mm: only set page->pfmemalloc when ALLOC_NO_WATERMARKS was used
__alloc_pages_slowpath() is called when the number of free pages is below
the low watermark. If the caller is entitled to use ALLOC_NO_WATERMARKS
then the page will be marked page->pfmemalloc. This protects more pages
than are strictly necessary as we only need to protect pages allocated
below the min watermark (the pfmemalloc reserves).
This patch only sets page->pfmemalloc when ALLOC_NO_WATERMARKS was
required to allocate the page.
[rientjes@google.com: David noticed the problem during review] Signed-off-by: Mel Gorman <mgorman@suse.de> Cc: David Miller <davem@davemloft.net> Cc: Neil Brown <neilb@suse.de> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Christie <michaelc@cs.wisc.edu> Cc: Eric B Munson <emunson@mgebm.net> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Sebastian Andrzej Siewior <sebastian@breakpoint.cc> Cc: Mel Gorman <mgorman@suse.de> Cc: Christoph Lameter <cl@linux.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This is needed to allow network softirq packet processing to make use of
PF_MEMALLOC.
Currently softirq context cannot use PF_MEMALLOC due to it not being
associated with a task, and therefore not having task flags to fiddle with
- thus the gfp to alloc flag mapping ignores the task flags when in
interrupts (hard or soft) context.
Allowing softirqs to make use of PF_MEMALLOC therefore requires some
trickery. This patch borrows the task flags from whatever process happens
to be preempted by the softirq. It then modifies the gfp to alloc flags
mapping to not exclude task flags in softirq context, and modify the
softirq code to save, clear and restore the PF_MEMALLOC flag.
The save and clear, ensures the preempted task's PF_MEMALLOC flag doesn't
leak into the softirq. The restore ensures a softirq's PF_MEMALLOC flag
cannot leak back into the preempted process. This should be safe due to
the following reasons
Softirqs can run on multiple CPUs sure but the same task should not be
executing the same softirq code. Neither should the softirq
handler be preempted by any other softirq handler so the flags
should not leak to an unrelated softirq.
Softirqs re-enable hardware interrupts in __do_softirq() so can be
preempted by hardware interrupts so PF_MEMALLOC is inherited
by the hard IRQ. However, this is similar to a process in
reclaim being preempted by a hardirq. While PF_MEMALLOC is
set, gfp_to_alloc_flags() distinguishes between hard and
soft irqs and avoids giving a hardirq the ALLOC_NO_WATERMARKS
flag.
If the softirq is deferred to ksoftirq then its flags may be used
instead of a normal tasks but as the softirq cannot be preempted,
the PF_MEMALLOC flag does not leak to other code by accident.
[davem@davemloft.net: Document why PF_MEMALLOC is safe] Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Mel Gorman <mgorman@suse.de> Cc: David Miller <davem@davemloft.net> Cc: Neil Brown <neilb@suse.de> Cc: Mike Christie <michaelc@cs.wisc.edu> Cc: Eric B Munson <emunson@mgebm.net> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Sebastian Andrzej Siewior <sebastian@breakpoint.cc> Cc: Mel Gorman <mgorman@suse.de> Cc: Christoph Lameter <cl@linux.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
mm: introduce __GFP_MEMALLOC to allow access to emergency reserves
__GFP_MEMALLOC will allow the allocation to disregard the watermarks, much
like PF_MEMALLOC. It allows one to pass along the memalloc state in
object related allocation flags as opposed to task related flags, such as
sk->sk_allocation. This removes the need for ALLOC_PFMEMALLOC as callers
using __GFP_MEMALLOC can get the ALLOC_NO_WATERMARK flag which is now
enough to identify allocations related to page reclaim.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Mel Gorman <mgorman@suse.de> Cc: David Miller <davem@davemloft.net> Cc: Neil Brown <neilb@suse.de> Cc: Mike Christie <michaelc@cs.wisc.edu> Cc: Eric B Munson <emunson@mgebm.net> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Sebastian Andrzej Siewior <sebastian@breakpoint.cc> Cc: Mel Gorman <mgorman@suse.de> Cc: Christoph Lameter <cl@linux.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
mm: slub: optimise the SLUB fast path to avoid pfmemalloc checks
This patch removes the check for pfmemalloc from the alloc hotpath and
puts the logic after the election of a new per cpu slab. For a pfmemalloc
page we do not use the fast path but force the use of the slow path which
is also used for the debug case.
This has the side-effect of weakening pfmemalloc processing in the
following way;
1. A process that is allocating for network swap calls __slab_alloc.
pfmemalloc_match is true so the freelist is loaded and c->freelist is
now pointing to a pfmemalloc page.
2. A process that is attempting normal allocations calls slab_alloc,
finds the pfmemalloc page on the freelist and uses it because it did
not check pfmemalloc_match()
The patch allows non-pfmemalloc allocations to use pfmemalloc pages with
the kmalloc slabs being the most vunerable caches on the grounds they
are most likely to have a mix of pfmemalloc and !pfmemalloc requests. A
later patch will still protect the system as processes will get throttled
if the pfmemalloc reserves get depleted but performance will not degrade
as smoothly.
[mgorman@suse.de: Expanded changelog] Signed-off-by: Christoph Lameter <cl@linux.com> Signed-off-by: Mel Gorman <mgorman@suse.de> Cc: David Miller <davem@davemloft.net> Cc: Neil Brown <neilb@suse.de> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Christie <michaelc@cs.wisc.edu> Cc: Eric B Munson <emunson@mgebm.net> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Sebastian Andrzej Siewior <sebastian@breakpoint.cc> Cc: Mel Gorman <mgorman@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
mm: sl[au]b: add knowledge of PFMEMALLOC reserve pages
When a user or administrator requires swap for their application, they
create a swap partition and file, format it with mkswap and activate it
with swapon. Swap over the network is considered as an option in diskless
systems. The two likely scenarios are when blade servers are used as part
of a cluster where the form factor or maintenance costs do not allow the
use of disks and thin clients.
The Linux Terminal Server Project recommends the use of the Network Block
Device (NBD) for swap according to the manual at
https://sourceforge.net/projects/ltsp/files/Docs-Admin-Guide/LTSPManual.pdf/download
There is also documentation and tutorials on how to setup swap over NBD at
places like https://help.ubuntu.com/community/UbuntuLTSP/EnableNBDSWAP The
nbd-client also documents the use of NBD as swap. Despite this, the fact
is that a machine using NBD for swap can deadlock within minutes if swap
is used intensively. This patch series addresses the problem.
The core issue is that network block devices do not use mempools like
normal block devices do. As the host cannot control where they receive
packets from, they cannot reliably work out in advance how much memory
they might need. Some years ago, Peter Zijlstra developed a series of
patches that supported swap over an NFS that at least one distribution is
carrying within their kernels. This patch series borrows very heavily
from Peter's work to support swapping over NBD as a pre-requisite to
supporting swap-over-NFS. The bulk of the complexity is concerned with
preserving memory that is allocated from the PFMEMALLOC reserves for use
by the network layer which is needed for both NBD and NFS.
Patch 1 adds knowledge of the PFMEMALLOC reserves to SLAB and SLUB to
preserve access to pages allocated under low memory situations
to callers that are freeing memory.
Patch 2 optimises the SLUB fast path to avoid pfmemalloc checks
Patch 3 introduces __GFP_MEMALLOC to allow access to the PFMEMALLOC
reserves without setting PFMEMALLOC.
Patch 4 opens the possibility for softirqs to use PFMEMALLOC reserves
for later use by network packet processing.
Patch 5 only sets page->pfmemalloc when ALLOC_NO_WATERMARKS was required
Patch 6 ignores memory policies when ALLOC_NO_WATERMARKS is set.
Patches 7-12 allows network processing to use PFMEMALLOC reserves when
the socket has been marked as being used by the VM to clean pages. If
packets are received and stored in pages that were allocated under
low-memory situations and are unrelated to the VM, the packets
are dropped.
Patch 11 reintroduces __skb_alloc_page which the networking
folk may object to but is needed in some cases to propogate
pfmemalloc from a newly allocated page to an skb. If there is a
strong objection, this patch can be dropped with the impact being
that swap-over-network will be slower in some cases but it should
not fail.
Patch 13 is a micro-optimisation to avoid a function call in the
common case.
Patch 14 tags NBD sockets as being SOCK_MEMALLOC so they can use
PFMEMALLOC if necessary.
Patch 15 notes that it is still possible for the PFMEMALLOC reserve
to be depleted. To prevent this, direct reclaimers get throttled on
a waitqueue if 50% of the PFMEMALLOC reserves are depleted. It is
expected that kswapd and the direct reclaimers already running
will clean enough pages for the low watermark to be reached and
the throttled processes are woken up.
Patch 16 adds a statistic to track how often processes get throttled
Some basic performance testing was run using kernel builds, netperf on
loopback for UDP and TCP, hackbench (pipes and sockets), iozone and
sysbench. Each of them were expected to use the sl*b allocators
reasonably heavily but there did not appear to be significant performance
variances.
For testing swap-over-NBD, a machine was booted with 2G of RAM with a
swapfile backed by NBD. 8*NUM_CPU processes were started that create
anonymous memory mappings and read them linearly in a loop. The total
size of the mappings were 4*PHYSICAL_MEMORY to use swap heavily under
memory pressure.
Without the patches and using SLUB, the machine locks up within minutes
and runs to completion with them applied. With SLAB, the story is
different as an unpatched kernel run to completion. However, the patched
kernel completed the test 45% faster.
MICRO
3.5.0-rc2 3.5.0-rc2
vanilla swapnbd
Unrecognised test vmscan-anon-mmap-write
MMTests Statistics: duration
Sys Time Running Test (seconds) 197.80 173.07
User+Sys Time Running Test (seconds) 206.96 182.03
Total Elapsed Time (seconds) 3240.70 1762.09
This patch: mm: sl[au]b: add knowledge of PFMEMALLOC reserve pages
Allocations of pages below the min watermark run a risk of the machine
hanging due to a lack of memory. To prevent this, only callers who have
PF_MEMALLOC or TIF_MEMDIE set and are not processing an interrupt are
allowed to allocate with ALLOC_NO_WATERMARKS. Once they are allocated to
a slab though, nothing prevents other callers consuming free objects
within those slabs. This patch limits access to slab pages that were
alloced from the PFMEMALLOC reserves.
When this patch is applied, pages allocated from below the low watermark
are returned with page->pfmemalloc set and it is up to the caller to
determine how the page should be protected. SLAB restricts access to any
page with page->pfmemalloc set to callers which are known to able to
access the PFMEMALLOC reserve. If one is not available, an attempt is
made to allocate a new page rather than use a reserve. SLUB is a bit more
relaxed in that it only records if the current per-CPU page was allocated
from PFMEMALLOC reserve and uses another partial slab if the caller does
not have the necessary GFP or process flags. This was found to be
sufficient in tests to avoid hangs due to SLUB generally maintaining
smaller lists than SLAB.
In low-memory conditions it does mean that !PFMEMALLOC allocators can fail
a slab allocation even though free objects are available because they are
being preserved for callers that are freeing pages.
[a.p.zijlstra@chello.nl: Original implementation]
[sebastian@breakpoint.cc: Correct order of page flag clearing] Signed-off-by: Mel Gorman <mgorman@suse.de> Cc: David Miller <davem@davemloft.net> Cc: Neil Brown <neilb@suse.de> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Christie <michaelc@cs.wisc.edu> Cc: Eric B Munson <emunson@mgebm.net> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Sebastian Andrzej Siewior <sebastian@breakpoint.cc> Cc: Mel Gorman <mgorman@suse.de> Cc: Christoph Lameter <cl@linux.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Minchan Kim [Sat, 21 Jul 2012 00:54:19 +0000 (10:54 +1000)]
memory-hotplug: fix kswapd looping forever problem
When hotplug offlining happens on zone A, it starts to mark freed page as
MIGRATE_ISOLATE type in buddy for preventing further allocation.
(MIGRATE_ISOLATE is very irony type because it's apparently on buddy but
we can't allocate them).
When the memory shortage happens during hotplug offlining, current task
starts to reclaim, then wake up kswapd. Kswapd checks watermark, then go
sleep because current zone_watermark_ok_safe doesn't consider
MIGRATE_ISOLATE freed page count. Current task continue to reclaim in
direct reclaim path without kswapd's helping. The problem is that
zone->all_unreclaimable is set by only kswapd so that current task would
be looping forever like below.
__alloc_pages_slowpath
restart:
wake_all_kswapd
rebalance:
__alloc_pages_direct_reclaim
do_try_to_free_pages
if global_reclaim && !all_unreclaimable
return 1; /* It means we did did_some_progress */
skip __alloc_pages_may_oom
should_alloc_retry
goto rebalance;
If we apply KOSAKI's patch[1] which doesn't depends on kswapd about
setting zone->all_unreclaimable, we can solve this problem by killing some
task in direct reclaim path. But it doesn't wake up kswapd, still. It
could be a problem still if other subsystem needs GFP_ATOMIC request. So
kswapd should consider MIGRATE_ISOLATE when it calculate free pages BEFORE
going sleep.
This patch counts the number of MIGRATE_ISOLATE page block and
zone_watermark_ok_safe will consider it if the system has such blocks
(fortunately, it's very rare so no problem in POV overhead and kswapd is
never hotpath).
Copy/modify from Mel's quote
"
Ideal solution would be "allocating" the pageblock.
It would keep the free space accounting as it is but historically,
memory hotplug didn't allocate pages because it would be difficult to
detect if a pageblock was isolated or if part of some balloon.
Allocating just full pageblocks would work around this, However,
it would play very badly with CMA.
"
[1] http://lkml.org/lkml/2012/6/14/74
Signed-off-by: Minchan Kim <minchan@kernel.org> Suggested-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Tested-by: Aaditya Kumar <aaditya.kumar.30@gmail.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Michal Hocko <mhocko@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Minchan Kim [Sat, 21 Jul 2012 00:54:19 +0000 (10:54 +1000)]
mm: fix free page check in zone_watermark_ok()
__zone_watermark_ok currently compares free_pages which is a signed type
with z->lowmem_reserve[classzone_idx] which is unsigned which might lead
to sign overflow if free_pages doesn't satisfy the given order (or it came
as negative already) and then we rely on the following order loop to fix
it (which doesn't work for order-0). Let's fix the type conversion and do
not rely on the given value of free_pages or follow up fixups.
This patch fixes it because "memory-hotplug: fix kswapd looping forever
problem" depends on this.
As benefit of this patch, it doesn't rely on the loop to exit
__zone_watermark_ok in case of high order check and make the first test
effective.(ie, if (free_pages <= min + lowmem_reserve))
Aaditya reported this problem when he test my hotplug patch.
Minchan Kim [Sat, 21 Jul 2012 00:54:19 +0000 (10:54 +1000)]
mm: factor out memory isolate functions
mm/page_alloc.c has some memory isolation functions but they are used only
when we enable CONFIG_{CMA|MEMORY_HOTPLUG|MEMORY_FAILURE}. So let's make
it configurable by new CONFIG_MEMORY_ISOLATION so that it can reduce
binary size and we can check it simple by CONFIG_MEMORY_ISOLATION, not if
defined CONFIG_{CMA|MEMORY_HOTPLUG|MEMORY_FAILURE}.
Signed-off-by: Minchan Kim <minchan@kernel.org> Cc: Andi Kleen <andi@firstfloor.org> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Michal Hocko <mhocko@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
David Rientjes [Sat, 21 Jul 2012 00:54:18 +0000 (10:54 +1000)]
mm, memcg: move all oom handling to memcontrol.c
By globally defining check_panic_on_oom(), the memcg oom handler can be
moved entirely to mm/memcontrol.c. This removes the ugly #ifdef in the
oom killer and cleans up the code.
Signed-off-by: David Rientjes <rientjes@google.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Michal Hocko <mhocko@suse.cz> Cc: Oleg Nesterov <oleg@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Slab poisoning gave me a General Protection Fault on the
atomic_dec(&__task_cred(p)->user->processes);
line of release_task() called from wait_task_zombie(),
every time my dd to USB testing generated a memcg OOM.
oom_kill_process() now does the put_task_struct(),
mem_cgroup_out_of_memory() should not repeat it.
Signed-off-by: Hugh Dickins <hughd@google.com> Cc: David Rientjes <rientjes@google.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Michal Hocko <mhocko@suse.cz> Cc: Oleg Nesterov <oleg@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
David Rientjes [Sat, 21 Jul 2012 00:54:18 +0000 (10:54 +1000)]
mm, oom: reduce dependency on tasklist_lock
Since exiting tasks require write_lock_irq(&tasklist_lock) several times,
try to reduce the amount of time the readside is held for oom kills. This
makes the interface with the memcg oom handler more consistent since it
now never needs to take tasklist_lock unnecessarily.
The only time the oom killer now takes tasklist_lock is when iterating the
children of the selected task, everything else is protected by
rcu_read_lock().
This requires that a reference to the selected process, p, is grabbed
before calling oom_kill_process(). It may release it and grab a reference
on another one of p's threads if !p->mm, but it also guarantees that it
will release the reference before returning.
Signed-off-by: David Rientjes <rientjes@google.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Cc: Oleg Nesterov <oleg@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
David Rientjes [Sat, 21 Jul 2012 00:54:17 +0000 (10:54 +1000)]
mm, memcg: introduce own oom handler to iterate only over its own threads
The global oom killer is serialized by the per-zonelist
try_set_zonelist_oom() which is used in the page allocator. Concurrent
oom kills are thus a rare event and only occur in systems using
mempolicies and with a large number of nodes.
Memory controller oom kills, however, can frequently be concurrent since
there is no serialization once the oom killer is called for oom conditions
in several different memcgs in parallel.
This creates a massive contention on tasklist_lock since the oom killer
requires the readside for the tasklist iteration. If several memcgs are
calling the oom killer, this lock can be held for a substantial amount of
time, especially if threads continue to enter it as other threads are
exiting.
Since the exit path grabs the writeside of the lock with irqs disabled in
a few different places, this can cause a soft lockup on cpus as a result
of tasklist_lock starvation.
The kernel lacks unfair writelocks, and successful calls to the oom killer
usually result in at least one thread entering the exit path, so an
alternative solution is needed.
This patch introduces a seperate oom handler for memcgs so that they do
not require tasklist_lock for as much time. Instead, it iterates only
over the threads attached to the oom memcg and grabs a reference to the
selected thread before calling oom_kill_process() to ensure it doesn't
prematurely exit.
This still requires tasklist_lock for the tasklist dump, iterating
children of the selected process, and killing all other threads on the
system sharing the same memory as the selected victim. So while this
isn't a complete solution to tasklist_lock starvation, it significantly
reduces the amount of time that it is held.
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Michal Hocko <mhocko@suse.cz> Signed-off-by: David Rientjes <rientjes@google.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Reviewed-by: Sha Zhengju <handai.szj@taobao.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
David Rientjes [Sat, 21 Jul 2012 00:54:17 +0000 (10:54 +1000)]
mm, oom: introduce helper function to process threads during scan
This patch introduces a helper function to process each thread during the
iteration over the tasklist. A new return type, enum oom_scan_t, is
defined to determine the future behavior of the iteration:
- OOM_SCAN_OK: continue scanning the thread and find its badness,
- OOM_SCAN_CONTINUE: do not consider this thread for oom kill, it's
ineligible,
- OOM_SCAN_ABORT: abort the iteration and return, or
- OOM_SCAN_SELECT: always select this thread with the highest badness
possible.
There is no functional change with this patch. This new helper function
will be used in the next patch in the memory controller.
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Signed-off-by: David Rientjes <rientjes@google.com> Cc: Oleg Nesterov <oleg@redhat.com> Reviewed-by: Sha Zhengju <handai.szj@taobao.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Jiang Liu [Sat, 21 Jul 2012 00:54:16 +0000 (10:54 +1000)]
mm/hotplug: free zone->pageset when a zone becomes empty
When a zone becomes empty after memory offlining, free zone->pageset.
Otherwise it will cause memory leak when adding memory to the empty zone
again because build_all_zonelists() will allocate zone->pageset for an
empty zone.
Signed-off-by: Jiang Liu <liuj97@gmail.com> Signed-off-by: Wei Wang <Bessel.Wang@huawei.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Michal Hocko <mhocko@suse.cz> Cc: Minchan Kim <minchan@kernel.org> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Yinghai Lu <yinghai@kernel.org> Cc: Tony Luck <tony.luck@intel.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: David Rientjes <rientjes@google.com> Cc: Keping Chen <chenkeping@huawei.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Jiang Liu [Sat, 21 Jul 2012 00:54:16 +0000 (10:54 +1000)]
mm/hotplug: correctly add new zone to all other nodes' zone lists
When online_pages() is called to add new memory to an empty zone, it
rebuilds all zone lists by calling build_all_zonelists(). But there's a
bug which prevents the new zone to be added to other nodes' zone lists.
Here the node of the zone is put into N_HIGH_MEMORY state after calling
build_all_zonelists(), but build_all_zonelists() only adds zones from
nodes in N_HIGH_MEMORY state to the fallback zone lists.
build_all_zonelists()
So memory in the new zone will never be used by other nodes, and it may
cause strange behavor when system is under memory pressure. So put node
into N_HIGH_MEMORY state before calling build_all_zonelists().
Signed-off-by: Jianguo Wu <wujianguo@huawei.com> Signed-off-by: Jiang Liu <liuj97@gmail.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Michal Hocko <mhocko@suse.cz> Cc: Minchan Kim <minchan@kernel.org> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Yinghai Lu <yinghai@kernel.org> Cc: Tony Luck <tony.luck@intel.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: David Rientjes <rientjes@google.com> Cc: Keping Chen <chenkeping@huawei.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Jiang Liu [Sat, 21 Jul 2012 00:54:16 +0000 (10:54 +1000)]
mm/hotplug: correctly setup fallback zonelists when creating new pgdat
When hotadd_new_pgdat() is called to create new pgdat for a new node, a
fallback zonelist should be created for the new node. There's code to try
to achieve that in hotadd_new_pgdat() as below:
/*
* The node we allocated has no zone fallback lists. For avoiding
* to access not-initialized zonelist, build here.
*/
mutex_lock(&zonelists_mutex);
build_all_zonelists(pgdat, NULL);
mutex_unlock(&zonelists_mutex);
But it doesn't work as expected. When hotadd_new_pgdat() is called, the
new node is still in offline state because node_set_online(nid) hasn't
been called yet. And build_all_zonelists() only builds zonelists for
online nodes as:
Though we hope to create zonelist for the new pgdat, but it doesn't. So
add a new parameter "pgdat" the build_all_zonelists() to build pgdat for
the new pgdat too.
Signed-off-by: Jiang Liu <liuj97@gmail.com> Signed-off-by: Xishi Qiu <qiuxishi@huawei.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Michal Hocko <mhocko@suse.cz> Cc: Minchan Kim <minchan@kernel.org> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Yinghai Lu <yinghai@kernel.org> Cc: Tony Luck <tony.luck@intel.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: David Rientjes <rientjes@google.com> Cc: Keping Chen <chenkeping@huawei.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
mm: setup pageblock_order before it's used by sparsemem
On architectures with CONFIG_HUGETLB_PAGE_SIZE_VARIABLE set, such as
Itanium, pageblock_order is a variable with default value of 0. It's set
to the right value by set_pageblock_order() in function
free_area_init_core().
But pageblock_order may be used by sparse_init() before free_area_init_core()
is called along path:
sparse_init()
->sparse_early_usemaps_alloc_node()
->usemap_size()
->SECTION_BLOCKFLAGS_BITS
->((1UL << (PFN_SECTION_SHIFT - pageblock_order)) *
NR_PAGEBLOCK_BITS)
The uninitialized pageblock_size will cause memory wasting because
usemap_size() returns a much bigger value then it's really needed.
For example, on an Itanium platform,
sparse_init() pageblock_order=0 usemap_size=24576
free_area_init_core() before pageblock_order=0, usemap_size=24576
free_area_init_core() after pageblock_order=12, usemap_size=8
That means 24K memory has been wasted for each section, so fix it by calling
set_pageblock_order() from sparse_init().
Signed-off-by: Xishi Qiu <qiuxishi@huawei.com> Signed-off-by: Jiang Liu <liuj97@gmail.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Yinghai Lu <yinghai@kernel.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: David Rientjes <rientjes@google.com> Cc: Keping Chen <chenkeping@huawei.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Minchan Kim [Sat, 21 Jul 2012 00:54:14 +0000 (10:54 +1000)]
vmscan: remove obsolete shrink_control comment
09f363c7 ("vmscan: fix shrinker callback bug in fs/super.c") fixed a
shrinker callback which was returning -1 when nr_to_scan is zero, which
caused excessive slab scanning. But 635697c6 ("vmscan: fix initial
shrinker size handling") fixed the problem, again so we can freely return
-1 although nr_to_scan is zero. So let's revert 09f363c7 because the
comment added in 09f363c7 made an unnecessary rule.
Signed-off-by: Minchan Kim <minchan@kernel.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Mikulas Patocka <mpatocka@redhat.com> Cc: Konstantin Khlebnikov <khlebnikov@openvz.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
0ee332c14518699 ("memblock: Kill early_node_map[]") wanted to replace
CONFIG_ARCH_POPULATES_NODE_MAP with CONFIG_HAVE_MEMBLOCK_NODE_MAP but
ended up replacing one occurence with a reference to the non-existent
symbol CONFIG_HAVE_MEMBLOCK_NODE.
The resulting omission of code would probably have been causing problems
to 32-bit machines with memory hotplug.
Signed-off-by: Rabin Vincent <rabin@rab.in> Cc: Tejun Heo <tj@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
mm: have order > 0 compaction start off where it left
This patch makes the comment for cc->wrapped longer, explaining what is
really going on. It also incorporates the comment fix pointed out by
Minchan.
Additionally, Minchan found that, when no pages get isolated, high_pfn
could be a value that is much lower than desired, which might potentially
cause compaction to skip a range of pages.
Only assign zone->compact_cache_free_pfn if we actually isolated free
pages for compaction.
Split out the calculation to get the start of the last page block in a
zone into its own, commented function.
Signed-off-by: Rik van Riel <riel@redhat.com> Cc: Jim Schutt <jaschut@sandia.gov> Acked-by: Minchan Kim <minchan@kernel.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Mel Gorman <mel@csn.ul.ie> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
mm: have order > 0 compaction start off where it left
Order > 0 compaction stops when enough free pages of the correct page
order have been coalesced. When doing subsequent higher order
allocations, it is possible for compaction to be invoked many times.
However, the compaction code always starts out looking for things to
compact at the start of the zone, and for free pages to compact things to
at the end of the zone.
This can cause quadratic behaviour, with isolate_freepages starting at the
end of the zone each time, even though previous invocations of the
compaction code already filled up all free memory on that end of the zone.
This can cause isolate_freepages to take enormous amounts of CPU with
certain workloads on larger memory systems.
The obvious solution is to have isolate_freepages remember where it left
off last time, and continue at that point the next time it gets invoked
for an order > 0 compaction. This could cause compaction to fail if
cc->free_pfn and cc->migrate_pfn are close together initially, in that
case we restart from the end of the zone and try once more.
Forced full (order == -1) compactions are left alone.
Signed-off-by: Rik van Riel <riel@redhat.com> Reported-by: Jim Schutt <jaschut@sandia.gov> Tested-by: Jim Schutt <jaschut@sandia.gov> Cc: Minchan Kim <minchan.kim@gmail.com> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Wanpeng Li [Sat, 21 Jul 2012 00:54:12 +0000 (10:54 +1000)]
memcg: rename mem_control_xxx to memcg_xxx
Replace memory_cgroup_xxx() with memcg_xxx()
Signed-off-by: Wanpeng Li <liwp.linux@gmail.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Michal Hocko <mhocko@suse.cz> Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Glauber Costa [Sat, 21 Jul 2012 00:54:12 +0000 (10:54 +1000)]
memcg: fix bad behavior in use_hierarchy file
I have an application that does the following:
* copy the state of all controllers attached to a hierarchy
* replicate it as a child of the current level.
I would expect writes to the files to mostly succeed, since they are
inheriting sane values from parents.
But that is not the case for use_hierarchy. If it is set to 0, we succeed
ok. If we're set to 1, the value of the file is automatically set to 1 in
the children, but if userspace tries to write the very same 1, it will
fail. That same situation happens if we set use_hierarchy, create a
child, and then try to write 1 again.
Now, there is no reason whatsoever for failing to write a value that is
already there. It doesn't even match the comments, that states:
/* If parent's use_hierarchy is set, we can't make any modifications
* in the child subtrees...
since we are not changing anything.
So test the new value against the one we're storing, and automatically
return 0 if we're not proposing a change.
Signed-off-by: Glauber Costa <glommer@parallels.com> Cc: Dhaval Giani <dhaval.giani@gmail.com> Acked-by: Michal Hocko <mhocko@suse.cz> Cc: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Minchan Kim [Sat, 21 Jul 2012 00:54:10 +0000 (10:54 +1000)]
mm: do not use page_count() without a page pin
d179e84ba ("mm: vmscan: do not use page_count without a page pin") fixed
this problem in vmscan.c but same problem is in __count_immobile_pages().
I copy and paste d179e84ba's contents for description.
"It is unsafe to run page_count during the physical pfn scan because
compound_head could trip on a dangling pointer when reading
page->first_page if the compound page is being freed by another CPU."
Signed-off-by: Minchan Kim <minchan@kernel.org> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Michal Hocko <mhocko@suse.cz> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Wanpeng Li <liwp.linux@gmail.com> Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
David Rientjes [Sat, 21 Jul 2012 00:54:09 +0000 (10:54 +1000)]
mm, oom: replace some information in tasklist dump
The number of ptes and swap entries are used in the oom killer's badness
heuristic, so they should be shown in the tasklist dump.
This patch adds those fields and replaces cpu and oom_adj values that are
currently emitted. Cpu isn't interesting and oom_adj is deprecated and
will be removed later this year, the same information is already displayed
as oom_score_adj which is used internally.
At the same time, make the documentation a little more clear to state this
information is helpful to determine why the oom killer chose the task it
did to kill.
Signed-off-by: David Rientjes <rientjes@google.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
David Rientjes [Sat, 21 Jul 2012 00:54:09 +0000 (10:54 +1000)]
mm, oom: fix potential killing of thread that is disabled from oom killing
/proc/sys/vm/oom_kill_allocating_task will immediately kill current when
the oom killer is called to avoid a potentially expensive tasklist scan
for large systems.
Currently, however, it is not checking current's oom_score_adj value which
may be OOM_SCORE_ADJ_MIN, meaning that it has been disabled from oom
killing.
This patch avoids killing current in such a condition and simply falls
back to the tasklist scan since memory still needs to be freed.
Signed-off-by: David Rientjes <rientjes@google.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
mm: clear pages_scanned only if draining a pcp adds pages to the buddy allocator again
commit 2ff754fa8f ("mm: clear pages_scanned only if draining a pcp adds
pages to the buddy allocator again") fixed one free_pcppages_bulk()
misuse. But two another miuse still exist.
This patch fixes it.
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Acked-by: David Rientjes <rientjes@google.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Minchan Kim <minchan@kernel.org> Cc: Wu Fengguang <fengguang.wu@intel.com> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
mm, fadvise: don't return -EINVAL when filesystem cannot implement fadvise()
Eric Wong reported his test suite failex when /tmp is tmpfs.
https://lkml.org/lkml/2012/2/24/479
Currentlt the input check of POSIX_FADV_WILLNEED has two problems.
- requires a_ops->readpage. But in fact, force_page_cache_readahead()
requires that the target filesystem has either ->readpage or ->readpages.
- returns -EINVAL when the filesystem doesn't have ->readpage. But
posix says that fadvise is merely a hint. Thus fadvise() should return
0 if filesystem has no means of implementing fadvise(). The userland
application should not know nor care whcih type of filesystem backs the
TMPDIR directory, as Eric pointed out. There is nothing which userspace
can do to solve this error.
So change the return value to 0 when filesytem doesn't support
readahead.
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Hugh Dickins <hughd@google.com> Cc: Hillf Danton <dhillf@gmail.com> Signed-off-by: Eric Wong <normalperson@yhbt.net> Tested-by: Eric Wong <normalperson@yhbt.net> Reviewed-by: Wanlong Gao <gaowanlong@cn.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
When CONFIG_COMPACTION is enabled, compaction_deferred() tries to
recalculate the deferred limit again, which isn't necessary.
When CONFIG_COMPACTION is disabled, compaction_deferred() should return
"true" or "false" since it has "bool" for its return value.
Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com> Acked-by: Minchan Kim <minchan@kernel.org> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
memcg: make mem_cgroup_force_empty_list() return bool
mem_cgroup_force_empty_list() just returns 0 or -EBUSY and -EBUSY
indicates 'you need to retry'. Make mem_cgroup_force_empty_list() return
a bool to simplify the logic.
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
memcg: clean up force_empty_list() return value check
After bf544fdc241da8 "memcg: move charges to root cgroup if
use_hierarchy=0 in mem_cgroup_move_hugetlb_parent()"
mem_cgroup_move_parent() returns only -EBUSY or -EINVAL. So we can remove
the -ENOMEM and -EINTR checks.
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Michal Hocko <mhocko@suse.cz> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
memcg: remove check for signal_pending() during rmdir()
After bf544fdc241da8 "memcg: move charges to root cgroup if
use_hierarchy=0 in mem_cgroup_move_hugetlb_parent()", no memory reclaim
will occur when removing a memory cgroup. If -EINTR is returned here,
cgroup will show a warning.
We don't need to handle any user interruption signal. Remove this.
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Michal Hocko <mhocko@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
David Rientjes [Sat, 21 Jul 2012 00:54:06 +0000 (10:54 +1000)]
mm, oom: do not schedule if current has been killed
The oom killer currently schedules away from current in an uninterruptible
sleep if it does not have access to memory reserves. It's possible that
current was killed because it shares memory with the oom killed thread or
because it was killed by the user in the interim, however.
This patch only schedules away from current if it does not have a pending
kill, i.e. if it does not share memory with the oom killed thread. It's
possible that it will immediately retry its memory allocation and fail,
but it will immediately be given access to memory reserves if it calls the
oom killer again.
This prevents the delay of memory freeing when threads that share memory
with the oom killed thread get unnecessarily scheduled.
Signed-off-by: David Rientjes <rientjes@google.com> Cc: Oleg Nesterov <oleg@redhat.com> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
hugetlb/cgroup: remove exclude and wakeup rmdir calls from migrate
We already hold the hugetlb_lock. That should prevent a parallel cgroup
rmdir from touching page's hugetlb cgroup. So remove the exclude and
wakeup calls.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
hugetlb/cgroup: assign the page hugetlb cgroup when we move the page to active list.
A page's hugetlb cgroup assignment and movement to the active list should
occur with hugetlb_lock held. Otherwise when we remove the hugetlb cgroup
we will iterate the active list and find pages with NULL hugetlb cgroup
values.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
When we fail to allocate pages from the reserve pool, hugetlb tries to
allocate huge pages using alloc_buddy_huge_page. Add these to the active
list. We also need to add the huge page we allocate when we soft offline
the oldpage to active list.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
hugetlb/cgroup: migrate hugetlb cgroup info from oldpage to new page during migration
With HugeTLB pages, hugetlb cgroup is uncharged in compound page
destructor. Since we are holding a hugepage reference, we can be sure
that old page won't get uncharged till the last put_page().
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Cc: David Rientjes <rientjes@google.com> Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Hillf Danton <dhillf@gmail.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Wanpeng Li [Sat, 21 Jul 2012 00:54:02 +0000 (10:54 +1000)]
mm/hugetlb_cgroup: Add huge_page_order check to avoid incorrectly uncharge
alloc_huge_page() will call hugetlb_cgroup_charge_cgroup() to charge
pages, the compound page have less than 3 pages will not charge to hugetlb
cgroup. When alloc_huge_page fails it will call
hugetlb_cgroup_uncharge_cgroup to uncharge pages, however,
hugetlb_cgroup_uncharge_cgroup doesn't have huge_page_order check. That
means it will uncharge pages even if the compound page have less than 3
pages. Add huge_page_order check to avoid this incorrectly uncharge.
Reviewed-by: Michal Hocko <mhocko@suse.cz> Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: Wanpeng Li <liwp.linux@gmail.com> Cc: David Rientjes <rientjes@google.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Hillf Danton <dhillf@gmail.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
cgroup_subsys_state can never be NULL, so don't check for that in
hugetlb_cgroup_from_css. Also current task will always be part of some
cgroup. So hugetlb_cgrop_from_task cannot return NULL.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
hugetlb/cgroup: add charge/uncharge routines for hugetlb cgroup
Add the charge and uncharge routines for hugetlb cgroup. We do cgroup
charging in page alloc and uncharge in compound page destructor.
Assigning page's hugetlb cgroup is protected by hugetlb_lock.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Cc: David Rientjes <rientjes@google.com> Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Hillf Danton <dhillf@gmail.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
hugetlb/cgroup: add the cgroup pointer to page lru
Add the hugetlb cgroup pointer to 3rd page lru.next. This limit the usage
to hugetlb cgroup to only hugepages with 3 or more normal pages. I guess
that is an acceptable limitation.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Cc: David Rientjes <rientjes@google.com> Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Hillf Danton <dhillf@gmail.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
cgroup_subsys_state can never be NULL, so don't check for that in
hugetlb_cgroup_from_css. Also current task will always be part of some
cgroup. So hugetlb_cgrop_from_task cannot return NULL.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Implement a new controller that allows us to control HugeTLB allocations.
The extension allows to limit the HugeTLB usage per control group and
enforces the controller limit during page fault. Since HugeTLB doesn't
support page reclaim, enforcing the limit at page fault time implies that,
the application will get SIGBUS signal if it tries to access HugeTLB pages
beyond its limit. This requires the application to know beforehand how
much HugeTLB pages it would require for its use.
The charge/uncharge calls will be added to HugeTLB code in later patch.
Support for cgroup removal will be added in later patches.
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Cc: David Rientjes <rientjes@google.com> Cc: Hillf Danton <dhillf@gmail.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
hugetlb: add a list for tracking in-use HugeTLB pages
hugepage_activelist will be used to track currently used HugeTLB pages.
We need to find the in-use HugeTLB pages to support HugeTLB cgroup removal.
On cgroup removal we update the page's HugeTLB cgroup to point to parent
cgroup.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: David Rientjes <rientjes@google.com> Cc: Hillf Danton <dhillf@gmail.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Since we migrate only one hugepage, don't use linked list for passing the
page around. Directly pass the page that need to be migrated as argument.
This also removes the usage of page->lru in the migrate path.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: David Rientjes <rientjes@google.com> Cc: Hillf Danton <dhillf@gmail.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
hugetlb: avoid taking i_mmap_mutex in unmap_single_vma() for hugetlb
i_mmap_mutex lock was added in unmap_single_vma by 502717f4e ("hugetlb:
fix linked list corruption in unmap_hugepage_range()") but we don't use
page->lru in unmap_hugepage_range any more. Also the lock was taken
higher up in the stack in some code path. That would result in deadlock.
For shared pagetable support for huge pages, since pagetable pages are ref
counted we don't need any lock during huge_pmd_unshare. We do take
i_mmap_mutex in huge_pmd_share while walking the vma_prio_tree in mapping.
(39dde65c9940c97f ("shared page table for hugetlb page")).
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Cc: David Rientjes <rientjes@google.com> Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Hillf Danton <dhillf@gmail.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
The current use of VM_FAULT_* codes with ERR_PTR requires us to ensure
VM_FAULT_* values will not exceed MAX_ERRNO value. Decouple the
VM_FAULT_* values from MAX_ERRNO.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Acked-by: Hillf Danton <dhillf@gmail.com> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: David Rientjes <rientjes@google.com> Cc: Michal Hocko <mhocko@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This patchset implements a cgroup resource controller for HugeTLB pages.
The controller allows to limit the HugeTLB usage per control group and
enforces the controller limit during page fault. Since HugeTLB doesn't
support page reclaim, enforcing the limit at page fault time implies that,
the application will get SIGBUS signal if it tries to access HugeTLB pages
beyond its limit. This requires the application to know beforehand how
much HugeTLB pages it would require for its use.
The goal is to control how many HugeTLB pages a group of task can
allocate. It can be looked at as an extension of the existing quota
interface which limits the number of HugeTLB pages per hugetlbfs
superblock. HPC job scheduler requires jobs to specify their resource
requirements in the job file. Once their requirements can be met, job
schedulers like (SLURM) will schedule the job. We need to make sure that
the jobs won't consume more resources than requested. If they do we
should either error out or kill the application.
This patch:
Rename max_hstate to hugetlb_max_hstate. We will be using this from other
subsystems like hugetlb controller in later patches.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Acked-by: David Rientjes <rientjes@google.com> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Hillf Danton <dhillf@gmail.com> Acked-by: Michal Hocko <mhocko@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Wanpeng Li [Sat, 21 Jul 2012 00:53:57 +0000 (10:53 +1000)]
mm: prepare for removal of obsolete /proc/sys/vm/nr_pdflush_threads
Since per-BDI flusher threads were introduced in 2.6, the pdflush
mechanism is not used any more. But the old interface exported through
/proc/sys/vm/nr_pdflush_threads still exists and is obviously useless.
For back-compatibility, printk warning information and return 2 to notify
the users that the interface is removed.
Signed-off-by: Wanpeng Li <liwp@linux.vnet.ibm.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Currently, function should_fail() has "bool" for its return value, so it's
reasonable to change the return value of function should_fail_alloc_page()
into "bool" as well.
The patch does cleanup on function should_fail_alloc_page() to have "bool"
for its return value.
Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
vm_stat_account() accounts the shared_vm, stack_vm and reserved_vm now.
But we can also account for total_vm in the vm_stat_account() which makes
the code tidy.
Even for mprotect_fixup(), we can get the right result in the end.
Signed-off-by: Huang Shijie <shijie8@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
documentation: update how page-cluster affects swap I/O
Fix of the documentation of /proc/sys/vm/page-cluster to match the
behavior of the code and add some comments about what the tunable will
change in that behavior.
Signed-off-by: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com> Acked-by: Jens Axboe <axboe@kernel.dk> Reviewed-by: Minchan Kim <minchan@kernel.org> Cc: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Swap readahead works fine, but the I/O to disk is almost always done in
page size requests, despite the fact that readahead submits
1<<page-cluster pages at a time.
On older kernels the old per device plugging behavior might have captured
this and merged the requests, but currently all comes down to much more
I/Os than required.
On a single device this might not be an issue, but as soon as a server
runs on shared san resources savin I/Os not only improves swapin
throughput but also provides a lower resource utilization.
With a load running KVM in a lot of memory overcommitment (the hot memory
is 1.5 times the host memory) swapping throughput improves significantly
and the lead feels more responsive as well as achieves more throughput.
In a test setup with 16 swap disks running blocktrace on one of those disks
shows the improved merging:
Prior:
Reads Queued: 560,888, 2,243MiB Writes Queued: 226,242, 904,968KiB
Read Dispatches: 544,701, 2,243MiB Write Dispatches: 159,318, 904,968KiB
Reads Requeued: 0 Writes Requeued: 0
Reads Completed: 544,716, 2,243MiB Writes Completed: 159,321, 904,980KiB
Read Merges: 16,187, 64,748KiB Write Merges: 61,744, 246,976KiB
IO unplugs: 149,614 Timer unplugs: 2,940
I got ~10% to ~40% more throughput in my cases and at the same time much
lower cpu consumption when broken down per transferred kilobyte (the
majority of that due to saved interrupts and better cache handling). In a
shared SAN others might get an additional benefit as well, because this
now causes less protocol overhead.
Signed-off-by: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com> Acked-by: Rik van Riel <riel@redhat.com> Acked-by: Jens Axboe <axboe@kernel.dk> Reviewed-by: Minchan Kim <minchan@kernel.org> Cc: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Jan Kara [Sat, 21 Jul 2012 00:53:54 +0000 (10:53 +1000)]
mm: make vb_alloc() more foolproof
If someone calls vb_alloc() (or vm_map_ram() for that matter) to allocate
0 bytes (0 pages), get_order() returns BITS_PER_LONG - PAGE_CACHE_SHIFT
and interesting stuff happens. So make debugging such problems easier and
warn about 0-size allocation.
Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
vmalloc: walk vmap_areas by sorted list instead of rb_next()
There's a walk by repeating rb_next to find a suitable hole. Could be
simply replaced by walk on the sorted vmap_area_list. More simpler and
efficient.
Mutation of the list and tree only happens in pair within
__insert_vmap_area and __free_vmap_area, under protection of
vmap_area_lock. The patch code is also under vmap_area_lock, so the list
walk is safe, and consistent with the tree walk.
Tested on SMP by repeating batch of vmalloc anf vfree for random sizes and
rounds for hours.
Signed-off-by: Hong Zhiguo <honkiko@gmail.com> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
slab: do not call compound_head() in page_get_cache()
page_get_cache() does not need to call compound_head(), as its unique
caller virt_to_slab() already makes sure to return a head page.
Additionally, removing the compound_head() call makes page_get_cache()
consistent with page_get_slab().
Signed-off-by: Michel Lespinasse <walken@google.com> Cc: Christoph Lameter <cl@linux-foundation.org> Cc: Pekka Enberg <penberg@kernel.org> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
While allocating pages using buddy allocator, the compound page is
probably split up to free pages. Under these circumstances, the compound
page should be destroyed by destroy_compound_page(). However, there is a
duplicate check to judge if the page is compound.
Remove the duplicate check since the compound_order() returns 0 when the
page doesn't have PG_head set in destroy_compound_page(). That is to say,
destroy_compound_page() needn't check PageHead().
Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Kautuk Consul [Sat, 21 Jul 2012 00:53:53 +0000 (10:53 +1000)]
xtensa/mm/fault.c: port OOM changes to do_page_fault
d065bd810b6de ("mm: retry page fault when blocking on disk transfer") and 37b23e0525d393d ("x86,mm: make pagefault killable")
The above commits introduced changes into the x86 pagefault handler for
making the page fault handler retryable as well as killable.
These changes reduce the mmap_sem hold time, which is crucial during OOM
killer invocation.
Port these changes to xtensa.
Signed-off-by: Kautuk Consul <consul.kautuk@gmail.com> Acked-by: Chris Zankel <chris@zankel.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
When the suid_dumpable sysctl is set to "2", and there is no core dump
pipe defined in the core_pattern sysctl, a local user can cause core files
to be written to root-writable directories, potentially with
user-controlled content. This means an admin can unknowningly reintroduce
a variation of CVE-2006-2451, allowing local users to gain root
privileges.
While cron has been fixed to abort reading a file when there is any parse
error, there are still other sensitive directories that will read any file
present and skip unparsable lines.
Instead of introducing a suid_dumpable=3 mode and breaking all users of
mode 2, this only disables the unsafe portion of mode 2 (writing to disk
via relative path). Most users of mode 2 (e.g. Chrome OS) already use a
core dump pipe handler, so this change will not break them. For the
situations where a pipe handler is not defined but mode 2 is still active,
crash dumps will only be written to fully qualified paths. If a relative
path is defined (e.g. the default "core" pattern), dump attempts will
trigger a printk yelling about the lack of a fully qualified path.
Signed-off-by: Kees Cook <keescook@chromium.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Alan Cox <alan@linux.intel.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Doug Ledford <dledford@redhat.com> Cc: Serge Hallyn <serge.hallyn@canonical.com> Cc: James Morris <james.l.morris@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
fs: push rcu_barrier() from deactivate_locked_super() to filesystems
There's no reason to call rcu_barrier() on every
deactivate_locked_super(). We only need to make sure that all delayed rcu
free inodes are flushed before we destroy related cache.
Removing rcu_barrier() from deactivate_locked_super() affects some fast
paths. E.g. on my machine exit_group() of a last process in IPC
namespace takes 0.07538s. rcu_barrier() takes 0.05188s of that time.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
When a file is truncated with truncate()/ftruncate() and then closed,
iversion is not updated. This patch uses ATTR_SIZE flag as an indication
to increment iversion.
Mimi said:
On fput(), i_version is used to detect and flag files that have changed
and need to be re-measured in the IMA measurement policy. When a file
is truncated with truncate()/ftruncate() and then closed, i_version is
not updated. As a result, although the file has changed, it will not be
re-measured and added to the IMA measurement list on subsequent access.
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com> Acked-by: Mimi Zohar <zohar@us.ibm.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Martin Michlmayr [Sat, 21 Jul 2012 00:53:51 +0000 (10:53 +1000)]
drivers/scsi/atp870u.c: fix bad use of udelay
The ACARD driver calls udelay() with a value > 2000, which leads to
to the following compilation error on ARM:
ERROR: "__bad_udelay" [drivers/scsi/atp870u.ko] undefined!
make[1]: *** [__modpost] Error 1
This is because udelay is defined on ARM, roughly speaking, as
The argument to __const_udelay is the number of jiffies to wait divided by
4, but this does not work unless the multiplication does not overflow, and
that is what the build error is designed to prevent. The intended
behavior can be achieved by using mdelay to call udelay multiple times in
a loop.
[jn: adding context] Signed-off-by: Martin Michlmayr <tbm@cyrius.com> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Akinobu Mita [Sat, 21 Jul 2012 00:53:50 +0000 (10:53 +1000)]
ocfs2: use bitmap_weight()
Use bitmap_weight() instead of reinventing the wheel.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
projects / karo-tx-linux.git / log