After commit 839a8e8660 ("writeback: replace custom worker pool
implementation with unbound workqueue") when device is removed while we
are writing to it we crash in bdi_writeback_workfn() ->
set_worker_desc() because bdi->dev is NULL.
This can happen because even though bdi_unregister() cancels all pending
flushing work, nothing really prevents new ones from being queued from
balance_dirty_pages() or other places.
Fix the problem by clearing BDI_registered bit in bdi_unregister() and
checking it before scheduling of any flushing work.
Fixes: 839a8e8660
Reviewed-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Cc: Derek Basehore <dbasehore@chromium.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
bdi_wakeup_thread_delayed() used the mod_delayed_work() function to
schedule work to writeback dirty inodes. The problem with this is that
it can delay work that is scheduled for immediate execution, such as the
work from sync_inodes_sb(). This can happen since mod_delayed_work()
can now steal work from a work_queue. This fixes the problem by using
queue_delayed_work() instead. This is a regression caused by commit
839a8e8660 ("writeback: replace custom worker pool implementation with
unbound workqueue").
The reason that this causes a problem is that laptop-mode will change
the delay, dirty_writeback_centisecs, to 60000 (10 minutes) by default.
In the case that bdi_wakeup_thread_delayed() races with
sync_inodes_sb(), sync will be stopped for 10 minutes and trigger a hung
task. Even if dirty_writeback_centisecs is not long enough to cause a
hung task, we still don't want to delay sync for that long.
We fix the problem by using queue_delayed_work() when we want to
schedule writeback sometime in future. This function doesn't change the
timer if it is already armed.
For the same reason, we also change bdi_writeback_workfn() to
immediately queue the work again in the case that the work_list is not
empty. The same problem can happen if the sync work is run on the
rescue worker.
[jack@suse.cz: update changelog, add comment, use bdi_wakeup_thread_delayed()]
Signed-off-by: Derek Basehore <dbasehore@chromium.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Alexander Viro <viro@zento.linux.org.uk>
Reviewed-by: Tejun Heo <tj@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: "Darrick J. Wong" <darrick.wong@oracle.com>
Cc: Derek Basehore <dbasehore@chromium.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Benson Leung <bleung@chromium.org>
Cc: Sonny Rao <sonnyrao@chromium.org>
Cc: Luigi Semenzato <semenzato@chromium.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Dave Chinner <david@fromorbit.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull x86 vdso changes from Peter Anvin:
"This is the revamp of the 32-bit vdso and the associated cleanups.
This adds timekeeping support to the 32-bit vdso that we already have
in the 64-bit vdso. Although 32-bit x86 is legacy, it is likely to
remain in the embedded space for a very long time to come.
This removes the traditional COMPAT_VDSO support; the configuration
variable is reused for simply removing the 32-bit vdso, which will
produce correct results but obviously suffer a performance penalty.
Only one beta version of glibc was affected, but that version was
unfortunately included in one OpenSUSE release.
This is not the end of the vdso cleanups. Stefani and Andy have
agreed to continue work for the next kernel cycle; in fact Andy has
already produced another set of cleanups that came too late for this
cycle.
An incidental, but arguably important, change is that this ensures
that unused space in the VVAR page is properly zeroed. It wasn't
before, and would contain whatever garbage was left in memory by BIOS
or the bootloader. Since the VVAR page is accessible to user space
this had the potential of information leaks"
* 'x86-vdso-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (23 commits)
x86, vdso: Fix the symbol versions on the 32-bit vDSO
x86, vdso, build: Don't rebuild 32-bit vdsos on every make
x86, vdso: Actually discard the .discard sections
x86, vdso: Fix size of get_unmapped_area()
x86, vdso: Finish removing VDSO32_PRELINK
x86, vdso: Move more vdso definitions into vdso.h
x86: Load the 32-bit vdso in place, just like the 64-bit vdsos
x86, vdso32: handle 32 bit vDSO larger one page
x86, vdso32: Disable stack protector, adjust optimizations
x86, vdso: Zero-pad the VVAR page
x86, vdso: Add 32 bit VDSO time support for 64 bit kernel
x86, vdso: Add 32 bit VDSO time support for 32 bit kernel
x86, vdso: Patch alternatives in the 32-bit VDSO
x86, vdso: Introduce VVAR marco for vdso32
x86, vdso: Cleanup __vdso_gettimeofday()
x86, vdso: Replace VVAR(vsyscall_gtod_data) by gtod macro
x86, vdso: __vdso_clock_gettime() cleanup
x86, vdso: Revamp vclock_gettime.c
mm: Add new func _install_special_mapping() to mmap.c
x86, vdso: Make vsyscall_gtod_data handling x86 generic
...
Pull percpu changes from Tejun Heo:
"The percpu allocation is now popular enough for the extremely naive
range allocator to cause scalability issues.
The existing allocator linearly scanned the allocation map on both
alloc and free without making use of hint or anything. Al
reimplemented the range allocator so that it can use binary search
instead of linear scan during free and alloc path uses simple hinting
to avoid scanning in common cases. Combined, the new allocator
resolves the scalability issue percpu allocator was showing during
container benchmark workload"
* 'for-3.15' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu:
percpu: renew the max_contig if we merge the head and previous block
percpu: allocation size should be even
percpu: speed alloc_pcpu_area() up
percpu: store offsets instead of lengths in ->map[]
perpcu: fold pcpu_split_block() into the only caller
Pull s390 updates from Martin Schwidefsky:
"There are two memory management related changes, the CMMA support for
KVM to avoid swap-in of freed pages and the split page table lock for
the PMD level. These two come with common code changes in mm/.
A fix for the long standing theoretical TLB flush problem, this one
comes with a common code change in kernel/sched/.
Another set of changes is Heikos uaccess work, included is the initial
set of patches with more to come.
And fixes and cleanups as usual"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (36 commits)
s390/con3270: optionally disable auto update
s390/mm: remove unecessary parameter from pgste_ipte_notify
s390/mm: remove unnecessary parameter from gmap_do_ipte_notify
s390/mm: fixing comment so that parameter name match
s390/smp: limit number of cpus in possible cpu mask
hypfs: Add clarification for "weight_min" attribute
s390: update defconfigs
s390/ptrace: add support for PTRACE_SINGLEBLOCK
s390/perf: make print_debug_cf() static
s390/topology: Remove call to update_cpu_masks()
s390/compat: remove compat exec domain
s390: select CONFIG_TTY for use of tty in unconditional keyboard driver
s390/appldata_os: fix cpu array size calculation
s390/checksum: remove memset() within csum_partial_copy_from_user()
s390/uaccess: remove copy_from_user_real()
s390/sclp_early: Return correct HSA block count also for zero
s390: add some drivers/subsystems to the MAINTAINERS file
s390: improve debug feature usage
s390/airq: add support for irq ranges
s390/mm: enable split page table lock for PMD level
...
Pull s390 compat wrapper rework from Heiko Carstens:
"S390 compat system call wrapper simplification work.
The intention of this work is to get rid of all hand written assembly
compat system call wrappers on s390, which perform proper sign or zero
extension, or pointer conversion of compat system call parameters.
Instead all of this should be done with C code eg by using Al's
COMPAT_SYSCALL_DEFINEx() macro.
Therefore all common code and s390 specific compat system calls have
been converted to the COMPAT_SYSCALL_DEFINEx() macro.
In order to generate correct code all compat system calls may only
have eg compat_ulong_t parameters, but no unsigned long parameters.
Those patches which change parameter types from unsigned long to
compat_ulong_t parameters are separate in this series, but shouldn't
cause any harm.
The only compat system calls which intentionally have 64 bit
parameters (preadv64 and pwritev64) in support of the x86/32 ABI
haven't been changed, but are now only available if an architecture
defines __ARCH_WANT_COMPAT_SYS_PREADV64/PWRITEV64.
System calls which do not have a compat variant but still need proper
zero extension on s390, like eg "long sys_brk(unsigned long brk)" will
get a proper wrapper function with the new s390 specific
COMPAT_SYSCALL_WRAPx() macro:
COMPAT_SYSCALL_WRAP1(brk, unsigned long, brk);
which generates the following code (simplified):
asmlinkage long sys_brk(unsigned long brk);
asmlinkage long compat_sys_brk(long brk)
{
return sys_brk((u32)brk);
}
Given that the C file which contains all the COMPAT_SYSCALL_WRAP lines
includes both linux/syscall.h and linux/compat.h, it will generate
build errors, if the declaration of sys_brk() doesn't match, or if
there exists a non-matching compat_sys_brk() declaration.
In addition this will intentionally result in a link error if
somewhere else a compat_sys_brk() function exists, which probably
should have been used instead. Two more BUILD_BUG_ONs make sure the
size and type of each compat syscall parameter can be handled
correctly with the s390 specific macros.
I converted the compat system calls step by step to verify the
generated code is correct and matches the previous code. In fact it
did not always match, however that was always a bug in the hand
written asm code.
In result we get less code, less bugs, and much more sanity checking"
* 'compat' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (44 commits)
s390/compat: add copyright statement
compat: include linux/unistd.h within linux/compat.h
s390/compat: get rid of compat wrapper assembly code
s390/compat: build error for large compat syscall args
mm/compat: convert to COMPAT_SYSCALL_DEFINE with changing parameter types
kexec/compat: convert to COMPAT_SYSCALL_DEFINE with changing parameter types
net/compat: convert to COMPAT_SYSCALL_DEFINE with changing parameter types
ipc/compat: convert to COMPAT_SYSCALL_DEFINE with changing parameter types
fs/compat: convert to COMPAT_SYSCALL_DEFINE with changing parameter types
ipc/compat: convert to COMPAT_SYSCALL_DEFINE
fs/compat: convert to COMPAT_SYSCALL_DEFINE
security/compat: convert to COMPAT_SYSCALL_DEFINE
mm/compat: convert to COMPAT_SYSCALL_DEFINE
net/compat: convert to COMPAT_SYSCALL_DEFINE
kernel/compat: convert to COMPAT_SYSCALL_DEFINE
fs/compat: optional preadv64/pwrite64 compat system calls
ipc/compat_sys_msgrcv: change msgtyp type from long to compat_long_t
s390/compat: partial parameter conversion within syscall wrappers
s390/compat: automatic zero, sign and pointer conversion of syscalls
s390/compat: add sync_file_range and fallocate compat syscalls
...
Pull scheduler changes from Ingo Molnar:
"Bigger changes:
- sched/idle restructuring: they are WIP preparation for deeper
integration between the scheduler and idle state selection, by
Nicolas Pitre.
- add NUMA scheduling pseudo-interleaving, by Rik van Riel.
- optimize cgroup context switches, by Peter Zijlstra.
- RT scheduling enhancements, by Thomas Gleixner.
The rest is smaller changes, non-urgnt fixes and cleanups"
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (68 commits)
sched: Clean up the task_hot() function
sched: Remove double calculation in fix_small_imbalance()
sched: Fix broken setscheduler()
sparc64, sched: Remove unused sparc64_multi_core
sched: Remove unused mc_capable() and smt_capable()
sched/numa: Move task_numa_free() to __put_task_struct()
sched/fair: Fix endless loop in idle_balance()
sched/core: Fix endless loop in pick_next_task()
sched/fair: Push down check for high priority class task into idle_balance()
sched/rt: Fix picking RT and DL tasks from empty queue
trace: Replace hardcoding of 19 with MAX_NICE
sched: Guarantee task priority in pick_next_task()
sched/idle: Remove stale old file
sched: Put rq's sched_avg under CONFIG_FAIR_GROUP_SCHED
cpuidle/arm64: Remove redundant cpuidle_idle_call()
cpuidle/powernv: Remove redundant cpuidle_idle_call()
sched, nohz: Exclude isolated cores from load balancing
sched: Fix select_task_rq_fair() description comments
workqueue: Replace hardcoding of -20 and 19 with MIN_NICE and MAX_NICE
sys: Replace hardcoding of -20 and 19 with MIN_NICE and MAX_NICE
...
During pcpu_alloc_area(), we might merge the current head with the
previous block. Since we have calculated the max_contig using the
size of previous block before we skip it, and now we update the size
of previous block, so we should renew the max_contig.
Signed-off-by: Jianyu Zhan <nasa4836@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Add remove_linear_migration_ptes_from_nonlinear(), to fix an interesting
little include/linux/swapops.h:131 BUG_ON(!PageLocked) found by trinity:
indicating that remove_migration_ptes() failed to find one of the
migration entries that was temporarily inserted.
The problem comes from remap_file_pages()'s switch from vma_interval_tree
(good for inserting the migration entry) to i_mmap_nonlinear list (no good
for locating it again); but can only be a problem if the remap_file_pages()
range does not cover the whole of the vma (zap_pte() clears the range).
remove_migration_ptes() needs a file_nonlinear method to go down the
i_mmap_nonlinear list, applying linear location to look for migration
entries in those vmas too, just in case there was this race.
The file_nonlinear method does need rmap_walk_control.arg to do this;
but it never needed vma passed in - vma comes from its own iteration.
Reported-and-tested-by: Dave Jones <davej@redhat.com>
Reported-and-tested-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix some "Bad rss-counter state" reports on exit, arising from the
interaction between page migration and remap_file_pages(): zap_pte()
must count a migration entry when zapping it.
And yes, it is possible (though very unusual) to find an anon page or
swap entry in a VM_SHARED nonlinear mapping: coming from that horrid
get_user_pages(write, force) case which COWs even in a shared mapping.
Signed-off-by: Hugh Dickins <hughd@google.com>
Tested-by: Sasha Levin sasha.levin@oracle.com>
Tested-by: Dave Jones davej@redhat.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The _install_special_mapping() is the new base function for
install_special_mapping(). This function will return a pointer of the
created VMA or a error code in an ERR_PTR()
This new function will be needed by the for the vdso 32 bit support to map the
additonal vvar and hpet pages into the 32 bit address space. This will be done
with io_remap_pfn_range() and remap_pfn_range, which requieres a vm_area_struct.
Reviewed-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Stefani Seibold <stefani@seibold.net>
Link: http://lkml.kernel.org/r/1395094933-14252-3-git-send-email-stefani@seibold.net
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
723ad1d90b ("percpu: store offsets instead of lengths in ->map[]")
updated percpu area allocator to use the lowest bit, instead of sign,
to signify whether the area is occupied and forced min align to 2;
unfortunately, it forgot to force the allocation size to be even
causing malfunctions for the very rare odd-sized allocations.
Always force the allocations to be even sized.
tj: Wrote patch description.
Original-patch-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Tejun Heo <tj@kernel.org>
The help text for CONFIG_PGTABLE_MAPPING has an incorrect URL. While
we're at it, remove the unnecessary footnote notation.
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Acked-by: Minchan Kim <minchan@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We received several reports of bad page state when freeing CMA pages
previously allocated with alloc_contig_range:
BUG: Bad page state in process Binder_A pfn:63202
page:d21130b0 count:0 mapcount:1 mapping: (null) index:0x7dfbf
page flags: 0x40080068(uptodate|lru|active|swapbacked)
Based on the page state, it looks like the page was still in use. The
page flags do not make sense for the use case though. Further debugging
showed that despite alloc_contig_range returning success, at least one
page in the range still remained in the buddy allocator.
There is an issue with isolate_freepages_block. In strict mode (which
CMA uses), if any pages in the range cannot be isolated,
isolate_freepages_block should return failure 0. The current check
keeps track of the total number of isolated pages and compares against
the size of the range:
if (strict && nr_strict_required > total_isolated)
total_isolated = 0;
After taking the zone lock, if one of the pages in the range is not in
the buddy allocator, we continue through the loop and do not increment
total_isolated. If in the last iteration of the loop we isolate more
than one page (e.g. last page needed is a higher order page), the check
for total_isolated may pass and we fail to detect that a page was
skipped. The fix is to bail out if the loop immediately if we are in
strict mode. There's no benfit to continuing anyway since we need all
pages to be isolated. Additionally, drop the error checking based on
nr_strict_required and just check the pfn ranges. This matches with
what isolate_freepages_range does.
Signed-off-by: Laura Abbott <lauraa@codeaurora.org>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Mel Gorman <mgorman@suse.de>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Acked-by: Michal Nazarewicz <mina86@mina86.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
GFP_THISNODE is for callers that implement their own clever fallback to
remote nodes. It restricts the allocation to the specified node and
does not invoke reclaim, assuming that the caller will take care of it
when the fallback fails, e.g. through a subsequent allocation request
without GFP_THISNODE set.
However, many current GFP_THISNODE users only want the node exclusive
aspect of the flag, without actually implementing their own fallback or
triggering reclaim if necessary. This results in things like page
migration failing prematurely even when there is easily reclaimable
memory available, unless kswapd happens to be running already or a
concurrent allocation attempt triggers the necessary reclaim.
Convert all callsites that don't implement their own fallback strategy
to __GFP_THISNODE. This restricts the allocation a single node too, but
at the same time allows the allocator to enter the slowpath, wake
kswapd, and invoke direct reclaim if necessary, to make the allocation
happen when memory is full.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Jan Stancek <jstancek@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If we know that first N areas are all in use, we can obviously skip
them when searching for a free one. And that kind of hint is very
easy to maintain.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Tejun Heo <tj@kernel.org>
Current code keeps +-length for each area in chunk->map[]. It has
several unpleasant consequences:
* even if we know that first 50 areas are all in use, allocation
still needs to go through all those areas just to sum their sizes, just
to get the offset of free one.
* freeing needs to find the array entry refering to the area
in question; again, the need to sum the sizes until we reach the offset
we are interested in. Note that offsets are monotonous, so simple
binary search would do here.
New data representation: array of <offset,in-use flag> pairs.
Each pair is represented by one int - we use offset|1 for <offset, in use>
and offset for <offset, free> (we make sure that all offsets are even).
In the end we put a sentry entry - <total size, in use>. The first
entry is <0, flag>; it would be possible to store together the flag
for Nth area and offset for N+1st, but that leads to much hairier code.
In other words, where the old variant would have
4, -8, -4, 4, -12, 100
(4 bytes free, 8 in use, 4 in use, 4 free, 12 in use, 100 free) we store
<0,0>, <4,1>, <12,1>, <16,0>, <20,1>, <32,0>, <132,1>
i.e.
0, 5, 13, 16, 21, 32, 133
This commit switches to new data representation and takes care of a couple
of low-hanging fruits in free_pcpu_area() - one is the switch to binary
search, another is not doing two memmove() when one would do. Speeding
the alloc side up (by keeping track of how many areas in the beginning are
known to be all in use) also becomes possible - that'll be done in the next
commit.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Tejun Heo <tj@kernel.org>
... and simplify the results a bit. Makes the next step easier
to deal with - we will be changing the data representation for
chunk->map[] and it's easier to do if the code in question is
not split between pcpu_alloc_area() and pcpu_split_block().
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Tejun Heo <tj@kernel.org>
In order to allow the COMPAT_SYSCALL_DEFINE macro generate code that
performs proper zero and sign extension convert all 64 bit parameters
to their corresponding 32 bit compat counterparts.
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Convert all compat system call functions where all parameter types
have a size of four or less than four bytes, or are pointer types
to COMPAT_SYSCALL_DEFINE.
The implicit casts within COMPAT_SYSCALL_DEFINE will perform proper
zero and sign extension to 64 bit of all parameters if needed.
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Jan Stancek reports manual page migration encountering allocation
failures after some pages when there is still plenty of memory free, and
bisected the problem down to commit 81c0a2bb51 ("mm: page_alloc: fair
zone allocator policy").
The problem is that GFP_THISNODE obeys the zone fairness allocation
batches on one hand, but doesn't reset them and wake kswapd on the other
hand. After a few of those allocations, the batches are exhausted and
the allocations fail.
Fixing this means either having GFP_THISNODE wake up kswapd, or
GFP_THISNODE not participating in zone fairness at all. The latter
seems safer as an acute bugfix, we can clean up later.
Reported-by: Jan Stancek <jstancek@redhat.com>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Rik van Riel <riel@redhat.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: <stable@kernel.org> [3.12+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Daniel Borkmann reported a VM_BUG_ON assertion failing:
------------[ cut here ]------------
kernel BUG at mm/mlock.c:528!
invalid opcode: 0000 [#1] SMP
Modules linked in: ccm arc4 iwldvm [...]
video
CPU: 3 PID: 2266 Comm: netsniff-ng Not tainted 3.14.0-rc2+ #8
Hardware name: LENOVO 2429BP3/2429BP3, BIOS G4ET37WW (1.12 ) 05/29/2012
task: ffff8801f87f9820 ti: ffff88002cb44000 task.ti: ffff88002cb44000
RIP: 0010:[<ffffffff81171ad0>] [<ffffffff81171ad0>] munlock_vma_pages_range+0x2e0/0x2f0
Call Trace:
do_munmap+0x18f/0x3b0
vm_munmap+0x41/0x60
SyS_munmap+0x22/0x30
system_call_fastpath+0x1a/0x1f
RIP munlock_vma_pages_range+0x2e0/0x2f0
---[ end trace a0088dcf07ae10f2 ]---
because munlock_vma_pages_range() thinks it's unexpectedly in the middle
of a THP page. This can be reproduced with default config since 3.11
kernels. A reproducer can be found in the kernel's selftest directory
for networking by running ./psock_tpacket.
The problem is that an order=2 compound page (allocated by
alloc_one_pg_vec_page() is part of the munlocked VM_MIXEDMAP vma (mapped
by packet_mmap()) and mistaken for a THP page and assumed to be order=9.
The checks for THP in munlock came with commit ff6a6da60b ("mm:
accelerate munlock() treatment of THP pages"), i.e. since 3.9, but did
not trigger a bug. It just makes munlock_vma_pages_range() skip such
compound pages until the next 512-pages-aligned page, when it encounters
a head page. This is however not a problem for vma's where mlocking has
no effect anyway, but it can distort the accounting.
Since commit 7225522bb4 ("mm: munlock: batch non-THP page isolation
and munlock+putback using pagevec") this can trigger a VM_BUG_ON in
PageTransHuge() check.
This patch fixes the issue by adding VM_MIXEDMAP flag to VM_SPECIAL, a
list of flags that make vma's non-mlockable and non-mergeable. The
reasoning is that VM_MIXEDMAP vma's are similar to VM_PFNMAP, which is
already on the VM_SPECIAL list, and both are intended for non-LRU pages
where mlocking makes no sense anyway. Related Lkml discussion can be
found in [2].
[1] tools/testing/selftests/net/psock_tpacket
[2] https://lkml.org/lkml/2014/1/10/427
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Reported-by: Daniel Borkmann <dborkman@redhat.com>
Tested-by: Daniel Borkmann <dborkman@redhat.com>
Cc: Thomas Hellstrom <thellstrom@vmware.com>
Cc: John David Anglin <dave.anglin@bell.net>
Cc: HATAYAMA Daisuke <d.hatayama@jp.fujitsu.com>
Cc: Konstantin Khlebnikov <khlebnikov@openvz.org>
Cc: Carsten Otte <cotte@de.ibm.com>
Cc: Jared Hulbert <jaredeh@gmail.com>
Tested-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: <stable@vger.kernel.org> [3.11.x+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Sometimes the cleanup after memcg hierarchy testing gets stuck in
mem_cgroup_reparent_charges(), unable to bring non-kmem usage down to 0.
There may turn out to be several causes, but a major cause is this: the
workitem to offline parent can get run before workitem to offline child;
parent's mem_cgroup_reparent_charges() circles around waiting for the
child's pages to be reparented to its lrus, but it's holding
cgroup_mutex which prevents the child from reaching its
mem_cgroup_reparent_charges().
Further testing showed that an ordered workqueue for cgroup_destroy_wq
is not always good enough: percpu_ref_kill_and_confirm's call_rcu_sched
stage on the way can mess up the order before reaching the workqueue.
Instead, when offlining a memcg, call mem_cgroup_reparent_charges() on
all its children (and grandchildren, in the correct order) to have their
charges reparented first.
Fixes: e5fca243ab ("cgroup: use a dedicated workqueue for cgroup destruction")
Signed-off-by: Filipe Brandenburger <filbranden@google.com>
Signed-off-by: Hugh Dickins <hughd@google.com>
Reviewed-by: Tejun Heo <tj@kernel.org>
Acked-by: Michal Hocko <mhocko@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: <stable@vger.kernel.org> [v3.10+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 0eef615665 ("memcg: fix css reference leak and endless loop in
mem_cgroup_iter") got the interaction with the commit a few before it
d8ad305597 ("mm/memcg: iteration skip memcgs not yet fully
initialized") slightly wrong, and we didn't notice at the time.
It's elusive, and harder to get than the original, but for a couple of
days before rc1, I several times saw a endless loop similar to that
supposedly being fixed.
This time it was a tighter loop in __mem_cgroup_iter_next(): because we
can get here when our root has already been offlined, and the ordering
of conditions was such that we then just cycled around forever.
Fixes: 0eef615665 ("memcg: fix css reference leak and endless loop in mem_cgroup_iter").
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Greg Thelen <gthelen@google.com>
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>
Commit bf6bddf192 ("mm: introduce compaction and migration for
ballooned pages") introduces page_count(page) into memory compaction
which dereferences page->first_page if PageTail(page).
This results in a very rare NULL pointer dereference on the
aforementioned page_count(page). Indeed, anything that does
compound_head(), including page_count() is susceptible to racing with
prep_compound_page() and seeing a NULL or dangling page->first_page
pointer.
This patch uses Andrea's implementation of compound_trans_head() that
deals with such a race and makes it the default compound_head()
implementation. This includes a read memory barrier that ensures that
if PageTail(head) is true that we return a head page that is neither
NULL nor dangling. The patch then adds a store memory barrier to
prep_compound_page() to ensure page->first_page is set.
This is the safest way to ensure we see the head page that we are
expecting, PageTail(page) is already in the unlikely() path and the
memory barriers are unfortunately required.
Hugetlbfs is the exception, we don't enforce a store memory barrier
during init since no race is possible.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Holger Kiehl <Holger.Kiehl@dwd.de>
Cc: Christoph Lameter <cl@linux.com>
Cc: Rafael Aquini <aquini@redhat.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Kirill has reported the following:
Task in /test killed as a result of limit of /test
memory: usage 10240kB, limit 10240kB, failcnt 51
memory+swap: usage 10240kB, limit 10240kB, failcnt 0
kmem: usage 0kB, limit 18014398509481983kB, failcnt 0
Memory cgroup stats for /test:
BUG: sleeping function called from invalid context at kernel/cpu.c:68
in_atomic(): 1, irqs_disabled(): 0, pid: 66, name: memcg_test
2 locks held by memcg_test/66:
#0: (memcg_oom_lock#2){+.+...}, at: [<ffffffff81131014>] pagefault_out_of_memory+0x14/0x90
#1: (oom_info_lock){+.+...}, at: [<ffffffff81197b2a>] mem_cgroup_print_oom_info+0x2a/0x390
CPU: 2 PID: 66 Comm: memcg_test Not tainted 3.14.0-rc1-dirty #745
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS Bochs 01/01/2011
Call Trace:
__might_sleep+0x16a/0x210
get_online_cpus+0x1c/0x60
mem_cgroup_read_stat+0x27/0xb0
mem_cgroup_print_oom_info+0x260/0x390
dump_header+0x88/0x251
? trace_hardirqs_on+0xd/0x10
oom_kill_process+0x258/0x3d0
mem_cgroup_oom_synchronize+0x656/0x6c0
? mem_cgroup_charge_common+0xd0/0xd0
pagefault_out_of_memory+0x14/0x90
mm_fault_error+0x91/0x189
__do_page_fault+0x48e/0x580
do_page_fault+0xe/0x10
page_fault+0x22/0x30
which complains that mem_cgroup_read_stat cannot be called from an atomic
context but mem_cgroup_print_oom_info takes a spinlock. Change
oom_info_lock to a mutex.
This was introduced by 947b3dd1a8 ("memcg, oom: lock
mem_cgroup_print_oom_info").
Signed-off-by: Michal Hocko <mhocko@suse.cz>
Reported-by: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Masayoshi Mizuma reported a bug with the hang of an application under
the memcg limit. It happens on write-protection fault to huge zero page
If we successfully allocate a huge page to replace zero page but hit the
memcg limit we need to split the zero page with split_huge_page_pmd()
and fallback to small pages.
The other part of the problem is that VM_FAULT_OOM has special meaning
in do_huge_pmd_wp_page() context. __handle_mm_fault() expects the page
to be split if it sees VM_FAULT_OOM and it will will retry page fault
handling. This causes an infinite loop if the page was not split.
do_huge_pmd_wp_zero_page_fallback() can return VM_FAULT_OOM if it failed
to allocate one small page, so fallback to small pages will not help.
The solution for this part is to replace VM_FAULT_OOM with
VM_FAULT_FALLBACK is fallback required.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reported-by: Masayoshi Mizuma <m.mizuma@jp.fujitsu.com>
Reviewed-by: Michal Hocko <mhocko@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: David Rientjes <rientjes@google.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 a virtualized environment and given an appropriate interface the guest
can mark pages as unused while they are free (for the s390 implementation
see git commit 45e576b1c3 "guest page hinting light"). For the host
the unused state is a property of the pte.
This patch adds the primitive 'pte_unused' and code to the host swap out
handler so that pages marked as unused by all mappers are not swapped out
but discarded instead, thus saving one IO for swap out and potentially
another one for swap in.
[ Martin Schwidefsky: patch reordering and simplification ]
Signed-off-by: Konstantin Weitz <konstantin.weitz@gmail.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
The finish_arch_post_lock_switch is called at the end of the task
switch after all locks have been released. In concept it is paired
with the switch_mm function, but the current code only does the
call in finish_task_switch. Add the call to idle_task_exit and
use_mm. One use case for the additional calls is s390 which will
use finish_arch_post_lock_switch to wait for the completion of
TLB flush operations.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Pull cgroup fixes from Tejun Heo:
"Quite a few fixes this time.
Three locking fixes, all marked for -stable. A couple error path
fixes and some misc fixes. Hugh found a bug in memcg offlining
sequence and we thought we could fix that from cgroup core side but
that turned out to be insufficient and got reverted. A different fix
has been applied to -mm"
* 'for-3.14-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
cgroup: update cgroup_enable_task_cg_lists() to grab siglock
Revert "cgroup: use an ordered workqueue for cgroup destruction"
cgroup: protect modifications to cgroup_idr with cgroup_mutex
cgroup: fix locking in cgroup_cfts_commit()
cgroup: fix error return from cgroup_create()
cgroup: fix error return value in cgroup_mount()
cgroup: use an ordered workqueue for cgroup destruction
nfs: include xattr.h from fs/nfs/nfs3proc.c
cpuset: update MAINTAINERS entry
arm, pm, vmpressure: add missing slab.h includes
Pull powerpc fixes from Ben Herrenschmidt:
"Here are some more powerpc fixes for 3.14
The main one is a nasty issue with the NUMA balancing support which
requires a small generic change and the addition of a new accessor to
set _PAGE_NUMA. Both have been reviewed and acked by Mel and Rik.
The changelog should have plenty of details but basically, without
this fix, we get random user segfaults and/or corruptions due to
missing TLB/hash flushes. Aneesh series of 3 patches fixes it.
We have some vDSO vs. perf fixes from Anton, some small EEH fixes
from Gavin, a ppc32 regression vs the stack overflow detector, and a
fix for the way we handle PCIe host bridge speed settings on pseries
(which is needed for proper operations of AMD graphics cards on
Power8)"
* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:
powerpc/eeh: Disable EEH on reboot
powerpc/eeh: Cleanup on eeh_subsystem_enabled
powerpc/powernv: Rework EEH reset
powerpc: Use unstripped VDSO image for more accurate profiling data
powerpc: Link VDSOs at 0x0
mm: Use ptep/pmdp_set_numa() for updating _PAGE_NUMA bit
mm: Dirty accountable change only apply to non prot numa case
powerpc/mm: Add new "set" flag argument to pte/pmd update function
powerpc/pseries: Add Gen3 definitions for PCIE link speed
powerpc/pseries: Fix regression on PCI link speed
powerpc: Set the correct ksp_limit on ppc32 when switching to irq stack
Archs like ppc64 doesn't do tlb flush in set_pte/pmd functions when using
a hash table MMU for various reasons (the flush is handled as part of
the PTE modification when necessary).
ppc64 thus doesn't implement flush_tlb_range for hash based MMUs.
Additionally ppc64 require the tlb flushing to be batched within ptl locks.
The reason to do that is to ensure that the hash page table is in sync with
linux page table.
We track the hpte index in linux pte and if we clear them without flushing
hash and drop the ptl lock, we can have another cpu update the pte and can
end up with duplicate entry in the hash table, which is fatal.
We also want to keep set_pte_at simpler by not requiring them to do hash
flush for performance reason. We do that by assuming that set_pte_at() is
never *ever* called on a PTE that is already valid.
This was the case until the NUMA code went in which broke that assumption.
Fix that by introducing a new pair of helpers to set _PAGE_NUMA in a
way similar to ptep/pmdp_set_wrprotect(), with a generic implementation
using set_pte_at() and a powerpc specific one using the appropriate
mechanism needed to keep the hash table in sync.
Acked-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
So move it within the if loop
Acked-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
mce-test detected a test failure when injecting error to a thp tail
page. This is because we take page refcount of the tail page in
madvise_hwpoison() while the fix in commit a3e0f9e47d
("mm/memory-failure.c: transfer page count from head page to tail page
after split thp") assumes that we always take refcount on the head page.
When a real memory error happens we take refcount on the head page where
memory_failure() is called without MF_COUNT_INCREASED set, so it seems
to me that testing memory error on thp tail page using madvise makes
little sense.
This patch cancels moving refcount in !MF_COUNT_INCREASED for valid
testing.
[akpm@linux-foundation.org: s/&&/&/]
Signed-off-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com>
Cc: Chen Gong <gong.chen@linux.intel.com>
Cc: <stable@vger.kernel.org> [3.9+: a3e0f9e47d]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Vladimir reported the following issue:
Commit c65c1877bd ("slub: use lockdep_assert_held") requires
remove_partial() to be called with n->list_lock held, but free_partial()
called from kmem_cache_close() on cache destruction does not follow this
rule, leading to a warning:
WARNING: CPU: 0 PID: 2787 at mm/slub.c:1536 __kmem_cache_shutdown+0x1b2/0x1f0()
Modules linked in:
CPU: 0 PID: 2787 Comm: modprobe Tainted: G W 3.14.0-rc1-mm1+ #1
Hardware name:
0000000000000600 ffff88003ae1dde8 ffffffff816d9583 0000000000000600
0000000000000000 ffff88003ae1de28 ffffffff8107c107 0000000000000000
ffff880037ab2b00 ffff88007c240d30 ffffea0001ee5280 ffffea0001ee52a0
Call Trace:
__kmem_cache_shutdown+0x1b2/0x1f0
kmem_cache_destroy+0x43/0xf0
xfs_destroy_zones+0x103/0x110 [xfs]
exit_xfs_fs+0x38/0x4e4 [xfs]
SyS_delete_module+0x19a/0x1f0
system_call_fastpath+0x16/0x1b
His solution was to add a spinlock in order to quiet lockdep. Although
there would be no contention to adding the lock, that lock also requires
disabling of interrupts which will have a larger impact on the system.
Instead of adding a spinlock to a location where it is not needed for
lockdep, make a __remove_partial() function that does not test if the
list_lock is held, as no one should have it due to it being freed.
Also added a __add_partial() function that does not do the lock
validation either, as it is not needed for the creation of the cache.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Reported-by: Vladimir Davydov <vdavydov@parallels.com>
Suggested-by: David Rientjes <rientjes@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Vladimir Davydov <vdavydov@parallels.com>
Acked-by: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit c65c1877bd ("slub: use lockdep_assert_held") incorrectly
required that add_full() and remove_full() hold n->list_lock. The lock
is only taken when kmem_cache_debug(s), since that's the only time it
actually does anything.
Require that the lock only be taken under such a condition.
Reported-by: Larry Finger <Larry.Finger@lwfinger.net>
Tested-by: Larry Finger <Larry.Finger@lwfinger.net>
Tested-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull vfs fixes from Al Viro:
"A couple of fixes, both -stable fodder. The O_SYNC bug is fairly
old..."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
fix a kmap leak in virtio_console
fix O_SYNC|O_APPEND syncing the wrong range on write()
It actually goes back to 2004 ([PATCH] Concurrent O_SYNC write support)
when sync_page_range() had been introduced; generic_file_write{,v}() correctly
synced
pos_after_write - written .. pos_after_write - 1
but generic_file_aio_write() synced
pos_before_write .. pos_before_write + written - 1
instead. Which is not the same thing with O_APPEND, obviously.
A couple of years later correct variant had been killed off when
everything switched to use of generic_file_aio_write().
All users of generic_file_aio_write() are affected, and the same bug
has been copied into other instances of ->aio_write().
The fix is trivial; the only subtle point is that generic_write_sync()
ought to be inlined to avoid calculations useless for the majority of
calls.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull x86 fixes from Peter Anvin:
"Quite a varied little collection of fixes. Most of them are
relatively small or isolated; the biggest one is Mel Gorman's fixes
for TLB range flushing.
A couple of AMD-related fixes (including not crashing when given an
invalid microcode image) and fix a crash when compiled with gcov"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86, microcode, AMD: Unify valid container checks
x86, hweight: Fix BUG when booting with CONFIG_GCOV_PROFILE_ALL=y
x86/efi: Allow mapping BGRT on x86-32
x86: Fix the initialization of physnode_map
x86, cpu hotplug: Fix stack frame warning in check_irq_vectors_for_cpu_disable()
x86/intel/mid: Fix X86_INTEL_MID dependencies
arch/x86/mm/srat: Skip NUMA_NO_NODE while parsing SLIT
mm, x86: Revisit tlb_flushall_shift tuning for page flushes except on IvyBridge
x86: mm: change tlb_flushall_shift for IvyBridge
x86/mm: Eliminate redundant page table walk during TLB range flushing
x86/mm: Clean up inconsistencies when flushing TLB ranges
mm, x86: Account for TLB flushes only when debugging
x86/AMD/NB: Fix amd_set_subcaches() parameter type
x86/quirks: Add workaround for AMD F16h Erratum792
x86, doc, kconfig: Fix dud URL for Microcode data
During aio stress test, we observed the following lockdep warning. This
mean AIO+numa_balancing is currently deadlockable.
The problem is, aio_migratepage disable interrupt, but
__set_page_dirty_nobuffers unintentionally enable it again.
Generally, all helper function should use spin_lock_irqsave() instead of
spin_lock_irq() because they don't know caller at all.
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(&(&ctx->completion_lock)->rlock);
<Interrupt>
lock(&(&ctx->completion_lock)->rlock);
*** DEADLOCK ***
dump_stack+0x19/0x1b
print_usage_bug+0x1f7/0x208
mark_lock+0x21d/0x2a0
mark_held_locks+0xb9/0x140
trace_hardirqs_on_caller+0x105/0x1d0
trace_hardirqs_on+0xd/0x10
_raw_spin_unlock_irq+0x2c/0x50
__set_page_dirty_nobuffers+0x8c/0xf0
migrate_page_copy+0x434/0x540
aio_migratepage+0xb1/0x140
move_to_new_page+0x7d/0x230
migrate_pages+0x5e5/0x700
migrate_misplaced_page+0xbc/0xf0
do_numa_page+0x102/0x190
handle_pte_fault+0x241/0x970
handle_mm_fault+0x265/0x370
__do_page_fault+0x172/0x5a0
do_page_fault+0x1a/0x70
page_fault+0x28/0x30
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Larry Woodman <lwoodman@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Johannes Weiner <jweiner@redhat.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
swapoff clear swap_info's SWP_USED flag prematurely and free its
resources after that. A concurrent swapon will reuse this swap_info
while its previous resources are not cleared completely.
These late freed resources are:
- p->percpu_cluster
- swap_cgroup_ctrl[type]
- block_device setting
- inode->i_flags &= ~S_SWAPFILE
This patch clears the SWP_USED flag after all its resources are freed,
so that swapon can reuse this swap_info by alloc_swap_info() safely.
[akpm@linux-foundation.org: tidy up code comment]
Signed-off-by: Weijie Yang <weijie.yang@samsung.com>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is a patch to improve swap readahead algorithm. It's from Hugh and
I slightly changed it.
Hugh's original changelog:
swapin readahead does a blind readahead, whether or not the swapin is
sequential. This may be ok on harddisk, because large reads have
relatively small costs, and if the readahead pages are unneeded they can
be reclaimed easily - though, what if their allocation forced reclaim of
useful pages? But on SSD devices large reads are more expensive than
small ones: if the readahead pages are unneeded, reading them in caused
significant overhead.
This patch adds very simplistic random read detection. Stealing the
PageReadahead technique from Konstantin Khlebnikov's patch, avoiding the
vma/anon_vma sophistications of Shaohua Li's patch, swapin_nr_pages()
simply looks at readahead's current success rate, and narrows or widens
its readahead window accordingly. There is little science to its
heuristic: it's about as stupid as can be whilst remaining effective.
The table below shows elapsed times (in centiseconds) when running a
single repetitive swapping load across a 1000MB mapping in 900MB ram
with 1GB swap (the harddisk tests had taken painfully too long when I
used mem=500M, but SSD shows similar results for that).
Vanilla is the 3.6-rc7 kernel on which I started; Shaohua denotes his
Sep 3 patch in mmotm and linux-next; HughOld denotes my Oct 1 patch
which Shaohua showed to be defective; HughNew this Nov 14 patch, with
page_cluster as usual at default of 3 (8-page reads); HughPC4 this same
patch with page_cluster 4 (16-page reads); HughPC0 with page_cluster 0
(1-page reads: no readahead).
HDD for swapping to harddisk, SSD for swapping to VertexII SSD. Seq for
sequential access to the mapping, cycling five times around; Rand for
the same number of random touches. Anon for a MAP_PRIVATE anon mapping;
Shmem for a MAP_SHARED anon mapping, equivalent to tmpfs.
One weakness of Shaohua's vma/anon_vma approach was that it did not
optimize Shmem: seen below. Konstantin's approach was perhaps mistuned,
50% slower on Seq: did not compete and is not shown below.
HDD Vanilla Shaohua HughOld HughNew HughPC4 HughPC0
Seq Anon 73921 76210 75611 76904 78191 121542
Seq Shmem 73601 73176 73855 72947 74543 118322
Rand Anon 895392 831243 871569 845197 846496 841680
Rand Shmem 1058375 1053486 827935 764955 764376 756489
SSD Vanilla Shaohua HughOld HughNew HughPC4 HughPC0
Seq Anon 24634 24198 24673 25107 21614 70018
Seq Shmem 24959 24932 25052 25703 22030 69678
Rand Anon 43014 26146 28075 25989 26935 25901
Rand Shmem 45349 45215 28249 24268 24138 24332
These tests are, of course, two extremes of a very simple case: under
heavier mixed loads I've not yet observed any consistent improvement or
degradation, and wider testing would be welcome.
Shaohua Li:
Test shows Vanilla is slightly better in sequential workload than Hugh's
patch. I observed with Hugh's patch sometimes the readahead size is
shrinked too fast (from 8 to 1 immediately) in sequential workload if
there is no hit. And in such case, continuing doing readahead is good
actually.
I don't prepare a sophisticated algorithm for the sequential workload
because so far we can't guarantee sequential accessed pages are swap out
sequentially. So I slightly change Hugh's heuristic - don't shrink
readahead size too fast.
Here is my test result (unit second, 3 runs average):
Vanilla Hugh New
Seq 356 370 360
Random 4525 2447 2444
Attached graph is the swapin/swapout throughput I collected with 'vmstat
2'. The first part is running a random workload (till around 1200 of
the x-axis) and the second part is running a sequential workload.
swapin and swapout throughput are almost identical in steady state in
both workloads. These are expected behavior. while in Vanilla, swapin
is much bigger than swapout especially in random workload (because wrong
readahead).
Original patches by: Shaohua Li and Konstantin Khlebnikov.
[fengguang.wu@intel.com: swapin_nr_pages() can be static]
Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Konstantin Khlebnikov <khlebnikov@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
arch/arm/mach-tegra/pm.c, kernel/power/console.c and mm/vmpressure.c
were somehow getting slab.h indirectly through cgroup.h which in turn
was getting it indirectly through xattr.h. A scheduled cgroup change
drops xattr.h inclusion from cgroup.h and breaks compilation of these
three files. Add explicit slab.h includes to the three files.
A pending cgroup patch depends on this change and it'd be great if
this can be routed through cgroup/for-3.14-fixes branch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: linux-tegra@vger.kernel.org
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: linux-pm@vger.kernel.org
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Balbir Singh <bsingharora@gmail.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: cgroups@vger.kernel.org
Pull SLAB changes from Pekka Enberg:
"Random bug fixes that have accumulated in my inbox over the past few
months"
* 'slab/next' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/linux:
mm: Fix warning on make htmldocs caused by slab.c
mm: slub: work around unneeded lockdep warning
mm: sl[uo]b: fix misleading comments
slub: Fix possible format string bug.
slub: use lockdep_assert_held
slub: Fix calculation of cpu slabs
slab.h: remove duplicate kmalloc declaration and fix kernel-doc warnings
Jan Kara