NUMA statistics are maintained on the zone level for hits, misses, foreign
etc but nothing relies on them being perfectly accurate for functional
correctness. The counters are used by userspace to get a general overview
of a workloads NUMA behaviour but the page allocator incurs a high cost to
maintain perfect accuracy similar to what is required for a vmstat like
NR_FREE_PAGES. There even is a sysctl vm.numa_stat to allow userspace to
turn off the collection of NUMA statistics like NUMA_HIT.
This patch converts NUMA_HIT and friends to be NUMA events with similar
accuracy to VM events. There is a possibility that slight errors will be
introduced but the overall trend as seen by userspace will be similar.
The counters are no longer updated from vmstat_refresh context as it is
unnecessary overhead for counters that may never be read by userspace.
Note that counters could be maintained at the node level to save space but
it would have a user-visible impact due to /proc/zoneinfo.
[lkp@intel.com: Fix misplaced closing brace for !CONFIG_NUMA]
Link: https://lkml.kernel.org/r/20210512095458.30632-4-mgorman@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Chuck Lever <chuck.lever@oracle.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The PCP (per-cpu page allocator in page_alloc.c) shares locking
requirements with vmstat and the zone lock which is inconvenient and
causes some issues. For example, the PCP list and vmstat share the same
per-cpu space meaning that it's possible that vmstat updates dirty cache
lines holding per-cpu lists across CPUs unless padding is used. Second,
PREEMPT_RT does not want to disable IRQs for too long in the page
allocator.
This series splits the locking requirements and uses locks types more
suitable for PREEMPT_RT, reduces the time when special locking is required
for stats and reduces the time when IRQs need to be disabled on
!PREEMPT_RT kernels.
Why local_lock? PREEMPT_RT considers the following sequence to be unsafe
as documented in Documentation/locking/locktypes.rst
local_irq_disable();
spin_lock(&lock);
The pcp allocator has this sequence for rmqueue_pcplist (local_irq_save)
-> __rmqueue_pcplist -> rmqueue_bulk (spin_lock). While it's possible to
separate this out, it generally means there are points where we enable
IRQs and reenable them again immediately. To prevent a migration and the
per-cpu pointer going stale, migrate_disable is also needed. That is a
custom lock that is similar, but worse, than local_lock. Furthermore, on
PREEMPT_RT, it's undesirable to leave IRQs disabled for too long. By
converting to local_lock which disables migration on PREEMPT_RT, the
locking requirements can be separated and start moving the protections for
PCP, stats and the zone lock to PREEMPT_RT-safe equivalent locking. As a
bonus, local_lock also means that PROVE_LOCKING does something useful.
After that, it's obvious that zone_statistics incurs too much overhead and
leaves IRQs disabled for longer than necessary on !PREEMPT_RT kernels.
zone_statistics uses perfectly accurate counters requiring IRQs be
disabled for parallel RMW sequences when inaccurate ones like vm_events
would do. The series makes the NUMA statistics (NUMA_HIT and friends)
inaccurate counters that then require no special protection on
!PREEMPT_RT.
The bulk page allocator can then do stat updates in bulk with IRQs enabled
which should improve the efficiency. Technically, this could have been
done without the local_lock and vmstat conversion work and the order
simply reflects the timing of when different series were implemented.
Finally, there are places where we conflate IRQs being disabled for the
PCP with the IRQ-safe zone spinlock. The remainder of the series reduces
the scope of what is protected by disabled IRQs on !PREEMPT_RT kernels.
By the end of the series, page_alloc.c does not call local_irq_save so the
locking scope is a bit clearer. The one exception is that modifying
NR_FREE_PAGES still happens in places where it's known the IRQs are
disabled as it's harmless for PREEMPT_RT and would be expensive to split
the locking there.
No performance data is included because despite the overhead of the stats,
it's within the noise for most workloads on !PREEMPT_RT. However, Jesper
Dangaard Brouer ran a page allocation microbenchmark on a E5-1650 v4 @
3.60GHz CPU on the first version of this series. Focusing on the array
variant of the bulk page allocator reveals the following.
(CPU: Intel(R) Xeon(R) CPU E5-1650 v4 @ 3.60GHz)
ARRAY variant: time_bulk_page_alloc_free_array: step=bulk size
Baseline Patched
1 56.383 54.225 (+3.83%)
2 40.047 35.492 (+11.38%)
3 37.339 32.643 (+12.58%)
4 35.578 30.992 (+12.89%)
8 33.592 29.606 (+11.87%)
16 32.362 28.532 (+11.85%)
32 31.476 27.728 (+11.91%)
64 30.633 27.252 (+11.04%)
128 30.596 27.090 (+11.46%)
While this is a positive outcome, the series is more likely to be
interesting to the RT people in terms of getting parts of the PREEMPT_RT
tree into mainline.
This patch (of 9):
The per-cpu page allocator lists and the per-cpu vmstat deltas are stored
in the same struct per_cpu_pages even though vmstats have no direct impact
on the per-cpu page lists. This is inconsistent because the vmstats for a
node are stored on a dedicated structure. The bigger issue is that the
per_cpu_pages structure is not cache-aligned and stat updates either cache
conflict with adjacent per-cpu lists incurring a runtime cost or padding
is required incurring a memory cost.
This patch splits the per-cpu pagelists and the vmstat deltas into
separate structures. It's mostly a mechanical conversion but some
variable renaming is done to clearly distinguish the per-cpu pages
structure (pcp) from the vmstats (pzstats).
Superficially, this appears to increase the size of the per_cpu_pages
structure but the movement of expire fills a structure hole so there is no
impact overall.
[mgorman@techsingularity.net: make it W=1 cleaner]
Link: https://lkml.kernel.org/r/20210514144622.GA3735@techsingularity.net
[mgorman@techsingularity.net: make it W=1 even cleaner]
Link: https://lkml.kernel.org/r/20210516140705.GB3735@techsingularity.net
[lkp@intel.com: check struct per_cpu_zonestat has a non-zero size]
[vbabka@suse.cz: Init zone->per_cpu_zonestats properly]
Link: https://lkml.kernel.org/r/20210512095458.30632-1-mgorman@techsingularity.net
Link: https://lkml.kernel.org/r/20210512095458.30632-2-mgorman@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Chuck Lever <chuck.lever@oracle.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
A customer experienced a low-memory situation and decided to issue a
SIGKILL (i.e. a fatal signal). Instead of promptly terminating as one
would expect, the aforementioned task remained unresponsive.
Further investigation indicated that the task was "stuck" in the
reclaim/compaction retry loop. Now, it does not make sense to retry
compaction when a fatal signal is pending.
In the context of try_to_compact_pages(), indeed COMPACT_SKIPPED can be
returned; albeit, not every zone, on the zone list, would be considered in
the case a fatal signal is found to be pending. Yet, in
should_compact_retry(), given the last known compaction result, each zone,
on the zone list, can be considered/or checked (see
compaction_zonelist_suitable()). For example, if a zone was found to
succeed, then reclaim/compaction would be tried again (notwithstanding the
above).
This patch ensures that compaction is not needlessly retried irrespective
of the last known compaction result e.g. if it was skipped, in the
unlikely case a fatal signal is found pending. So, OOM is at least
attempted.
Link: https://lkml.kernel.org/r/20210520142901.3371299-1-atomlin@redhat.com
Signed-off-by: Aaron Tomlin <atomlin@redhat.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "Constify struct page arguments".
While working on various solutions to the 32-bit struct page size
regression, one of the problems I found was the networking stack expects
to be able to pass const struct page pointers around, and the mm doesn't
provide a lot of const-friendly functions to call. The root tangle of
problems is that a lot of functions call VM_BUG_ON_PAGE(), which calls
dump_page(), which calls a lot of functions which don't take a const
struct page (but could be const).
This patch (of 6):
The only caller of __dump_page() now opencodes dump_page(), so remove it
as an externally visible symbol.
Link: https://lkml.kernel.org/r/20210416231531.2521383-1-willy@infradead.org
Link: https://lkml.kernel.org/r/20210416231531.2521383-2-willy@infradead.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Anshuman Khandual <anshuman.khandual@arm.com>
Reviewed-by: William Kucharski <william.kucharski@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Recently there has been introduced a page bulk allocator for users which
need to get number of pages per one call request.
For order-0 pages switch to an alloc_pages_bulk_array_node() instead of
alloc_pages_node(), the reason is the former is not capable of allocating
set of pages, thus a one call is per one page.
Second, according to my tests the bulk allocator uses less cycles even for
scenarios when only one page is requested. Running the "perf" on same
test case shows below difference:
<default>
- 45.18% __vmalloc_node
- __vmalloc_node_range
- 35.60% __alloc_pages
- get_page_from_freelist
3.36% __list_del_entry_valid
3.00% check_preemption_disabled
1.42% prep_new_page
<default>
<patch>
- 31.00% __vmalloc_node
- __vmalloc_node_range
- 14.48% __alloc_pages_bulk
3.22% __list_del_entry_valid
- 0.83% __alloc_pages
get_page_from_freelist
<patch>
The "test_vmalloc.sh" also shows performance improvements:
fix_size_alloc_test_4MB loops: 1000000 avg: 89105095 usec
fix_size_alloc_test loops: 1000000 avg: 513672 usec
full_fit_alloc_test loops: 1000000 avg: 748900 usec
long_busy_list_alloc_test loops: 1000000 avg: 8043038 usec
random_size_alloc_test loops: 1000000 avg: 4028582 usec
fix_align_alloc_test loops: 1000000 avg: 1457671 usec
fix_size_alloc_test_4MB loops: 1000000 avg: 62083711 usec
fix_size_alloc_test loops: 1000000 avg: 449207 usec
full_fit_alloc_test loops: 1000000 avg: 735985 usec
long_busy_list_alloc_test loops: 1000000 avg: 5176052 usec
random_size_alloc_test loops: 1000000 avg: 2589252 usec
fix_align_alloc_test loops: 1000000 avg: 1365009 usec
For example 4MB allocations illustrates ~30% gain, all the
rest is also better.
Link: https://lkml.kernel.org/r/20210516202056.2120-3-urezki@gmail.com
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Oleksiy Avramchenko <oleksiy.avramchenko@sonymobile.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
set_active_memcg() worked for kernel allocations but was silently ignored
for user pages.
This patch establishes a precedence order for who gets charged:
1. If there is a memcg associated with the page already, that memcg is
charged. This happens during swapin.
2. If an explicit mm is passed, mm->memcg is charged. This happens
during page faults, which can be triggered in remote VMs (eg gup).
3. Otherwise consult the current process context. If there is an
active_memcg, use that. Otherwise, current->mm->memcg.
Previously, if a NULL mm was passed to mem_cgroup_charge (case 3) it would
always charge the root cgroup. Now it looks up the active_memcg first
(falling back to charging the root cgroup if not set).
Link: https://lkml.kernel.org/r/20210610173944.1203706-3-schatzberg.dan@gmail.com
Signed-off-by: Dan Schatzberg <schatzberg.dan@gmail.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Tejun Heo <tj@kernel.org>
Acked-by: Chris Down <chris@chrisdown.name>
Acked-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Reviewed-by: Michal Koutný <mkoutny@suse.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The obj_cgroup_release() and memcg_reparent_objcgs() are serialized by the
css_set_lock. We do not need to care about objcg->memcg being released in
the process of obj_cgroup_release(). So there is no need to pin memcg
before releasing objcg. Remove those pinning logic to simplfy the code.
There are only two places that modifies the objcg->memcg. One is the
initialization to objcg->memcg in the memcg_online_kmem(), another is
objcgs reparenting in the memcg_reparent_objcgs(). It is also impossible
for the two to run in parallel. So xchg() is unnecessary and it is enough
to use WRITE_ONCE().
Link: https://lkml.kernel.org/r/20210417043538.9793-7-songmuchun@bytedance.com
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Roman Gushchin <guro@fb.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Xiongchun Duan <duanxiongchun@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The below scenario can cause the page counters of the root_mem_cgroup to
be out of balance.
CPU0: CPU1:
objcg = get_obj_cgroup_from_current()
obj_cgroup_charge_pages(objcg)
memcg_reparent_objcgs()
// reparent to root_mem_cgroup
WRITE_ONCE(iter->memcg, parent)
// memcg == root_mem_cgroup
memcg = get_mem_cgroup_from_objcg(objcg)
// do not charge to the root_mem_cgroup
try_charge(memcg)
obj_cgroup_uncharge_pages(objcg)
memcg = get_mem_cgroup_from_objcg(objcg)
// uncharge from the root_mem_cgroup
refill_stock(memcg)
drain_stock(memcg)
page_counter_uncharge(&memcg->memory)
get_obj_cgroup_from_current() never returns a root_mem_cgroup's objcg, so
we never explicitly charge the root_mem_cgroup. And it's not going to
change. It's all about a race when we got an obj_cgroup pointing at some
non-root memcg, but before we were able to charge it, the cgroup was gone,
objcg was reparented to the root and so we're skipping the charging. Then
we store the objcg pointer and later use to uncharge the root_mem_cgroup.
This can cause the page counter to be less than the actual value.
Although we do not display the value (mem_cgroup_usage) so there shouldn't
be any actual problem, but there is a WARN_ON_ONCE in the
page_counter_cancel(). Who knows if it will trigger? So it is better to
fix it.
Link: https://lkml.kernel.org/r/20210425075410.19255-1-songmuchun@bytedance.com
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Acked-by: Roman Gushchin <guro@fb.com>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Cc: Xiongchun Duan <duanxiongchun@bytedance.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There are currently two problems in the way the objcg pointer array
(memcg_data) in the page structure is being allocated and freed.
On its allocation, it is possible that the allocated objcg pointer
array comes from the same slab that requires memory accounting. If this
happens, the slab will never become empty again as there is at least
one object left (the obj_cgroup array) in the slab.
When it is freed, the objcg pointer array object may be the last one
in its slab and hence causes kfree() to be called again. With the
right workload, the slab cache may be set up in a way that allows the
recursive kfree() calling loop to nest deep enough to cause a kernel
stack overflow and panic the system.
One way to solve this problem is to split the kmalloc-<n> caches
(KMALLOC_NORMAL) into two separate sets - a new set of kmalloc-<n>
(KMALLOC_NORMAL) caches for unaccounted objects only and a new set of
kmalloc-cg-<n> (KMALLOC_CGROUP) caches for accounted objects only. All
the other caches can still allow a mix of accounted and unaccounted
objects.
With this change, all the objcg pointer array objects will come from
KMALLOC_NORMAL caches which won't have their objcg pointer arrays. So
both the recursive kfree() problem and non-freeable slab problem are
gone.
Since both the KMALLOC_NORMAL and KMALLOC_CGROUP caches no longer have
mixed accounted and unaccounted objects, this will slightly reduce the
number of objcg pointer arrays that need to be allocated and save a bit
of memory. On the other hand, creating a new set of kmalloc caches does
have the effect of reducing cache utilization. So it is properly a wash.
The new KMALLOC_CGROUP is added between KMALLOC_NORMAL and
KMALLOC_RECLAIM so that the first for loop in create_kmalloc_caches()
will include the newly added caches without change.
[vbabka@suse.cz: don't create kmalloc-cg caches with cgroup.memory=nokmem]
Link: https://lkml.kernel.org/r/20210512145107.6208-1-longman@redhat.com
[akpm@linux-foundation.org: un-fat-finger v5 delta creation]
[longman@redhat.com: disable cache merging for KMALLOC_NORMAL caches]
Link: https://lkml.kernel.org/r/20210505200610.13943-4-longman@redhat.com
Link: https://lkml.kernel.org/r/20210512145107.6208-1-longman@redhat.com
Link: https://lkml.kernel.org/r/20210505200610.13943-3-longman@redhat.com
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Suggested-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Roman Gushchin <guro@fb.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
[longman@redhat.com: fix for CONFIG_ZONE_DMA=n]
Suggested-by: Roman Gushchin <guro@fb.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "mm: memcg/slab: Fix objcg pointer array handling problem", v4.
Since the merging of the new slab memory controller in v5.9, the page
structure stores a pointer to objcg pointer array for slab pages. When
the slab has no used objects, it can be freed in free_slab() which will
call kfree() to free the objcg pointer array in
memcg_alloc_page_obj_cgroups(). If it happens that the objcg pointer
array is the last used object in its slab, that slab may then be freed
which may caused kfree() to be called again.
With the right workload, the slab cache may be set up in a way that allows
the recursive kfree() calling loop to nest deep enough to cause a kernel
stack overflow and panic the system. In fact, we have a reproducer that
can cause kernel stack overflow on a s390 system involving kmalloc-rcl-256
and kmalloc-rcl-128 slabs with the following kfree() loop recursively
called 74 times:
[ 285.520739] [<000000000ec432fc>] kfree+0x4bc/0x560 [ 285.520740]
[<000000000ec43466>] __free_slab+0xc6/0x228 [ 285.520741]
[<000000000ec41fc2>] __slab_free+0x3c2/0x3e0 [ 285.520742]
[<000000000ec432fc>] kfree+0x4bc/0x560 : While investigating this issue, I
also found an issue on the allocation side. If the objcg pointer array
happen to come from the same slab or a circular dependency linkage is
formed with multiple slabs, those affected slabs can never be freed again.
This patch series addresses these two issues by introducing a new set of
kmalloc-cg-<n> caches split from kmalloc-<n> caches. The new set will
only contain non-reclaimable and non-dma objects that are accounted in
memory cgroups whereas the old set are now for unaccounted objects only.
By making this split, all the objcg pointer arrays will come from the
kmalloc-<n> caches, but those caches will never hold any objcg pointer
array. As a result, deeply nested kfree() call and the unfreeable slab
problems are now gone.
This patch (of 4):
Since the merging of the new slab memory controller in v5.9, the page
structure may store a pointer to obj_cgroup pointer array for slab pages.
Currently, only the __GFP_ACCOUNT bit is masked off. However, the array
is not readily reclaimable and doesn't need to come from the DMA buffer.
So those GFP bits should be masked off as well.
Do the flag bit clearing at memcg_alloc_page_obj_cgroups() to make sure
that it is consistently applied no matter where it is called.
Link: https://lkml.kernel.org/r/20210505200610.13943-1-longman@redhat.com
Link: https://lkml.kernel.org/r/20210505200610.13943-2-longman@redhat.com
Fixes: 286e04b8ed ("mm: memcg/slab: allocate obj_cgroups for non-root slab pages")
Signed-off-by: Waiman Long <longman@redhat.com>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Roman Gushchin <guro@fb.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There are two issues with the current refill_obj_stock() code. First of
all, when nr_bytes reaches over PAGE_SIZE, it calls drain_obj_stock() to
atomically flush out remaining bytes to obj_cgroup, clear cached_objcg and
do a obj_cgroup_put(). It is likely that the same obj_cgroup will be used
again which leads to another call to drain_obj_stock() and
obj_cgroup_get() as well as atomically retrieve the available byte from
obj_cgroup. That is costly. Instead, we should just uncharge the excess
pages, reduce the stock bytes and be done with it. The drain_obj_stock()
function should only be called when obj_cgroup changes.
Secondly, when charging an object of size not less than a page in
obj_cgroup_charge(), it is possible that the remaining bytes to be
refilled to the stock will overflow a page and cause refill_obj_stock() to
uncharge 1 page. To avoid the additional uncharge in this case, a new
allow_uncharge flag is added to refill_obj_stock() which will be set to
false when called from obj_cgroup_charge() so that an uncharge_pages()
call won't be issued right after a charge_pages() call unless the objcg
changes.
A multithreaded kmalloc+kfree microbenchmark on a 2-socket 48-core
96-thread x86-64 system with 96 testing threads were run. Before this
patch, the total number of kilo kmalloc+kfree operations done for a 4k
large object by all the testing threads per second were 4,304 kops/s
(cgroup v1) and 8,478 kops/s (cgroup v2). After applying this patch, the
number were 4,731 (cgroup v1) and 418,142 (cgroup v2) respectively. This
represents a performance improvement of 1.10X (cgroup v1) and 49.3X
(cgroup v2).
Link: https://lkml.kernel.org/r/20210506150007.16288-4-longman@redhat.com
Signed-off-by: Waiman Long <longman@redhat.com>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Cc: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Chris Down <chris@chrisdown.name>
Cc: Christoph Lameter <cl@linux.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Masayoshi Mizuma <msys.mizuma@gmail.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Roman Gushchin <guro@fb.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Wei Yang <richard.weiyang@gmail.com>
Cc: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Cc: Yafang Shao <laoar.shao@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Before the new slab memory controller with per object byte charging,
charging and vmstat data update happen only when new slab pages are
allocated or freed. Now they are done with every kmem_cache_alloc() and
kmem_cache_free(). This causes additional overhead for workloads that
generate a lot of alloc and free calls.
The memcg_stock_pcp is used to cache byte charge for a specific obj_cgroup
to reduce that overhead. To further reducing it, this patch makes the
vmstat data cached in the memcg_stock_pcp structure as well until it
accumulates a page size worth of update or when other cached data change.
Caching the vmstat data in the per-cpu stock eliminates two writes to
non-hot cachelines for memcg specific as well as memcg-lruvecs specific
vmstat data by a write to a hot local stock cacheline.
On a 2-socket Cascade Lake server with instrumentation enabled and this
patch applied, it was found that about 20% (634400 out of 3243830) of the
time when mod_objcg_state() is called leads to an actual call to
__mod_objcg_state() after initial boot. When doing parallel kernel build,
the figure was about 17% (24329265 out of 142512465). So caching the
vmstat data reduces the number of calls to __mod_objcg_state() by more
than 80%.
Link: https://lkml.kernel.org/r/20210506150007.16288-3-longman@redhat.com
Signed-off-by: Waiman Long <longman@redhat.com>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Cc: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Chris Down <chris@chrisdown.name>
Cc: Christoph Lameter <cl@linux.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Masayoshi Mizuma <msys.mizuma@gmail.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Roman Gushchin <guro@fb.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Wei Yang <richard.weiyang@gmail.com>
Cc: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Cc: Yafang Shao <laoar.shao@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "mm/memcg: Reduce kmemcache memory accounting overhead", v6.
With the recent introduction of the new slab memory controller, we
eliminate the need for having separate kmemcaches for each memory cgroup
and reduce overall kernel memory usage. However, we also add additional
memory accounting overhead to each call of kmem_cache_alloc() and
kmem_cache_free().
For workloads that require a lot of kmemcache allocations and
de-allocations, they may experience performance regression as illustrated
in [1] and [2].
A simple kernel module that performs repeated loop of 100,000,000
kmem_cache_alloc() and kmem_cache_free() of either a small 32-byte object
or a big 4k object at module init time with a batch size of 4 (4 kmalloc's
followed by 4 kfree's) is used for benchmarking. The benchmarking tool
was run on a kernel based on linux-next-20210419. The test was run on a
CascadeLake server with turbo-boosting disable to reduce run-to-run
variation.
The small object test exercises mainly the object stock charging and
vmstat update code paths. The large object test also exercises the
refill_obj_stock() and __memcg_kmem_charge()/__memcg_kmem_uncharge() code
paths.
With memory accounting disabled, the run time was 3.130s with both small
object big object tests.
With memory accounting enabled, both cgroup v1 and v2 showed similar
results in the small object test. The performance results of the large
object test, however, differed between cgroup v1 and v2.
The execution times with the application of various patches in the
patchset were:
Applied patches Run time Accounting overhead %age 1 %age 2
--------------- -------- ------------------- ------ ------
Small 32-byte object:
None 11.634s 8.504s 100.0% 271.7%
1-2 9.425s 6.295s 74.0% 201.1%
1-3 9.708s 6.578s 77.4% 210.2%
1-4 8.062s 4.932s 58.0% 157.6%
Large 4k object (v2):
None 22.107s 18.977s 100.0% 606.3%
1-2 20.960s 17.830s 94.0% 569.6%
1-3 14.238s 11.108s 58.5% 354.9%
1-4 11.329s 8.199s 43.2% 261.9%
Large 4k object (v1):
None 36.807s 33.677s 100.0% 1075.9%
1-2 36.648s 33.518s 99.5% 1070.9%
1-3 22.345s 19.215s 57.1% 613.9%
1-4 18.662s 15.532s 46.1% 496.2%
N.B. %age 1 = overhead/unpatched overhead
%age 2 = overhead/accounting disabled time
Patch 2 (vmstat data stock caching) helps in both the small object test
and the large v2 object test. It doesn't help much in v1 big object test.
Patch 3 (refill_obj_stock improvement) does help the small object test
but offer significant performance improvement for the large object test
(both v1 and v2).
Patch 4 (eliminating irq disable/enable) helps in all test cases.
To test for the extreme case, a multi-threaded kmalloc/kfree
microbenchmark was run on the 2-socket 48-core 96-thread system with
96 testing threads in the same memcg doing kmalloc+kfree of a 4k object
with accounting enabled for 10s. The total number of kmalloc+kfree done
in kilo operations per second (kops/s) were as follows:
Applied patches v1 kops/s v1 change v2 kops/s v2 change
--------------- --------- --------- --------- ---------
None 3,520 1.00X 6,242 1.00X
1-2 4,304 1.22X 8,478 1.36X
1-3 4,731 1.34X 418,142 66.99X
1-4 4,587 1.30X 438,838 70.30X
With memory accounting disabled, the kmalloc/kfree rate was 1,481,291
kop/s. This test shows how significant the memory accouting overhead
can be in some extreme situations.
For this multithreaded test, the improvement from patch 2 mainly
comes from the conditional atomic xchg of objcg->nr_charged_bytes in
mod_objcg_state(). By using an unconditional xchg, the operation rates
were similar to the unpatched kernel.
Patch 3 elminates the single highly contended cacheline of
objcg->nr_charged_bytes for cgroup v2 leading to a huge performance
improvement. Cgroup v1, however, still has another highly contended
cacheline in the shared page counter &memcg->kmem. So the improvement
is only modest.
Patch 4 helps in cgroup v2, but performs worse in cgroup v1 as
eliminating the irq_disable/irq_enable overhead seems to aggravate the
cacheline contention.
[1] https://lore.kernel.org/linux-mm/20210408193948.vfktg3azh2wrt56t@gabell/T/#u
[2] https://lore.kernel.org/lkml/20210114025151.GA22932@xsang-OptiPlex-9020/
This patch (of 4):
mod_objcg_state() is moved from mm/slab.h to mm/memcontrol.c so that
further optimization can be done to it in later patches without exposing
unnecessary details to other mm components.
Link: https://lkml.kernel.org/r/20210506150007.16288-1-longman@redhat.com
Link: https://lkml.kernel.org/r/20210506150007.16288-2-longman@redhat.com
Signed-off-by: Waiman Long <longman@redhat.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Roman Gushchin <guro@fb.com>
Cc: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Chris Down <chris@chrisdown.name>
Cc: Christoph Lameter <cl@linux.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Masayoshi Mizuma <msys.mizuma@gmail.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Wei Yang <richard.weiyang@gmail.com>
Cc: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Cc: Yafang Shao <laoar.shao@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
