~~^~
net/netfilter/nft_bitwise.c:138:6: note: did you mean to compare the
result of 'memcmp' instead?
if (memcmp(&priv->xor, &zero, sizeof(priv->xor) ||
^
)
net/netfilter/nft_bitwise.c:138:32: note: explicitly cast the argument
to size_t to silence this warning
if (memcmp(&priv->xor, &zero, sizeof(priv->xor) ||
^
(size_t)(
1 error generated.
Adjust the parentheses so that the result of the sizeof is used for the
size argument in memcmp, rather than the result of the comparison (which
would always be true because sizeof is a non-zero number).
Fixes: bd8699e9e292 ("netfilter: nft_bitwise: add offload support")
Link: https://github.com/ClangBuiltLinux/linux/issues/638
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
There are two netfilter userspace headers which contain deprecation
warnings. While these headers are not used within the kernel, they are
compiled stand-alone for header-testing.
Pablo informs me that userspace iptables still refer to these headers,
and the intention was to use xt_LOG.h instead and remove these, but
userspace was never updated.
Remove the warnings.
Fixes: 2a475c409fe8 ("kbuild: remove all netfilter headers from header-test blacklist.")
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs
Pull afs fixes from David Howells:
- Fix the CB.ProbeUuid handler to generate its reply correctly.
- Fix a mix up in indices when parsing a Volume Location entry record.
- Fix a potential NULL-pointer deref when cleaning up a read request.
- Fix the expected data version of the destination directory in
afs_rename().
- Fix afs_d_revalidate() to only update d_fsdata if it's not the same
as the directory data version to reduce the likelihood of overwriting
the result of a competing operation. (d_fsdata carries the directory
DV or the least-significant word thereof).
- Fix the tracking of the data-version on a directory and make sure
that dentry objects get properly initialised, updated and
revalidated.
Also fix rename to update d_fsdata to match the new directory's DV if
the dentry gets moved over and unhash the dentry to stop
afs_d_revalidate() from interfering.
* tag 'afs-fixes-20190814' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs:
afs: Fix missing dentry data version updating
afs: Only update d_fsdata if different in afs_d_revalidate()
afs: Fix off-by-one in afs_rename() expected data version calculation
fs: afs: Fix a possible null-pointer dereference in afs_put_read()
afs: Fix loop index mixup in afs_deliver_vl_get_entry_by_name_u()
afs: Fix the CB.ProbeUuid service handler to reply correctly
|
|
The spsc_queue_peek function is accessing queue->head which belongs to
the consumer thread and shouldn't be accessed by the producer
This is fixing a rare race condition when destroying entities.
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Monk.liu@amd.com
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
Make the __fstate_clean() function correctly set the
state of sstatus.FS in pt_regs to SR_FS_CLEAN.
Fixes: 7db91e57a0acd ("RISC-V: Task implementation")
Cc: linux-stable <stable@vger.kernel.org>
Signed-off-by: Vincent Chen <vincent.chen@sifive.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
[paul.walmsley@sifive.com: expanded "Fixes" commit ID]
Signed-off-by: Paul Walmsley <paul.walmsley@sifive.com>
|
|
The following two reasons cause FP registers are sometimes not
initialized before starting the user program.
1. Currently, the FP context is initialized in flush_thread() function
and we expect these initial values to be restored to FP register when
doing FP context switch. However, the FP context switch only occurs in
switch_to function. Hence, if this process does not be scheduled out
and scheduled in before entering the user space, the FP registers
have no chance to initialize.
2. In flush_thread(), the state of reg->sstatus.FS inherits from the
parent. Hence, the state of reg->sstatus.FS may be dirty. If this
process is scheduled out during flush_thread() and initializing the
FP register, the fstate_save() in switch_to will corrupt the FP context
which has been initialized until flush_thread().
To solve the 1st case, the initialization of the FP register will be
completed in start_thread(). It makes sure all FP registers are initialized
before starting the user program. For the 2nd case, the state of
reg->sstatus.FS in start_thread will be set to SR_FS_OFF to prevent this
process from corrupting FP context in doing context save. The FP state is
set to SR_FS_INITIAL in start_trhead().
Signed-off-by: Vincent Chen <vincent.chen@sifive.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Fixes: 7db91e57a0acd ("RISC-V: Task implementation")
Cc: stable@vger.kernel.org
[paul.walmsley@sifive.com: fixed brace alignment issue reported by
checkpatch]
Signed-off-by: Paul Walmsley <paul.walmsley@sifive.com>
|
|
Pull rdma fixes from Doug Ledford:
"Fairly small pull request for -rc3. I'm out of town the rest of this
week, so I made sure to clean out as much as possible from patchworks
in enough time for 0-day to chew through it (Yay! for 0-day being back
online! :-)). Jason might send through any emergency stuff that could
pop up, otherwise I'm back next week.
The only real thing of note is the siw ABI change. Since we just
merged siw *this* release, there are no prior kernel releases to
maintain kernel ABI with. I told Bernard that if there is anything
else about the siw ABI he thinks he might want to change before it
goes set in stone, he should get it in ASAP. The siw module was around
for several years outside the kernel tree, and it had to be revamped
considerably for inclusion upstream, so we are making no attempts to
be backward compatible with the out of tree version. Once 5.3 is
actually released, we will have our baseline ABI to maintain.
Summary:
- Fix a memory registration release flow issue that was causing a
WARN_ON (mlx5)
- If the counters for a port aren't allocated, then we can't do
operations on the non-existent counters (core)
- Check the right variable for error code result (mlx5)
- Fix a use after free issue (mlx5)
- Fix an off by one memory leak (siw)
- Actually return an error code on error (core)
- Allow siw to be built on 32bit arches (siw, ABI change, but OK
since siw was just merged this merge window and there is no prior
released kernel to maintain compatibility with and we also updated
the rdma-core user space package to match)"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
RDMA/siw: Change CQ flags from 64->32 bits
RDMA/core: Fix error code in stat_get_doit_qp()
RDMA/siw: Fix a memory leak in siw_init_cpulist()
IB/mlx5: Fix use-after-free error while accessing ev_file pointer
IB/mlx5: Check the correct variable in error handling code
RDMA/counter: Prevent QP counter binding if counters unsupported
IB/mlx5: Fix implicit MR release flow
|
|
The `uac_mixer_unit_descriptor` shown as below is read from the
device side. In `parse_audio_mixer_unit`, `baSourceID` field is
accessed from index 0 to `bNrInPins` - 1, the current implementation
assumes that descriptor is always valid (the length of descriptor
is no shorter than 5 + `bNrInPins`). If a descriptor read from
the device side is invalid, it may trigger out-of-bound memory
access.
```
struct uac_mixer_unit_descriptor {
__u8 bLength;
__u8 bDescriptorType;
__u8 bDescriptorSubtype;
__u8 bUnitID;
__u8 bNrInPins;
__u8 baSourceID[];
}
```
This patch fixes the bug by add a sanity check on the length of
the descriptor.
Reported-by: Hui Peng <benquike@gmail.com>
Reported-by: Mathias Payer <mathias.payer@nebelwelt.net>
Cc: <stable@vger.kernel.org>
Signed-off-by: Hui Peng <benquike@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
|
|
Pull dma-mapping fixes from Christoph Hellwig:
- fix the handling of the bus_dma_mask in dma_get_required_mask, which
caused a regression in this merge window (Lucas Stach)
- fix a regression in the handling of DMA_ATTR_NO_KERNEL_MAPPING (me)
- fix dma_mmap_coherent to not cause page attribute mismatches on
coherent architectures like x86 (me)
* tag 'dma-mapping-5.3-4' of git://git.infradead.org/users/hch/dma-mapping:
dma-mapping: fix page attributes for dma_mmap_*
dma-direct: don't truncate dma_required_mask to bus addressing capabilities
dma-direct: fix DMA_ATTR_NO_KERNEL_MAPPING
|
|
Realtek provided information on how the new NIC-integrated PHY's
expose whether they support 2.5G/5G/10G. This allows to automatically
differentiate 1Gbps and 2.5Gbps PHY's, and therefore allows to
remove the fake PHY ID mechanism for RTL8125.
So far RTL8125 supports 2.5Gbps only, but register layout for faster
modes has been defined already, so let's use this information to be
future-proof.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
Pull iommu fixes from Joerg Roedel:
- A couple more fixes for the Intel VT-d driver for bugs introduced
during the recent conversion of this driver to use IOMMU core default
domains.
- Fix for common dma-iommu code to make sure MSI mappings happen in the
correct domain for a device.
- Fix a corner case in the handling of sg-lists in dma-iommu code that
might cause dma_length to be truncated.
- Mark a switch as fall-through in arm-smmu code.
* tag 'iommu-fixes-v5.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
iommu/vt-d: Fix possible use-after-free of private domain
iommu/vt-d: Detach domain before using a private one
iommu/dma: Handle SG length overflow better
iommu/vt-d: Correctly check format of page table in debugfs
iommu/vt-d: Detach domain when move device out of group
iommu/arm-smmu: Mark expected switch fall-through
iommu/dma: Handle MSI mappings separately
|
|
Merge misc VM fixes from Andrew Morton:
"A bunch of hotfixes, all affecting mm/.
The two-patch series from Andrea may be controversial. This restores
patches which were reverted in Dec 2018 due to a regression report [*].
After extensive discussion it is evident that the problems which these
patches solved were significantly more serious than the problems they
introduced. I am told that major distros are already carrying these
two patches for this reason"
[*] See
https://lore.kernel.org/lkml/alpine.DEB.2.21.1812061343240.144733@chino.kir.corp.google.com/
https://lore.kernel.org/lkml/alpine.DEB.2.21.1812031545560.161134@chino.kir.corp.google.com/
for the google-specific issues brought up by David Rijentes. And as
Andrew says:
"I'm unaware of anyone else who will be adversely affected by this,
and google already carries over a thousand kernel patches - another
won't kill them.
There has been sporadic discussion about fixing these things for
real but it's clear that nobody apart from David is particularly
motivated"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
hugetlbfs: fix hugetlb page migration/fault race causing SIGBUS
mm, vmscan: do not special-case slab reclaim when watermarks are boosted
Revert "mm, thp: restore node-local hugepage allocations"
Revert "Revert "mm, thp: consolidate THP gfp handling into alloc_hugepage_direct_gfpmask""
include/asm-generic/5level-fixup.h: fix variable 'p4d' set but not used
seq_file: fix problem when seeking mid-record
mm: workingset: fix vmstat counters for shadow nodes
mm/usercopy: use memory range to be accessed for wraparound check
mm: kmemleak: disable early logging in case of error
mm/vmalloc.c: fix percpu free VM area search criteria
mm/memcontrol.c: fix use after free in mem_cgroup_iter()
mm/z3fold.c: fix z3fold_destroy_pool() race condition
mm/z3fold.c: fix z3fold_destroy_pool() ordering
mm: mempolicy: handle vma with unmovable pages mapped correctly in mbind
mm: mempolicy: make the behavior consistent when MPOL_MF_MOVE* and MPOL_MF_STRICT were specified
mm/hmm: fix bad subpage pointer in try_to_unmap_one
mm/hmm: fix ZONE_DEVICE anon page mapping reuse
mm: document zone device struct page field usage
|
|
Holger reported sporadic transmit timeouts and it turned out that one
path misses ringing the doorbell. Fix was suggested by Eric.
Fixes: ef14358546b1 ("r8169: make use of xmit_more")
Suggested-by: Eric Dumazet <edumazet@google.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Don't fall through to print error message when receive sleep indication
in HCI_IBS_RX_ASLEEP state, this is allowed behavior.
Signed-off-by: Rocky Liao <rjliao@codeaurora.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
|
|
This patch corrects the SPDX License Identifier style
in header file related to STM32 Driver for I2C hardware
bus support.
For C header files Documentation/process/license-rules.rst
mandates C-like comments (opposed to C source files where
C++ style should be used)
Changes made by using a script provided by Joe Perches here:
https://lkml.org/lkml/2019/2/7/46
Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Nishad Kamdar <nishadkamdar@gmail.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
|
|
After we disabled interrupts, there might still be an active one
running. Sync before clearing the pointer to the slave device.
Fixes: c31d0a00021d ("i2c: emev2: add slave support")
Reported-by: Krzysztof Adamski <krzysztof.adamski@nokia.com>
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Krzysztof Adamski <krzysztof.adamski@nokia.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
|
|
After we disabled interrupts, there might still be an active one
running. Sync before clearing the pointer to the slave device.
Fixes: de20d1857dd6 ("i2c: rcar: add slave support")
Reported-by: Krzysztof Adamski <krzysztof.adamski@nokia.com>
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Krzysztof Adamski <krzysztof.adamski@nokia.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
|
|
rxrpc_queue_local() attempts to queue the local endpoint it is given and
then, if successful, prints a trace line. The trace line includes the
current usage count - but we're not allowed to look at the local endpoint
at this point as we passed our ref on it to the workqueue.
Fix this by reading the usage count before queuing the work item.
Also fix the reading of local->debug_id for trace lines, which must be done
with the same consideration as reading the usage count.
Fixes: 09d2bf595db4 ("rxrpc: Add a tracepoint to track rxrpc_local refcounting")
Reported-by: syzbot+78e71c5bab4f76a6a719@syzkaller.appspotmail.com
Signed-off-by: David Howells <dhowells@redhat.com>
|
|
When a local endpoint (struct rxrpc_local) ceases to be in use by any
AF_RXRPC sockets, it starts the process of being destroyed, but this
doesn't cause it to be removed from the namespace endpoint list immediately
as tearing it down isn't trivial and can't be done in softirq context, so
it gets deferred.
If a new socket comes along that wants to bind to the same endpoint, a new
rxrpc_local object will be allocated and rxrpc_lookup_local() will use
list_replace() to substitute the new one for the old.
Then, when the dying object gets to rxrpc_local_destroyer(), it is removed
unconditionally from whatever list it is on by calling list_del_init().
However, list_replace() doesn't reset the pointers in the replaced
list_head and so the list_del_init() will likely corrupt the local
endpoints list.
Fix this by using list_replace_init() instead.
Fixes: 730c5fd42c1e ("rxrpc: Fix local endpoint refcounting")
Reported-by: syzbot+193e29e9387ea5837f1d@syzkaller.appspotmail.com
Signed-off-by: David Howells <dhowells@redhat.com>
|
|
I would like to maintain the i2c-imx driver. Since I work with
different i.MX variants and have access to the hardware, I can spend
some time on the reviewing of this driver.
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
|
|
Since commit e1ab9a468e3b ("i2c: imx: improve the error handling in
i2c_imx_dma_request()") when booting with the DMA driver as module (such
as CONFIG_FSL_EDMA=m) the following endless clk warnings are seen:
[ 153.077831] ------------[ cut here ]------------
[ 153.082528] WARNING: CPU: 0 PID: 15 at drivers/clk/clk.c:924 clk_core_disable_lock+0x18/0x24
[ 153.093077] i2c0 already disabled
[ 153.096416] Modules linked in:
[ 153.099521] CPU: 0 PID: 15 Comm: kworker/0:1 Tainted: G W 5.2.0+ #321
[ 153.107290] Hardware name: Freescale Vybrid VF5xx/VF6xx (Device Tree)
[ 153.113772] Workqueue: events deferred_probe_work_func
[ 153.118979] [<c0019560>] (unwind_backtrace) from [<c0014734>] (show_stack+0x10/0x14)
[ 153.126778] [<c0014734>] (show_stack) from [<c083f8dc>] (dump_stack+0x9c/0xd4)
[ 153.134051] [<c083f8dc>] (dump_stack) from [<c0031154>] (__warn+0xf8/0x124)
[ 153.141056] [<c0031154>] (__warn) from [<c0031248>] (warn_slowpath_fmt+0x38/0x48)
[ 153.148580] [<c0031248>] (warn_slowpath_fmt) from [<c040fde0>] (clk_core_disable_lock+0x18/0x24)
[ 153.157413] [<c040fde0>] (clk_core_disable_lock) from [<c058f520>] (i2c_imx_probe+0x554/0x6ec)
[ 153.166076] [<c058f520>] (i2c_imx_probe) from [<c04b9178>] (platform_drv_probe+0x48/0x98)
[ 153.174297] [<c04b9178>] (platform_drv_probe) from [<c04b7298>] (really_probe+0x1d8/0x2c0)
[ 153.182605] [<c04b7298>] (really_probe) from [<c04b7554>] (driver_probe_device+0x5c/0x174)
[ 153.190909] [<c04b7554>] (driver_probe_device) from [<c04b58c8>] (bus_for_each_drv+0x44/0x8c)
[ 153.199480] [<c04b58c8>] (bus_for_each_drv) from [<c04b746c>] (__device_attach+0xa0/0x108)
[ 153.207782] [<c04b746c>] (__device_attach) from [<c04b65a4>] (bus_probe_device+0x88/0x90)
[ 153.215999] [<c04b65a4>] (bus_probe_device) from [<c04b6a04>] (deferred_probe_work_func+0x60/0x90)
[ 153.225003] [<c04b6a04>] (deferred_probe_work_func) from [<c004f190>] (process_one_work+0x204/0x634)
[ 153.234178] [<c004f190>] (process_one_work) from [<c004f618>] (worker_thread+0x20/0x484)
[ 153.242315] [<c004f618>] (worker_thread) from [<c0055c2c>] (kthread+0x118/0x150)
[ 153.249758] [<c0055c2c>] (kthread) from [<c00090b4>] (ret_from_fork+0x14/0x20)
[ 153.257006] Exception stack(0xdde43fb0 to 0xdde43ff8)
[ 153.262095] 3fa0: 00000000 00000000 00000000 00000000
[ 153.270306] 3fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[ 153.278520] 3fe0: 00000000 00000000 00000000 00000000 00000013 00000000
[ 153.285159] irq event stamp: 3323022
[ 153.288787] hardirqs last enabled at (3323021): [<c0861c4c>] _raw_spin_unlock_irq+0x24/0x2c
[ 153.297261] hardirqs last disabled at (3323022): [<c040d7a0>] clk_enable_lock+0x10/0x124
[ 153.305392] softirqs last enabled at (3322092): [<c000a504>] __do_softirq+0x344/0x540
[ 153.313352] softirqs last disabled at (3322081): [<c00385c0>] irq_exit+0x10c/0x128
[ 153.320946] ---[ end trace a506731ccd9bd703 ]---
This endless clk warnings behaviour is well explained by Andrey Smirnov:
"Allocating DMA after registering I2C adapter can lead to infinite
probing loop, for example, consider the following scenario:
1. i2c_imx_probe() is called and successfully registers an I2C
adapter via i2c_add_numbered_adapter()
2. As a part of i2c_add_numbered_adapter() new I2C slave devices
are added from DT which results in a call to
driver_deferred_probe_trigger()
3. i2c_imx_probe() continues and calls i2c_imx_dma_request() which
due to lack of proper DMA driver returns -EPROBE_DEFER
4. i2c_imx_probe() fails, removes I2C adapter and returns
-EPROBE_DEFER, which places it into deferred probe list
5. Deferred probe work triggered in #2 above kicks in and calls
i2c_imx_probe() again thus bringing us to step #1"
So revert commit e1ab9a468e3b ("i2c: imx: improve the error handling in
i2c_imx_dma_request()") and restore the old behaviour, in order to
avoid regressions on existing setups.
Cc: <stable@vger.kernel.org>
Reported-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Reported-by: Russell King <linux@armlinux.org.uk>
Fixes: e1ab9a468e3b ("i2c: imx: improve the error handling in i2c_imx_dma_request()")
Signed-off-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
|
|
TCP rst and fin packets do not qualify to place a flow into the
flowtable. Most likely there will be no more packets after connection
closure. Without this patch, this flow entry expires and connection
tracking picks up the entry in ESTABLISHED state using the fixup
timeout, which makes this look inconsistent to the user for a connection
that is actually already closed.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Make codec enter D3 before rebooting or poweroff can fix the noise
issue on some laptops. And in theory it is harmless for all codecs
to enter D3 before rebooting or poweroff, let us add a generic
reboot_notify, then realtek and conexant drivers can call this
function.
Cc: stable@vger.kernel.org
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
|
|
We have 3 new lenovo laptops which have conexant codec 0x14f11f86,
these 3 laptops also have the noise issue when rebooting, after
letting the codec enter D3 before rebooting or poweroff, the noise
disappers.
Instead of adding a new ID again in the reboot_notify(), let us make
this function apply to all conexant codec. In theory make codec enter
D3 before rebooting or poweroff is harmless, and I tested this change
on a couple of other Lenovo laptops which have different conexant
codecs, there is no side effect so far.
Cc: stable@vger.kernel.org
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
|
|
If the stream outq is not empty, need to kfree nstr_list.
Fixes: d570a59c5b5f ("sctp: only allow the out stream reset when the stream outq is empty")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
Eric reported a syzbot warning:
BUG: KMSAN: uninit-value in nh_valid_get_del_req+0x6f1/0x8c0 net/ipv4/nexthop.c:1510
CPU: 0 PID: 11812 Comm: syz-executor444 Not tainted 5.3.0-rc3+ #17
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0x191/0x1f0 lib/dump_stack.c:113
kmsan_report+0x162/0x2d0 mm/kmsan/kmsan_report.c:109
__msan_warning+0x75/0xe0 mm/kmsan/kmsan_instr.c:294
nh_valid_get_del_req+0x6f1/0x8c0 net/ipv4/nexthop.c:1510
rtm_del_nexthop+0x1b1/0x610 net/ipv4/nexthop.c:1543
rtnetlink_rcv_msg+0x115a/0x1580 net/core/rtnetlink.c:5223
netlink_rcv_skb+0x431/0x620 net/netlink/af_netlink.c:2477
rtnetlink_rcv+0x50/0x60 net/core/rtnetlink.c:5241
netlink_unicast_kernel net/netlink/af_netlink.c:1302 [inline]
netlink_unicast+0xf6c/0x1050 net/netlink/af_netlink.c:1328
netlink_sendmsg+0x110f/0x1330 net/netlink/af_netlink.c:1917
sock_sendmsg_nosec net/socket.c:637 [inline]
sock_sendmsg net/socket.c:657 [inline]
___sys_sendmsg+0x14ff/0x1590 net/socket.c:2311
__sys_sendmmsg+0x53a/0xae0 net/socket.c:2413
__do_sys_sendmmsg net/socket.c:2442 [inline]
__se_sys_sendmmsg+0xbd/0xe0 net/socket.c:2439
__x64_sys_sendmmsg+0x56/0x70 net/socket.c:2439
do_syscall_64+0xbc/0xf0 arch/x86/entry/common.c:297
entry_SYSCALL_64_after_hwframe+0x63/0xe7
The root cause is nlmsg_parse calling __nla_parse which means the
header struct size is not checked.
nlmsg_parse should be a wrapper around __nlmsg_parse with
NL_VALIDATE_STRICT for the validate argument very much like
nlmsg_parse_deprecated is for NL_VALIDATE_LIBERAL.
Fixes: 3de6440354465 ("netlink: re-add parse/validate functions in strict mode")
Reported-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
After configuring and restarting aneg we immediately try to read the
link status. On some systems the PHY may not yet have cleared the
"aneg complete" and "link up" bits, resulting in a false link-up
signal. See [0] for a report.
Clause 22 and 45 both require the PHY to keep the AN_RESTART
bit set until the PHY actually starts auto-negotiation.
Let's consider this in the generic functions for reading link status.
The commit marked as fixed is the first one where the patch applies
cleanly.
[0] https://marc.info/?t=156518400300003&r=1&w=2
Fixes: c1164bb1a631 ("net: phy: check PMAPMD link status only in genphy_c45_read_link")
Tested-by: Yonglong Liu <liuyonglong@huawei.com>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
In mlx4_en_config_rss_steer(), 'rss_map->indir_qp' is allocated through
kzalloc(). After that, mlx4_qp_alloc() is invoked to configure RSS
indirection. However, if mlx4_qp_alloc() fails, the allocated
'rss_map->indir_qp' is not deallocated, leading to a memory leak bug.
To fix the above issue, add the 'qp_alloc_err' label to free
'rss_map->indir_qp'.
Fixes: 4931c6ef04b4 ("net/mlx4_en: Optimized single ring steering")
Signed-off-by: Wenwen Wang <wenwen@cs.uga.edu>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
The ibm,mac-address-filters property defines the maximum number of
addresses the hypervisor's multicast filter list can support. It is
encoded as a big-endian integer in the OF device tree, but the virtual
ethernet driver does not convert it for use by little-endian systems.
As a result, the driver is not behaving as it should on affected systems
when a large number of multicast addresses are assigned to the device.
Reported-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: Thomas Falcon <tlfalcon@linux.ibm.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
Callbacks for a cmd reply run outside the protection of card->lock, to
allow for additional cmds to be issued & enqueued in parallel.
When qeth_send_control_data() bails out for a cmd without having
received a reply (eg. due to timeout), its callback may concurrently be
processing a reply that just arrived. In this case, the callback
potentially accesses a stale reply->reply_param area that eg. was
on-stack and has already been released.
To avoid this race, add some locking so that qeth_send_control_data()
can (1) wait for a concurrently running callback, and (2) zap any
pending callback that still wants to run.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
Update the defconfig:
- Add CONFIG_HW_RANDOM=y and CONFIG_HW_RANDOM_VIRTIO=y to enable
VirtIORNG when running on QEMU
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Paul Walmsley <paul.walmsley@sifive.com>
|
|
Update the rv32_defconfig:
- Add 'CONFIG_DEVTMPFS_MOUNT=y' to match the RISC-V defconfig
- Add CONFIG_HW_RANDOM=y and CONFIG_HW_RANDOM_VIRTIO=y to enable
VirtIORNG when running on QEMU
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Paul Walmsley <paul.walmsley@sifive.com>
|
|
As the annotation says in sctp_do_8_2_transport_strike():
"If the transport error count is greater than the pf_retrans
threshold, and less than pathmaxrtx ..."
It should be transport->error_count checked with pathmaxrxt,
instead of asoc->pf_retrans.
Fixes: 5aa93bcf66f4 ("sctp: Implement quick failover draft from tsvwg")
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
Pablo Neira Ayuso says:
====================
Netfilter/IPVS updates for net-next
The following patchset contains Netfilter/IPVS updates for net-next:
1) Rename mss field to mss_option field in synproxy, from Fernando Mancera.
2) Use SYSCTL_{ZERO,ONE} definitions in conntrack, from Matteo Croce.
3) More strict validation of IPVS sysctl values, from Junwei Hu.
4) Remove unnecessary spaces after on the right hand side of assignments,
from yangxingwu.
5) Add offload support for bitwise operation.
6) Extend the nft_offload_reg structure to store immediate date.
7) Collapse several ip_set header files into ip_set.h, from
Jeremy Sowden.
8) Make netfilter headers compile with CONFIG_KERNEL_HEADER_TEST=y,
from Jeremy Sowden.
9) Fix several sparse warnings due to missing prototypes, from
Valdis Kletnieks.
10) Use static lock initialiser to ensure connlabel spinlock is
initialized on boot time to fix sched/act_ct.c, patch
from Florian Westphal.
====================
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
Hayes says:
====================
v2:
For patch #2, replace list_for_each_safe with list_for_each_entry_safe.
Remove unlikely in WARN_ON. Adjust the coding style.
For patch #4, replace list_for_each_safe with list_for_each_entry_safe.
Remove "else" after "continue".
For patch #5. replace sysfs with ethtool to modify rx_copybreak and
rx_pending.
v1:
The different chips use different rx buffer size.
Use skb_add_rx_frag() to reduce memory copy for RX.
====================
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
Let the rx_copybreak and rx_pending could be modified by
ethtool.
Signed-off-by: Hayes Wang <hayeswang@realtek.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
Use skb_add_rx_frag() to reduce the memory copy for rx data.
Use a new list of rx_used to store the rx buffer which couldn't be
reused yet.
Besides, the total number of rx buffer may be increased or decreased
dynamically. And it is limited by RTL8152_MAX_RX_AGG.
Signed-off-by: Hayes Wang <hayeswang@realtek.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
Replace kmalloc_node() with alloc_pages() for rx buffer.
Signed-off-by: Hayes Wang <hayeswang@realtek.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
The original method uses an array to store the rx information. The
new one uses a list to link each rx structure. Then, it is possible
to increase/decrease the number of rx structure dynamically.
Signed-off-by: Hayes Wang <hayeswang@realtek.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
The different chips may accept different rx buffer sizes. The RTL8152
supports 16K bytes, and RTL8153 support 32K bytes.
Signed-off-by: Hayes Wang <hayeswang@realtek.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
Heiner says:
====================
So far phy_speed_down/up can be used up to 1Gbps only. Remove this
restriction and add needed helpers to phy-core.c
v2:
- remove unused parameter in patch 1
- rename __phy_speed_down to phy_speed_down_core in patch 2
====================
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
So far phy_speed_down/up can be used up to 1Gbps only. Remove this
restriction by using new helper __phy_speed_down. New member adv_old
in struct phy_device is used by phy_speed_up to restore the advertised
modes before calling phy_speed_down. Don't simply advertise what is
supported because a user may have intentionally removed modes from
advertisement.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
phy_speed_down_core provides most of the functionality for
phy_speed_down. It makes use of new helper phy_resolve_min_speed that is
based on the sorting of the settings[] array. In certain cases it may be
helpful to be able to exclude legacy half duplex modes, therefore
prepare phy_resolve_min_speed() for it.
v2:
- rename __phy_speed_down to phy_speed_down_core
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
We will need the functionality of __set_linkmode_max_speed also for
linkmode bitmaps other than phydev->supported. Therefore split it.
v2:
- remove unused parameter from __set_linkmode_max_speed
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
It is enough for caller of devlink_compat_switch_id_get() to hold the net
device to guarantee that devlink port is not destroyed concurrently. Remove
rtnl lock assertion and modify comment to warn user that they must hold
either rtnl lock or reference to net device. This is necessary to
accommodate future implementation of rtnl-unlocked TC offloads driver
callbacks.
Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
Daniel Borkmann says:
====================
The following pull-request contains BPF updates for your *net-next* tree.
There is a small merge conflict in libbpf (Cc Andrii so he's in the loop
as well):
for (i = 1; i <= btf__get_nr_types(btf); i++) {
t = (struct btf_type *)btf__type_by_id(btf, i);
if (!has_datasec && btf_is_var(t)) {
/* replace VAR with INT */
t->info = BTF_INFO_ENC(BTF_KIND_INT, 0, 0);
<<<<<<< HEAD
/*
* using size = 1 is the safest choice, 4 will be too
* big and cause kernel BTF validation failure if
* original variable took less than 4 bytes
*/
t->size = 1;
*(int *)(t+1) = BTF_INT_ENC(0, 0, 8);
} else if (!has_datasec && kind == BTF_KIND_DATASEC) {
=======
t->size = sizeof(int);
*(int *)(t + 1) = BTF_INT_ENC(0, 0, 32);
} else if (!has_datasec && btf_is_datasec(t)) {
>>>>>>> 72ef80b5ee131e96172f19e74b4f98fa3404efe8
/* replace DATASEC with STRUCT */
Conflict is between the two commits 1d4126c4e119 ("libbpf: sanitize VAR to
conservative 1-byte INT") and b03bc6853c0e ("libbpf: convert libbpf code to
use new btf helpers"), so we need to pick the sanitation fixup as well as
use the new btf_is_datasec() helper and the whitespace cleanup. Looks like
the following:
[...]
if (!has_datasec && btf_is_var(t)) {
/* replace VAR with INT */
t->info = BTF_INFO_ENC(BTF_KIND_INT, 0, 0);
/*
* using size = 1 is the safest choice, 4 will be too
* big and cause kernel BTF validation failure if
* original variable took less than 4 bytes
*/
t->size = 1;
*(int *)(t + 1) = BTF_INT_ENC(0, 0, 8);
} else if (!has_datasec && btf_is_datasec(t)) {
/* replace DATASEC with STRUCT */
[...]
The main changes are:
1) Addition of core parts of compile once - run everywhere (co-re) effort,
that is, relocation of fields offsets in libbpf as well as exposure of
kernel's own BTF via sysfs and loading through libbpf, from Andrii.
More info on co-re: http://vger.kernel.org/bpfconf2019.html#session-2
and http://vger.kernel.org/lpc-bpf2018.html#session-2
2) Enable passing input flags to the BPF flow dissector to customize parsing
and allowing it to stop early similar to the C based one, from Stanislav.
3) Add a BPF helper function that allows generating SYN cookies from XDP and
tc BPF, from Petar.
4) Add devmap hash-based map type for more flexibility in device lookup for
redirects, from Toke.
5) Improvements to XDP forwarding sample code now utilizing recently enabled
devmap lookups, from Jesper.
6) Add support for reporting the effective cgroup progs in bpftool, from Jakub
and Takshak.
7) Fix reading kernel config from bpftool via /proc/config.gz, from Peter.
8) Fix AF_XDP umem pages mapping for 32 bit architectures, from Ivan.
9) Follow-up to add two more BPF loop tests for the selftest suite, from Alexei.
10) Add perf event output helper also for other skb-based program types, from Allan.
11) Fix a co-re related compilation error in selftests, from Yonghong.
====================
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
Fix sparse warning:
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c:3190:5:
warning: symbol 'hclge_func_reset_sync_vf' was not declared. Should it be static?
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
|
|
Li Wang discovered that LTP/move_page12 V2 sometimes triggers SIGBUS in
the kernel-v5.2.3 testing. This is caused by a race between hugetlb
page migration and page fault.
If a hugetlb page can not be allocated to satisfy a page fault, the task
is sent SIGBUS. This is normal hugetlbfs behavior. A hugetlb fault
mutex exists to prevent two tasks from trying to instantiate the same
page. This protects against the situation where there is only one
hugetlb page, and both tasks would try to allocate. Without the mutex,
one would fail and SIGBUS even though the other fault would be
successful.
There is a similar race between hugetlb page migration and fault.
Migration code will allocate a page for the target of the migration. It
will then unmap the original page from all page tables. It does this
unmap by first clearing the pte and then writing a migration entry. The
page table lock is held for the duration of this clear and write
operation. However, the beginnings of the hugetlb page fault code
optimistically checks the pte without taking the page table lock. If
clear (as it can be during the migration unmap operation), a hugetlb
page allocation is attempted to satisfy the fault. Note that the page
which will eventually satisfy this fault was already allocated by the
migration code. However, the allocation within the fault path could
fail which would result in the task incorrectly being sent SIGBUS.
Ideally, we could take the hugetlb fault mutex in the migration code
when modifying the page tables. However, locks must be taken in the
order of hugetlb fault mutex, page lock, page table lock. This would
require significant rework of the migration code. Instead, the issue is
addressed in the hugetlb fault code. After failing to allocate a huge
page, take the page table lock and check for huge_pte_none before
returning an error. This is the same check that must be made further in
the code even if page allocation is successful.
Link: http://lkml.kernel.org/r/20190808000533.7701-1-mike.kravetz@oracle.com
Fixes: 290408d4a250 ("hugetlb: hugepage migration core")
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
Reported-by: Li Wang <liwang@redhat.com>
Tested-by: Li Wang <liwang@redhat.com>
Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Cyril Hrubis <chrubis@suse.cz>
Cc: Xishi Qiu <xishi.qiuxishi@alibaba-inc.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Dave Chinner reported a problem pointing a finger at commit 1c30844d2dfe
("mm: reclaim small amounts of memory when an external fragmentation
event occurs").
The report is extensive:
https://lore.kernel.org/linux-mm/20190807091858.2857-1-david@fromorbit.com/
and it's worth recording the most relevant parts (colorful language and
typos included).
When running a simple, steady state 4kB file creation test to
simulate extracting tarballs larger than memory full of small
files into the filesystem, I noticed that once memory fills up
the cache balance goes to hell.
The workload is creating one dirty cached inode for every dirty
page, both of which should require a single IO each to clean and
reclaim, and creation of inodes is throttled by the rate at which
dirty writeback runs at (via balance dirty pages). Hence the ingest
rate of new cached inodes and page cache pages is identical and
steady. As a result, memory reclaim should quickly find a steady
balance between page cache and inode caches.
The moment memory fills, the page cache is reclaimed at a much
faster rate than the inode cache, and evidence suggests that
the inode cache shrinker is not being called when large batches
of pages are being reclaimed. In roughly the same time period
that it takes to fill memory with 50% pages and 50% slab caches,
memory reclaim reduces the page cache down to just dirty pages
and slab caches fill the entirety of memory.
The LRU is largely full of dirty pages, and we're getting spikes
of random writeback from memory reclaim so it's all going to shit.
Behaviour never recovers, the page cache remains pinned at just
dirty pages, and nothing I could tune would make any difference.
vfs_cache_pressure makes no difference - I would set it so high
it should trim the entire inode caches in a single pass, yet it
didn't do anything. It was clear from tracing and live telemetry
that the shrinkers were pretty much not running except when
there was absolutely no memory free at all, and then they did
the minimum necessary to free memory to make progress.
So I went looking at the code, trying to find places where pages
got reclaimed and the shrinkers weren't called. There's only one
- kswapd doing boosted reclaim as per commit 1c30844d2dfe ("mm:
reclaim small amounts of memory when an external fragmentation
event occurs").
The watermark boosting introduced by the commit is triggered in response
to an allocation "fragmentation event". The boosting was not intended
to target THP specifically and triggers even if THP is disabled.
However, with Dave's perfectly reasonable workload, fragmentation events
can be very common given the ratio of slab to page cache allocations so
boosting remains active for long periods of time.
As high-order allocations might use compaction and compaction cannot
move slab pages the decision was made in the commit to special-case
kswapd when watermarks are boosted -- kswapd avoids reclaiming slab as
reclaiming slab does not directly help compaction.
As Dave notes, this decision means that slab can be artificially
protected for long periods of time and messes up the balance with slab
and page caches.
Removing the special casing can still indirectly help avoid
fragmentation by avoiding fragmentation-causing events due to slab
allocation as pages from a slab pageblock will have some slab objects
freed. Furthermore, with the special casing, reclaim behaviour is
unpredictable as kswapd sometimes examines slab and sometimes does not
in a manner that is tricky to tune or analyse.
This patch removes the special casing. The downside is that this is not
a universal performance win. Some benchmarks that depend on the
residency of data when rereading metadata may see a regression when slab
reclaim is restored to its original behaviour. Similarly, some
benchmarks that only read-once or write-once may perform better when
page reclaim is too aggressive. The primary upside is that slab
shrinker is less surprising (arguably more sane but that's a matter of
opinion), behaves consistently regardless of the fragmentation state of
the system and properly obeys VM sysctls.
A fsmark benchmark configuration was constructed similar to what Dave
reported and is codified by the mmtest configuration
config-io-fsmark-small-file-stream. It was evaluated on a 1-socket
machine to avoid dealing with NUMA-related issues and the timing of
reclaim. The storage was an SSD Samsung Evo and a fresh trimmed XFS
filesystem was used for the test data.
This is not an exact replication of Dave's setup. The configuration
scales its parameters depending on the memory size of the SUT to behave
similarly across machines. The parameters mean the first sample
reported by fs_mark is using 50% of RAM which will barely be throttled
and look like a big outlier. Dave used fake NUMA to have multiple
kswapd instances which I didn't replicate. Finally, the number of
iterations differ from Dave's test as the target disk was not large
enough. While not identical, it should be representative.
fsmark
5.3.0-rc3 5.3.0-rc3
vanilla shrinker-v1r1
Min 1-files/sec 4444.80 ( 0.00%) 4765.60 ( 7.22%)
1st-qrtle 1-files/sec 5005.10 ( 0.00%) 5091.70 ( 1.73%)
2nd-qrtle 1-files/sec 4917.80 ( 0.00%) 4855.60 ( -1.26%)
3rd-qrtle 1-files/sec 4667.40 ( 0.00%) 4831.20 ( 3.51%)
Max-1 1-files/sec 11421.50 ( 0.00%) 9999.30 ( -12.45%)
Max-5 1-files/sec 11421.50 ( 0.00%) 9999.30 ( -12.45%)
Max-10 1-files/sec 11421.50 ( 0.00%) 9999.30 ( -12.45%)
Max-90 1-files/sec 4649.60 ( 0.00%) 4780.70 ( 2.82%)
Max-95 1-files/sec 4491.00 ( 0.00%) 4768.20 ( 6.17%)
Max-99 1-files/sec 4491.00 ( 0.00%) 4768.20 ( 6.17%)
Max 1-files/sec 11421.50 ( 0.00%) 9999.30 ( -12.45%)
Hmean 1-files/sec 5004.75 ( 0.00%) 5075.96 ( 1.42%)
Stddev 1-files/sec 1778.70 ( 0.00%) 1369.66 ( 23.00%)
CoeffVar 1-files/sec 33.70 ( 0.00%) 26.05 ( 22.71%)
BHmean-99 1-files/sec 5053.72 ( 0.00%) 5101.52 ( 0.95%)
BHmean-95 1-files/sec 5053.72 ( 0.00%) 5101.52 ( 0.95%)
BHmean-90 1-files/sec 5107.05 ( 0.00%) 5131.41 ( 0.48%)
BHmean-75 1-files/sec 5208.45 ( 0.00%) 5206.68 ( -0.03%)
BHmean-50 1-files/sec 5405.53 ( 0.00%) 5381.62 ( -0.44%)
BHmean-25 1-files/sec 6179.75 ( 0.00%) 6095.14 ( -1.37%)
5.3.0-rc3 5.3.0-rc3
vanillashrinker-v1r1
Duration User 501.82 497.29
Duration System 4401.44 4424.08
Duration Elapsed 8124.76 8358.05
This is showing a slight skew for the max result representing a large
outlier for the 1st, 2nd and 3rd quartile are similar indicating that
the bulk of the results show little difference. Note that an earlier
version of the fsmark configuration showed a regression but that
included more samples taken while memory was still filling.
Note that the elapsed time is higher. Part of this is that the
configuration included time to delete all the test files when the test
completes -- the test automation handles the possibility of testing
fsmark with multiple thread counts. Without the patch, many of these
objects would be memory resident which is part of what the patch is
addressing.
There are other important observations that justify the patch.
1. With the vanilla kernel, the number of dirty pages in the system is
very low for much of the test. With this patch, dirty pages is
generally kept at 10% which matches vm.dirty_background_ratio which
is normal expected historical behaviour.
2. With the vanilla kernel, the ratio of Slab/Pagecache is close to
0.95 for much of the test i.e. Slab is being left alone and
dominating memory consumption. With the patch applied, the ratio
varies between 0.35 and 0.45 with the bulk of the measured ratios
roughly half way between those values. This is a different balance to
what Dave reported but it was at least consistent.
3. Slabs are scanned throughout the entire test with the patch applied.
The vanille kernel has periods with no scan activity and then
relatively massive spikes.
4. Without the patch, kswapd scan rates are very variable. With the
patch, the scan rates remain quite steady.
4. Overall vmstats are closer to normal expectations
5.3.0-rc3 5.3.0-rc3
vanilla shrinker-v1r1
Ops Direct pages scanned 99388.00 328410.00
Ops Kswapd pages scanned 45382917.00 33451026.00
Ops Kswapd pages reclaimed 30869570.00 25239655.00
Ops Direct pages reclaimed 74131.00 5830.00
Ops Kswapd efficiency % 68.02 75.45
Ops Kswapd velocity 5585.75 4002.25
Ops Page reclaim immediate 1179721.00 430927.00
Ops Slabs scanned 62367361.00 73581394.00
Ops Direct inode steals 2103.00 1002.00
Ops Kswapd inode steals 570180.00 5183206.00
o Vanilla kernel is hitting direct reclaim more frequently,
not very much in absolute terms but the fact the patch
reduces it is interesting
o "Page reclaim immediate" in the vanilla kernel indicates
dirty pages are being encountered at the tail of the LRU.
This is generally bad and means in this case that the LRU
is not long enough for dirty pages to be cleaned by the
background flush in time. This is much reduced by the
patch.
o With the patch, kswapd is reclaiming 10 times more slab
pages than with the vanilla kernel. This is indicative
of the watermark boosting over-protecting slab
A more complete set of tests were run that were part of the basis for
introducing boosting and while there are some differences, they are well
within tolerances.
Bottom line, the special casing kswapd to avoid slab behaviour is
unpredictable and can lead to abnormal results for normal workloads.
This patch restores the expected behaviour that slab and page cache is
balanced consistently for a workload with a steady allocation ratio of
slab/pagecache pages. It also means that if there are workloads that
favour the preservation of slab over pagecache that it can be tuned via
vm.vfs_cache_pressure where as the vanilla kernel effectively ignores
the parameter when boosting is active.
Link: http://lkml.kernel.org/r/20190808182946.GM2739@techsingularity.net
Fixes: 1c30844d2dfe ("mm: reclaim small amounts of memory when an external fragmentation event occurs")
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: <stable@vger.kernel.org> [5.0+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
This reverts commit 2f0799a0ffc033b ("mm, thp: restore node-local
hugepage allocations").
commit 2f0799a0ffc033b was rightfully applied to avoid the risk of a
severe regression that was reported by the kernel test robot at the end
of the merge window. Now we understood the regression was a false
positive and was caused by a significant increase in fairness during a
swap trashing benchmark. So it's safe to re-apply the fix and continue
improving the code from there. The benchmark that reported the
regression is very useful, but it provides a meaningful result only when
there is no significant alteration in fairness during the workload. The
removal of __GFP_THISNODE increased fairness.
__GFP_THISNODE cannot be used in the generic page faults path for new
memory allocations under the MPOL_DEFAULT mempolicy, or the allocation
behavior significantly deviates from what the MPOL_DEFAULT semantics are
supposed to be for THP and 4k allocations alike.
Setting THP defrag to "always" or using MADV_HUGEPAGE (with THP defrag
set to "madvise") has never meant to provide an implicit MPOL_BIND on
the "current" node the task is running on, causing swap storms and
providing a much more aggressive behavior than even zone_reclaim_node =
3.
Any workload who could have benefited from __GFP_THISNODE has now to
enable zone_reclaim_mode=1||2||3. __GFP_THISNODE implicitly provided
the zone_reclaim_mode behavior, but it only did so if THP was enabled:
if THP was disabled, there would have been no chance to get any 4k page
from the current node if the current node was full of pagecache, which
further shows how this __GFP_THISNODE was misplaced in MADV_HUGEPAGE.
MADV_HUGEPAGE has never been intended to provide any zone_reclaim_mode
semantics, in fact the two are orthogonal, zone_reclaim_mode = 1|2|3
must work exactly the same with MADV_HUGEPAGE set or not.
The performance characteristic of memory depends on the hardware
details. The numbers below are obtained on Naples/EPYC architecture and
the N/A projection extends them to show what we should aim for in the
future as a good THP NUMA locality default. The benchmark used
exercises random memory seeks (note: the cost of the page faults is not
part of the measurement).
D0 THP | D0 4k | D1 THP | D1 4k | D2 THP | D2 4k | D3 THP | D3 4k | ...
0% | +43% | +45% | +106% | +131% | +224% | N/A | N/A
D0 means distance zero (i.e. local memory), D1 means distance one (i.e.
intra socket memory), D2 means distance two (i.e. inter socket memory),
etc...
For the guest physical memory allocated by qemu and for guest mode
kernel the performance characteristic of RAM is more complex and an
ideal default could be:
D0 THP | D1 THP | D0 4k | D2 THP | D1 4k | D3 THP | D2 4k | D3 4k | ...
0% | +58% | +101% | N/A | +222% | N/A | N/A | N/A
NOTE: the N/A are projections and haven't been measured yet, the
measurement in this case is done on a 1950x with only two NUMA nodes.
The THP case here means THP was used both in the host and in the guest.
After applying this commit the THP NUMA locality order that we'll get
out of MADV_HUGEPAGE is this:
D0 THP | D1 THP | D2 THP | D3 THP | ... | D0 4k | D1 4k | D2 4k | D3 4k | ...
Before this commit it was:
D0 THP | D0 4k | D1 4k | D2 4k | D3 4k | ...
Even if we ignore the breakage of large workloads that can't fit in a
single node that the __GFP_THISNODE implicit "current node" mbind
caused, the THP NUMA locality order provided by __GFP_THISNODE was still
not the one we shall aim for in the long term (i.e. the first one at
the top).
After this commit is applied, we can introduce a new allocator multi
order API and to replace those two alloc_pages_vmas calls in the page
fault path, with a single multi order call:
unsigned int order = (1 << HPAGE_PMD_ORDER) | (1 << 0);
page = alloc_pages_multi_order(..., &order);
if (!page)
goto out;
if (!(order & (1 << 0))) {
VM_WARN_ON(order != 1 << HPAGE_PMD_ORDER);
/* THP fault */
} else {
VM_WARN_ON(order != 1 << 0);
/* 4k fallback */
}
The page allocator logic has to be altered so that when it fails on any
zone with order 9, it has to try again with a order 0 before falling
back to the next zone in the zonelist.
After that we need to do more measurements and evaluate if adding an
opt-in feature for guest mode is worth it, to swap "DN 4k | DN+1 THP"
with "DN+1 THP | DN 4k" at every NUMA distance crossing.
Link: http://lkml.kernel.org/r/20190503223146.2312-3-aarcange@redhat.com
Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: David Rientjes <rientjes@google.com>
Cc: Zi Yan <zi.yan@cs.rutgers.edu>
Cc: Stefan Priebe - Profihost AG <s.priebe@profihost.ag>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
