If the freeing page and its buddy page are not at the same zone, the
current holding zone->lock for the freeing page cann't prevent buddy page
getting allocated, this could trigger VM_BUG_ON_PAGE in page_is_buddy() at
a very tiny chance, such as:
cpu 0: cpu 1:
hold zone_1 lock
check page and it buddy
PageBuddy(buddy) is true hold zone_2 lock
page_order(buddy) == order is true alloc buddy
trigger VM_BUG_ON_PAGE(page_count(buddy) != 0)
zone_1->lock prevents the freeing page getting allocated
zone_2->lock prevents the buddy page getting allocated
they are not the same zone->lock.
If we can't remove the zone_id check statement, it's better handle this
rare race. This patch fixes this by placing the zone_id check before the
VM_BUG_ON_PAGE check.
Signed-off-by: Weijie Yang <weijie.yang@samsung.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The OOM killing invocation does a lot of duplicative checks against the
task's allocation context. Rework it to take advantage of the existing
checks in the allocator slowpath.
The OOM killer is invoked when the allocator is unable to reclaim any
pages but the allocation has to keep looping. Instead of having a check
for __GFP_NORETRY hidden in oom_gfp_allowed(), just move the OOM
invocation to the true branch of should_alloc_retry(). The __GFP_FS
check from oom_gfp_allowed() can then be moved into the OOM avoidance
branch in __alloc_pages_may_oom(), along with the PF_DUMPCORE test.
__alloc_pages_may_oom() can then signal to the caller whether the OOM
killer was invoked, instead of requiring it to duplicate the order and
high_zoneidx checks to guess this when deciding whether to continue.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.cz>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When the system boots up, in the dmesg logs we can see the memory
statistics along with total reserved as below. Memory: 458840k/458840k
available, 65448k reserved, 0K highmem
When CMA is enabled, still the total reserved memory remains the same.
However, the CMA memory is not considered as reserved. But, when we see
/proc/meminfo, the CMA memory is part of free memory. This creates
confusion. This patch corrects the problem by properly subtracting the
CMA reserved memory from the total reserved memory in dmesg logs.
Below is the dmesg snapshot from an arm based device with 512MB RAM and
12MB single CMA region.
Before this change:
Memory: 458840k/458840k available, 65448k reserved, 0K highmem
After this change:
Memory: 458840k/458840k available, 53160k reserved, 12288k cma-reserved, 0K highmem
Signed-off-by: Pintu Kumar <pintu.k@samsung.com>
Signed-off-by: Vishnu Pratap Singh <vishnu.ps@samsung.com>
Acked-by: Michal Nazarewicz <mina86@mina86.com>
Cc: Rafael Aquini <aquini@redhat.com>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.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>
Since 01cefaef40 ("mm: provide more accurate estimation
of pages occupied by memmap") allocate the pages from lowmem for the
highmem zones' memmap. So It is not need to reserver the memmap's for
the highmem.
A 2G DDR3 for the arm platform:
On node 0 totalpages: 524288
free_area_init_node: node 0, pgdat 80ccd380, node_mem_map 80d38000
DMA zone: 3568 pages used for memmap
DMA zone: 0 pages reserved
DMA zone: 456704 pages, LIFO batch:31
HighMem zone: 528 pages used for memmap
HighMem zone: 67584 pages, LIFO batch:15
On node 0 totalpages: 524288
free_area_init_node: node 0, pgdat 80cd6f40, node_mem_map 80d42000
DMA zone: 3568 pages used for memmap
DMA zone: 0 pages reserved
DMA zone: 456704 pages, LIFO batch:31
HighMem zone: 67584 pages, LIFO batch:15
Signed-off-by: Hongbo Zhong <hongbo.zhong@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The slab shrinkers are currently invoked from the zonelist walkers in
kswapd, direct reclaim, and zone reclaim, all of which roughly gauge the
eligible LRU pages and assemble a nodemask to pass to NUMA-aware
shrinkers, which then again have to walk over the nodemask. This is
redundant code, extra runtime work, and fairly inaccurate when it comes to
the estimation of actually scannable LRU pages. The code duplication will
only get worse when making the shrinkers cgroup-aware and requiring them
to have out-of-band cgroup hierarchy walks as well.
Instead, invoke the shrinkers from shrink_zone(), which is where all
reclaimers end up, to avoid this duplication.
Take the count for eligible LRU pages out of get_scan_count(), which
considers many more factors than just the availability of swap space, like
zone_reclaimable_pages() currently does. Accumulate the number over all
visited lruvecs to get the per-zone value.
Some nodes have multiple zones due to memory addressing restrictions. To
avoid putting too much pressure on the shrinkers, only invoke them once
for each such node, using the class zone of the allocation as the pivot
zone.
For now, this integrates the slab shrinking better into the reclaim logic
and gets rid of duplicative invocations from kswapd, direct reclaim, and
zone reclaim. It also prepares for cgroup-awareness, allowing
memcg-capable shrinkers to be added at the lruvec level without much
duplication of both code and runtime work.
This changes kswapd behavior, which used to invoke the shrinkers for each
zone, but with scan ratios gathered from the entire node, resulting in
meaningless pressure quantities on multi-zone nodes.
Zone reclaim behavior also changes. It used to shrink slabs until the
same amount of pages were shrunk as were reclaimed from the LRUs. Now it
merely invokes the shrinkers once with the zone's scan ratio, which makes
the shrinkers go easier on caches that implement aging and would prefer
feeding back pressure from recently used slab objects to unused LRU pages.
[vdavydov@parallels.com: assure class zone is populated]
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Dave Chinner <david@fromorbit.com>
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is the page owner tracking code which is introduced so far ago. It
is resident on Andrew's tree, though, nobody tried to upstream so it
remain as is. Our company uses this feature actively to debug memory leak
or to find a memory hogger so I decide to upstream this feature.
This functionality help us to know who allocates the page. When
allocating a page, we store some information about allocation in extra
memory. Later, if we need to know status of all pages, we can get and
analyze it from this stored information.
In previous version of this feature, extra memory is statically defined in
struct page, but, in this version, extra memory is allocated outside of
struct page. It enables us to turn on/off this feature at boottime
without considerable memory waste.
Although we already have tracepoint for tracing page allocation/free,
using it to analyze page owner is rather complex. We need to enlarge the
trace buffer for preventing overlapping until userspace program launched.
And, launched program continually dump out the trace buffer for later
analysis and it would change system behaviour with more possibility rather
than just keeping it in memory, so bad for debug.
Moreover, we can use page_owner feature further for various purposes. For
example, we can use it for fragmentation statistics implemented in this
patch. And, I also plan to implement some CMA failure debugging feature
using this interface.
I'd like to give the credit for all developers contributed this feature,
but, it's not easy because I don't know exact history. Sorry about that.
Below is people who has "Signed-off-by" in the patches in Andrew's tree.
Contributor:
Alexander Nyberg <alexn@dsv.su.se>
Mel Gorman <mgorman@suse.de>
Dave Hansen <dave@linux.vnet.ibm.com>
Minchan Kim <minchan@kernel.org>
Michal Nazarewicz <mina86@mina86.com>
Andrew Morton <akpm@linux-foundation.org>
Jungsoo Son <jungsoo.son@lge.com>
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Dave Hansen <dave@sr71.net>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Jungsoo Son <jungsoo.son@lge.com>
Cc: Ingo Molnar <mingo@redhat.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>
Now, we have prepared to avoid using debug-pagealloc in boottime. So
introduce new kernel-parameter to disable debug-pagealloc in boottime, and
makes related functions to be disabled in this case.
Only non-intuitive part is change of guard page functions. Because guard
page is effective only if debug-pagealloc is enabled, turning off
according to debug-pagealloc is reasonable thing to do.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Dave Hansen <dave@sr71.net>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Jungsoo Son <jungsoo.son@lge.com>
Cc: Ingo Molnar <mingo@redhat.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>
Until now, debug-pagealloc needs extra flags in struct page, so we need to
recompile whole source code when we decide to use it. This is really
painful, because it takes some time to recompile and sometimes rebuild is
not possible due to third party module depending on struct page. So, we
can't use this good feature in many cases.
Now, we have the page extension feature that allows us to insert extra
flags to outside of struct page. This gets rid of third party module
issue mentioned above. And, this allows us to determine if we need extra
memory for this page extension in boottime. With these property, we can
avoid using debug-pagealloc in boottime with low computational overhead in
the kernel built with CONFIG_DEBUG_PAGEALLOC. This will help our
development process greatly.
This patch is the preparation step to achive above goal. debug-pagealloc
originally uses extra field of struct page, but, after this patch, it will
use field of struct page_ext. Because memory for page_ext is allocated
later than initialization of page allocator in CONFIG_SPARSEMEM, we should
disable debug-pagealloc feature temporarily until initialization of
page_ext. This patch implements this.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Dave Hansen <dave@sr71.net>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Jungsoo Son <jungsoo.son@lge.com>
Cc: Ingo Molnar <mingo@redhat.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>
When we debug something, we'd like to insert some information to every
page. For this purpose, we sometimes modify struct page itself. But,
this has drawbacks. First, it requires re-compile. This makes us
hesitate to use the powerful debug feature so development process is
slowed down. And, second, sometimes it is impossible to rebuild the
kernel due to third party module dependency. At third, system behaviour
would be largely different after re-compile, because it changes size of
struct page greatly and this structure is accessed by every part of
kernel. Keeping this as it is would be better to reproduce errornous
situation.
This feature is intended to overcome above mentioned problems. This
feature allocates memory for extended data per page in certain place
rather than the struct page itself. This memory can be accessed by the
accessor functions provided by this code. During the boot process, it
checks whether allocation of huge chunk of memory is needed or not. If
not, it avoids allocating memory at all. With this advantage, we can
include this feature into the kernel in default and can avoid rebuild and
solve related problems.
Until now, memcg uses this technique. But, now, memcg decides to embed
their variable to struct page itself and it's code to extend struct page
has been removed. I'd like to use this code to develop debug feature, so
this patch resurrect it.
To help these things to work well, this patch introduces two callbacks for
clients. One is the need callback which is mandatory if user wants to
avoid useless memory allocation at boot-time. The other is optional, init
callback, which is used to do proper initialization after memory is
allocated. Detailed explanation about purpose of these functions is in
code comment. Please refer it.
Others are completely same with previous extension code in memcg.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Dave Hansen <dave@sr71.net>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Jungsoo Son <jungsoo.son@lge.com>
Cc: Ingo Molnar <mingo@redhat.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>
Page guard is used by debug-pagealloc feature. Currently, it is
open-coded, but, I think that more abstraction of it makes core page
allocator code more readable.
There is no functional difference.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Gioh Kim <gioh.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull cgroup update from Tejun Heo:
"cpuset got simplified a bit. cgroup core got a fix on unified
hierarchy and grew some effective css related interfaces which will be
used for blkio support for writeback IO traffic which is currently
being worked on"
* 'for-3.19' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
cgroup: implement cgroup_get_e_css()
cgroup: add cgroup_subsys->css_e_css_changed()
cgroup: add cgroup_subsys->css_released()
cgroup: fix the async css offline wait logic in cgroup_subtree_control_write()
cgroup: restructure child_subsys_mask handling in cgroup_subtree_control_write()
cgroup: separate out cgroup_calc_child_subsys_mask() from cgroup_refresh_child_subsys_mask()
cpuset: lock vs unlock typo
cpuset: simplify cpuset_node_allowed API
cpuset: convert callback_mutex to a spinlock
Now that the external page_cgroup data structure and its lookup is
gone, let the generic bad_page() check for page->mem_cgroup sanity.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.cz>
Acked-by: Vladimir Davydov <vdavydov@parallels.com>
Acked-by: David S. Miller <davem@davemloft.net>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Tejun Heo <tj@kernel.org>
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>
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead during runtime, those pointers were
allocated in a separate array, with a translation layer between them and
struct page.
There is now only one page pointer remaining: the memcg pointer, that
indicates which cgroup the page is associated with when charged. The
complexity of runtime allocation and the runtime translation overhead is
no longer justified to save that *potential* 0.19% of memory. With
CONFIG_SLUB, page->mem_cgroup actually sits in the doubleword padding
after the page->private member and doesn't even increase struct page,
and then this patch actually saves space. Remaining users that care can
still compile their kernels without CONFIG_MEMCG.
text data bss dec hex filename
8828345 1725264 983040 11536649 b00909 vmlinux.old
8827425 1725264 966656 11519345 afc571 vmlinux.new
[mhocko@suse.cz: update Documentation/cgroups/memory.txt]
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.cz>
Acked-by: Vladimir Davydov <vdavydov@parallels.com>
Acked-by: David S. Miller <davem@davemloft.net>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Konstantin Khlebnikov <koct9i@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The goal of memory compaction is to create high-order freepages through
page migration. Page migration however puts pages on the per-cpu lru_add
cache, which is later flushed to per-cpu pcplists, and only after pcplists
are drained the pages can actually merge. This can happen due to the
per-cpu caches becoming full through further freeing, or explicitly.
During direct compaction, it is useful to do the draining explicitly so
that pages merge as soon as possible and compaction can detect success
immediately and keep the latency impact at minimum. However the current
implementation is far from ideal. Draining is done only in
__alloc_pages_direct_compact(), after all zones were already compacted,
and the decisions to continue or stop compaction in individual zones was
done without the last batch of migrations being merged. It is also
missing the draining of lru_add cache before the pcplists.
This patch moves the draining for direct compaction into compact_zone().
It adds the missing lru_cache draining and uses the newly introduced
single zone pcplists draining to reduce overhead and avoid impact on
unrelated zones. Draining is only performed when it can actually lead to
merging of a page of desired order (passed by cc->order). This means it
is only done when migration occurred in the previously scanned cc->order
aligned block(s) and the migration scanner is now pointing to the next
cc->order aligned block.
The patch has been tested with stress-highalloc benchmark from mmtests.
Although overal allocation success rates of the benchmark were not
affected, the number of detected compaction successes has doubled. This
suggests that allocations were previously successful due to implicit
merging caused by background activity, making a later allocation attempt
succeed immediately, but not attributing the success to compaction. Since
stress-highalloc always tries to allocate almost the whole memory, it
cannot show the improvement in its reported success rate metric. However
after this patch, compaction should detect success and terminate earlier,
reducing the direct compaction latencies in a real scenario.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Christoph Lameter <cl@linux.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Since commit 53853e2d2b ("mm, compaction: defer each zone individually
instead of preferred zone"), compaction is deferred for each zone where
sync direct compaction fails, and reset where it succeeds. However, it
was observed that for DMA zone compaction often appeared to succeed
while subsequent allocation attempt would not, due to different outcome
of watermark check.
In order to properly defer compaction in this zone, the candidate zone
has to be passed back to __alloc_pages_direct_compact() and compaction
deferred in the zone after the allocation attempt fails.
The large source of mismatch between watermark check in compaction and
allocation was the lack of alloc_flags and classzone_idx values in
compaction, which has been fixed in the previous patch. So with this
problem fixed, we can simplify the code by removing the candidate_zone
parameter and deferring in __alloc_pages_direct_compact().
After this patch, the compaction activity during stress-highalloc
benchmark is still somewhat increased, but it's negligible compared to the
increase that occurred without the better watermark checking. This
suggests that it is still possible to apparently succeed in compaction but
fail to allocate, possibly due to parallel allocation activity.
[akpm@linux-foundation.org: fix build]
Suggested-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Compaction relies on zone watermark checks for decisions such as if it's
worth to start compacting in compaction_suitable() or whether compaction
should stop in compact_finished(). The watermark checks take
classzone_idx and alloc_flags parameters, which are related to the memory
allocation request. But from the context of compaction they are currently
passed as 0, including the direct compaction which is invoked to satisfy
the allocation request, and could therefore know the proper values.
The lack of proper values can lead to mismatch between decisions taken
during compaction and decisions related to the allocation request. Lack
of proper classzone_idx value means that lowmem_reserve is not taken into
account. This has manifested (during recent changes to deferred
compaction) when DMA zone was used as fallback for preferred Normal zone.
compaction_suitable() without proper classzone_idx would think that the
watermarks are already satisfied, but watermark check in
get_page_from_freelist() would fail. Because of this problem, deferring
compaction has extra complexity that can be removed in the following
patch.
The issue (not confirmed in practice) with missing alloc_flags is opposite
in nature. For allocations that include ALLOC_HIGH, ALLOC_HIGHER or
ALLOC_CMA in alloc_flags (the last includes all MOVABLE allocations on
CMA-enabled systems) the watermark checking in compaction with 0 passed
will be stricter than in get_page_from_freelist(). In these cases
compaction might be running for a longer time than is really needed.
Another issue compaction_suitable() is that the check for "does the zone
need compaction at all?" comes only after the check "does the zone have
enough free free pages to succeed compaction". The latter considers extra
pages for migration and can therefore in some situations fail and return
COMPACT_SKIPPED, although the high-order allocation would succeed and we
should return COMPACT_PARTIAL.
This patch fixes these problems by adding alloc_flags and classzone_idx to
struct compact_control and related functions involved in direct compaction
and watermark checking. Where possible, all other callers of
compaction_suitable() pass proper values where those are known. This is
currently limited to classzone_idx, which is sometimes known in kswapd
context. However, the direct reclaim callers should_continue_reclaim()
and compaction_ready() do not currently know the proper values, so the
coordination between reclaim and compaction may still not be as accurate
as it could. This can be fixed later, if it's shown to be an issue.
Additionaly the checks in compact_suitable() are reordered to address the
second issue described above.
The effect of this patch should be slightly better high-order allocation
success rates and/or less compaction overhead, depending on the type of
allocations and presence of CMA. It allows simplifying deferred
compaction code in a followup patch.
When testing with stress-highalloc, there was some slight improvement
(which might be just due to variance) in success rates of non-THP-like
allocations.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Christoph Lameter <cl@linux.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This allows us to catch the bug fixed in the previous patch (mm: free
compound page with correct order).
Here we also verify whether a page is tail page or not -- tail pages are
supposed to be freed along with their head, not by themselves.
Signed-off-by: Yu Zhao <yuzhao@google.com>
Reviewed-by: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: David Rientjes <rientjes@google.com>
Cc: Bob Liu <lliubbo@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
CMA allocation drains pcplists so that pages can merge back to buddy
allocator. Since it operates on a single zone, we can reduce the
pcplists drain to the single zone, which is now possible.
The change should make CMA allocations faster and not disturbing
unrelated pcplists anymore.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Xishi Qiu <qiuxishi@huawei.com>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The functions for draining per-cpu pages back to buddy allocators
currently always operate on all zones. There are however several cases
where the drain is only needed in the context of a single zone, and
spilling other pcplists is a waste of time both due to the extra
spilling and later refilling.
This patch introduces new zone pointer parameter to drain_all_pages()
and changes the dummy parameter of drain_local_pages() to be also a zone
pointer. When NULL is passed, the functions operate on all zones as
usual. Passing a specific zone pointer reduces the work to the single
zone.
All callers are updated to pass the NULL pointer in this patch.
Conversion to single zone (where appropriate) is done in further
patches.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Xishi Qiu <qiuxishi@huawei.com>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
One thing I did in this patch is fixing freepage accounting. If we
clear guard page and link it onto isolate buddy list, we should not
increase freepage count. This patch adds conditional branch to skip
counting in this case. Without this patch, this overcounting happens
frequently if guard order is set and CMA is used.
Another thing fixed in this patch is the target to reset order. In
__free_one_page(), we check the buddy page whether it is a guard page or
not. And, if so, we should clear guard attribute on the buddy page and
reset order of it to 0. But, current code resets original page's order
rather than buddy one's. Maybe, this doesn't have any problem, because
whole merged page's order will be re-assigned soon. But, it is better
to correct code.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Gioh Kim <gioh.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Having test_pages_isolated failure message as a warning confuses users
into thinking that it is more serious than it really is. In reality, if
called via CMA, allocation will be retried so a single
test_pages_isolated failure does not prevent allocation from succeeding.
Demote the warning message to an info message and reformat it such that
the text "failed" does not appear and instead a less worrying "PFNS
busy" is used.
This message is trivially reproducible on a 10GB x86 machine on 3.16.y
kernels configured with CONFIG_DMA_CMA.
Signed-off-by: Michal Nazarewicz <mina86@mina86.com>
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: Peter Hurley <peter@hurleysoftware.com>
Cc: Minchan Kim <minchan@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Current pageblock isolation logic could isolate each pageblock
individually. This causes freepage accounting problem if freepage with
pageblock order on isolate pageblock is merged with other freepage on
normal pageblock. We can prevent merging by restricting max order of
merging to pageblock order if freepage is on isolate pageblock.
A side-effect of this change is that there could be non-merged buddy
freepage even if finishing pageblock isolation, because undoing
pageblock isolation is just to move freepage from isolate buddy list to
normal buddy list rather than to consider merging. So, the patch also
makes undoing pageblock isolation consider freepage merge. When
un-isolation, freepage with more than pageblock order and it's buddy are
checked. If they are on normal pageblock, instead of just moving, we
isolate the freepage and free it in order to get merged.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Tang Chen <tangchen@cn.fujitsu.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Cc: Wen Congyang <wency@cn.fujitsu.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Laura Abbott <lauraa@codeaurora.org>
Cc: Heesub Shin <heesub.shin@samsung.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: Ritesh Harjani <ritesh.list@gmail.com>
Cc: Gioh Kim <gioh.kim@lge.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
All the caller of __free_one_page() has similar freepage counting logic,
so we can move it to __free_one_page(). This reduce line of code and
help future maintenance.
This is also preparation step for "mm/page_alloc: restrict max order of
merging on isolated pageblock" which fix the freepage counting problem
on freepage with more than pageblock order.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Tang Chen <tangchen@cn.fujitsu.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Cc: Wen Congyang <wency@cn.fujitsu.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Laura Abbott <lauraa@codeaurora.org>
Cc: Heesub Shin <heesub.shin@samsung.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: Ritesh Harjani <ritesh.list@gmail.com>
Cc: Gioh Kim <gioh.kim@lge.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In free_pcppages_bulk(), we use cached migratetype of freepage to
determine type of buddy list where freepage will be added. This
information is stored when freepage is added to pcp list, so if
isolation of pageblock of this freepage begins after storing, this
cached information could be stale. In other words, it has original
migratetype rather than MIGRATE_ISOLATE.
There are two problems caused by this stale information.
One is that we can't keep these freepages from being allocated.
Although this pageblock is isolated, freepage will be added to normal
buddy list so that it could be allocated without any restriction. And
the other problem is incorrect freepage accounting. Freepages on
isolate pageblock should not be counted for number of freepage.
Following is the code snippet in free_pcppages_bulk().
/* MIGRATE_MOVABLE list may include MIGRATE_RESERVEs */
__free_one_page(page, page_to_pfn(page), zone, 0, mt);
trace_mm_page_pcpu_drain(page, 0, mt);
if (likely(!is_migrate_isolate_page(page))) {
__mod_zone_page_state(zone, NR_FREE_PAGES, 1);
if (is_migrate_cma(mt))
__mod_zone_page_state(zone, NR_FREE_CMA_PAGES, 1);
}
As you can see above snippet, current code already handle second
problem, incorrect freepage accounting, by re-fetching pageblock
migratetype through is_migrate_isolate_page(page).
But, because this re-fetched information isn't used for
__free_one_page(), first problem would not be solved. This patch try to
solve this situation to re-fetch pageblock migratetype before
__free_one_page() and to use it for __free_one_page().
In addition to move up position of this re-fetch, this patch use
optimization technique, re-fetching migratetype only if there is isolate
pageblock. Pageblock isolation is rare event, so we can avoid
re-fetching in common case with this optimization.
This patch also correct migratetype of the tracepoint output.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Acked-by: Michal Nazarewicz <mina86@mina86.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Tang Chen <tangchen@cn.fujitsu.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Cc: Wen Congyang <wency@cn.fujitsu.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Laura Abbott <lauraa@codeaurora.org>
Cc: Heesub Shin <heesub.shin@samsung.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: Ritesh Harjani <ritesh.list@gmail.com>
Cc: Gioh Kim <gioh.kim@lge.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Before describing bugs itself, I first explain definition of freepage.
1. pages on buddy list are counted as freepage.
2. pages on isolate migratetype buddy list are *not* counted as freepage.
3. pages on cma buddy list are counted as CMA freepage, too.
Now, I describe problems and related patch.
Patch 1: There is race conditions on getting pageblock migratetype that
it results in misplacement of freepages on buddy list, incorrect
freepage count and un-availability of freepage.
Patch 2: Freepages on pcp list could have stale cached information to
determine migratetype of buddy list to go. This causes misplacement of
freepages on buddy list and incorrect freepage count.
Patch 4: Merging between freepages on different migratetype of
pageblocks will cause freepages accouting problem. This patch fixes it.
Without patchset [3], above problem doesn't happens on my CMA allocation
test, because CMA reserved pages aren't used at all. So there is no
chance for above race.
With patchset [3], I did simple CMA allocation test and get below
result:
- Virtual machine, 4 cpus, 1024 MB memory, 256 MB CMA reservation
- run kernel build (make -j16) on background
- 30 times CMA allocation(8MB * 30 = 240MB) attempts in 5 sec interval
- Result: more than 5000 freepage count are missed
With patchset [3] and this patchset, I found that no freepage count are
missed so that I conclude that problems are solved.
On my simple memory offlining test, these problems also occur on that
environment, too.
This patch (of 4):
There are two paths to reach core free function of buddy allocator,
__free_one_page(), one is free_one_page()->__free_one_page() and the
other is free_hot_cold_page()->free_pcppages_bulk()->__free_one_page().
Each paths has race condition causing serious problems. At first, this
patch is focused on first type of freepath. And then, following patch
will solve the problem in second type of freepath.
In the first type of freepath, we got migratetype of freeing page
without holding the zone lock, so it could be racy. There are two cases
of this race.
1. pages are added to isolate buddy list after restoring orignal
migratetype
CPU1 CPU2
get migratetype => return MIGRATE_ISOLATE
call free_one_page() with MIGRATE_ISOLATE
grab the zone lock
unisolate pageblock
release the zone lock
grab the zone lock
call __free_one_page() with MIGRATE_ISOLATE
freepage go into isolate buddy list,
although pageblock is already unisolated
This may cause two problems. One is that we can't use this page anymore
until next isolation attempt of this pageblock, because freepage is on
isolate buddy list. The other is that freepage accouting could be wrong
due to merging between different buddy list. Freepages on isolate buddy
list aren't counted as freepage, but ones on normal buddy list are
counted as freepage. If merge happens, buddy freepage on normal buddy
list is inevitably moved to isolate buddy list without any consideration
of freepage accouting so it could be incorrect.
2. pages are added to normal buddy list while pageblock is isolated.
It is similar with above case.
This also may cause two problems. One is that we can't keep these
freepages from being allocated. Although this pageblock is isolated,
freepage would be added to normal buddy list so that it could be
allocated without any restriction. And the other problem is same as
case 1, that it, incorrect freepage accouting.
This race condition would be prevented by checking migratetype again
with holding the zone lock. Because it is somewhat heavy operation and
it isn't needed in common case, we want to avoid rechecking as much as
possible. So this patch introduce new variable, nr_isolate_pageblock in
struct zone to check if there is isolated pageblock. With this, we can
avoid to re-check migratetype in common case and do it only if there is
isolated pageblock or migratetype is MIGRATE_ISOLATE. This solve above
mentioned problems.
Changes from v3:
Add one more check in free_one_page() that checks whether migratetype is
MIGRATE_ISOLATE or not. Without this, abovementioned case 1 could happens.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Acked-by: Michal Nazarewicz <mina86@mina86.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Tang Chen <tangchen@cn.fujitsu.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Cc: Wen Congyang <wency@cn.fujitsu.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Laura Abbott <lauraa@codeaurora.org>
Cc: Heesub Shin <heesub.shin@samsung.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: Ritesh Harjani <ritesh.list@gmail.com>
Cc: Gioh Kim <gioh.kim@lge.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Current cpuset API for checking if a zone/node is allowed to allocate
from looks rather awkward. We have hardwall and softwall versions of
cpuset_node_allowed with the softwall version doing literally the same
as the hardwall version if __GFP_HARDWALL is passed to it in gfp flags.
If it isn't, the softwall version may check the given node against the
enclosing hardwall cpuset, which it needs to take the callback lock to
do.
Such a distinction was introduced by commit 02a0e53d82 ("cpuset:
rework cpuset_zone_allowed api"). Before, we had the only version with
the __GFP_HARDWALL flag determining its behavior. The purpose of the
commit was to avoid sleep-in-atomic bugs when someone would mistakenly
call the function without the __GFP_HARDWALL flag for an atomic
allocation. The suffixes introduced were intended to make the callers
think before using the function.
However, since the callback lock was converted from mutex to spinlock by
the previous patch, the softwall check function cannot sleep, and these
precautions are no longer necessary.
So let's simplify the API back to the single check.
Suggested-by: David Rientjes <rientjes@google.com>
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Acked-by: Christoph Lameter <cl@linux.com>
Acked-by: Zefan Li <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
PM freezer relies on having all tasks frozen by the time devices are
getting frozen so that no task will touch them while they are getting
frozen. But OOM killer is allowed to kill an already frozen task in
order to handle OOM situtation. In order to protect from late wake ups
OOM killer is disabled after all tasks are frozen. This, however, still
keeps a window open when a killed task didn't manage to die by the time
freeze_processes finishes.
Reduce the race window by checking all tasks after OOM killer has been
disabled. This is still not race free completely unfortunately because
oom_killer_disable cannot stop an already ongoing OOM killer so a task
might still wake up from the fridge and get killed without
freeze_processes noticing. Full synchronization of OOM and freezer is,
however, too heavy weight for this highly unlikely case.
Introduce and check oom_kills counter which gets incremented early when
the allocator enters __alloc_pages_may_oom path and only check all the
tasks if the counter changes during the freezing attempt. The counter
is updated so early to reduce the race window since allocator checked
oom_killer_disabled which is set by PM-freezing code. A false positive
will push the PM-freezer into a slow path but that is not a big deal.
Changes since v1
- push the re-check loop out of freeze_processes into
check_frozen_processes and invert the condition to make the code more
readable as per Rafael
Fixes: f660daac47 (oom: thaw threads if oom killed thread is frozen before deferring)
Cc: 3.2+ <stable@vger.kernel.org> # 3.2+
Signed-off-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Pull x86 mm updates from Ingo Molnar:
"This tree includes the following changes:
- fix memory hotplug
- fix hibernation bootup memory layout assumptions
- fix hyperv numa guest kernel messages
- remove dead code
- update documentation"
* 'x86-mm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mm: Update memory map description to list hypervisor-reserved area
x86/mm, hibernate: Do not assume the first e820 area to be RAM
x86/mm/numa: Drop dead code and rename setup_node_data() to setup_alloc_data()
x86/mm/hotplug: Modify PGD entry when removing memory
x86/mm/hotplug: Pass sync_global_pgds() a correct argument in remove_pagetable()
x86: Remove set_pmd_pfn
dump_page() and dump_vma() are not specific to page_alloc.c, move them out
so page_alloc.c won't turn into the unofficial debug repository.
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Zones are allocated by the page allocator in either node or zone order.
Node ordering is preferred in terms of locality and is applied
automatically in one of three cases:
1. If a node has only low memory
2. If DMA/DMA32 is a high percentage of memory
3. If low memory on a single node is greater than 70% of the node size
Otherwise zone ordering is used to preserve low memory for devices that
require it. Unfortunately a consequence of this is that applications
running on a machine with balanced NUMA nodes will experience different
performance characteristics depending on which node they happen to start
from.
The point of zone ordering is to protect lower zones for devices that
require DMA/DMA32 memory. When NUMA was first introduced, this was
critical as 32-bit NUMA machines existed and exhausting low memory
triggered OOMs easily as so many allocations required low memory. On
64-bit machines the primary concern is devices that are 32-bit only which
is less severe than the low memory exhaustion problem on 32-bit NUMA. It
seems there are really few devices that depends on it.
AGP -- I assume this is getting more rare but even then I think the allocations
happen early in boot time where lowmem pressure is less of a problem
DRM -- If the device is 32-bit only then there may be low pressure. I didn't
evaluate these in detail but it looks like some of these are mobile
graphics card. Not many NUMA laptops out there. DRM folk should know
better though.
Some TV cards -- Much demand for 32-bit capable TV cards on NUMA machines?
B43 wireless card -- again not really a NUMA thing.
I cannot find a good reason to incur a performance penalty on all 64-bit NUMA
machines in case someone throws a brain damanged TV or graphics card in there.
This patch defaults to node-ordering on 64-bit NUMA machines. I was tempted
to make it default everywhere but I understand that some embedded arches may
be using 32-bit NUMA where I cannot predict the consequences.
The performance impact depends on the workload and the characteristics of the
machine and the machine I tested on had a large Normal zone on node 0 so the
impact is within the noise for the majority of tests. The allocation stats
show more allocation requests were from DMA32 and local node. Running SpecJBB
with multiple JVMs and automatic NUMA balancing disabled the results were
specjbb
3.17.0-rc2 3.17.0-rc2
vanilla nodeorder-v1r1
Min 1 29534.00 ( 0.00%) 30020.00 ( 1.65%)
Min 10 115717.00 ( 0.00%) 134038.00 ( 15.83%)
Min 19 109718.00 ( 0.00%) 114186.00 ( 4.07%)
Min 28 104459.00 ( 0.00%) 103639.00 ( -0.78%)
Min 37 98245.00 ( 0.00%) 103756.00 ( 5.61%)
Min 46 97198.00 ( 0.00%) 96197.00 ( -1.03%)
Mean 1 30953.25 ( 0.00%) 31917.75 ( 3.12%)
Mean 10 124432.50 ( 0.00%) 140904.00 ( 13.24%)
Mean 19 116033.50 ( 0.00%) 119294.75 ( 2.81%)
Mean 28 108365.25 ( 0.00%) 106879.50 ( -1.37%)
Mean 37 102984.75 ( 0.00%) 106924.25 ( 3.83%)
Mean 46 100783.25 ( 0.00%) 105368.50 ( 4.55%)
Stddev 1 1260.38 ( 0.00%) 1109.66 ( 11.96%)
Stddev 10 7434.03 ( 0.00%) 5171.91 ( 30.43%)
Stddev 19 8453.84 ( 0.00%) 5309.59 ( 37.19%)
Stddev 28 4184.55 ( 0.00%) 2906.63 ( 30.54%)
Stddev 37 5409.49 ( 0.00%) 3192.12 ( 40.99%)
Stddev 46 4521.95 ( 0.00%) 7392.52 (-63.48%)
Max 1 32738.00 ( 0.00%) 32719.00 ( -0.06%)
Max 10 136039.00 ( 0.00%) 148614.00 ( 9.24%)
Max 19 130566.00 ( 0.00%) 127418.00 ( -2.41%)
Max 28 115404.00 ( 0.00%) 111254.00 ( -3.60%)
Max 37 112118.00 ( 0.00%) 111732.00 ( -0.34%)
Max 46 108541.00 ( 0.00%) 116849.00 ( 7.65%)
TPut 1 123813.00 ( 0.00%) 127671.00 ( 3.12%)
TPut 10 497730.00 ( 0.00%) 563616.00 ( 13.24%)
TPut 19 464134.00 ( 0.00%) 477179.00 ( 2.81%)
TPut 28 433461.00 ( 0.00%) 427518.00 ( -1.37%)
TPut 37 411939.00 ( 0.00%) 427697.00 ( 3.83%)
TPut 46 403133.00 ( 0.00%) 421474.00 ( 4.55%)
3.17.0-rc2 3.17.0-rc2
vanillanodeorder-v1r1
DMA allocs 0 0
DMA32 allocs 57 1491992
Normal allocs 32543566 30026383
Movable allocs 0 0
Direct pages scanned 0 0
Kswapd pages scanned 0 0
Kswapd pages reclaimed 0 0
Direct pages reclaimed 0 0
Kswapd efficiency 100% 100%
Kswapd velocity 0.000 0.000
Direct efficiency 100% 100%
Direct velocity 0.000 0.000
Percentage direct scans 0% 0%
Zone normal velocity 0.000 0.000
Zone dma32 velocity 0.000 0.000
Zone dma velocity 0.000 0.000
THP fault alloc 55164 52987
THP collapse alloc 139 147
THP splits 26 21
NUMA alloc hit 4169066 4250692
NUMA alloc miss 0 0
Note that there were more DMA32 allocations with the patch applied. In this
particular case there was no difference in numa_hit and numa_miss. The
expectation is that DMA32 was being used at the low watermark instead of
falling into the slow path. kswapd was not woken but it's not worken for
THP allocations.
On 32-bit, this patch defaults to zone-ordering as low memory depletion
can be a serious problem on 32-bit large memory machines. If the default
ordering was node then processes on node 0 will deplete the Normal zone
due to normal activity. The problem is worse if CONFIG_HIGHPTE is not
set. If combined with large amounts of dirty/writeback pages in Normal
zone then there is also a high risk of OOM. The heuristics are removed
as it's not clear they were ever important on 32-bit. They were only
relevant for setting node-ordering on 64-bit.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Since 2.6.24 there has been a paranoid check in move_freepages that looks
up the zone of two pages. This is a very slow path and the only time I've
seen this bug trigger recently is when memory initialisation was broken
during patch development. Despite the fact it's a slow path, this patch
converts the check to a VM_BUG_ON anyway as it has served its purpose by
now.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Rik van Riel <riel@redhat.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Page reclaim tests zone_is_reclaim_dirty(), but the site that actually
sets this state does zone_set_flag(zone, ZONE_TAIL_LRU_DIRTY), sending the
reader through layers indirection just to track down a simple bit.
Remove all zone flag wrappers and just use bitops against zone->flags
directly. It's just as readable and the lines are barely any longer.
Also rename ZONE_TAIL_LRU_DIRTY to ZONE_DIRTY to match ZONE_WRITEBACK, and
remove the zone_flags_t typedef.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When entering the page_alloc slowpath, we wakeup kswapd on every pgdat
according to the zonelist and high_zoneidx. However, this doesn't take
nodemask into account, and could prematurely wakeup kswapd on some
unintended nodes.
This patch uses for_each_zone_zonelist_nodemask() instead of
for_each_zone_zonelist() in wake_all_kswapds() to avoid the above
situation.
Signed-off-by: Weijie Yang <weijie.yang@samsung.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Introduce a helper to dump information about a VMA, this also makes
dump_page_flags more generic and re-uses that so the output looks very
similar to dump_page:
[ 61.903437] vma ffff88070f88be00 start 00007fff25970000 end 00007fff25992000
[ 61.903437] next ffff88070facd600 prev ffff88070face400 mm ffff88070fade000
[ 61.903437] prot 8000000000000025 anon_vma ffff88070fa1e200 vm_ops (null)
[ 61.903437] pgoff 7ffffffdd file (null) private_data (null)
[ 61.909129] flags: 0x100173(read|write|mayread|maywrite|mayexec|growsdown|account)
[akpm@linux-foundation.org: make dump_vma() require CONFIG_DEBUG_VM]
[swarren@nvidia.com: fix dump_vma() compilation]
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Konstantin Khlebnikov <khlebnikov@openvz.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Michel Lespinasse <walken@google.com>
Cc: Minchan Kim <minchan@kernel.org>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The page allocator has gfp flags (like __GFP_WAIT) and alloc flags (like
ALLOC_CPUSET) that have separate semantics.
The function allocflags_to_migratetype() actually takes gfp flags, not
alloc flags, and returns a migratetype. Rename it to
gfpflags_to_migratetype().
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Async compaction aborts when it detects zone lock contention or
need_resched() is true. David Rientjes has reported that in practice,
most direct async compactions for THP allocation abort due to
need_resched(). This means that a second direct compaction is never
attempted, which might be OK for a page fault, but khugepaged is intended
to attempt a sync compaction in such case and in these cases it won't.
This patch replaces "bool contended" in compact_control with an int that
distinguishes between aborting due to need_resched() and aborting due to
lock contention. This allows propagating the abort through all compaction
functions as before, but passing the abort reason up to
__alloc_pages_slowpath() which decides when to continue with direct
reclaim and another compaction attempt.
Another problem is that try_to_compact_pages() did not act upon the
reported contention (both need_resched() or lock contention) immediately
and would proceed with another zone from the zonelist. When
need_resched() is true, that means initializing another zone compaction,
only to check again need_resched() in isolate_migratepages() and aborting.
For zone lock contention, the unintended consequence is that the lock
contended status reported back to the allocator is detrmined from the last
zone where compaction was attempted, which is rather arbitrary.
This patch fixes the problem in the following way:
- async compaction of a zone aborting due to need_resched() or fatal signal
pending means that further zones should not be tried. We report
COMPACT_CONTENDED_SCHED to the allocator.
- aborting zone compaction due to lock contention means we can still try
another zone, since it has different set of locks. We report back
COMPACT_CONTENDED_LOCK only if *all* zones where compaction was attempted,
it was aborted due to lock contention.
As a result of these fixes, khugepaged will proceed with second sync
compaction as intended, when the preceding async compaction aborted due to
need_resched(). Page fault compactions aborting due to need_resched()
will spare some cycles previously wasted by initializing another zone
compaction only to abort again. Lock contention will be reported only
when compaction in all zones aborted due to lock contention, and therefore
it's not a good idea to try again after reclaim.
In stress-highalloc from mmtests configured to use __GFP_NO_KSWAPD, this
has improved number of THP collapse allocations by 10%, which shows
positive effect on khugepaged. The benchmark's success rates are
unchanged as it is not recognized as khugepaged. Numbers of compact_stall
and compact_fail events have however decreased by 20%, with
compact_success still a bit improved, which is good. With benchmark
configured not to use __GFP_NO_KSWAPD, there is 6% improvement in THP
collapse allocations, and only slight improvement in stalls and failures.
[akpm@linux-foundation.org: fix warnings]
Reported-by: David Rientjes <rientjes@google.com>
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Minchan Kim <minchan@kernel.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
isolate_migratepages_range() is the main function of the compaction
scanner, called either on a single pageblock by isolate_migratepages()
during regular compaction, or on an arbitrary range by CMA's
__alloc_contig_migrate_range(). It currently perfoms two pageblock-wide
compaction suitability checks, and because of the CMA callpath, it tracks
if it crossed a pageblock boundary in order to repeat those checks.
However, closer inspection shows that those checks are always true for CMA:
- isolation_suitable() is true because CMA sets cc->ignore_skip_hint to true
- migrate_async_suitable() check is skipped because CMA uses sync compaction
We can therefore move the compaction-specific checks to
isolate_migratepages() and simplify isolate_migratepages_range().
Furthermore, we can mimic the freepage scanner family of functions, which
has isolate_freepages_block() function called both by compaction from
isolate_freepages() and by CMA from isolate_freepages_range(), where each
use-case adds own specific glue code. This allows further code
simplification.
Thus, we rename isolate_migratepages_range() to
isolate_migratepages_block() and limit its functionality to a single
pageblock (or its subset). For CMA, a new different
isolate_migratepages_range() is created as a CMA-specific wrapper for the
_block() function. The checks specific to compaction are moved to
isolate_migratepages(). As part of the unification of these two families
of functions, we remove the redundant zone parameter where applicable,
since zone pointer is already passed in cc->zone.
Furthermore, going back to compact_zone() and compact_finished() when
pageblock is found unsuitable (now by isolate_migratepages()) is wasteful
- the checks are meant to skip pageblocks quickly. The patch therefore
also introduces a simple loop into isolate_migratepages() so that it does
not return immediately on failed pageblock checks, but keeps going until
isolate_migratepages_range() gets called once. Similarily to
isolate_freepages(), the function periodically checks if it needs to
reschedule or abort async compaction.
[iamjoonsoo.kim@lge.com: fix isolated page counting bug in compaction]
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Minchan Kim <minchan@kernel.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The compact_stall vmstat counter counts the number of allocations stalled
by direct compaction. It does not count when all attempted zones had
deferred compaction, but it does count when all zones skipped compaction.
The skipping is decided based on very early check of
compaction_suitable(), based on watermarks and memory fragmentation.
Therefore it makes sense not to count skipped compactions as stalls.
Moreover, compact_success or compact_fail is also already not being
counted when compaction was skipped, so this patch changes the
compact_stall counting to match the other two.
Additionally, restructure __alloc_pages_direct_compact() code for better
readability.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Minchan Kim <minchan@kernel.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Rik van Riel <riel@redhat.com>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When direct sync compaction is often unsuccessful, it may become deferred
for some time to avoid further useless attempts, both sync and async.
Successful high-order allocations un-defer compaction, while further
unsuccessful compaction attempts prolong the compaction deferred period.
Currently the checking and setting deferred status is performed only on
the preferred zone of the allocation that invoked direct compaction. But
compaction itself is attempted on all eligible zones in the zonelist, so
the behavior is suboptimal and may lead both to scenarios where 1)
compaction is attempted uselessly, or 2) where it's not attempted despite
good chances of succeeding, as shown on the examples below:
1) A direct compaction with Normal preferred zone failed and set
deferred compaction for the Normal zone. Another unrelated direct
compaction with DMA32 as preferred zone will attempt to compact DMA32
zone even though the first compaction attempt also included DMA32 zone.
In another scenario, compaction with Normal preferred zone failed to
compact Normal zone, but succeeded in the DMA32 zone, so it will not
defer compaction. In the next attempt, it will try Normal zone which
will fail again, instead of skipping Normal zone and trying DMA32
directly.
2) Kswapd will balance DMA32 zone and reset defer status based on
watermarks looking good. A direct compaction with preferred Normal
zone will skip compaction of all zones including DMA32 because Normal
was still deferred. The allocation might have succeeded in DMA32, but
won't.
This patch makes compaction deferring work on individual zone basis
instead of preferred zone. For each zone, it checks compaction_deferred()
to decide if the zone should be skipped. If watermarks fail after
compacting the zone, defer_compaction() is called. The zone where
watermarks passed can still be deferred when the allocation attempt is
unsuccessful. When allocation is successful, compaction_defer_reset() is
called for the zone containing the allocated page. This approach should
approximate calling defer_compaction() only on zones where compaction was
attempted and did not yield allocated page. There might be corner cases
but that is inevitable as long as the decision to stop compacting dues not
guarantee that a page will be allocated.
Due to a new COMPACT_DEFERRED return value, some functions relying
implicitly on COMPACT_SKIPPED = 0 had to be updated, with comments made
more accurate. The did_some_progress output parameter of
__alloc_pages_direct_compact() is removed completely, as the caller
actually does not use it after compaction sets it - it is only considered
when direct reclaim sets it.
During testing on a two-node machine with a single very small Normal zone
on node 1, this patch has improved success rates in stress-highalloc
mmtests benchmark. The success here were previously made worse by commit
3a025760fc ("mm: page_alloc: spill to remote nodes before waking
kswapd") as kswapd was no longer resetting often enough the deferred
compaction for the Normal zone, and DMA32 zones on both nodes were thus
not considered for compaction. On different machine, success rates were
improved with __GFP_NO_KSWAPD allocations.
[akpm@linux-foundation.org: fix CONFIG_COMPACTION=n build]
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Minchan Kim <minchan@kernel.org>
Reviewed-by: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The check for ALLOC_CMA in __alloc_pages_nodemask() derives migratetype
from gfp_mask in each retry pass, although the migratetype variable
already has the value determined and it does not change. Use the variable
and perform the check only once. Also convert #ifdef CONFIG_CMA to
IS_ENABLED.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Minchan Kim <minchan@kernel.org>
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>
Anton noticed (http://www.spinics.net/lists/linux-mm/msg67489.html) that
on ppc LPARs with memoryless nodes, a large amount of memory was consumed
by slabs and was marked unreclaimable. He tracked it down to slab
deactivations in the SLUB core when we allocate remotely, leading to poor
efficiency always when memoryless nodes are present.
After much discussion, Joonsoo provided a few patches that help
significantly. They don't resolve the problem altogether:
- memory hotplug still needs testing, that is when a memoryless node
becomes memory-ful, we want to dtrt
- there are other reasons for going off-node than memoryless nodes,
e.g., fully exhausted local nodes
Neither case is resolved with this series, but I don't think that should
block their acceptance, as they can be explored/resolved with follow-on
patches.
The series consists of:
[1/3] topology: add support for node_to_mem_node() to determine the
fallback node
[2/3] slub: fallback to node_to_mem_node() node if allocating on
memoryless node
- Joonsoo's patches to cache the nearest node with memory for each
NUMA node
[3/3] Partial revert of 81c98869fa (""kthread: ensure locality of
task_struct allocations")
- At Tejun's request, keep the knowledge of memoryless node fallback
to the allocator core.
This patch (of 3):
We need to determine the fallback node in slub allocator if the allocation
target node is memoryless node. Without it, the SLUB wrongly select the
node which has no memory and can't use a partial slab, because of node
mismatch. Introduced function, node_to_mem_node(X), will return a node Y
with memory that has the nearest distance. If X is memoryless node, it
will return nearest distance node, but, if X is normal node, it will
return itself.
We will use this function in following patch to determine the fallback
node.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Nishanth Aravamudan <nacc@linux.vnet.ibm.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Han Pingtian <hanpt@linux.vnet.ibm.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Anton Blanchard <anton@samba.org>
Cc: Christoph Lameter <cl@linux.com>
Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The zone allocation batches can easily underflow due to higher-order
allocations or spills to remote nodes. On SMP that's fine, because
underflows are expected from concurrency and dealt with by returning 0.
But on UP, zone_page_state will just return a wrapped unsigned long,
which will get past the <= 0 check and then consider the zone eligible
until its watermarks are hit.
Commit 3a025760fc ("mm: page_alloc: spill to remote nodes before
waking kswapd") already made the counter-resetting use
atomic_long_read() to accomodate underflows from remote spills, but it
didn't go all the way with it.
Make it clear that these batches are expected to go negative regardless
of concurrency, and use atomic_long_read() everywhere.
Fixes: 81c0a2bb51 ("mm: page_alloc: fair zone allocator policy")
Reported-by: Vlastimil Babka <vbabka@suse.cz>
Reported-by: Leon Romanovsky <leon@leon.nu>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: <stable@vger.kernel.org> [3.12+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The setup_node_data() function allocates a pg_data_t object,
inserts it into the node_data[] array and initializes the
following fields: node_id, node_start_pfn and
node_spanned_pages.
However, a few function calls later during the kernel boot,
free_area_init_node() re-initializes those fields, possibly with
setup_node_data() is not used.
This causes a small glitch when running Linux as a hyperv numa
guest:
SRAT: PXM 0 -> APIC 0x00 -> Node 0
SRAT: PXM 0 -> APIC 0x01 -> Node 0
SRAT: PXM 1 -> APIC 0x02 -> Node 1
SRAT: PXM 1 -> APIC 0x03 -> Node 1
SRAT: Node 0 PXM 0 [mem 0x00000000-0x7fffffff]
SRAT: Node 1 PXM 1 [mem 0x80200000-0xf7ffffff]
SRAT: Node 1 PXM 1 [mem 0x100000000-0x1081fffff]
NUMA: Node 1 [mem 0x80200000-0xf7ffffff] + [mem 0x100000000-0x1081fffff] -> [mem 0x80200000-0x1081fffff]
Initmem setup node 0 [mem 0x00000000-0x7fffffff]
NODE_DATA [mem 0x7ffdc000-0x7ffeffff]
Initmem setup node 1 [mem 0x80800000-0x1081fffff]
NODE_DATA [mem 0x1081ea000-0x1081fdfff]
crashkernel: memory value expected
[ffffea0000000000-ffffea0001ffffff] PMD -> [ffff88007de00000-ffff88007fdfffff] on node 0
[ffffea0002000000-ffffea00043fffff] PMD -> [ffff880105600000-ffff8801077fffff] on node 1
Zone ranges:
DMA [mem 0x00001000-0x00ffffff]
DMA32 [mem 0x01000000-0xffffffff]
Normal [mem 0x100000000-0x1081fffff]
Movable zone start for each node
Early memory node ranges
node 0: [mem 0x00001000-0x0009efff]
node 0: [mem 0x00100000-0x7ffeffff]
node 1: [mem 0x80200000-0xf7ffffff]
node 1: [mem 0x100000000-0x1081fffff]
On node 0 totalpages: 524174
DMA zone: 64 pages used for memmap
DMA zone: 21 pages reserved
DMA zone: 3998 pages, LIFO batch:0
DMA32 zone: 8128 pages used for memmap
DMA32 zone: 520176 pages, LIFO batch:31
On node 1 totalpages: 524288
DMA32 zone: 7672 pages used for memmap
DMA32 zone: 491008 pages, LIFO batch:31
Normal zone: 520 pages used for memmap
Normal zone: 33280 pages, LIFO batch:7
In this dmesg, the SRAT table reports that the memory range for
node 1 starts at 0x80200000. However, the line starting with
"Initmem" reports that node 1 memory range starts at 0x80800000.
The "Initmem" line is reported by setup_node_data() and is
wrong, because the kernel ends up using the range as reported in
the SRAT table.
This commit drops all that dead code from setup_node_data(),
renames it to alloc_node_data() and adds a printk() to
free_area_init_node() so that we report a node's memory range
accurately.
Here's the same dmesg section with this patch applied:
SRAT: PXM 0 -> APIC 0x00 -> Node 0
SRAT: PXM 0 -> APIC 0x01 -> Node 0
SRAT: PXM 1 -> APIC 0x02 -> Node 1
SRAT: PXM 1 -> APIC 0x03 -> Node 1
SRAT: Node 0 PXM 0 [mem 0x00000000-0x7fffffff]
SRAT: Node 1 PXM 1 [mem 0x80200000-0xf7ffffff]
SRAT: Node 1 PXM 1 [mem 0x100000000-0x1081fffff]
NUMA: Node 1 [mem 0x80200000-0xf7ffffff] + [mem 0x100000000-0x1081fffff] -> [mem 0x80200000-0x1081fffff]
NODE_DATA(0) allocated [mem 0x7ffdc000-0x7ffeffff]
NODE_DATA(1) allocated [mem 0x1081ea000-0x1081fdfff]
crashkernel: memory value expected
[ffffea0000000000-ffffea0001ffffff] PMD -> [ffff88007de00000-ffff88007fdfffff] on node 0
[ffffea0002000000-ffffea00043fffff] PMD -> [ffff880105600000-ffff8801077fffff] on node 1
Zone ranges:
DMA [mem 0x00001000-0x00ffffff]
DMA32 [mem 0x01000000-0xffffffff]
Normal [mem 0x100000000-0x1081fffff]
Movable zone start for each node
Early memory node ranges
node 0: [mem 0x00001000-0x0009efff]
node 0: [mem 0x00100000-0x7ffeffff]
node 1: [mem 0x80200000-0xf7ffffff]
node 1: [mem 0x100000000-0x1081fffff]
Initmem setup node 0 [mem 0x00001000-0x7ffeffff]
On node 0 totalpages: 524174
DMA zone: 64 pages used for memmap
DMA zone: 21 pages reserved
DMA zone: 3998 pages, LIFO batch:0
DMA32 zone: 8128 pages used for memmap
DMA32 zone: 520176 pages, LIFO batch:31
Initmem setup node 1 [mem 0x80200000-0x1081fffff]
On node 1 totalpages: 524288
DMA32 zone: 7672 pages used for memmap
DMA32 zone: 491008 pages, LIFO batch:31
Normal zone: 520 pages used for memmap
Normal zone: 33280 pages, LIFO batch:7
This commit was tested on a two node bare-metal NUMA machine and
Linux as a numa guest on hyperv and qemu/kvm.
PS: The wrong memory range reported by setup_node_data() seems to be
harmless in the current kernel because it's just not used. However,
that bad range is used in kernel 2.6.32 to initialize the old boot
memory allocator, which causes a crash during boot.
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
__GFP_NO_KSWAPD, once the way to determine if an allocation was for thp
or not, has gained more users. Their use is not necessarily wrong, they
are trying to do a memory allocation that can easily fail without
disturbing kswapd, so the bit has gained additional usecases.
This restructures the check to determine whether MIGRATE_SYNC_LIGHT
should be used for memory compaction in the page allocator. Rather than
testing solely for __GFP_NO_KSWAPD, test for all bits that must be set
for thp allocations.
This also moves the check to be done only after the page allocator is
aborted for deferred or contended memory compaction since setting
migration_mode for this case is pointless.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
try_set_zonelist_oom() and clear_zonelist_oom() are not named properly
to imply that they require locking semantics to avoid out_of_memory()
being reordered.
zone_scan_lock is required for both functions to ensure that there is
proper locking synchronization.
Rename try_set_zonelist_oom() to oom_zonelist_trylock() and rename
clear_zonelist_oom() to oom_zonelist_unlock() to imply there is proper
locking semantics.
At the same time, convert oom_zonelist_trylock() to return bool instead
of int since only success and failure are tested.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The fair zone allocation policy round-robins allocations between zones
within a node to avoid age inversion problems during reclaim. If the
first allocation fails, the batch counts are reset and a second attempt
made before entering the slow path.
One assumption made with this scheme is that batches expire at roughly
the same time and the resets each time are justified. This assumption
does not hold when zones reach their low watermark as the batches will
be consumed at uneven rates. Allocation failure due to watermark
depletion result in additional zonelist scans for the reset and another
watermark check before hitting the slowpath.
On UMA, the benefit is negligible -- around 0.25%. On 4-socket NUMA
machine it's variable due to the variability of measuring overhead with
the vmstat changes. The system CPU overhead comparison looks like
3.16.0-rc3 3.16.0-rc3 3.16.0-rc3
vanilla vmstat-v5 lowercost-v5
User 746.94 774.56 802.00
System 65336.22 32847.27 40852.33
Elapsed 27553.52 27415.04 27368.46
However it is worth noting that the overall benchmark still completed
faster and intuitively it makes sense to take as few passes as possible
through the zonelists.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The purpose of numa_zonelist_order=zone is to preserve lower zones for
use with 32-bit devices. If locality is preferred then the
numa_zonelist_order=node policy should be used.
Unfortunately, the fair zone allocation policy overrides this by
skipping zones on remote nodes until the lower one is found. While this
makes sense from a page aging and performance perspective, it breaks the
expected zonelist policy. This patch restores the expected behaviour
for zone-list ordering.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
zone->pages_scanned is a write-intensive cache line during page reclaim
and it's also updated during page free. Move the counter into vmstat to
take advantage of the per-cpu updates and do not update it in the free
paths unless necessary.
On a small UMA machine running tiobench the difference is marginal. On
a 4-node machine the overhead is more noticable. Note that automatic
NUMA balancing was disabled for this test as otherwise the system CPU
overhead is unpredictable.
3.16.0-rc3 3.16.0-rc3 3.16.0-rc3
vanillarearrange-v5 vmstat-v5
User 746.94 759.78 774.56
System 65336.22 58350.98 32847.27
Elapsed 27553.52 27282.02 27415.04
Note that the overhead reduction will vary depending on where exactly
pages are allocated and freed.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>