- 03 May, 2017 40 commits
-
-
Yisheng Xie authored
By reviewing code, I find that if the migrate target is a large free page and we ignore suitable, it may splite large target free page into smaller block which is not good for defrag. So move the ignore block suitable after check large free page. As Vlastimil pointed out in RFC version that this patch is just based on logical analyses which might be better for future-proofing the function and it is most likely won't have any visible effect right now, for direct compaction shouldn't have to be called if there's a >=pageblock_order page already available. Link: http://lkml.kernel.org/r/1489490743-5364-1-git-send-email-xieyisheng1@huawei.comSigned-off-by:
Yisheng Xie <xieyisheng1@huawei.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Michal Hocko <mhocko@suse.com> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: David Rientjes <rientjes@google.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Hanjun Guo <guohanjun@huawei.com> Cc: Xishi Qiu <qiuxishi@huawei.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Wei Yang authored
The current implementation calculates usemap_size in two steps: * calculate number of bytes to cover these bits * calculate number of "unsigned long" to cover these bytes It would be more clear by: * calculate number of "unsigned long" to cover these bits * multiple it with sizeof(unsigned long) This patch refine usemap_size() a little to make it more easy to understand. Link: http://lkml.kernel.org/r/20170310043713.96871-1-richard.weiyang@gmail.comSigned-off-by:Wei Yang <richard.weiyang@gmail.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by:
-
Johannes Weiner authored
__GFP_NOWARN, which is usually added to avoid warnings from callsites that expect to fail and have fallbacks, currently also suppresses allocation stall warnings. These trigger when an allocation is stuck inside the allocator for 10 seconds or longer. But there is no class of allocations that can get legitimately stuck in the allocator for this long. This always indicates a problem. Always emit stall warnings. Restrict __GFP_NOWARN to alloc failures. Link: http://lkml.kernel.org/r/20170125181150.GA16398@cmpxchg.orgSigned-off-by:
Johannes Weiner <hannes@cmpxchg.org> Acked-by:
Minchan Kim <minchan@kernel.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by:
-
Mel Gorman authored
kswapd is woken to reclaim a node based on a failed allocation request from any eligible zone. Once reclaiming in balance_pgdat(), it will continue reclaiming until there is an eligible zone available for the zone it was woken for. kswapd tracks what zone it was recently woken for in pgdat->kswapd_classzone_idx. If it has not been woken recently, this zone will be 0. However, the decision on whether to sleep is made on kswapd_classzone_idx which is 0 without a recent wakeup request and that classzone does not account for lowmem reserves. This allows kswapd to sleep when a low small zone such as ZONE_DMA is balanced for a GFP_DMA request even if a stream of allocations cannot use that zone. While kswapd may be woken again shortly in the near future there are two consequences -- the pgdat bits that control congestion are cleared prematurely and direct reclaim is more likely as kswapd slept prematurely. This patch flips kswapd_classzone_idx to default to MAX_NR_ZONES (an invalid index) when there has been no recent wakeups. If there are no wakeups, it'll decide whether to sleep based on the highest possible zone available (MAX_NR_ZONES - 1). It then becomes critical that the "pgdat balanced" decisions during reclaim and when deciding to sleep are the same. If there is a mismatch, kswapd can stay awake continually trying to balance tiny zones. simoop was used to evaluate it again. Two of the preparation patches regressed the workload so they are included as the second set of results. Otherwise this patch looks artifically excellent 4.11.0-rc1 4.11.0-rc1 4.11.0-rc1 vanilla clear-v2 keepawake-v2 Amean p50-Read 21670074.18 ( 0.00%) 19786774.76 ( 8.69%) 22668332.52 ( -4.61%) Amean p95-Read 25456267.64 ( 0.00%) 24101956.27 ( 5.32%) 26738688.00 ( -5.04%) Amean p99-Read 29369064.73 ( 0.00%) 27691872.71 ( 5.71%) 30991404.52 ( -5.52%) Amean p50-Write 1390.30 ( 0.00%) 1011.91 ( 27.22%) 924.91 ( 33.47%) Amean p95-Write 412901.57 ( 0.00%) 34874.98 ( 91.55%) 1362.62 ( 99.67%) Amean p99-Write 6668722.09 ( 0.00%) 575449.60 ( 91.37%) 16854.04 ( 99.75%) Amean p50-Allocation 78714.31 ( 0.00%) 84246.26 ( -7.03%) 74729.74 ( 5.06%) Amean p95-Allocation 175533.51 ( 0.00%) 400058.43 (-127.91%) 101609.74 ( 42.11%) Amean p99-Allocation 247003.02 ( 0.00%) 10905600.00 (-4315.17%) 125765.57 ( 49.08%) With this patch on top, write and allocation latencies are massively improved. The read latencies are slightly impaired but it's worth noting that this is mostly due to the IO scheduler and not directly related to reclaim. The vmstats are a bit of a mix but the relevant ones are as follows; 4.10.0-rc7 4.10.0-rc7 4.10.0-rc7 mmots-20170209 clear-v1r25keepawake-v1r25 Swap Ins 0 0 0 Swap Outs 0 608 0 Direct pages scanned 6910672 3132699 6357298 Kswapd pages scanned 57036946 82488665 56986286 Kswapd pages reclaimed 55993488 63474329 55939113 Direct pages reclaimed 6905990 2964843 6352115 Kswapd efficiency 98% 76% 98% Kswapd velocity 12494.375 17597.507 12488.065 Direct efficiency 99% 94% 99% Direct velocity 1513.835 668.306 1393.148 Page writes by reclaim 0.000 4410243.000 0.000 Page writes file 0 4409635 0 Page writes anon 0 608 0 Page reclaim immediate 1036792 14175203 1042571 4.11.0-rc1 4.11.0-rc1 4.11.0-rc1 vanilla clear-v2 keepawake-v2 Swap Ins 0 12 0 Swap Outs 0 838 0 Direct pages scanned 6579706 3237270 6256811 Kswapd pages scanned 61853702 79961486 54837791 Kswapd pages reclaimed 60768764 60755788 53849586 Direct pages reclaimed 6579055 2987453 6256151 Kswapd efficiency 98% 75% 98% Page writes by reclaim 0.000 4389496.000 0.000 Page writes file 0 4388658 0 Page writes anon 0 838 0 Page reclaim immediate 1073573 14473009 982507 Swap-outs are equivalent to baseline. Direct reclaim is reduced but not eliminated. It's worth noting that there are two periods of direct reclaim for this workload. The first is when it switches from preparing the files for the actual test itself. It's a lot of file IO followed by a lot of allocs that reclaims heavily for a brief window. While direct reclaim is lower with clear-v2, it is due to kswapd scanning aggressively and trying to reclaim the world which is not the right thing to do. With the patches applied, there is still direct reclaim but the phase change from "creating work files" to starting multiple threads that allocate a lot of anonymous memory faster than kswapd can reclaim. Scanning/reclaim efficiency is restored by this patch. Page writes from reclaim context are back at 0 which is ideal. Pages immediately reclaimed after IO completes is slightly improved but it is expected this will vary slightly. On UMA, there is almost no change so this is not expected to be a universal win. [mgorman@suse.de: fix ->kswapd_classzone_idx initialization] Link: http://lkml.kernel.org/r/20170406174538.5msrznj6nt6qpbx5@suse.de Link: http://lkml.kernel.org/r/20170309075657.25121-4-mgorman@techsingularity.netSigned-off-by:Mel Gorman <mgorman@techsingularity.net> Acked-by:
Vlastimil Babka <vbabka@suse.cz> Acked-by:
Michal Hocko <mhocko@suse.com> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Shantanu Goel <sgoel01@yahoo.com> Signed-off-by:
-
Mel Gorman authored
A pgdat tracks if recent reclaim encountered too many dirty, writeback or congested pages. The flags control whether kswapd writes pages back from reclaim context, tags pages for immediate reclaim when IO completes, whether processes block on wait_iff_congested and whether kswapd blocks when too many pages marked for immediate reclaim are encountered. The state is cleared in a check function with side-effects. With the patch "mm, vmscan: fix zone balance check in prepare_kswapd_sleep", the timing of when the bits get cleared changed. Due to the way the check works, it'll clear the bits if ZONE_DMA is balanced for a GFP_DMA allocation because it does not account for lowmem reserves properly. For the simoop workload, kswapd is not stalling when it should due to the premature clearing, writing pages from reclaim context like crazy and generally being unhelpful. This patch resets the pgdat bits related to page reclaim only when kswapd is going to sleep. The comparison with simoop is then 4.11.0-rc1 4.11.0-rc1 4.11.0-rc1 vanilla fixcheck-v2 clear-v2 Amean p50-Read 21670074.18 ( 0.00%) 20464344.18 ( 5.56%) 19786774.76 ( 8.69%) Amean p95-Read 25456267.64 ( 0.00%) 25721423.64 ( -1.04%) 24101956.27 ( 5.32%) Amean p99-Read 29369064.73 ( 0.00%) 30174230.76 ( -2.74%) 27691872.71 ( 5.71%) Amean p50-Write 1390.30 ( 0.00%) 1395.28 ( -0.36%) 1011.91 ( 27.22%) Amean p95-Write 412901.57 ( 0.00%) 37737.74 ( 90.86%) 34874.98 ( 91.55%) Amean p99-Write 6668722.09 ( 0.00%) 666489.04 ( 90.01%) 575449.60 ( 91.37%) Amean p50-Allocation 78714.31 ( 0.00%) 86286.22 ( -9.62%) 84246.26 ( -7.03%) Amean p95-Allocation 175533.51 ( 0.00%) 351812.27 (-100.42%) 400058.43 (-127.91%) Amean p99-Allocation 247003.02 ( 0.00%) 6291171.56 (-2447.00%) 10905600.00 (-4315.17%) Read latency is improved, write latency is mostly improved but allocation latency is regressed. kswapd is still reclaiming inefficiently, pages are being written back from writeback context and a host of other issues. However, given the change, it needed to be spelled out why the side-effect was moved. Link: http://lkml.kernel.org/r/20170309075657.25121-3-mgorman@techsingularity.netSigned-off-by: -
Shantanu Goel authored
Patch series "Reduce amount of time kswapd sleeps prematurely", v2. The series is unusual in that the first patch fixes one problem and introduces other issues that are noted in the changelog. Patch 2 makes a minor modification that is worth considering on its own but leaves the kernel in a state where it behaves badly. It's not until patch 3 that there is an improvement against baseline. This was mostly motivated by examining Chris Mason's "simoop" benchmark which puts the VM under similar pressure to HADOOP. It has been reported that the benchmark has regressed severely during the last number of releases. While I cannot reproduce all the same problems Chris experienced due to hardware limitations, there was a number of problems on a 2-socket machine with a single disk. simoop latencies 4.11.0-rc1 4.11.0-rc1 vanilla keepawake-v2 Amean p50-Read 21670074.18 ( 0.00%) 22668332.52 ( -4.61%) Amean p95-Read 25456267.64 ( 0.00%) 26738688.00 ( -5.04%) Amean p99-Read 29369064.73 ( 0.00%) 30991404.52 ( -5.52%) Amean p50-Write 1390.30 ( 0.00%) 924.91 ( 33.47%) Amean p95-Write 412901.57 ( 0.00%) 1362.62 ( 99.67%) Amean p99-Write 6668722.09 ( 0.00%) 16854.04 ( 99.75%) Amean p50-Allocation 78714.31 ( 0.00%) 74729.74 ( 5.06%) Amean p95-Allocation 175533.51 ( 0.00%) 101609.74 ( 42.11%) Amean p99-Allocation 247003.02 ( 0.00%) 125765.57 ( 49.08%) These are latencies. Read/write are threads reading fixed-size random blocks from a simulated database. The allocation latency is mmaping and faulting regions of memory. The p50, 95 and p99 reports the worst latencies for 50% of the samples, 95% and 99% respectively. For example, the report indicates that while the test was running 99% of writes completed 99.75% faster. It's worth noting that on a UMA machine that no difference in performance with simoop was observed so milage will vary. It's noted that there is a slight impact to read latencies but it's mostly due to IO scheduler decisions and offset by the large reduction in other latencies. This patch (of 3): The check in prepare_kswapd_sleep needs to match the one in balance_pgdat since the latter will return as soon as any one of the zones in the classzone is above the watermark. This is specially important for higher order allocations since balance_pgdat will typically reset the order to zero relying on compaction to create the higher order pages. Without this patch, prepare_kswapd_sleep fails to wake up kcompactd since the zone balance check fails. It was first reported against 4.9.7 that kswapd is failing to wake up kcompactd due to a mismatch in the zone balance check between balance_pgdat() and prepare_kswapd_sleep(). balance_pgdat() returns as soon as a single zone satisfies the allocation but prepare_kswapd_sleep() requires all zones to do +the same. This causes prepare_kswapd_sleep() to never succeed except in the order == 0 case and consequently, wakeup_kcompactd() is never called. For the machine that originally motivated this patch, the state of compaction from /proc/vmstat looked this way after a day and a half +of uptime: compact_migrate_scanned 240496 compact_free_scanned 76238632 compact_isolated 123472 compact_stall 1791 compact_fail 29 compact_success 1762 compact_daemon_wake 0 After applying the patch and about 10 hours of uptime the state looks like this: compact_migrate_scanned 59927299 compact_free_scanned 2021075136 compact_isolated 640926 compact_stall 4 compact_fail 2 compact_success 2 compact_daemon_wake 5160 Further notes from Mel that motivated him to pick this patch up and resend it; It was observed for the simoop workload (pressures the VM similar to HADOOP) that kswapd was failing to keep ahead of direct reclaim. The investigation noted that there was a need to rationalise kswapd decisions to reclaim with kswapd decisions to sleep. With this patch on a 2-socket box, there was a 49% reduction in direct reclaim scanning. However, the impact otherwise is extremely negative. Kswapd reclaim efficiency dropped from 98% to 76%. simoop has three latency-related metrics for read, write and allocation (an anonymous mmap and fault). 4.11.0-rc1 4.11.0-rc1 vanilla fixcheck-v2 Amean p50-Read 21670074.18 ( 0.00%) 20464344.18 ( 5.56%) Amean p95-Read 25456267.64 ( 0.00%) 25721423.64 ( -1.04%) Amean p99-Read 29369064.73 ( 0.00%) 30174230.76 ( -2.74%) Amean p50-Write 1390.30 ( 0.00%) 1395.28 ( -0.36%) Amean p95-Write 412901.57 ( 0.00%) 37737.74 ( 90.86%) Amean p99-Write 6668722.09 ( 0.00%) 666489.04 ( 90.01%) Amean p50-Allocation 78714.31 ( 0.00%) 86286.22 ( -9.62%) Amean p95-Allocation 175533.51 ( 0.00%) 351812.27 (-100.42%) Amean p99-Allocation 247003.02 ( 0.00%) 6291171.56 (-2447.00%) Of greater concern is that the patch causes swapping and page writes from kswapd context rose from 0 pages to 4189753 pages during the hour the workload ran for. By and large, the patch has very bad behaviour but easily missed as the impact on a UMA machine is negligible. This patch is included with the data in case a bisection leads to this area. This patch is also a pre-requisite for the rest of the series. Link: http://lkml.kernel.org/r/20170309075657.25121-2-mgorman@techsingularity.netSigned-off-by:Shantanu Goel <sgoel01@yahoo.com> Signed-off-by:
Hillf Danton <hillf.zj@alibaba-inc.com> Acked-by:
-
Minchan Kim authored
With the discussion[1], I found it seems there are every PageFlags functions return bool at this moment so we don't need double negation any more. Although it's not a problem to keep it, it makes future users confused to use double negation for them, too. Remove such possibility. [1] https://marc.info/?l=linux-kernel&m=148881578820434 Frankly sepaking, I like every PageFlags to return bool instead of int. It will make it clear. AFAIR, Chen Gang had tried it but don't know why it was not merged at that time. http://lkml.kernel.org/r/1469336184-1904-1-git-send-email-chengang@emindsoft.com.cn Link: http://lkml.kernel.org/r/1488868597-32222-1-git-send-email-minchan@kernel.orgSigned-off-by:
-
Kees Cook authored
Since commit 3ad38ceb ("x86/mm: Remove CONFIG_DEBUG_NX_TEST"), nothing is using the exported rodata_test_data variable, so drop the export. This additionally updates the pr_fmt to avoid redundant strings and adjusts some whitespace. Link: http://lkml.kernel.org/r/20170307005313.GA85809@beastSigned-off-by:
Kees Cook <keescook@chromium.org> Cc: Jinbum Park <jinb.park7@gmail.com> Cc: Arjan van de Ven <arjan@linux.intel.com> Signed-off-by:
-
Matthew Wilcox authored
The round_up() macro generates a couple of unnecessary instructions in this usage: 48cd: 49 8b 47 50 mov 0x50(%r15),%rax 48d1: 48 83 e8 01 sub $0x1,%rax 48d5: 48 0d ff 0f 00 00 or $0xfff,%rax 48db: 48 83 c0 01 add $0x1,%rax 48df: 48 c1 f8 0c sar $0xc,%rax 48e3: 48 39 c3 cmp %rax,%rbx 48e6: 72 2e jb 4916 <filemap_fault+0x96> If we change round_up() to ((x) + __round_mask(x, y)) & ~__round_mask(x, y) then GCC can see through it and remove the mask (because that would be dead code given the subsequent shift): 48cd: 49 8b 47 50 mov 0x50(%r15),%rax 48d1: 48 05 ff 0f 00 00 add $0xfff,%rax 48d7: 48 c1 e8 0c shr $0xc,%rax 48db: 48 39 c3 cmp %rax,%rbx 48de: 72 2e jb 490e <filemap_fault+0x8e> But that's problematic because we'd evaluate 'y' twice. Converting round_up into an inline function prevents it from being used in other definitions. The easiest thing to do is just change these three usages of round_up to use DIV_ROUND_UP. Also add an unlikely() because GCC's heuristic is wrong in this case. Link: http://lkml.kernel.org/r/20170207192812.5281-1-willy@infradead.orgSigned-off-by:Matthew Wilcox <mawilcox@microsoft.com> Signed-off-by:
-
Michal Hocko authored
kjournald2 is central to the transaction commit processing. As such any potential allocation from this kernel thread has to be GFP_NOFS. Make sure to mark the whole kernel thread GFP_NOFS by the memalloc_nofs_save. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/20170306131408.9828-8-mhocko@kernel.orgSigned-off-by:
Jan Kara <jack@suse.cz> Reviewed-by:
-
Michal Hocko authored
now that we have memalloc_nofs_{save,restore} api we can mark the whole transaction context as implicitly GFP_NOFS. All allocations will automatically inherit GFP_NOFS this way. This means that we do not have to mark any of those requests with GFP_NOFS and moreover all the ext4_kv[mz]alloc(GFP_NOFS) are also safe now because even the hardcoded GFP_KERNEL allocations deep inside the vmalloc will be NOFS now. [akpm@linux-foundation.org: tweak comments] Link: http://lkml.kernel.org/r/20170306131408.9828-7-mhocko@kernel.orgSigned-off-by: -
Michal Hocko authored
kmem_zalloc_large and _xfs_buf_map_pages use memalloc_noio_{save,restore} API to prevent from reclaim recursion into the fs because vmalloc can invoke unconditional GFP_KERNEL allocations and these functions might be called from the NOFS contexts. The memalloc_noio_save will enforce GFP_NOIO context which is even weaker than GFP_NOFS and that seems to be unnecessary. Let's use memalloc_nofs_{save,restore} instead as it should provide exactly what we need here - implicit GFP_NOFS context. Link: http://lkml.kernel.org/r/20170306131408.9828-6-mhocko@kernel.orgSigned-off-by:Brian Foster <bfoster@redhat.com> Reviewed-by:
Darrick J. Wong <darrick.wong@oracle.com> Acked-by:
-
Michal Hocko authored
GFP_NOFS context is used for the following 5 reasons currently: - to prevent from deadlocks when the lock held by the allocation context would be needed during the memory reclaim - to prevent from stack overflows during the reclaim because the allocation is performed from a deep context already - to prevent lockups when the allocation context depends on other reclaimers to make a forward progress indirectly - just in case because this would be safe from the fs POV - silence lockdep false positives Unfortunately overuse of this allocation context brings some problems to the MM. Memory reclaim is much weaker (especially during heavy FS metadata workloads), OOM killer cannot be invoked because the MM layer doesn't have enough information about how much memory is freeable by the FS layer. In many cases it is far from clear why the weaker context is even used and so it might be used unnecessarily. We would like to get rid of those as much as possible. One way to do that is to use the flag in scopes rather than isolated cases. Such a scope is declared when really necessary, tracked per task and all the allocation requests from within the context will simply inherit the GFP_NOFS semantic. Not only this is easier to understand and maintain because there are much less problematic contexts than specific allocation requests, this also helps code paths where FS layer interacts with other layers (e.g. crypto, security modules, MM etc...) and there is no easy way to convey the allocation context between the layers. Introduce memalloc_nofs_{save,restore} API to control the scope of GFP_NOFS allocation context. This is basically copying memalloc_noio_{save,restore} API we have for other restricted allocation context GFP_NOIO. The PF_MEMALLOC_NOFS flag already exists and it is just an alias for PF_FSTRANS which has been xfs specific until recently. There are no more PF_FSTRANS users anymore so let's just drop it. PF_MEMALLOC_NOFS is now checked in the MM layer and drops __GFP_FS implicitly same as PF_MEMALLOC_NOIO drops __GFP_IO. memalloc_noio_flags is renamed to current_gfp_context because it now cares about both PF_MEMALLOC_NOFS and PF_MEMALLOC_NOIO contexts. Xfs code paths preserve their semantic. kmem_flags_convert() doesn't need to evaluate the flag anymore. This patch shouldn't introduce any functional changes. Let's hope that filesystems will drop direct GFP_NOFS (resp. ~__GFP_FS) usage as much as possible and only use a properly documented memalloc_nofs_{save,restore} checkpoints where they are appropriate. [akpm@linux-foundation.org: fix comment typo, reflow comment] Link: http://lkml.kernel.org/r/20170306131408.9828-5-mhocko@kernel.orgSigned-off-by: -
Michal Hocko authored
xfs has defined PF_FSTRANS to declare a scope GFP_NOFS semantic quite some time ago. We would like to make this concept more generic and use it for other filesystems as well. Let's start by giving the flag a more generic name PF_MEMALLOC_NOFS which is in line with an exiting PF_MEMALLOC_NOIO already used for the same purpose for GFP_NOIO contexts. Replace all PF_FSTRANS usage from the xfs code in the first step before we introduce a full API for it as xfs uses the flag directly anyway. This patch doesn't introduce any functional change. Link: http://lkml.kernel.org/r/20170306131408.9828-4-mhocko@kernel.orgSigned-off-by:
-
Michal Hocko authored
The current implementation of the reclaim lockup detection can lead to false positives and those even happen and usually lead to tweak the code to silence the lockdep by using GFP_NOFS even though the context can use __GFP_FS just fine. See http://lkml.kernel.org/r/20160512080321.GA18496@dastard as an example. ================================= [ INFO: inconsistent lock state ] 4.5.0-rc2+ #4 Tainted: G O --------------------------------- inconsistent {RECLAIM_FS-ON-R} -> {IN-RECLAIM_FS-W} usage. kswapd0/543 [HC0[0]:SC0[0]:HE1:SE1] takes: (&xfs_nondir_ilock_class){++++-+}, at: xfs_ilock+0x177/0x200 [xfs] {RECLAIM_FS-ON-R} state was registered at: mark_held_locks+0x79/0xa0 lockdep_trace_alloc+0xb3/0x100 kmem_cache_alloc+0x33/0x230 kmem_zone_alloc+0x81/0x120 [xfs] xfs_refcountbt_init_cursor+0x3e/0xa0 [xfs] __xfs_refcount_find_shared+0x75/0x580 [xfs] xfs_refcount_find_shared+0x84/0xb0 [xfs] xfs_getbmap+0x608/0x8c0 [xfs] xfs_vn_fiemap+0xab/0xc0 [xfs] do_vfs_ioctl+0x498/0x670 SyS_ioctl+0x79/0x90 entry_SYSCALL_64_fastpath+0x12/0x6f CPU0 ---- lock(&xfs_nondir_ilock_class); <Interrupt> lock(&xfs_nondir_ilock_class); *** DEADLOCK *** 3 locks held by kswapd0/543: stack backtrace: CPU: 0 PID: 543 Comm: kswapd0 Tainted: G O 4.5.0-rc2+ #4 Call Trace: lock_acquire+0xd8/0x1e0 down_write_nested+0x5e/0xc0 xfs_ilock+0x177/0x200 [xfs] xfs_reflink_cancel_cow_range+0x150/0x300 [xfs] xfs_fs_evict_inode+0xdc/0x1e0 [xfs] evict+0xc5/0x190 dispose_list+0x39/0x60 prune_icache_sb+0x4b/0x60 super_cache_scan+0x14f/0x1a0 shrink_slab.part.63.constprop.79+0x1e9/0x4e0 shrink_zone+0x15e/0x170 kswapd+0x4f1/0xa80 kthread+0xf2/0x110 ret_from_fork+0x3f/0x70 To quote Dave: "Ignoring whether reflink should be doing anything or not, that's a "xfs_refcountbt_init_cursor() gets called both outside and inside transactions" lockdep false positive case. The problem here is lockdep has seen this allocation from within a transaction, hence a GFP_NOFS allocation, and now it's seeing it in a GFP_KERNEL context. Also note that we have an active reference to this inode. So, because the reclaim annotations overload the interrupt level detections and it's seen the inode ilock been taken in reclaim ("interrupt") context, this triggers a reclaim context warning where it thinks it is unsafe to do this allocation in GFP_KERNEL context holding the inode ilock..." This sounds like a fundamental problem of the reclaim lock detection. It is really impossible to annotate such a special usecase IMHO unless the reclaim lockup detection is reworked completely. Until then it is much better to provide a way to add "I know what I am doing flag" and mark problematic places. This would prevent from abusing GFP_NOFS flag which has a runtime effect even on configurations which have lockdep disabled. Introduce __GFP_NOLOCKDEP flag which tells the lockdep gfp tracking to skip the current allocation request. While we are at it also make sure that the radix tree doesn't accidentaly override tags stored in the upper part of the gfp_mask. Link: http://lkml.kernel.org/r/20170306131408.9828-3-mhocko@kernel.orgSigned-off-by:
Peter Zijlstra <peterz@infradead.org> Acked-by:
-
Nikolay Borisov authored
Patch series "scope GFP_NOFS api", v5. This patch (of 7): Commit 21caf2fc ("mm: teach mm by current context info to not do I/O during memory allocation") added the memalloc_noio_(save|restore) functions to enable people to modify the MM behavior by disabling I/O during memory allocation. This was further extended in commit 934f3072 ("mm: clear __GFP_FS when PF_MEMALLOC_NOIO is set"). memalloc_noio_* functions prevent allocation paths recursing back into the filesystem without explicitly changing the flags for every allocation site. However, lockdep hasn't been keeping up with the changes and it entirely misses handling the memalloc_noio adjustments. Instead, it is left to the callers of __lockdep_trace_alloc to call the function after they have shaven the respective GFP flags which can lead to false positives: ================================= [ INFO: inconsistent lock state ] 4.10.0-nbor #134 Not tainted --------------------------------- inconsistent {IN-RECLAIM_FS-W} -> {RECLAIM_FS-ON-W} usage. fsstress/3365 [HC0[0]:SC0[0]:HE1:SE1] takes: (&xfs_nondir_ilock_class){++++?.}, at: xfs_ilock+0x141/0x230 {IN-RECLAIM_FS-W} state was registered at: __lock_acquire+0x62a/0x17c0 lock_acquire+0xc5/0x220 down_write_nested+0x4f/0x90 xfs_ilock+0x141/0x230 xfs_reclaim_inode+0x12a/0x320 xfs_reclaim_inodes_ag+0x2c8/0x4e0 xfs_reclaim_inodes_nr+0x33/0x40 xfs_fs_free_cached_objects+0x19/0x20 super_cache_scan+0x191/0x1a0 shrink_slab+0x26f/0x5f0 shrink_node+0xf9/0x2f0 kswapd+0x356/0x920 kthread+0x10c/0x140 ret_from_fork+0x31/0x40 irq event stamp: 173777 hardirqs last enabled at (173777): __local_bh_enable_ip+0x70/0xc0 hardirqs last disabled at (173775): __local_bh_enable_ip+0x37/0xc0 softirqs last enabled at (173776): _xfs_buf_find+0x67a/0xb70 softirqs last disabled at (173774): _xfs_buf_find+0x5db/0xb70 other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock(&xfs_nondir_ilock_class); <Interrupt> lock(&xfs_nondir_ilock_class); *** DEADLOCK *** 4 locks held by fsstress/3365: #0: (sb_writers#10){++++++}, at: mnt_want_write+0x24/0x50 #1: (&sb->s_type->i_mutex_key#12){++++++}, at: vfs_setxattr+0x6f/0xb0 #2: (sb_internal#2){++++++}, at: xfs_trans_alloc+0xfc/0x140 #3: (&xfs_nondir_ilock_class){++++?.}, at: xfs_ilock+0x141/0x230 stack backtrace: CPU: 0 PID: 3365 Comm: fsstress Not tainted 4.10.0-nbor #134 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Ubuntu-1.8.2-1ubuntu1 04/01/2014 Call Trace: kmem_cache_alloc_node_trace+0x3a/0x2c0 vm_map_ram+0x2a1/0x510 _xfs_buf_map_pages+0x77/0x140 xfs_buf_get_map+0x185/0x2a0 xfs_attr_rmtval_set+0x233/0x430 xfs_attr_leaf_addname+0x2d2/0x500 xfs_attr_set+0x214/0x420 xfs_xattr_set+0x59/0xb0 __vfs_setxattr+0x76/0xa0 __vfs_setxattr_noperm+0x5e/0xf0 vfs_setxattr+0xae/0xb0 setxattr+0x15e/0x1a0 path_setxattr+0x8f/0xc0 SyS_lsetxattr+0x11/0x20 entry_SYSCALL_64_fastpath+0x23/0xc6 Let's fix this by making lockdep explicitly do the shaving of respective GFP flags. Fixes: 934f3072 ("mm: clear __GFP_FS when PF_MEMALLOC_NOIO is set") Link: http://lkml.kernel.org/r/20170306131408.9828-2-mhocko@kernel.orgSigned-off-by:
Nikolay Borisov <nborisov@suse.com> Signed-off-by:
-
David Rientjes authored
After "mm, vmstat: print non-populated zones in zoneinfo", /proc/zoneinfo will show unpopulated zones. The per-cpu pageset statistics are not relevant for unpopulated zones and can be potentially lengthy, so supress them when they are not interesting. Also moves lowmem reserve protection information above pcp stats since it is relevant for all zones per vm.lowmem_reserve_ratio. Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1703061400500.46428@chino.kir.corp.google.comSigned-off-by:
David Rientjes <rientjes@google.com> Cc: Anshuman Khandual <khandual@linux.vnet.ibm.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by:
-
David Rientjes authored
Initscripts can use the information (protection levels) from /proc/zoneinfo to configure vm.lowmem_reserve_ratio at boot. vm.lowmem_reserve_ratio is an array of ratios for each configured zone on the system. If a zone is not populated on an arch, /proc/zoneinfo suppresses its output. This results in there not being a 1:1 mapping between the set of zones emitted by /proc/zoneinfo and the zones configured by vm.lowmem_reserve_ratio. This patch shows statistics for non-populated zones in /proc/zoneinfo. The zones exist and hold a spot in the vm.lowmem_reserve_ratio array. Without this patch, it is not possible to determine which index in the array controls which zone if one or more zones on the system are not populated. Remaining users of walk_zones_in_node() are unchanged. Files such as /proc/pagetypeinfo require certain zone data to be initialized properly for display, which is not done for unpopulated zones. Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1703031451310.98023@chino.kir.corp.google.comSigned-off-by:
Anshuman Khandual <khandual@linux.vnet.ibm.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Signed-off-by:
-
Xishi Qiu authored
Use is_migrate_isolate_page() to simplify the code, no functional changes. Link: http://lkml.kernel.org/r/58B94FB1.8020802@huawei.comSigned-off-by:
Xishi Qiu <qiuxishi@huawei.com> Acked-by:
-
Xishi Qiu authored
Introduce two helpers, is_migrate_highatomic() and is_migrate_highatomic_page(). Simplify the code, no functional changes. [akpm@linux-foundation.org: use static inlines rather than macros, per mhocko] Link: http://lkml.kernel.org/r/58B94F15.6060606@huawei.comSigned-off-by:
-
Huang Ying authored
Before using cluster lock in free_swap_and_cache(), the swap_info_struct->lock will be held during freeing the swap entry and acquiring page lock, so the page swap count will not change when testing page information later. But after using cluster lock, the cluster lock (or swap_info_struct->lock) will be held only during freeing the swap entry. So before acquiring the page lock, the page swap count may be changed in another thread. If the page swap count is not 0, we should not delete the page from the swap cache. This is fixed via checking page swap count again after acquiring the page lock. I found the race when I review the code, so I didn't trigger the race via a test program. If the race occurs for an anonymous page shared by multiple processes via fork, multiple pages will be allocated and swapped in from the swap device for the previously shared one page. That is, the user-visible runtime effect is more memory will be used and the access latency for the page will be higher, that is, the performance regression. Link: http://lkml.kernel.org/r/20170301143905.12846-1-ying.huang@intel.comSigned-off-by:
"Huang, Ying" <ying.huang@intel.com> Cc: Hugh Dickins <hughd@google.com> Cc: Shaohua Li <shli@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Rik van Riel <riel@redhat.com> Cc: Tim Chen <tim.c.chen@intel.com> Signed-off-by:
-
Johannes Weiner authored
Cgroups currently don't report how much shmem they use, which can be useful data to have, in particular since shmem is included in the cache/file item while being reclaimed like anonymous memory. Add a counter to track shmem pages during charging and uncharging. Link: http://lkml.kernel.org/r/20170221164343.32252-1-hannes@cmpxchg.orgSigned-off-by:
Chris Down <cdown@fb.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by:
-
Shaohua Li authored
Show MADV_FREE pages info of each vma in smaps. The interface is for diganose or monitoring purpose, userspace could use it to understand what happens in the application. Since userspace could dirty MADV_FREE pages without notice from kernel, this interface is the only place we can get accurate accounting info about MADV_FREE pages. [mhocko@kernel.org: update Documentation/filesystems/proc.txt] Link: http://lkml.kernel.org/r/89efde633559de1ec07444f2ef0f4963a97a2ce8.1487965799.git.shli@fb.comSigned-off-by:
Shaohua Li <shli@fb.com> Acked-by:
-
Shaohua Li authored
Now MADV_FREE pages can be easily reclaimed even for swapless system. We can safely enable MADV_FREE for all systems. Link: http://lkml.kernel.org/r/155648585589300bfae1d45078e7aebb3d988b87.1487965799.git.shli@fb.comSigned-off-by:
-
Minchan Kim authored
If a page is swapbacked, it means it should be in swapcache in try_to_unmap_one's path. If a page is !swapbacked, it mean it shouldn't be in swapcache in try_to_unmap_one's path. Check both two cases all at once and if it fails, warn and return SWAP_FAIL. Such bug never mean we should shut down the kernel. [minchan@kernel.org: do not use VM_WARN_ON_ONCE as if condition[ Link: http://lkml.kernel.org/r/20170309060226.GB854@bbox Link: http://lkml.kernel.org/r/20170307055551.GC29458@bboxSigned-off-by:
-
Shaohua Li authored
When memory pressure is high, we free MADV_FREE pages. If the pages are not dirty in pte, the pages could be freed immediately. Otherwise we can't reclaim them. We put the pages back to anonumous LRU list (by setting SwapBacked flag) and the pages will be reclaimed in normal swapout way. We use normal page reclaim policy. Since MADV_FREE pages are put into inactive file list, such pages and inactive file pages are reclaimed according to their age. This is expected, because we don't want to reclaim too many MADV_FREE pages before used once pages. Based on Minchan's original patch [minchan@kernel.org: clean up lazyfree page handling] Link: http://lkml.kernel.org/r/20170303025237.GB3503@bbox Link: http://lkml.kernel.org/r/14b8eb1d3f6bf6cc492833f183ac8c304e560484.1487965799.git.shli@fb.comSigned-off-by:
-
Shaohua Li authored
madv()'s MADV_FREE indicate pages are 'lazyfree'. They are still anonymous pages, but they can be freed without pageout. To distinguish these from normal anonymous pages, we clear their SwapBacked flag. MADV_FREE pages could be freed without pageout, so they pretty much like used once file pages. For such pages, we'd like to reclaim them once there is memory pressure. Also it might be unfair reclaiming MADV_FREE pages always before used once file pages and we definitively want to reclaim the pages before other anonymous and file pages. To speed up MADV_FREE pages reclaim, we put the pages into LRU_INACTIVE_FILE list. The rationale is LRU_INACTIVE_FILE list is tiny nowadays and should be full of used once file pages. Reclaiming MADV_FREE pages will not have much interfere of anonymous and active file pages. And the inactive file pages and MADV_FREE pages will be reclaimed according to their age, so we don't reclaim too many MADV_FREE pages too. Putting the MADV_FREE pages into LRU_INACTIVE_FILE_LIST also means we can reclaim the pages without swap support. This idea is suggested by Johannes. This patch doesn't move MADV_FREE pages to LRU_INACTIVE_FILE list yet to avoid bisect failure, next patch will do it. The patch is based on Minchan's original patch. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/2f87063c1e9354677b7618c647abde77b07561e5.1487965799.git.shli@fb.comSigned-off-by:
-
Shaohua Li authored
There are a few places the code assumes anonymous pages should have SwapBacked flag set. MADV_FREE pages are anonymous pages but we are going to add them to LRU_INACTIVE_FILE list and clear SwapBacked flag for them. The assumption doesn't hold any more, so fix them. Link: http://lkml.kernel.org/r/3945232c0df3dd6c4ef001976f35a95f18dcb407.1487965799.git.shli@fb.comSigned-off-by:
-
Shaohua Li authored
Patch series "mm: fix some MADV_FREE issues", v5. We are trying to use MADV_FREE in jemalloc. Several issues are found. Without solving the issues, jemalloc can't use the MADV_FREE feature. - Doesn't support system without swap enabled. Because if swap is off, we can't or can't efficiently age anonymous pages. And since MADV_FREE pages are mixed with other anonymous pages, we can't reclaim MADV_FREE pages. In current implementation, MADV_FREE will fallback to MADV_DONTNEED without swap enabled. But in our environment, a lot of machines don't enable swap. This will prevent our setup using MADV_FREE. - Increases memory pressure. page reclaim bias file pages reclaim against anonymous pages. This doesn't make sense for MADV_FREE pages, because those pages could be freed easily and refilled with very slight penality. Even page reclaim doesn't bias file pages, there is still an issue, because MADV_FREE pages and other anonymous pages are mixed together. To reclaim a MADV_FREE page, we probably must scan a lot of other anonymous pages, which is inefficient. In our test, we usually see oom with MADV_FREE enabled and nothing without it. - Accounting. There are two accounting problems. We don't have a global accounting. If the system is abnormal, we don't know if it's a problem from MADV_FREE side. The other problem is RSS accounting. MADV_FREE pages are accounted as normal anon pages and reclaimed lazily, so application's RSS becomes bigger. This confuses our workloads. We have monitoring daemon running and if it finds applications' RSS becomes abnormal, the daemon will kill the applications even kernel can reclaim the memory easily. To address the first the two issues, we can either put MADV_FREE pages into a separate LRU list (Minchan's previous patches and V1 patches), or put them into LRU_INACTIVE_FILE list (suggested by Johannes). The patchset use the second idea. The reason is LRU_INACTIVE_FILE list is tiny nowadays and should be full of used once file pages. So we can still efficiently reclaim MADV_FREE pages there without interference with other anon and active file pages. Putting the pages into inactive file list also has an advantage which allows page reclaim to prioritize MADV_FREE pages and used once file pages. MADV_FREE pages are put into the lru list and clear SwapBacked flag, so PageAnon(page) && !PageSwapBacked(page) will indicate a MADV_FREE pages. These pages will directly freed without pageout if they are clean, otherwise normal swap will reclaim them. For the third issue, the previous post adds global accounting and a separate RSS count for MADV_FREE pages. The problem is we never get accurate accounting for MADV_FREE pages. The pages are mapped to userspace, can be dirtied without notice from kernel side. To get accurate accounting, we could write protect the page, but then there is extra page fault overhead, which people don't want to pay. Jemalloc guys have concerns about the inaccurate accounting, so this post drops the accounting patches temporarily. The info exported to /proc/pid/smaps for MADV_FREE pages are kept, which is the only place we can get accurate accounting right now. This patch (of 6): Johannes pointed out TTU_LZFREE is unnecessary. It's true because we always have the flag set if we want to do an unmap. For cases we don't do an unmap, the TTU_LZFREE part of code should never run. Also the TTU_UNMAP is unnecessary. If no other flags set (for example, TTU_MIGRATION), an unmap is implied. The patch includes Johannes's cleanup and dead TTU_ACTION macro removal code Link: http://lkml.kernel.org/r/4be3ea1bc56b26fd98a54d0a6f70bec63f6d8980.1487965799.git.shli@fb.comSigned-off-by:
-
Geliang Tang authored
Use setup_deferrable_timer() instead of init_timer_deferrable() to simplify the code. Link: http://lkml.kernel.org/r/e8e3d4280a34facbc007346f31df833cec28801e.1488070291.git.geliangtang@gmail.comSigned-off-by:
Geliang Tang <geliangtang@gmail.com> Signed-off-by:
-
Johannes Weiner authored
The backoff mechanism is not needed. If we have MAX_RECLAIM_RETRIES loops without progress, we'll OOM anyway; backing off might cut one or two iterations off that in the rare OOM case. If we have intermittent success reclaiming a few pages, the backoff function gets reset also, and so is of little help in these scenarios. We might want a backoff function for when there IS progress, but not enough to be satisfactory. But this isn't that. Remove it. Link: http://lkml.kernel.org/r/20170228214007.5621-10-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
This reverts commit d7f05528. Now that reclaimability of a node is no longer based on the ratio between pages scanned and theoretically reclaimable pages, we can remove accounting tricks for pages skipped due to zone constraints. Link: http://lkml.kernel.org/r/20170228214007.5621-9-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
NR_PAGES_SCANNED counts number of pages scanned since the last page free event in the allocator. This was used primarily to measure the reclaimability of zones and nodes, and determine when reclaim should give up on them. In that role, it has been replaced in the preceding patches by a different mechanism. Being implemented as an efficient vmstat counter, it was automatically exported to userspace as well. It's however unlikely that anyone outside the kernel is using this counter in any meaningful way. Remove the counter and the unused pgdat_reclaimable(). Link: http://lkml.kernel.org/r/20170228214007.5621-8-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
Commit 246e87a9 ("memcg: fix get_scan_count() for small targets") sought to avoid high reclaim priorities for memcg by forcing it to scan a minimum amount of pages when lru_pages >> priority yielded nothing. This was done at a time when reclaim decisions like dirty throttling were tied to the priority level. Nowadays, the only meaningful thing still tied to priority dropping below DEF_PRIORITY - 2 is gating whether laptop_mode=1 is generally allowed to write. But that is from an era where direct reclaim was still allowed to call ->writepage, and kswapd nowadays avoids writes until it's scanned every clean page in the system. Potential changes to how quick sc->may_writepage could trigger are of little concern. Remove the force_scan stuff, as well as the ugly multi-pass target calculation that it necessitated. Link: http://lkml.kernel.org/r/20170228214007.5621-7-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
Commit 246e87a9 ("memcg: fix get_scan_count() for small targets") sought to avoid high reclaim priorities for kswapd by forcing it to scan a minimum amount of pages when lru_pages >> priority yielded nothing. Commit b95a2f2d ("mm: vmscan: convert global reclaim to per-memcg LRU lists"), due to switching global reclaim to a round-robin scheme over all cgroups, had to restrict this forceful behavior to unreclaimable zones in order to prevent massive overreclaim with many cgroups. The latter patch effectively neutered the behavior completely for all but extreme memory pressure. But in those situations we might as well drop the reclaimers to lower priority levels. Remove the check. Link: http://lkml.kernel.org/r/20170228214007.5621-6-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
NUMA balancing already checks the watermarks of the target node to decide whether it's a suitable balancing target. Whether the node is reclaimable or not is irrelevant when we don't intend to reclaim. Link: http://lkml.kernel.org/r/20170228214007.5621-5-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
Commit 1d82de61 ("mm, vmscan: make kswapd reclaim in terms of nodes") allowed laptop_mode=1 to start writing not just when the priority drops to DEF_PRIORITY - 2 but also when the node is unreclaimable. That appears to be a spurious change in this patch as I doubt the series was tested with laptop_mode, and neither is that particular change mentioned in the changelog. Remove it, it's still recent. Link: http://lkml.kernel.org/r/20170228214007.5621-4-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
PF_MEMALLOC direct reclaimers get throttled on a node when the sum of all free pages in each zone fall below half the min watermark. During the summation, we want to exclude zones that don't have reclaimables. Checking the same pgdat over and over again doesn't make sense. Fixes: 599d0c95 ("mm, vmscan: move LRU lists to node") Link: http://lkml.kernel.org/r/20170228214007.5621-3-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
Patch series "mm: kswapd spinning on unreclaimable nodes - fixes and cleanups". Jia reported a scenario in which the kswapd of a node indefinitely spins at 100% CPU usage. We have seen similar cases at Facebook. The kernel's current method of judging its ability to reclaim a node (or whether to back off and sleep) is based on the amount of scanned pages in proportion to the amount of reclaimable pages. In Jia's and our scenarios, there are no reclaimable pages in the node, however, and the condition for backing off is never met. Kswapd busyloops in an attempt to restore the watermarks while having nothing to work with. This series reworks the definition of an unreclaimable node based not on scanning but on whether kswapd is able to actually reclaim pages in MAX_RECLAIM_RETRIES (16) consecutive runs. This is the same criteria the page allocator uses for giving up on direct reclaim and invoking the OOM killer. If it cannot free any pages, kswapd will go to sleep and leave further attempts to direct reclaim invocations, which will either make progress and re-enable kswapd, or invoke the OOM killer. Patch #1 fixes the immediate problem Jia reported, the remainder are smaller fixlets, cleanups, and overall phasing out of the old method. Patch #6 is the odd one out. It's a nice cleanup to get_scan_count(), and directly related to #5, but in itself not relevant to the series. If the whole series is too ambitious for 4.11, I would consider the first three patches fixes, the rest cleanups. This patch (of 9): Jia He reports a problem with kswapd spinning at 100% CPU when requesting more hugepages than memory available in the system: $ echo 4000 >/proc/sys/vm/nr_hugepages top - 13:42:59 up 3:37, 1 user, load average: 1.09, 1.03, 1.01 Tasks: 1 total, 1 running, 0 sleeping, 0 stopped, 0 zombie %Cpu(s): 0.0 us, 12.5 sy, 0.0 ni, 85.5 id, 2.0 wa, 0.0 hi, 0.0 si, 0.0 st KiB Mem: 31371520 total, 30915136 used, 456384 free, 320 buffers KiB Swap: 6284224 total, 115712 used, 6168512 free. 48192 cached Mem PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 76 root 20 0 0 0 0 R 100.0 0.000 217:17.29 kswapd3 At that time, there are no reclaimable pages left in the node, but as kswapd fails to restore the high watermarks it refuses to go to sleep. Kswapd needs to back away from nodes that fail to balance. Up until commit 1d82de61 ("mm, vmscan: make kswapd reclaim in terms of nodes") kswapd had such a mechanism. It considered zones whose theoretically reclaimable pages it had reclaimed six times over as unreclaimable and backed away from them. This guard was erroneously removed as the patch changed the definition of a balanced node. However, simply restoring this code wouldn't help in the case reported here: there *are* no reclaimable pages that could be scanned until the threshold is met. Kswapd would stay awake anyway. Introduce a new and much simpler way of backing off. If kswapd runs through MAX_RECLAIM_RETRIES (16) cycles without reclaiming a single page, make it back off from the node. This is the same number of shots direct reclaim takes before declaring OOM. Kswapd will go to sleep on that node until a direct reclaimer manages to reclaim some pages, thus proving the node reclaimable again. [hannes@cmpxchg.org: check kswapd failure against the cumulative nr_reclaimed count] Link: http://lkml.kernel.org/r/20170306162410.GB2090@cmpxchg.org [shakeelb@google.com: fix condition for throttle_direct_reclaim] Link: http://lkml.kernel.org/r/20170314183228.20152-1-shakeelb@google.com Link: http://lkml.kernel.org/r/20170228214007.5621-2-hannes@cmpxchg.orgSigned-off-by:
Shakeel Butt <shakeelb@google.com> Reported-by:
Jia He <hejianet@gmail.com> Tested-by:
-
Greg Thelen authored
Each slab kmem cache has per cpu array caches. The array caches are created when the kmem_cache is created, either via kmem_cache_create() or lazily when the first object is allocated in context of a kmem enabled memcg. Array caches are replaced by writing to /proc/slabinfo. Array caches are protected by holding slab_mutex or disabling interrupts. Array cache allocation and replacement is done by __do_tune_cpucache() which holds slab_mutex and calls kick_all_cpus_sync() to interrupt all remote processors which confirms there are no references to the old array caches. IPIs are needed when replacing array caches. But when creating a new array cache, there's no need to send IPIs because there cannot be any references to the new cache. Outside of memcg kmem accounting these IPIs occur at boot time, so they're not a problem. But with memcg kmem accounting each container can create kmem caches, so the IPIs are wasteful. Avoid unnecessary IPIs when creating array caches. Test which reports the IPI count of allocating slab in 10000 memcg: import os def ipi_count(): with open("/proc/interrupts") as f: for l in f: if 'Function call interrupts' in l: return int(l.split()[1]) def echo(val, path): with open(path, "w") as f: f.write(val) n = 10000 os.chdir("/mnt/cgroup/memory") pid = str(os.getpid()) a = ipi_count() for i in range(n): os.mkdir(str(i)) echo("1G\n", "%d/memory.limit_in_bytes" % i) echo("1G\n", "%d/memory.kmem.limit_in_bytes" % i) echo(pid, "%d/cgroup.procs" % i) open("/tmp/x", "w").close() os.unlink("/tmp/x") b = ipi_count() print "%d loops: %d => %d (+%d ipis)" % (n, a, b, b-a) echo(pid, "cgroup.procs") for i in range(n): os.rmdir(str(i)) patched: 10000 loops: 1069 => 1170 (+101 ipis) unpatched: 10000 loops: 1192 => 48933 (+47741 ipis) Link: http://lkml.kernel.org/r/20170416214544.109476-1-gthelen@google.comSigned-off-by:Greg Thelen <gthelen@google.com> Acked-by:
Joonsoo Kim <iamjoonsoo.kim@lge.com> Acked-by:
-
