Patch series "reintroduce compaction feedback for OOM decisions".
After several people reported OOM's for order-2 allocations in 4.7 due
to Michal Hocko's OOM rework, he reverted the part that considered
compaction feedback [1] in the decisions to retry reclaim/compaction.
This was to provide a fix quickly for 4.8 rc and 4.7 stable series,
while mmotm had an almost complete solution that instead improved
compaction reliability.
This series completes the mmotm solution and reintroduces the compaction
feedback into OOM decisions. The first two patches restore the state of
mmotm before the temporary solution was merged, the last patch should be
the missing piece for reliability. The third patch restricts the
hardened compaction to non-costly orders, since costly orders don't
result in OOMs in the first place.
[1] http://marc.info/?i=20160822093249.GA14916%40dhcp22.suse.cz%3E
This patch (of 4):
Commit 6b4e3181d7 ("mm, oom: prevent premature OOM killer invocation
for high order request") was intended as a quick fix of OOM regressions
for 4.8 and stable 4.7.x kernels. For a better long-term solution, we
still want to consider compaction feedback, which should be possible
after some more improvements in the following patches.
This reverts commit 6b4e3181d7.
Link: http://lkml.kernel.org/r/20160906135258.18335-2-vbabka@suse.cz
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch is to improve the performance of swap cache operations when
the type of the swap device is not 0. Originally, the whole swap entry
value is used as the key of the swap cache, even though there is one
radix tree for each swap device. If the type of the swap device is not
0, the height of the radix tree of the swap cache will be increased
unnecessary, especially on 64bit architecture. For example, for a 1GB
swap device on the x86_64 architecture, the height of the radix tree of
the swap cache is 11. But if the offset of the swap entry is used as
the key of the swap cache, the height of the radix tree of the swap
cache is 4. The increased height causes unnecessary radix tree
descending and increased cache footprint.
This patch reduces the height of the radix tree of the swap cache via
using the offset of the swap entry instead of the whole swap entry value
as the key of the swap cache. In 32 processes sequential swap out test
case on a Xeon E5 v3 system with RAM disk as swap, the lock contention
for the spinlock of the swap cache is reduced from 20.15% to 12.19%,
when the type of the swap device is 1.
Use the whole swap entry as key,
perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.__add_to_swap_cache.add_to_swap_cache.add_to_swap.shrink_page_list: 10.37,
perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__remove_mapping.shrink_page_list.shrink_inactive_list.shrink_node_memcg: 9.78,
Use the swap offset as key,
perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.__add_to_swap_cache.add_to_swap_cache.add_to_swap.shrink_page_list: 6.25,
perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__remove_mapping.shrink_page_list.shrink_inactive_list.shrink_node_memcg: 5.94,
Link: http://lkml.kernel.org/r/1473270649-27229-1-git-send-email-ying.huang@intel.com
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Aaron Lu <aaron.lu@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The global zero page is used to satisfy an anonymous read fault. If
THP(Transparent HugePage) is enabled then the global huge zero page is
used. The global huge zero page uses an atomic counter for reference
counting and is allocated/freed dynamically according to its counter
value.
CPU time spent on that counter will greatly increase if there are a lot
of processes doing anonymous read faults. This patch proposes a way to
reduce the access to the global counter so that the CPU load can be
reduced accordingly.
To do this, a new flag of the mm_struct is introduced:
MMF_USED_HUGE_ZERO_PAGE. With this flag, the process only need to touch
the global counter in two cases:
1 The first time it uses the global huge zero page;
2 The time when mm_user of its mm_struct reaches zero.
Note that right now, the huge zero page is eligible to be freed as soon
as its last use goes away. With this patch, the page will not be
eligible to be freed until the exit of the last process from which it
was ever used.
And with the use of mm_user, the kthread is not eligible to use huge
zero page either. Since no kthread is using huge zero page today, there
is no difference after applying this patch. But if that is not desired,
I can change it to when mm_count reaches zero.
Case used for test on Haswell EP:
usemem -n 72 --readonly -j 0x200000 100G
Which spawns 72 processes and each will mmap 100G anonymous space and
then do read only access to that space sequentially with a step of 2MB.
CPU cycles from perf report for base commit:
54.03% usemem [kernel.kallsyms] [k] get_huge_zero_page
CPU cycles from perf report for this commit:
0.11% usemem [kernel.kallsyms] [k] mm_get_huge_zero_page
Performance(throughput) of the workload for base commit: 1784430792
Performance(throughput) of the workload for this commit: 4726928591
164% increase.
Runtime of the workload for base commit: 707592 us
Runtime of the workload for this commit: 303970 us
50% drop.
Link: http://lkml.kernel.org/r/fe51a88f-446a-4622-1363-ad1282d71385@intel.com
Signed-off-by: Aaron Lu <aaron.lu@intel.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Ebru Akagunduz <ebru.akagunduz@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Trying to walk all of virtual memory requires architecture specific
knowledge. On x86_64, addresses must be sign extended from bit 48,
whereas on arm64 the top VA_BITS of address space have their own set of
page tables.
clear_refs_write() calls walk_page_range() on the range 0 to ~0UL, it
provides a test_walk() callback that only expects to be walking over
VMAs. Currently walk_pmd_range() will skip memory regions that don't
have a VMA, reporting them as a hole.
As this call only expects to walk user address space, make it walk 0 to
'highest_vm_end'.
Link: http://lkml.kernel.org/r/1472655792-22439-1-git-send-email-james.morse@arm.com
Signed-off-by: James Morse <james.morse@arm.com>
Acked-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In current kernel code, we only call node_set_state(cpu_to_node(cpu),
N_CPU) when a cpu is hot plugged. But we do not set the node state for
N_CPU when the cpus are brought online during boot.
So this could lead to failure when we check to see if a node contains
cpu with node_state(node_id, N_CPU).
One use case is in the node_reclaime function:
/*
* Only run node reclaim on the local node or on nodes that do
* not
* have associated processors. This will favor the local
* processor
* over remote processors and spread off node memory allocations
* as wide as possible.
*/
if (node_state(pgdat->node_id, N_CPU) && pgdat->node_id !=
numa_node_id())
return NODE_RECLAIM_NOSCAN;
I instrumented the kernel to call this function after boot and it always
returns 0 on a x86 desktop machine until I apply the attached patch.
int num_cpu_node(void)
{
int i, nr_cpu_nodes = 0;
for_each_node(i) {
if (node_state(i, N_CPU))
++ nr_cpu_nodes;
}
return nr_cpu_nodes;
}
Fix this by checking each node for online CPU when we initialize
vmstat that's responsible for maintaining node state.
Link: http://lkml.kernel.org/r/20160829175922.GA21775@linux.intel.com
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tim Chen <tim.c.chen@linux.intel.com>
Cc: <Huang@linux.intel.com>
Cc: Ying <ying.huang@intel.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When CONFIG_FS_DAX_PMD is set, DAX supports mmap() using pmd page size.
This feature relies on both mmap virtual address and FS block (i.e.
physical address) to be aligned by the pmd page size. Users can use
mkfs options to specify FS to align block allocations. However,
aligning mmap address requires code changes to existing applications for
providing a pmd-aligned address to mmap().
For instance, fio with "ioengine=mmap" performs I/Os with mmap() [1].
It calls mmap() with a NULL address, which needs to be changed to
provide a pmd-aligned address for testing with DAX pmd mappings.
Changing all applications that call mmap() with NULL is undesirable.
Add thp_get_unmapped_area(), which can be called by filesystem's
get_unmapped_area to align an mmap address by the pmd size for a DAX
file. It calls the default handler, mm->get_unmapped_area(), to find a
range and then aligns it for a DAX file.
The patch is based on Matthew Wilcox's change that allows adding support
of the pud page size easily.
[1]: https://github.com/axboe/fio/blob/master/engines/mmap.c
Link: http://lkml.kernel.org/r/1472497881-9323-2-git-send-email-toshi.kani@hpe.com
Signed-off-by: Toshi Kani <toshi.kani@hpe.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Since the lumpy reclaim is gone there is no source of higher order pages
if CONFIG_COMPACTION=n except for the order-0 pages reclaim which is
unreliable for that purpose to say the least. Hitting an OOM for
!costly higher order requests is therefore all not that hard to imagine.
We are trying hard to not invoke OOM killer as much as possible but
there is simply no reliable way to detect whether more reclaim retries
make sense.
Disabling COMPACTION is not widespread but it seems that some users
might have disable the feature without realizing full consequences
(mostly along with disabling THP because compaction used to be THP
mainly thing). This patch just adds a note if the OOM killer was
triggered by higher order request with compaction disabled. This will
help us identifying possible misconfiguration right from the oom report
which is easier than to always keep in mind that somebody might have
disabled COMPACTION without a good reason.
Link: http://lkml.kernel.org/r/20160830111632.GD23963@dhcp22.suse.cz
Signed-off-by: Michal Hocko <mhocko@suse.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
File pages use a set of radix tree tags (DIRTY, TOWRITE, WRITEBACK,
etc.) to accelerate finding the pages with a specific tag in the radix
tree during inode writeback. But for anonymous pages in the swap cache,
there is no inode writeback. So there is no need to find the pages with
some writeback tags in the radix tree. It is not necessary to touch
radix tree writeback tags for pages in the swap cache.
Per Rik van Riel's suggestion, a new flag AS_NO_WRITEBACK_TAGS is
introduced for address spaces which don't need to update the writeback
tags. The flag is set for swap caches. It may be used for DAX file
systems, etc.
With this patch, the swap out bandwidth improved 22.3% (from ~1.2GB/s to
~1.48GBps) in the vm-scalability swap-w-seq test case with 8 processes.
The test is done on a Xeon E5 v3 system. The swap device used is a RAM
simulated PMEM (persistent memory) device. The improvement comes from
the reduced contention on the swap cache radix tree lock. To test
sequential swapping out, the test case uses 8 processes, which
sequentially allocate and write to the anonymous pages until RAM and
part of the swap device is used up.
Details of comparison is as follow,
base base+patch
---------------- --------------------------
%stddev %change %stddev
\ | \
2506952 ± 2% +28.1% 3212076 ± 7% vm-scalability.throughput
1207402 ± 7% +22.3% 1476578 ± 6% vmstat.swap.so
10.86 ± 12% -23.4% 8.31 ± 16% perf-profile.cycles-pp._raw_spin_lock_irq.__add_to_swap_cache.add_to_swap_cache.add_to_swap.shrink_page_list
10.82 ± 13% -33.1% 7.24 ± 14% perf-profile.cycles-pp._raw_spin_lock_irqsave.__remove_mapping.shrink_page_list.shrink_inactive_list.shrink_zone_memcg
10.36 ± 11% -100.0% 0.00 ± -1% perf-profile.cycles-pp._raw_spin_lock_irqsave.__test_set_page_writeback.bdev_write_page.__swap_writepage.swap_writepage
10.52 ± 12% -100.0% 0.00 ± -1% perf-profile.cycles-pp._raw_spin_lock_irqsave.test_clear_page_writeback.end_page_writeback.page_endio.pmem_rw_page
Link: http://lkml.kernel.org/r/1472578089-5560-1-git-send-email-ying.huang@intel.com
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Shaohua Li <shli@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Tejun Heo <tj@kernel.org>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently significant amount of memory is reserved only in kernel booted
to capture kernel dump using the fa_dump method.
Kernels compiled with CONFIG_DEFERRED_STRUCT_PAGE_INIT will initialize
only certain size memory per node. The certain size takes into account
the dentry and inode cache sizes. Currently the cache sizes are
calculated based on the total system memory including the reserved
memory. However such a kernel when booting the same kernel as fadump
kernel will not be able to allocate the required amount of memory to
suffice for the dentry and inode caches. This results in crashes like
Hence only implement arch_reserved_kernel_pages() for CONFIG_FA_DUMP
configurations. The amount reserved will be reduced while calculating
the large caches and will avoid crashes like the below on large systems
such as 32 TB systems.
Dentry cache hash table entries: 536870912 (order: 16, 4294967296 bytes)
vmalloc: allocation failure, allocated 4097114112 of 17179934720 bytes
swapper/0: page allocation failure: order:0, mode:0x2080020(GFP_ATOMIC)
CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.6-master+ #3
Call Trace:
dump_stack+0xb0/0xf0 (unreliable)
warn_alloc_failed+0x114/0x160
__vmalloc_node_range+0x304/0x340
__vmalloc+0x6c/0x90
alloc_large_system_hash+0x1b8/0x2c0
inode_init+0x94/0xe4
vfs_caches_init+0x8c/0x13c
start_kernel+0x50c/0x578
start_here_common+0x20/0xa8
Link: http://lkml.kernel.org/r/1472476010-4709-4-git-send-email-srikar@linux.vnet.ibm.com
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Suggested-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Cc: Hari Bathini <hbathini@linux.vnet.ibm.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Balbir Singh <bsingharora@gmail.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Vlastimil Babka
Huang Ying
Dan Williams
Reza Arbab
James Morse
Aaron Lu
Tim Chen
Toshi Kani
Simon Guo
Michal Hocko
zijun_hu
Srikar Dronamraju
Aneesh Kumar K.V