Since commit 077fcf116c ("mm/thp: allocate transparent hugepages on
local node"), we handle THP allocations on page fault in a special way -
for non-interleave memory policies, the allocation is only attempted on
the node local to the current CPU, if the policy's nodemask allows the
node.
This is motivated by the assumption that THP benefits cannot offset the
cost of remote accesses, so it's better to fallback to base pages on the
local node (which might still be available, while huge pages are not due
to fragmentation) than to allocate huge pages on a remote node.
The nodemask check prevents us from violating e.g. MPOL_BIND policies
where the local node is not among the allowed nodes. However, the
current implementation can still give surprising results for the
MPOL_PREFERRED policy when the preferred node is different than the
current CPU's local node.
In such case we should honor the preferred node and not use the local
node, which is what this patch does. If hugepage allocation on the
preferred node fails, we fall back to base pages and don't try other
nodes, with the same motivation as is done for the local node hugepage
allocations. The patch also moves the MPOL_INTERLEAVE check around to
simplify the hugepage specific test.
The difference can be demonstrated using in-tree transhuge-stress test
on the following 2-node machine where half memory on one node was
occupied to show the difference.
> numactl --hardware
available: 2 nodes (0-1)
node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 24 25 26 27 28 29 30 31 32 33 34 35
node 0 size: 7878 MB
node 0 free: 3623 MB
node 1 cpus: 12 13 14 15 16 17 18 19 20 21 22 23 36 37 38 39 40 41 42 43 44 45 46 47
node 1 size: 8045 MB
node 1 free: 7818 MB
node distances:
node 0 1
0: 10 21
1: 21 10
Before the patch:
> numactl -p0 -C0 ./transhuge-stress
transhuge-stress: 2.197 s/loop, 0.276 ms/page, 7249.168 MiB/s 7962 succeed, 0 failed, 1786 different pages
> numactl -p0 -C12 ./transhuge-stress
transhuge-stress: 2.962 s/loop, 0.372 ms/page, 5376.172 MiB/s 7962 succeed, 0 failed, 3873 different pages
Number of successful THP allocations corresponds to free memory on node 0 in
the first case and node 1 in the second case, i.e. -p parameter is ignored and
cpu binding "wins".
After the patch:
> numactl -p0 -C0 ./transhuge-stress
transhuge-stress: 2.183 s/loop, 0.274 ms/page, 7295.516 MiB/s 7962 succeed, 0 failed, 1760 different pages
> numactl -p0 -C12 ./transhuge-stress
transhuge-stress: 2.878 s/loop, 0.361 ms/page, 5533.638 MiB/s 7962 succeed, 0 failed, 1750 different pages
> numactl -p1 -C0 ./transhuge-stress
transhuge-stress: 4.628 s/loop, 0.581 ms/page, 3440.893 MiB/s 7962 succeed, 0 failed, 3918 different pages
The -p parameter is respected regardless of cpu binding.
> numactl -C0 ./transhuge-stress
transhuge-stress: 2.202 s/loop, 0.277 ms/page, 7230.003 MiB/s 7962 succeed, 0 failed, 1750 different pages
> numactl -C12 ./transhuge-stress
transhuge-stress: 3.020 s/loop, 0.379 ms/page, 5273.324 MiB/s 7962 succeed, 0 failed, 3916 different pages
Without -p parameter, hugepage restriction to CPU-local node works as before.
Fixes: 077fcf116c ("mm/thp: allocate transparent hugepages on local node")
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: <stable@vger.kernel.org> [4.0+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
NUMA balancing is meant to be disabled by default on UMA machines but
the check is using nr_node_ids (highest node) instead of
num_online_nodes (online nodes).
The consequences are that a UMA machine with a node ID of 1 or higher
will enable NUMA balancing. This will incur useless overhead due to
minor faults with the impact depending on the workload. These are the
impact on the stats when running a kernel build on a single node machine
whose node ID happened to be 1:
vanilla patched
NUMA base PTE updates 5113158 0
NUMA huge PMD updates 643 0
NUMA page range updates 5442374 0
NUMA hint faults 2109622 0
NUMA hint local faults 2109622 0
NUMA hint local percent 100 100
NUMA pages migrated 0 0
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: <stable@vger.kernel.org> [3.8+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 077fcf116c ("mm/thp: allocate transparent hugepages on local
node") restructured alloc_hugepage_vma() with the intent of only
allocating transparent hugepages locally when there was not an effective
interleave mempolicy.
alloc_pages_exact_node() does not limit the allocation to the single node,
however, but rather prefers it. This is because __GFP_THISNODE is not set
which would cause the node-local nodemask to be passed. Without it, only
a nodemask that prefers the local node is passed.
Fix this by passing __GFP_THISNODE and falling back to small pages when
the allocation fails.
Commit 9f1b868a13 ("mm: thp: khugepaged: add policy for finding target
node") suffers from a similar problem for khugepaged, which is also fixed.
Fixes: 077fcf116c ("mm/thp: allocate transparent hugepages on local node")
Fixes: 9f1b868a13 ("mm: thp: khugepaged: add policy for finding target node")
Signed-off-by: David Rientjes <rientjes@google.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Pravin Shelar <pshelar@nicira.com>
Cc: Jarno Rajahalme <jrajahalme@nicira.com>
Cc: Li Zefan <lizefan@huawei.com>
Cc: Greg Thelen <gthelen@google.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>
migrate_to_node() is intended to migrate a page from one source node to
a target node.
Today, migrate_to_node() could end up migrating to any node, not only
the target node. This is because the page migration allocator,
new_node_page() does not pass __GFP_THISNODE to
alloc_pages_exact_node(). This causes the target node to be preferred
but allows fallback to any other node in order of affinity.
Prevent this by allocating with __GFP_THISNODE. If memory is not
available, -ENOMEM will be returned as appropriate.
Signed-off-by: David Rientjes <rientjes@google.com>
Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
printk and friends can now format bitmaps using '%*pb[l]'. cpumask
and nodemask also provide cpumask_pr_args() and nodemask_pr_args()
respectively which can be used to generate the two printf arguments
necessary to format the specified cpu/nodemask.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
walk_page_range() silently skips vma having VM_PFNMAP set, which leads to
undesirable behaviour at client end (who called walk_page_range). For
example for pagemap_read(), when no callbacks are called against VM_PFNMAP
vma, pagemap_read() may prepare pagemap data for next virtual address
range at wrong index. That could confuse and/or break userspace
applications.
This patch avoid this misbehavior caused by vma(VM_PFNMAP) like follows:
- for pagemap_read() which has its own ->pte_hole(), call the ->pte_hole()
over vma(VM_PFNMAP),
- for clear_refs and queue_pages which have their own ->tests_walk,
just return 1 and skip vma(VM_PFNMAP). This is no problem because
these are not interested in hole regions,
- for other callers, just skip the vma(VM_PFNMAP) as a default behavior.
Signed-off-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Signed-off-by: Shiraz Hashim <shashim@codeaurora.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
queue_pages_range() does page table walking in its own way now, but there
is some code duplicate. This patch applies page table walker to reduce
lines of code.
queue_pages_range() has to do some precheck to determine whether we really
walk over the vma or just skip it. Now we have test_walk() callback in
mm_walk for this purpose, so we can do this replacement cleanly.
queue_pages_test_walk() depends on not only the current vma but also the
previous one, so queue_pages->prev is introduced to remember it.
Signed-off-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Pavel Emelyanov <xemul@parallels.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>
The previous commit ("mm/thp: Allocate transparent hugepages on local
node") introduced alloc_hugepage_vma() to mm/mempolicy.c to perform a
special policy for THP allocations. The function has the same interface
as alloc_pages_vma(), shares a lot of boilerplate code and a long
comment.
This patch merges the hugepage special case into alloc_pages_vma. The
extra if condition should be cheap enough price to pay. We also prevent
a (however unlikely) race with parallel mems_allowed update, which could
make hugepage allocation restart only within the fallback call to
alloc_hugepage_vma() and not reconsider the special rule in
alloc_hugepage_vma().
Also by making sure mpol_cond_put(pol) is always called before actual
allocation attempt, we can use a single exit path within the function.
Also update the comment for missing node parameter and obsolete reference
to mm_sem.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: David Rientjes <rientjes@google.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This make sure that we try to allocate hugepages from local node if
allowed by mempolicy. If we can't, we fallback to small page allocation
based on mempolicy. This is based on the observation that allocating
pages on local node is more beneficial than allocating hugepages on remote
node.
With this patch applied we may find transparent huge page allocation
failures if the current node doesn't have enough freee hugepages. Before
this patch such failures result in us retrying the allocation on other
nodes in the numa node mask.
[akpm@linux-foundation.org: fix comment, add CONFIG_TRANSPARENT_HUGEPAGE dependency]
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: David Rientjes <rientjes@google.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull vfs pile #3 from Al Viro:
"Assorted fixes and patches from the last cycle"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
[regression] chunk lost from bd9b51
vfs: make mounts and mountstats honor root dir like mountinfo does
vfs: cleanup show_mountinfo
init: fix read-write root mount
unfuck binfmt_misc.c (broken by commit e6084d4)
vm_area_operations: kill ->migrate()
new helper: iter_is_iovec()
move_extent_per_page(): get rid of unused w_flags
lustre: get rid of playing with ->fs
btrfs: filp_open() returns ERR_PTR() on failure, not NULL...
the only instance this method has ever grown was one in kernfs -
one that call ->migrate() of another vm_ops if it exists.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
PROT_NUMA VMAs are skipped to avoid problems distinguishing between
present, prot_none and special entries. MPOL_MF_LAZY is not visible from
userspace since commit a720094ded ("mm: mempolicy: Hide MPOL_NOOP and
MPOL_MF_LAZY from userspace for now") but it should still skip VMAs the
same way task_numa_work does.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: Rik van Riel <riel@redhat.com>
Acked-by: Hugh Dickins <hughd@google.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull cgroup fixes from Tejun Heo:
"Mostly fixes for the fallouts from the recent cgroup core changes.
The decoupled nature of cgroup dynamic hierarchy management
(hierarchies are created dynamically on mount but may or may not be
reused once unmounted depending on remaining usages) led to more
ugliness being added to kernfs.
Hopefully, this is the last of it"
* 'for-3.16-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
cpuset: break kernfs active protection in cpuset_write_resmask()
cgroup: fix a race between cgroup_mount() and cgroup_kill_sb()
kernfs: introduce kernfs_pin_sb()
cgroup: fix mount failure in a corner case
cpuset,mempolicy: fix sleeping function called from invalid context
cgroup: fix broken css_has_online_children()
When runing with the kernel(3.15-rc7+), the follow bug occurs:
[ 9969.258987] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:586
[ 9969.359906] in_atomic(): 1, irqs_disabled(): 0, pid: 160655, name: python
[ 9969.441175] INFO: lockdep is turned off.
[ 9969.488184] CPU: 26 PID: 160655 Comm: python Tainted: G A 3.15.0-rc7+ #85
[ 9969.581032] Hardware name: FUJITSU-SV PRIMEQUEST 1800E/SB, BIOS PRIMEQUEST 1000 Series BIOS Version 1.39 11/16/2012
[ 9969.706052] ffffffff81a20e60 ffff8803e941fbd0 ffffffff8162f523 ffff8803e941fd18
[ 9969.795323] ffff8803e941fbe0 ffffffff8109995a ffff8803e941fc58 ffffffff81633e6c
[ 9969.884710] ffffffff811ba5dc ffff880405c6b480 ffff88041fdd90a0 0000000000002000
[ 9969.974071] Call Trace:
[ 9970.003403] [<ffffffff8162f523>] dump_stack+0x4d/0x66
[ 9970.065074] [<ffffffff8109995a>] __might_sleep+0xfa/0x130
[ 9970.130743] [<ffffffff81633e6c>] mutex_lock_nested+0x3c/0x4f0
[ 9970.200638] [<ffffffff811ba5dc>] ? kmem_cache_alloc+0x1bc/0x210
[ 9970.272610] [<ffffffff81105807>] cpuset_mems_allowed+0x27/0x140
[ 9970.344584] [<ffffffff811b1303>] ? __mpol_dup+0x63/0x150
[ 9970.409282] [<ffffffff811b1385>] __mpol_dup+0xe5/0x150
[ 9970.471897] [<ffffffff811b1303>] ? __mpol_dup+0x63/0x150
[ 9970.536585] [<ffffffff81068c86>] ? copy_process.part.23+0x606/0x1d40
[ 9970.613763] [<ffffffff810bf28d>] ? trace_hardirqs_on+0xd/0x10
[ 9970.683660] [<ffffffff810ddddf>] ? monotonic_to_bootbased+0x2f/0x50
[ 9970.759795] [<ffffffff81068cf0>] copy_process.part.23+0x670/0x1d40
[ 9970.834885] [<ffffffff8106a598>] do_fork+0xd8/0x380
[ 9970.894375] [<ffffffff81110e4c>] ? __audit_syscall_entry+0x9c/0xf0
[ 9970.969470] [<ffffffff8106a8c6>] SyS_clone+0x16/0x20
[ 9971.030011] [<ffffffff81642009>] stub_clone+0x69/0x90
[ 9971.091573] [<ffffffff81641c29>] ? system_call_fastpath+0x16/0x1b
The cause is that cpuset_mems_allowed() try to take
mutex_lock(&callback_mutex) under the rcu_read_lock(which was hold in
__mpol_dup()). And in cpuset_mems_allowed(), the access to cpuset is
under rcu_read_lock, so in __mpol_dup, we can reduce the rcu_read_lock
protection region to protect the access to cpuset only in
current_cpuset_is_being_rebound(). So that we can avoid this bug.
This patch is a temporary solution that just addresses the bug
mentioned above, can not fix the long-standing issue about cpuset.mems
rebinding on fork():
"When the forker's task_struct is duplicated (which includes
->mems_allowed) and it races with an update to cpuset_being_rebound
in update_tasks_nodemask() then the task's mems_allowed doesn't get
updated. And the child task's mems_allowed can be wrong if the
cpuset's nodemask changes before the child has been added to the
cgroup's tasklist."
Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Acked-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: stable <stable@vger.kernel.org>
Now that 3.15 is released, this merges the 'next' branch into 'master',
bringing us to the normal situation where my 'master' branch is the
merge window.
* accumulated work in next: (6809 commits)
ufs: sb mutex merge + mutex_destroy
powerpc: update comments for generic idle conversion
cris: update comments for generic idle conversion
idle: remove cpu_idle() forward declarations
nbd: zero from and len fields in NBD_CMD_DISCONNECT.
mm: convert some level-less printks to pr_*
MAINTAINERS: adi-buildroot-devel is moderated
MAINTAINERS: add linux-api for review of API/ABI changes
mm/kmemleak-test.c: use pr_fmt for logging
fs/dlm/debug_fs.c: replace seq_printf by seq_puts
fs/dlm/lockspace.c: convert simple_str to kstr
fs/dlm/config.c: convert simple_str to kstr
mm: mark remap_file_pages() syscall as deprecated
mm: memcontrol: remove unnecessary memcg argument from soft limit functions
mm: memcontrol: clean up memcg zoneinfo lookup
mm/memblock.c: call kmemleak directly from memblock_(alloc|free)
mm/mempool.c: update the kmemleak stack trace for mempool allocations
lib/radix-tree.c: update the kmemleak stack trace for radix tree allocations
mm: introduce kmemleak_update_trace()
mm/kmemleak.c: use %u to print ->checksum
...
Vlastimil Babka
Mel Gorman
David Rientjes
Tejun Heo
Naoya Horiguchi
Aneesh Kumar K.V
Linus Torvalds
Zhihui Zhang
Al Viro
Oleg Nesterov
Gu Zheng
Hugh Dickins