Skip to content

K
Kernel

openeuler / Kernel
接近 2 年 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 12 9月, 2022 5 次提交
  3. 30 7月, 2022 2 次提交
  5. 18 7月, 2022 9 次提交
    • U
      mm/page_alloc: use try_cmpxchg in set_pfnblock_flags_mask · 04ec0061
      由 Uros Bizjak 提交于 
      Use try_cmpxchg instead of cmpxchg in set_pfnblock_flags_mask.  x86
CMPXCHG instruction returns success in ZF flag, so this change saves a
compare after cmpxchg (and related move instruction in front of cmpxchg). 
The main loop improves from:

    1c5d:	48 89 c2             	mov    %rax,%rdx
    1c60:	48 89 c1             	mov    %rax,%rcx
    1c63:	48 21 fa             	and    %rdi,%rdx
    1c66:	4c 09 c2             	or     %r8,%rdx
    1c69:	f0 48 0f b1 16       	lock cmpxchg %rdx,(%rsi)
    1c6e:	48 39 c1             	cmp    %rax,%rcx
    1c71:	75 ea                	jne    1c5d <...>

to:

    1c60:	48 89 ca             	mov    %rcx,%rdx
    1c63:	48 21 c2             	and    %rax,%rdx
    1c66:	4c 09 c2             	or     %r8,%rdx
    1c69:	f0 48 0f b1 16       	lock cmpxchg %rdx,(%rsi)
    1c6e:	75 f0                	jne    1c60 <...>

Link: https://lkml.kernel.org/r/20220708140736.8737-1-ubizjak@gmail.comSigned-off-by: NUros Bizjak <ubizjak@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      04ec0061
    • G
      mm, hugetlb: skip irrelevant nodes in show_free_areas() · dcadcf1c
      由 Gang Li 提交于 
      show_free_areas() allows to filter out node specific data which is
irrelevant to the allocation request.  But hugetlb_show_meminfo() still
shows hugetlb on all nodes, which is redundant and unnecessary.

Use show_mem_node_skip() to skip irrelevant nodes.  And replace
hugetlb_show_meminfo() with hugetlb_show_meminfo_node(nid).

before-and-after sample output of OOM:

before:
```
[  214.362453] Node 1 active_anon:148kB inactive_anon:4050920kB active_file:112kB inactive_file:100kB
[  214.375429] Node 1 Normal free:45100kB boost:0kB min:45576kB low:56968kB high:68360kB reserved_hig
[  214.388334] lowmem_reserve[]: 0 0 0 0 0
[  214.390251] Node 1 Normal: 423*4kB (UE) 320*8kB (UME) 187*16kB (UE) 117*32kB (UE) 57*64kB (UME) 20
[  214.397626] Node 0 hugepages_total=0 hugepages_free=0 hugepages_surp=0 hugepages_size=2048kB
[  214.401518] Node 1 hugepages_total=0 hugepages_free=0 hugepages_surp=0 hugepages_size=2048kB
```

after:
```
[  145.069705] Node 1 active_anon:128kB inactive_anon:4049412kB active_file:56kB inactive_file:84kB u
[  145.110319] Node 1 Normal free:45424kB boost:0kB min:45576kB low:56968kB high:68360kB reserved_hig
[  145.152315] lowmem_reserve[]: 0 0 0 0 0
[  145.155244] Node 1 Normal: 470*4kB (UME) 373*8kB (UME) 247*16kB (UME) 168*32kB (UE) 86*64kB (UME)
[  145.164119] Node 1 hugepages_total=0 hugepages_free=0 hugepages_surp=0 hugepages_size=2048kB
```

Link: https://lkml.kernel.org/r/20220706034655.1834-1-ligang.bdlg@bytedance.comSigned-off-by: NGang Li <ligang.bdlg@bytedance.com>
Reviewed-by: NMike Kravetz <mike.kravetz@oracle.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      dcadcf1c
    • M
      mm/page_alloc: replace local_lock with normal spinlock · 01b44456
      由 Mel Gorman 提交于 
      struct per_cpu_pages is no longer strictly local as PCP lists can be
drained remotely using a lock for protection.  While the use of local_lock
works, it goes against the intent of local_lock which is for "pure CPU
local concurrency control mechanisms and not suited for inter-CPU
concurrency control" (Documentation/locking/locktypes.rst)

local_lock protects against migration between when the percpu pointer is
accessed and the pcp->lock acquired.  The lock acquisition is a preemption
point so in the worst case, a task could migrate to another NUMA node and
accidentally allocate remote memory.  The main requirement is to pin the
task to a CPU that is suitable for PREEMPT_RT and !PREEMPT_RT.

Replace local_lock with helpers that pin a task to a CPU, lookup the
per-cpu structure and acquire the embedded lock.  It's similar to
local_lock without breaking the intent behind the API.  It is not a
complete API as only the parts needed for PCP-alloc are implemented but in
theory, the generic helpers could be promoted to a general API if there
was demand for an embedded lock within a per-cpu struct with a guarantee
that the per-cpu structure locked matches the running CPU and cannot use
get_cpu_var due to RT concerns.  PCP requires these semantics to avoid
accidentally allocating remote memory.

[mgorman@techsingularity.net: use pcp_spin_trylock_irqsave instead of pcpu_spin_trylock_irqsave]
  Link: https://lkml.kernel.org/r/20220627084645.GA27531@techsingularity.net
Link: https://lkml.kernel.org/r/20220624125423.6126-8-mgorman@techsingularity.netSigned-off-by: NMel Gorman <mgorman@techsingularity.net>
Tested-by: NYu Zhao <yuzhao@google.com>
Reviewed-by: NNicolas Saenz Julienne <nsaenzju@redhat.com>
Tested-by: NNicolas Saenz Julienne <nsaenzju@redhat.com>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Tested-by: NYu Zhao <yuzhao@google.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      01b44456
    • N
      mm/page_alloc: remotely drain per-cpu lists · 443c2acc
      由 Nicolas Saenz Julienne 提交于 
      Some setups, notably NOHZ_FULL CPUs, are too busy to handle the per-cpu
drain work queued by __drain_all_pages().  So introduce a new mechanism to
remotely drain the per-cpu lists.  It is made possible by remotely locking
'struct per_cpu_pages' new per-cpu spinlocks.  A benefit of this new
scheme is that drain operations are now migration safe.

There was no observed performance degradation vs.  the previous scheme. 
Both netperf and hackbench were run in parallel to triggering the
__drain_all_pages(NULL, true) code path around ~100 times per second.  The
new scheme performs a bit better (~5%), although the important point here
is there are no performance regressions vs.  the previous mechanism. 
Per-cpu lists draining happens only in slow paths.

Minchan Kim tested an earlier version and reported;

	My workload is not NOHZ CPUs but run apps under heavy memory
	pressure so they goes to direct reclaim and be stuck on
	drain_all_pages until work on workqueue run.

	unit: nanosecond
	max(dur)        avg(dur)                count(dur)
	166713013       487511.77786438033      1283

	From traces, system encountered the drain_all_pages 1283 times and
	worst case was 166ms and avg was 487us.

	The other problem was alloc_contig_range in CMA. The PCP draining
	takes several hundred millisecond sometimes though there is no
	memory pressure or a few of pages to be migrated out but CPU were
	fully booked.

	Your patch perfectly removed those wasted time.

Link: https://lkml.kernel.org/r/20220624125423.6126-7-mgorman@techsingularity.netSigned-off-by: NNicolas Saenz Julienne <nsaenzju@redhat.com>
Signed-off-by: NMel Gorman <mgorman@techsingularity.net>
Tested-by: NYu Zhao <yuzhao@google.com>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      443c2acc
    • M
      mm/page_alloc: protect PCP lists with a spinlock · 4b23a68f
      由 Mel Gorman 提交于 
      Currently the PCP lists are protected by using local_lock_irqsave to
prevent migration and IRQ reentrancy but this is inconvenient.  Remote
draining of the lists is impossible and a workqueue is required and every
task allocation/free must disable then enable interrupts which is
expensive.

As preparation for dealing with both of those problems, protect the
lists with a spinlock.  The IRQ-unsafe version of the lock is used
because IRQs are already disabled by local_lock_irqsave.  spin_trylock
is used in combination with local_lock_irqsave() but later will be
replaced with a spin_trylock_irqsave when the local_lock is removed.

The per_cpu_pages still fits within the same number of cache lines after
this patch relative to before the series.

struct per_cpu_pages {
        spinlock_t                 lock;                 /*     0     4 */
        int                        count;                /*     4     4 */
        int                        high;                 /*     8     4 */
        int                        batch;                /*    12     4 */
        short int                  free_factor;          /*    16     2 */
        short int                  expire;               /*    18     2 */

        /* XXX 4 bytes hole, try to pack */

        struct list_head           lists[13];            /*    24   208 */

        /* size: 256, cachelines: 4, members: 7 */
        /* sum members: 228, holes: 1, sum holes: 4 */
        /* padding: 24 */
} __attribute__((__aligned__(64)));

There is overhead in the fast path due to acquiring the spinlock even
though the spinlock is per-cpu and uncontended in the common case.  Page
Fault Test (PFT) running on a 1-socket reported the following results on a
1 socket machine.

                                     5.19.0-rc3               5.19.0-rc3
                                        vanilla      mm-pcpspinirq-v5r16
Hmean     faults/sec-1   869275.7381 (   0.00%)   874597.5167 *   0.61%*
Hmean     faults/sec-3  2370266.6681 (   0.00%)  2379802.0362 *   0.40%*
Hmean     faults/sec-5  2701099.7019 (   0.00%)  2664889.7003 *  -1.34%*
Hmean     faults/sec-7  3517170.9157 (   0.00%)  3491122.8242 *  -0.74%*
Hmean     faults/sec-8  3965729.6187 (   0.00%)  3939727.0243 *  -0.66%*

There is a small hit in the number of faults per second but given that the
results are more stable, it's borderline noise.

[akpm@linux-foundation.org: add missing local_unlock_irqrestore() on contention path]
Link: https://lkml.kernel.org/r/20220624125423.6126-6-mgorman@techsingularity.netSigned-off-by: NMel Gorman <mgorman@techsingularity.net>
Tested-by: NYu Zhao <yuzhao@google.com>
Reviewed-by: NNicolas Saenz Julienne <nsaenzju@redhat.com>
Tested-by: NNicolas Saenz Julienne <nsaenzju@redhat.com>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      4b23a68f
    • M
      mm/page_alloc: remove mistaken page == NULL check in rmqueue · e2a66c21
      由 Mel Gorman 提交于 
      If a page allocation fails, the ZONE_BOOSTER_WATERMARK should be tested,
cleared and kswapd woken whether the allocation attempt was via the PCP or
directly via the buddy list.

Remove the page == NULL so the ZONE_BOOSTED_WATERMARK bit is checked
unconditionally.  As it is unlikely that ZONE_BOOSTED_WATERMARK is set,
mark the branch accordingly.

Link: https://lkml.kernel.org/r/20220624125423.6126-5-mgorman@techsingularity.netSigned-off-by: NMel Gorman <mgorman@techsingularity.net>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Tested-by: NYu Zhao <yuzhao@google.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nicolas Saenz Julienne <nsaenzju@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      e2a66c21
    • M
      mm/page_alloc: split out buddy removal code from rmqueue into separate helper · 589d9973
      由 Mel Gorman 提交于 
      This is a preparation page to allow the buddy removal code to be reused in
a later patch.

No functional change.

Link: https://lkml.kernel.org/r/20220624125423.6126-4-mgorman@techsingularity.netSigned-off-by: NMel Gorman <mgorman@techsingularity.net>
Tested-by: NMinchan Kim <minchan@kernel.org>
Acked-by: NMinchan Kim <minchan@kernel.org>
Reviewed-by: NNicolas Saenz Julienne <nsaenzju@redhat.com>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Tested-by: NYu Zhao <yuzhao@google.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      589d9973
    • M
      mm/page_alloc: use only one PCP list for THP-sized allocations · 5d0a661d
      由 Mel Gorman 提交于 
      The per_cpu_pages is cache-aligned on a standard x86-64 distribution
configuration but a later patch will add a new field which would push the
structure into the next cache line.  Use only one list to store THP-sized
pages on the per-cpu list.  This assumes that the vast majority of
THP-sized allocations are GFP_MOVABLE but even if it was another type, it
would not contribute to serious fragmentation that potentially causes a
later THP allocation failure.  Align per_cpu_pages on the cacheline
boundary to ensure there is no false cache sharing.

After this patch, the structure sizing is;

struct per_cpu_pages {
        int                        count;                /*     0     4 */
        int                        high;                 /*     4     4 */
        int                        batch;                /*     8     4 */
        short int                  free_factor;          /*    12     2 */
        short int                  expire;               /*    14     2 */
        struct list_head           lists[13];            /*    16   208 */

        /* size: 256, cachelines: 4, members: 6 */
        /* padding: 32 */
} __attribute__((__aligned__(64)));

Link: https://lkml.kernel.org/r/20220624125423.6126-3-mgorman@techsingularity.netSigned-off-by: NMel Gorman <mgorman@techsingularity.net>
Tested-by: NMinchan Kim <minchan@kernel.org>
Acked-by: NMinchan Kim <minchan@kernel.org>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Tested-by: NYu Zhao <yuzhao@google.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Nicolas Saenz Julienne <nsaenzju@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      5d0a661d
    • M
      mm/page_alloc: add page->buddy_list and page->pcp_list · bf75f200
      由 Mel Gorman 提交于 
      Patch series "Drain remote per-cpu directly", v5.

Some setups, notably NOHZ_FULL CPUs, may be running realtime or
latency-sensitive applications that cannot tolerate interference due to
per-cpu drain work queued by __drain_all_pages().  Introduce a new
mechanism to remotely drain the per-cpu lists.  It is made possible by
remotely locking 'struct per_cpu_pages' new per-cpu spinlocks.  This has
two advantages, the time to drain is more predictable and other unrelated
tasks are not interrupted.

This series has the same intent as Nicolas' series "mm/page_alloc: Remote
per-cpu lists drain support" -- avoid interference of a high priority task
due to a workqueue item draining per-cpu page lists.  While many workloads
can tolerate a brief interruption, it may cause a real-time task running
on a NOHZ_FULL CPU to miss a deadline and at minimum, the draining is
non-deterministic.

Currently an IRQ-safe local_lock protects the page allocator per-cpu
lists.  The local_lock on its own prevents migration and the IRQ disabling
protects from corruption due to an interrupt arriving while a page
allocation is in progress.

This series adjusts the locking.  A spinlock is added to struct
per_cpu_pages to protect the list contents while local_lock_irq is
ultimately replaced by just the spinlock in the final patch.  This allows
a remote CPU to safely.  Follow-on work should allow the spin_lock_irqsave
to be converted to spin_lock to avoid IRQs being disabled/enabled in most
cases.  The follow-on patch will be one kernel release later as it is
relatively high risk and it'll make bisections more clear if there are any
problems.

Patch 1 is a cosmetic patch to clarify when page->lru is storing buddy pages
	and when it is storing per-cpu pages.

Patch 2 shrinks per_cpu_pages to make room for a spin lock. Strictly speaking
	this is not necessary but it avoids per_cpu_pages consuming another
	cache line.

Patch 3 is a preparation patch to avoid code duplication.

Patch 4 is a minor correction.

Patch 5 uses a spin_lock to protect the per_cpu_pages contents while still
	relying on local_lock to prevent migration, stabilise the pcp
	lookup and prevent IRQ reentrancy.

Patch 6 remote drains per-cpu pages directly instead of using a workqueue.

Patch 7 uses a normal spinlock instead of local_lock for remote draining


This patch (of 7):

The page allocator uses page->lru for storing pages on either buddy or PCP
lists.  Create page->buddy_list and page->pcp_list as a union with
page->lru.  This is simply to clarify what type of list a page is on in
the page allocator.

No functional change intended.

[minchan@kernel.org: fix page lru fields in macros]
Link: https://lkml.kernel.org/r/20220624125423.6126-2-mgorman@techsingularity.netSigned-off-by: NMel Gorman <mgorman@techsingularity.net>
Tested-by: NMinchan Kim <minchan@kernel.org>
Acked-by: NMinchan Kim <minchan@kernel.org>
Reviewed-by: NNicolas Saenz Julienne <nsaenzju@redhat.com>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Tested-by: NYu Zhao <yuzhao@google.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      bf75f200
  7. 07 7月, 2022 2 次提交
  9. 04 7月, 2022 4 次提交
  11. 17 6月, 2022 1 次提交
  13. 15 6月, 2022 1 次提交
  15. 28 5月, 2022 2 次提交
  17. 27 5月, 2022 1 次提交
    • M
      mm/page_alloc: always attempt to allocate at least one page during bulk allocation · c572e488
      由 Mel Gorman 提交于 
      Peter Pavlisko reported the following problem on kernel bugzilla 216007.

	When I try to extract an uncompressed tar archive (2.6 milion
	files, 760.3 GiB in size) on newly created (empty) XFS file system,
	after first low tens of gigabytes extracted the process hangs in
	iowait indefinitely. One CPU core is 100% occupied with iowait,
	the other CPU core is idle (on 2-core Intel Celeron G1610T).

It was bisected to c9fa5630 ("xfs: use alloc_pages_bulk_array() for
buffers") but XFS is only the messenger.  The problem is that nothing is
waking kswapd to reclaim some pages at a time the PCP lists cannot be
refilled until some reclaim happens.  The bulk allocator checks that there
are some pages in the array and the original intent was that a bulk
allocator did not necessarily need all the requested pages and it was best
to return as quickly as possible.

This was fine for the first user of the API but both NFS and XFS require
the requested number of pages be available before making progress.  Both
could be adjusted to call the page allocator directly if a bulk allocation
fails but it puts a burden on users of the API.  Adjust the semantics to
attempt at least one allocation via __alloc_pages() before returning so
kswapd is woken if necessary.

It was reported via bugzilla that the patch addressed the problem and that
the tar extraction completed successfully.  This may also address bug
215975 but has yet to be confirmed.

BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=216007
BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=215975
Link: https://lkml.kernel.org/r/20220526091210.GC3441@techsingularity.net
Fixes: 387ba26f ("mm/page_alloc: add a bulk page allocator")
Signed-off-by: NMel Gorman <mgorman@techsingularity.net>
Cc: "Darrick J. Wong" <djwong@kernel.org>
Cc: Dave Chinner <dchinner@redhat.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Chuck Lever <chuck.lever@oracle.com>
Cc: <stable@vger.kernel.org>	[5.13+]
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      c572e488
  19. 26 5月, 2022 1 次提交
    • Z
      mm: fix a potential infinite loop in start_isolate_page_range() · 88ee1343
      由 Zi Yan 提交于 
      In isolate_single_pageblock() called by start_isolate_page_range(), there
are some pageblock isolation issues causing a potential infinite loop when
isolating a page range.  This is reported by Qian Cai.

1. the pageblock was isolated by just changing pageblock migratetype
   without checking unmovable pages. Calling set_migratetype_isolate() to
   isolate pageblock properly.
2. an off-by-one error caused migrating pages unnecessarily, since the page
   is not crossing pageblock boundary.
3. migrating a compound page across pageblock boundary then splitting the
   free page later has a small race window that the free page might be
   allocated again, so that the code will try again, causing an potential
   infinite loop. Temporarily set the to-be-migrated page's pageblock to
   MIGRATE_ISOLATE to prevent that and bail out early if no free page is
   found after page migration.

An additional fix to split_free_page() aims to avoid crashing in
__free_one_page().  When the free page is split at the specified
split_pfn_offset, free_page_order should check both the first bit of
free_page_pfn and the last bit of split_pfn_offset and use the smaller
one.  For example, if free_page_pfn=0x10000, split_pfn_offset=0xc000,
free_page_order should first be 0x8000 then 0x4000, instead of 0x4000 then
0x8000, which the original algorithm did.

[akpm@linux-foundation.org: suppress min() warning]
Link: https://lkml.kernel.org/r/20220524194756.1698351-1-zi.yan@sent.com
Fixes: b2c9e2fb ("mm: make alloc_contig_range work at pageblock granularity")
Signed-off-by: NZi Yan <ziy@nvidia.com>
Reported-by: NQian Cai <quic_qiancai@quicinc.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Hildenbrand <david@redhat.com>
Cc: Eric Ren <renzhengeek@gmail.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
      88ee1343
  21. 20 5月, 2022 2 次提交
  23. 13 5月, 2022 6 次提交
  25. 10 5月, 2022 1 次提交
  27. 30 4月, 2022 1 次提交
  29. 29 4月, 2022 2 次提交