Cyrill Gorcunov [Sat, 17 May 2014 13:18:05 +0000 (23:18 +1000)]
mm: pgtable -- Require X86_64 for soft-dirty tracker, v2
v2 (by akpm@):
- guard helpers with CONFIG_HAVE_ARCH_SOFT_DIRTY on i386, otherwise
it fails to build because we've a generic definitions in
asm-generic/pgtable.h
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cyrill Gorcunov [Sat, 17 May 2014 13:18:05 +0000 (23:18 +1000)]
mm: x86 pgtable: require X86_64 for soft-dirty tracker
Tracking dirty status on 2 level pages requires very ugly macros and
taking into account how old the machines who can operate without PAE mode
only are, lets drop soft dirty tracker from them for code simplicity (note
I can't drop all the macros from 2 level pages by now since
_PAGE_BIT_PROTNONE and _PAGE_BIT_FILE are still used even without
tracker).
Linus proposed to completely rip off softdirty support on x86-32 (even
with PAE) and since for CRIU we're not planning to support native x86-32
mode, lets do that.
(Softdirty tracker is relatively new feature which is mostly used by CRIU
so I don't expect if such API change would cause problems for userspace).
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Peter Anvin <hpa@zytor.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Steven Noonan <steven@uplinklabs.net> Cc: Rik van Riel <riel@redhat.com> Cc: David Vrabel <david.vrabel@citrix.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Pavel Emelyanov <xemul@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cleanups:
- move pte-related code to separate function. It's about half of the
function;
- get rid of some goto-logic;
- use 'return NULL' instead of 'return page' where page can only be
NULL;
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Dave Hansen [Sat, 17 May 2014 13:18:03 +0000 (23:18 +1000)]
mm: shrinker: add nid to tracepoint output
Now that we are doing NUMA-aware shrinking, and can have
shrinkers running in parallel, or working on individual nodes, it
seems like we should also be sticking the node in the output.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Acked-by: Dave Chinner <david@fromorbit.com> Cc: Konstantin Khlebnikov <khlebnikov@openvz.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
and noticed that "total_scan" can go negative in some cases. We
used to dump out the "total_scan" variable directly, but some of
the shrinker modifications along the way changed that.
This patch just dumps it out directly, again. It doesn't make
any sense to derive it from new_nr and nr any more since there
are now other shrinkers that can be running in parallel and
mucking with those values.
Here's an example of the negative numbers in the output:
> kswapd0-840 [000] 160.869398: mm_shrink_slab_end: i915_gem_inactive_scan+0x0 0xffff8800037cbc68: unused scan count 10 new scan count 39 total_scan 29 last shrinker return val 256
> kswapd0-840 [000] 160.869618: mm_shrink_slab_end: i915_gem_inactive_scan+0x0 0xffff8800037cbc68: unused scan count 39 new scan count 102 total_scan 63 last shrinker return val 256
> kswapd0-840 [000] 160.870031: mm_shrink_slab_end: i915_gem_inactive_scan+0x0 0xffff8800037cbc68: unused scan count 102 new scan count 47 total_scan -55 last shrinker return val 768
> kswapd0-840 [000] 160.870464: mm_shrink_slab_end: i915_gem_inactive_scan+0x0 0xffff8800037cbc68: unused scan count 47 new scan count 45 total_scan -2 last shrinker return val 768
> kswapd0-840 [000] 163.384144: mm_shrink_slab_end: i915_gem_inactive_scan+0x0 0xffff8800037cbc68: unused scan count 45 new scan count 56 total_scan 11 last shrinker return val 0
> kswapd0-840 [000] 163.384297: mm_shrink_slab_end: i915_gem_inactive_scan+0x0 0xffff8800037cbc68: unused scan count 56 new scan count 15 total_scan -41 last shrinker return val 256
> kswapd0-840 [000] 163.384414: mm_shrink_slab_end: i915_gem_inactive_scan+0x0 0xffff8800037cbc68: unused scan count 15 new scan count 117 total_scan 102 last shrinker return val 0
> kswapd0-840 [000] 163.384657: mm_shrink_slab_end: i915_gem_inactive_scan+0x0 0xffff8800037cbc68: unused scan count 117 new scan count 36 total_scan -81 last shrinker return val 512
> kswapd0-840 [000] 163.384880: mm_shrink_slab_end: i915_gem_inactive_scan+0x0 0xffff8800037cbc68: unused scan count 36 new scan count 111 total_scan 75 last shrinker return val 256
> kswapd0-840 [000] 163.385256: mm_shrink_slab_end: i915_gem_inactive_scan+0x0 0xffff8800037cbc68: unused scan count 111 new scan count 34 total_scan -77 last shrinker return val 768
> kswapd0-840 [000] 163.385598: mm_shrink_slab_end: i915_gem_inactive_scan+0x0 0xffff8800037cbc68: unused scan count 34 new scan count 122 total_scan 88 last shrinker return val 512
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Acked-by: Dave Chinner <david@fromorbit.com> Cc: Konstantin Khlebnikov <khlebnikov@openvz.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Daeseok Youn [Sat, 17 May 2014 13:18:02 +0000 (23:18 +1000)]
mm/dmapool.c: remove redundant NULL check for dev in dma_pool_create()
"dev" cannot be NULL because it is already checked before calling
dma_pool_create().
If dev ever was NULL, the code would oops in dev_to_node() after enabling
CONFIG_NUMA.
It is possible that some driver is using dev==NULL and has never been run
on a NUMA machine. Such a driver is probably outdated, possibly buggy and
will need some attention if it starts triggering NULL derefs.
Signed-off-by: Daeseok Youn <daeseok.youn@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Jianyu Zhan [Sat, 17 May 2014 13:18:02 +0000 (23:18 +1000)]
mm: introdule compound_head_by_tail()
Currently, in put_compound_page(), we have
======
if (likely(!PageTail(page))) { <------ (1)
if (put_page_testzero(page)) {
/*
¦* By the time all refcounts have been released
¦* split_huge_page cannot run anymore from under us.
¦*/
if (PageHead(page))
__put_compound_page(page);
else
__put_single_page(page);
}
return;
}
/* __split_huge_page_refcount can run under us */
page_head = compound_head(page); <------------ (2)
======
if at (1) , we fail the check, this means page is *likely* a tail page.
Then at (2), as compoud_head(page) is inlined, it is :
smp_rmb();
if (likely(PageTail(page)))
return head;
}
return page;
}
======
here, the (3) unlikely in the case is a negative hint, because it is
*likely* a tail page. So the check (3) in this case is not good, so I
introduce a helper for this case.
So this patch introduces compound_head_by_tail() which deals with a
possible tail page(though it could be spilt by a racy thread), and make
compound_head() a wrapper on it.
This patch has no functional change, and it reduces the object
size slightly:
text data bss dec hex filename
11003 1328 16 12347 303b mm/swap.o.orig
10971 1328 16 12315 301b mm/swap.o.patched
I've ran "perf top -e branch-miss" to observe branch-miss in this case.
As Michael points out, it's a slow path, so only very few times this case
happens. But I grep'ed the code base, and found there still are some
other call sites could be benifited from this helper. And given that it
only bloating up the source by only 5 lines, but with a reduced object
size. I still believe this helper deserves to exsit.
Signed-off-by: Jianyu Zhan <nasa4836@gmail.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Rik van Riel <riel@redhat.com> Cc: Jiang Liu <liuj97@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Sasha Levin <sasha.levin@oracle.com> Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com> Cc: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Jianyu Zhan [Sat, 17 May 2014 13:18:02 +0000 (23:18 +1000)]
mm/swap.c: split put_compound_page()
Currently, put_compound_page() carefully handles tricky cases to avoid
racing with compound page releasing or splitting, which makes it quite
lenthy (about 200+ lines) and needs deep tab indention, which makes it
quite hard to follow and maintain.
Now based on two helpers introduced in the previous patch ("mm/swap.c:
introduce put_[un]refcounted_compound_page helpers for spliting
put_compound_page"), this patch replaces those two lengthy code paths with
these two helpers, respectively. Also, it has some comment rephrasing.
After this patch, the put_compound_page() is very compact, thus easy to
read and maintain.
After splitting, the object file is of same size as the original one.
Actually, I've diff'ed put_compound_page()'s orginal disassemble code and
the patched disassemble code, the are 100% the same!
This fact shows that this splitting has no functional change, but it
brings readability.
This patch and the previous one blow the code by 32 lines, mostly due to
comments.
Signed-off-by: Jianyu Zhan <nasa4836@gmail.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Rik van Riel <riel@redhat.com> Cc: Jiang Liu <liuj97@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Sasha Levin <sasha.levin@oracle.com> Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com> Cc: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Jianyu Zhan [Sat, 17 May 2014 13:18:01 +0000 (23:18 +1000)]
mm/swap.c: introduce put_[un]refcounted_compound_page helpers for splitting put_compound_page()
Currently, put_compound_page() carefully handles tricky cases to avoid
racing with compound page releasing or splitting, which makes it quite
lenthy (about 200+ lines) and needs deep tab indention, which makes it
quite hard to follow and maintain.
This patch and the next patch refactor this function.
Based on the code skeleton of put_compound_page:
put_compound_pge:
if !PageTail(page)
put head page fastpath;
return;
/* else PageTail */
page_head = compound_head(page)
if !__compound_tail_refcounted(page_head)
put head page optimal path; <---(1)
return;
else
put head page slowpath; <--- (2)
return;
This patch introduces two helpers, put_[un]refcounted_compound_page,
handling the code path (1) and code path (2), respectively. They both are
tagged __always_inline, thus elmiating function call overhead, making them
operating the same way as before.
They are almost copied verbatim(except one place, a "goto out_put_single"
is expanded), with some comments rephrasing.
Signed-off-by: Jianyu Zhan <nasa4836@gmail.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Rik van Riel <riel@redhat.com> Cc: Jiang Liu <liuj97@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Sasha Levin <sasha.levin@oracle.com> Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com> Cc: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Rasmus Villemoes [Sat, 17 May 2014 13:18:01 +0000 (23:18 +1000)]
mm: constify nmask argument to set_mempolicy()
The nmask argument to set_mempolicy() is const according to the user-space
header numaif.h, and since the kernel does indeed not modify it, it might
as well be declared const in the kernel.
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Rasmus Villemoes [Sat, 17 May 2014 13:18:01 +0000 (23:18 +1000)]
mm: constify nmask argument to mbind()
The nmask argument to mbind() is const according to the userspace header
numaif.h, and since the kernel does indeed not modify it, it might as well
be declared const in the kernel.
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk> Acked-by: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This doesn't really hurt but unnecessary and misleading. init_task is the
"swapper" thread == current, its ->mm is always NULL. And init_mm can
only be used as ->active_mm, not as ->mm.
mm_init_owner() has a single caller with this patch, perhaps it should
die. mm_init() can initialize ->owner under #ifdef.
Signed-off-by: Oleg Nesterov <oleg@redhat.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Cc: Balbir Singh <bsingharora@gmail.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: Peter Chiang <pchiang@nvidia.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Oleg Nesterov [Sat, 17 May 2014 13:18:00 +0000 (23:18 +1000)]
memcg: optimize the "Search everything else" loop in mm_update_next_owner()
for_each_process_thread() is sub-optimal. All threads share the same
->mm, we can swicth to the next process once we found a thread with
->mm != NULL and ->mm != mm.
Signed-off-by: Oleg Nesterov <oleg@redhat.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Cc: Balbir Singh <bsingharora@gmail.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: Peter Chiang <pchiang@nvidia.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Oleg Nesterov [Sat, 17 May 2014 13:18:00 +0000 (23:18 +1000)]
memcg: mm_update_next_owner() should skip kthreads
"Search through everything else" in mm_update_next_owner() can hit a
kthread which adopted this "mm" via use_mm(), it should not be used as
mm->owner. Add the PF_KTHREAD check.
While at it, change this code to use for_each_process_thread()
instead of deprecated do_each_thread/while_each_thread.
Signed-off-by: Oleg Nesterov <oleg@redhat.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Cc: Balbir Singh <bsingharora@gmail.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: Peter Chiang <pchiang@nvidia.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Matthew Wilcox [Sat, 17 May 2014 13:17:59 +0000 (23:17 +1000)]
brd: return -ENOSPC rather than -ENOMEM on page allocation failure
brd is effectively a thinly provisioned device. Thinly provisioned
devices return -ENOSPC when they can't write a new block. -ENOMEM is an
implementation detail that callers shouldn't know.
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com> Acked-by: Dave Chinner <david@fromorbit.com> Cc: Dheeraj Reddy <dheeraj.reddy@intel.com> Cc: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Matthew Wilcox [Sat, 17 May 2014 13:17:58 +0000 (23:17 +1000)]
fs/block_dev.c: add bdev_read_page() and bdev_write_page()
A block device driver may choose to provide a rw_page operation. These
will be called when the filesystem is attempting to do page sized I/O to
page cache pages (ie not for direct I/O). This does preclude I/Os that
are larger than page size, so this may only be a performance gain for some
devices.
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com> Tested-by: Dheeraj Reddy <dheeraj.reddy@intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Matthew Wilcox [Sat, 17 May 2014 13:17:58 +0000 (23:17 +1000)]
fs/mpage.c: factor page_endio() out of mpage_end_io()
page_endio() takes care of updating all the appropriate page flags once
I/O has finished to a page. Switch to using mapping_set_error() instead
of setting AS_EIO directly; this will handle thin-provisioned devices
correctly.
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dheeraj Reddy <dheeraj.reddy@intel.com> Cc: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Matthew Wilcox [Sat, 17 May 2014 13:17:58 +0000 (23:17 +1000)]
fs/mpage.c: factor clean_buffers() out of __mpage_writepage()
__mpage_writepage() is over 200 lines long, has 20 local variables, four
goto labels and could desperately use simplification. Splitting
clean_buffers() into a helper function improves matters a little, removing
20+ lines from it.
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dheeraj Reddy <dheeraj.reddy@intel.com> Cc: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Matthew Wilcox [Sat, 17 May 2014 13:17:58 +0000 (23:17 +1000)]
fs/buffer.c: remove block_write_full_page_endio()
The last in-tree caller of block_write_full_page_endio() was removed in
January 2013. It's time to remove the EXPORT_SYMBOL, which leaves
block_write_full_page() as the only caller of
block_write_full_page_endio(), so inline block_write_full_page_endio()
into block_write_full_page().
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com> Cc: Hugh Dickins <hughd@google.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dheeraj Reddy <dheeraj.reddy@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
NeilBrown [Sat, 17 May 2014 13:17:57 +0000 (23:17 +1000)]
mm/vmscan.c: avoid throttling reclaim for loop-back nfsd threads
When a loopback NFS mount is active and the backing device for the NFS
mount becomes congested, that can impose throttling delays on the nfsd
threads.
These delays significantly reduce throughput and so the NFS mount remains
congested.
This results in a livelock and the reduced throughput persists.
This livelock has been found in testing with the 'wait_iff_congested'
call, and could possibly be caused by the 'congestion_wait' call.
This livelock is similar to the deadlock which justified the introduction
of PF_LESS_THROTTLE, and the same flag can be used to remove this
livelock.
To minimise the impact of the change, we still throttle nfsd when the
filesystem it is writing to is congested, but not when some separate
filesystem (e.g. the NFS filesystem) is congested.
Signed-off-by: NeilBrown <neilb@suse.de> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Mel Gorman [Sat, 17 May 2014 13:17:57 +0000 (23:17 +1000)]
mm: numa: add migrated transhuge pages to LRU the same way as base pages
Migration of misplaced transhuge pages uses page_add_new_anon_rmap() when
putting the page back as it avoided an atomic operations and added the new
page to the correct LRU. A side-effect is that the page gets marked
activated as part of the migration meaning that transhuge and base pages
are treated differently from an aging perspective than base page
migration.
This patch uses page_add_anon_rmap() and putback_lru_page() on completion
of a transhuge migration similar to base page migration. It would require
fewer atomic operations to use lru_cache_add without taking an additional
reference to the page. The downside would be that it's still different to
base page migration and unevictable pages may be added to the wrong LRU
for cleaning up later. Testing of the usual workloads did not show any
adverse impact to the change.
Vladimir Davydov [Sat, 17 May 2014 13:17:57 +0000 (23:17 +1000)]
memcg, slab: simplify synchronization scheme
At present, we have the following mutexes protecting data related to per
memcg kmem caches:
- slab_mutex. This one is held during the whole kmem cache creation
and destruction paths. We also take it when updating per root cache
memcg_caches arrays (see memcg_update_all_caches). As a result, taking
it guarantees there will be no changes to any kmem cache (including per
memcg). Why do we need something else then? The point is it is
private to slab implementation and has some internal dependencies with
other mutexes (get_online_cpus). So we just don't want to rely upon it
and prefer to introduce additional mutexes instead.
- activate_kmem_mutex. Initially it was added to synchronize
initializing kmem limit (memcg_activate_kmem). However, since we can
grow per root cache memcg_caches arrays only on kmem limit
initialization (see memcg_update_all_caches), we also employ it to
protect against memcg_caches arrays relocation (e.g. see
__kmem_cache_destroy_memcg_children).
- We have a convention not to take slab_mutex in memcontrol.c, but we
want to walk over per memcg memcg_slab_caches lists there (e.g. for
destroying all memcg caches on offline). So we have per memcg
slab_caches_mutex's protecting those lists.
Such a syncrhonization scheme has a number of flaws, for instance:
- We can't call kmem_cache_{destroy,shrink} while walking over a
memcg::memcg_slab_caches list due to locking order. As a result, in
mem_cgroup_destroy_all_caches we schedule the
memcg_cache_params::destroy work shrinking and destroying the cache.
- We don't have a mutex to synchronize per memcg caches destruction
between memcg offline (mem_cgroup_destroy_all_caches) and root cache
destruction (__kmem_cache_destroy_memcg_children). Currently we just
don't bother about it.
This patch simplifies it by substituting per memcg slab_caches_mutex's
with the global memcg_slab_mutex. It will be held whenever a new per
memcg cache is created or destroyed, so it protects per root cache
memcg_caches arrays and per memcg memcg_slab_caches lists. The locking
order is following:
This allows us to call kmem_cache_{create,shrink,destroy} under the
memcg_slab_mutex. As a result, we don't need memcg_cache_params::destroy
work any more - we can simply destroy caches while iterating over a per
memcg slab caches list.
Also using the global mutex simplifies synchronization between concurrent
per memcg caches creation/destruction, e.g. mem_cgroup_destroy_all_caches
vs __kmem_cache_destroy_memcg_children.
The downside of this is that we substitute per-memcg slab_caches_mutex's
with a hummer-like global mutex, but since we already take either the
slab_mutex or the cgroup_mutex along with a memcg::slab_caches_mutex, it
shouldn't hurt concurrency a lot.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.cz> Cc: Glauber Costa <glommer@gmail.com> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Vladimir Davydov [Sat, 17 May 2014 13:17:56 +0000 (23:17 +1000)]
memcg, slab: merge memcg_{bind,release}_pages to memcg_{un}charge_slab
Currently we have two pairs of kmemcg-related functions that are called on
slab alloc/free. The first is memcg_{bind,release}_pages that count the
total number of pages allocated on a kmem cache. The second is
memcg_{un}charge_slab that {un}charge slab pages to kmemcg resource
counter. Let's just merge them to keep the code clean.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.cz> Cc: Glauber Costa <glommer@gmail.com> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Vladimir Davydov [Sat, 17 May 2014 13:17:56 +0000 (23:17 +1000)]
memcg, slab: do not schedule cache destruction when last page goes away
This patchset is a part of preparations for kmemcg re-parenting. It
targets at simplifying kmemcg work-flows and synchronization.
First, it removes async per memcg cache destruction (see patches 1, 2).
Now caches are only destroyed on memcg offline. That means the caches
that are not empty on memcg offline will be leaked. However, they are
already leaked, because memcg_cache_params::nr_pages normally never drops
to 0 so the destruction work is never scheduled except kmem_cache_shrink
is called explicitly. In the future I'm planning reaping such dead caches
on vmpressure or periodically.
Second, it substitutes per memcg slab_caches_mutex's with the global
memcg_slab_mutex, which should be taken during the whole per memcg cache
creation/destruction path before the slab_mutex (see patch 3). This
greatly simplifies synchronization among various per memcg cache
creation/destruction paths.
I'm still not quite sure about the end picture, in particular I don't know
whether we should reap dead memcgs' kmem caches periodically or try to
merge them with their parents (see https://lkml.org/lkml/2014/4/20/38 for
more details), but whichever way we choose, this set looks like a
reasonable change to me, because it greatly simplifies kmemcg work-flows
and eases further development.
This patch (of 3):
After a memcg is offlined, we mark its kmem caches that cannot be deleted
right now due to pending objects as dead by setting the
memcg_cache_params::dead flag, so that memcg_release_pages will schedule
cache destruction (memcg_cache_params::destroy) as soon as the last slab
of the cache is freed (memcg_cache_params::nr_pages drops to zero).
I guess the idea was to destroy the caches as soon as possible, i.e.
immediately after freeing the last object. However, it just doesn't work
that way, because kmem caches always preserve some pages for the sake of
performance, so that nr_pages never gets to zero unless the cache is
shrunk explicitly using kmem_cache_shrink. Of course, we could account
the total number of objects on the cache or check if all the slabs
allocated for the cache are empty on kmem_cache_free and schedule
destruction if so, but that would be too costly.
Thus we have a piece of code that works only when we explicitly call
kmem_cache_shrink, but complicates the whole picture a lot. Moreover,
it's racy in fact. For instance, kmem_cache_shrink may free the last slab
and thus schedule cache destruction before it finishes checking that the
cache is empty, which can lead to use-after-free.
So I propose to remove this async cache destruction from
memcg_release_pages, and check if the cache is empty explicitly after
calling kmem_cache_shrink instead. This will simplify things a lot w/o
introducing any functional changes.
And regarding dead memcg caches (i.e. those that are left hanging around
after memcg offline for they have objects), I suppose we should reap them
either periodically or on vmpressure as Glauber suggested initially. I'm
going to implement this later.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.cz> Cc: Glauber Costa <glommer@gmail.com> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Michal Hocko [Sat, 17 May 2014 13:17:56 +0000 (23:17 +1000)]
memcg: do not hang on OOM when killed by userspace OOM access to memory reserves
Eric has reported that he can see task(s) stuck in memcg OOM handler
regularly. The only way out is to
echo 0 > $GROUP/memory.oom_control
His usecase is:
- Setup a hierarchy with memory and the freezer (disable kernel oom and
have a process watch for oom).
- In that memory cgroup add a process with one thread per cpu.
- In one thread slowly allocate once per second I think it is 16M of ram
and mlock and dirty it (just to force the pages into ram and stay
there).
- When oom is achieved loop:
* attempt to freeze all of the tasks.
* if frozen send every task SIGKILL, unfreeze, remove the directory in
cgroupfs.
Eric has then pinpointed the issue to be memcg specific.
All tasks are sitting on the memcg_oom_waitq when memcg oom is disabled.
Those that have received fatal signal will bypass the charge and should
continue on their way out. The tricky part is that the exit path might
trigger a page fault (e.g. exit_robust_list), thus the memcg charge,
while its memcg is still under OOM because nobody has released any charges
yet.
Unlike with the in-kernel OOM handler the exiting task doesn't get
TIF_MEMDIE set so it doesn't shortcut further charges of the killed task
and falls to the memcg OOM again without any way out of it as there are no
fatal signals pending anymore.
This patch fixes the issue by checking PF_EXITING early in
mem_cgroup_try_charge and bypass the charge same as if it had fatal
signal pending or TIF_MEMDIE set.
Normally exiting tasks (aka not killed) will bypass the charge now but
this should be OK as the task is leaving and will release memory and
increasing the memory pressure just to release it in a moment seems
dubious wasting of cycles. Besides that charges after exit_signals should
be rare.
I am bringing this patch again (rebased on the current mmotm tree). I
hope we can move forward finally. If there is still an opposition then
I would really appreciate a concurrent approach so that we can discuss
alternatives.
http://comments.gmane.org/gmane.linux.kernel.stable/77650 is a reference
to the followup discussion when the patch has been dropped from the mmotm
last time.
Reported-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Michal Hocko <mhocko@suse.cz> Acked-by: David Rientjes <rientjes@google.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Mel Gorman [Sat, 17 May 2014 13:17:55 +0000 (23:17 +1000)]
mm: vmscan: do not throttle based on pfmemalloc reserves if node has no ZONE_NORMAL
throttle_direct_reclaim() is meant to trigger during swap-over-network
during which the min watermark is treated as a pfmemalloc reserve. It
throttes on the first node in the zonelist but this is flawed.
The user-visible impact is that a process running on CPU whose local
memory node has no ZONE_NORMAL will stall for prolonged periods of time,
possibly indefintely. This is due to throttle_direct_reclaim thinking the
pfmemalloc reserves are depleted when in fact they don't exist on that
node.
On a NUMA machine running a 32-bit kernel (I know) allocation requests
from CPUs on node 1 would detect no pfmemalloc reserves and the process
gets throttled. This patch adjusts throttling of direct reclaim to
throttle based on the first node in the zonelist that has a usable
ZONE_NORMAL or lower zone.
Signed-off-by: Mel Gorman <mgorman@suse.de> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Oleg Nesterov [Sat, 17 May 2014 13:17:55 +0000 (23:17 +1000)]
memcg: kill CONFIG_MM_OWNER
CONFIG_MM_OWNER makes no sense. It is not user-selectable, it is only
selected by CONFIG_MEMCG automatically. So we can kill this option in
init/Kconfig and do s/CONFIG_MM_OWNER/CONFIG_MEMCG/ globally.
Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Michal Hocko <mhocko@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Jianyu Zhan [Sat, 17 May 2014 13:17:54 +0000 (23:17 +1000)]
mm/swap.c: clean up *lru_cache_add* functions
In mm/swap.c, __lru_cache_add() is exported, but actually there are no
users outside this file.
This patch unexports __lru_cache_add(), and makes it static. It also
exports lru_cache_add_file(), as it is use by cifs and fuse, which can
loaded as modules.
Signed-off-by: Jianyu Zhan <nasa4836@gmail.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Shaohua Li <shli@kernel.org> Cc: Bob Liu <bob.liu@oracle.com> Cc: Seth Jennings <sjenning@linux.vnet.ibm.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Rafael Aquini <aquini@redhat.com> Cc: Mel Gorman <mgorman@suse.de> Acked-by: Rik van Riel <riel@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Khalid Aziz <khalid.aziz@oracle.com> Cc: Christoph Hellwig <hch@lst.de> Reviewed-by: Zhang Yanfei <zhangyanfei@cn.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Performing vma lookups without taking the mm->mmap_sem is asking for
trouble. While doing the search, the vma in question can be modified or
even removed before returning to the caller. Take the lock (exclusively)
in order to avoid races while iterating through the vmacache and/or
rbtree.
Signed-off-by: Jonathan Gonzalez V <zeus@gnu.org> Signed-off-by: Davidlohr Bueso <davidlohr@hp.com> Cc: Inki Dae <inki.dae@samsung.com> Cc: Joonyoung Shim <jy0922.shim@samsung.com> Cc: David Airlie <airlied@linux.ie> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Davidlohr Bueso [Sat, 17 May 2014 13:17:54 +0000 (23:17 +1000)]
arc: call find_vma with the mmap_sem held
Performing vma lookups without taking the mm->mmap_sem is asking for
trouble. While doing the search, the vma in question can be modified or
even removed before returning to the caller. Take the lock (shared) in
order to avoid races while iterating through the vmacache and/or rbtree.
Davidlohr Bueso [Sat, 17 May 2014 13:17:53 +0000 (23:17 +1000)]
mips: call find_vma with the mmap_sem held
Performing vma lookups without taking the mm->mmap_sem is asking for
trouble. While doing the search, the vma in question can be modified or
even removed before returning to the caller. Take the lock (exclusively)
in order to avoid races while iterating through the vmacache and/or
rbtree.
Updates two functions:
- process_fpemu_return()
- cteon_flush_cache_sigtramp()
Signed-off-by: Davidlohr Bueso <davidlohr@hp.com> Cc: Ralf Baechle <ralf@linux-mips.org> Tested-by: Andreas Herrmann <andreas.herrmann@caviumnetworks.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Davidlohr Bueso [Sat, 17 May 2014 13:17:53 +0000 (23:17 +1000)]
m68k: call find_vma with the mmap_sem held in sys_cacheflush()
Performing vma lookups without taking the mm->mmap_sem is asking for
trouble. While doing the search, the vma in question can be modified or
even removed before returning to the caller. Take the lock (shared) in
order to avoid races while iterating through the vmacache and/or rbtree.
Vladimir Davydov [Sat, 17 May 2014 13:17:53 +0000 (23:17 +1000)]
Documentation/memcg: warn about incomplete kmemcg state
Kmemcg is currently under development and lacks some important features.
In particular, it does not have support of kmem reclaim on memory pressure
inside cgroup, which practically makes it unusable in real life. Let's
warn about it in both Kconfig and Documentation to prevent complaints
arising.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com> Acked-by: Michal Hocko <mhocko@suse.cz> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Dave Hansen [Sat, 17 May 2014 13:17:53 +0000 (23:17 +1000)]
mm: debug: make bad_range() output more usable and readable
Nobody outputs memory addresses in decimal. PFNs are essentially
addresses, and they're gibberish in decimal. Output them in hex.
Also, add the nid and zone name to give a little more context to the
message.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Joonsoo Kim [Sat, 17 May 2014 13:17:53 +0000 (23:17 +1000)]
mm-compaction-cleanup-isolate_freepages-fix3
What I did here is taking end_pfn out of the loop and considering zone
boundary once. After then, we can just set previous pfn to end_pfn on
every iteration to move scanning window. With this change, we can remove
local variable, z_end_pfn.
Another things I did are removing max() operation and un-needed assignment
to isolate variable.
In addition, I change both the variable names, from pfn and end_pfn to
block_start_pfn and block_end_pfn, respectively. They represent their
meaning perfectly.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com> Acked-by: Vlastimil Babka <vbabka@suse.de> Acked-by: Michal Nazarewicz <mina86@mina86.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: Christoph Lameter <cl@linux.com> Cc: Rik van Riel <riel@redhat.com> Cc: Dongjun Shin <d.j.shin@samsung.com> Cc: Sunghwan Yun <sunghwan.yun@samsung.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Vlastimil Babka [Sat, 17 May 2014 13:17:52 +0000 (23:17 +1000)]
mm-compaction-cleanup-isolate_freepages-fix 2
Cleanup detection of compaction scanners crossing in isolate_freepages().
To make sure compact_finished() observes scanners crossing, we can just
set free_pfn to migrate_pfn instead of confusing max() construct.
Suggested-by: Minchan Kim <minchan@kernel.org> Signed-off-by: Vlastimil Babka <vbabka@suse.cz> Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Cc: Christoph Lameter <cl@linux.com> Cc: Dongjun Shin <d.j.shin@samsung.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Michal Nazarewicz <mina86@mina86.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: Rik van Riel <riel@redhat.com> Cc: Sunghwan Yun <sunghwan.yun@samsung.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Vlastimil Babka [Sat, 17 May 2014 13:17:52 +0000 (23:17 +1000)]
mm/compaction: cleanup isolate_freepages()
isolate_freepages() is currently somewhat hard to follow thanks to many
looks like it is related to the 'low_pfn' variable, but in fact it is not.
This patch renames the 'high_pfn' variable to a hopefully less confusing name,
and slightly changes its handling without a functional change. A comment made
obsolete by recent changes is also updated.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz> Cc: Minchan Kim <minchan@kernel.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Cc: Michal Nazarewicz <mina86@mina86.com> Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: Christoph Lameter <cl@linux.com> Cc: Rik van Riel <riel@redhat.com> Cc: Dongjun Shin <d.j.shin@samsung.com> Cc: Sunghwan Yun <sunghwan.yun@samsung.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Vlastimil Babka [Sat, 17 May 2014 13:17:51 +0000 (23:17 +1000)]
mm/page_alloc: prevent MIGRATE_RESERVE pages from being misplaced
For the MIGRATE_RESERVE pages, it is useful when they do not get misplaced
on free_list of other migratetype, otherwise they might get allocated
prematurely and e.g. fragment the MIGRATE_RESEVE pageblocks. While this
cannot be avoided completely when allocating new MIGRATE_RESERVE
pageblocks in min_free_kbytes sysctl handler, we should prevent the
misplacement where possible.
Currently, it is possible for the misplacement to happen when a
MIGRATE_RESERVE page is allocated on pcplist through rmqueue_bulk() as a
fallback for other desired migratetype, and then later freed back through
free_pcppages_bulk() without being actually used. This happens because
free_pcppages_bulk() uses get_freepage_migratetype() to choose the
free_list, and rmqueue_bulk() calls set_freepage_migratetype() with the
*desired* migratetype and not the page's original MIGRATE_RESERVE
migratetype.
This patch fixes the problem by moving the call to
set_freepage_migratetype() from rmqueue_bulk() down to
__rmqueue_smallest() and __rmqueue_fallback() where the actual page's
migratetype (e.g. from which free_list the page is taken from) is used.
Note that this migratetype might be different from the pageblock's
migratetype due to freepage stealing decisions. This is OK, as page
stealing never uses MIGRATE_RESERVE as a fallback, and also takes care to
leave all MIGRATE_CMA pages on the correct freelist.
Therefore, as an additional benefit, the call to
get_pageblock_migratetype() from rmqueue_bulk() when CMA is enabled, can
be removed completely. This relies on the fact that MIGRATE_CMA
pageblocks are created only during system init, and the above. The
related is_migrate_isolate() check is also unnecessary, as memory
isolation has other ways to move pages between freelists, and drain pcp
lists containing pages that should be isolated. The buffered_rmqueue()
can also benefit from calling get_freepage_migratetype() instead of
get_pageblock_migratetype().
Signed-off-by: Vlastimil Babka <vbabka@suse.cz> Reported-by: Yong-Taek Lee <ytk.lee@samsung.com> Reported-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Suggested-by: Joonsoo Kim <iamjoonsoo.kim@lge.com> Acked-by: Joonsoo Kim <iamjoonsoo.kim@lge.com> Suggested-by: Mel Gorman <mgorman@suse.de> Acked-by: Minchan Kim <minchan@kernel.org> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Hugh Dickins <hughd@google.com> Cc: Rik van Riel <riel@redhat.com> Cc: Michal Nazarewicz <mina86@mina86.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Vladimir Davydov [Sat, 17 May 2014 13:17:51 +0000 (23:17 +1000)]
slab: get_online_mems for kmem_cache_{create,destroy,shrink}
When we create a sl[au]b cache, we allocate kmem_cache_node structures for
each online NUMA node. To handle nodes taken online/offline, we register
memory hotplug notifier and allocate/free kmem_cache_node corresponding to
the node that changes its state for each kmem cache.
To synchronize between the two paths we hold the slab_mutex during both
the cache creationg/destruction path and while tuning per-node parts of
kmem caches in memory hotplug handler, but that's not quite right, because
it does not guarantee that a newly created cache will have all
kmem_cache_nodes initialized in case it races with memory hotplug. For
instance, in case of slub:
CPU0 CPU1
---- ----
kmem_cache_create: online_pages:
__kmem_cache_create: slab_memory_callback:
slab_mem_going_online_callback:
lock slab_mutex
for each slab_caches list entry
allocate kmem_cache node
unlock slab_mutex
lock slab_mutex
init_kmem_cache_nodes:
for_each_node_state(node, N_NORMAL_MEMORY)
allocate kmem_cache node
add kmem_cache to slab_caches list
unlock slab_mutex
online_pages (continued):
node_states_set_node
As a result we'll get a kmem cache with not all kmem_cache_nodes
allocated.
To avoid issues like that we should hold get/put_online_mems() during the
whole kmem cache creation/destruction/shrink paths, just like we deal with
cpu hotplug. This patch does the trick.
Note, that after it's applied, there is no need in taking the slab_mutex
for kmem_cache_shrink any more, so it is removed from there.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: Tang Chen <tangchen@cn.fujitsu.com> Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com> Cc: Toshi Kani <toshi.kani@hp.com> Cc: Xishi Qiu <qiuxishi@huawei.com> Cc: Jiang Liu <liuj97@gmail.com> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Cc: David Rientjes <rientjes@google.com> Cc: Wen Congyang <wency@cn.fujitsu.com> Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Vladimir Davydov [Sat, 17 May 2014 13:17:50 +0000 (23:17 +1000)]
mem-hotplug: implement get/put_online_mems
kmem_cache_{create,destroy,shrink} need to get a stable value of cpu/node
online mask, because they init/destroy/access per-cpu/node kmem_cache
parts, which can be allocated or destroyed on cpu/mem hotplug. To protect
against cpu hotplug, these functions use {get,put}_online_cpus. However,
they do nothing to synchronize with memory hotplug - taking the slab_mutex
does not eliminate the possibility of race as described in patch 2.
What we need there is something like get_online_cpus, but for memory. We
already have lock_memory_hotplug, which serves for the purpose, but it's a
bit of a hammer right now, because it's backed by a mutex. As a result,
it imposes some limitations to locking order, which are not desirable, and
can't be used just like get_online_cpus. That's why in patch 1 I
substitute it with get/put_online_mems, which work exactly like
get/put_online_cpus except they block not cpu, but memory hotplug.
[ v1 can be found at https://lkml.org/lkml/2014/4/6/68. I NAK'ed it by
myself, because it used an rw semaphore for get/put_online_mems, making
them dead lock prune. ]
This patch (of 2):
{un}lock_memory_hotplug, which is used to synchronize against memory
hotplug, is currently backed by a mutex, which makes it a bit of a hammer
- threads that only want to get a stable value of online nodes mask won't
be able to proceed concurrently. Also, it imposes some strong locking
ordering rules on it, which narrows down the set of its usage scenarios.
This patch introduces get/put_online_mems, which are the same as
get/put_online_cpus, but for memory hotplug, i.e. executing a code inside
a get/put_online_mems section will guarantee a stable value of online
nodes, present pages, etc.
lock_memory_hotplug()/unlock_memory_hotplug() are removed altogether.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: Tang Chen <tangchen@cn.fujitsu.com> Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com> Cc: Toshi Kani <toshi.kani@hp.com> Cc: Xishi Qiu <qiuxishi@huawei.com> Cc: Jiang Liu <liuj97@gmail.com> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Cc: David Rientjes <rientjes@google.com> Cc: Wen Congyang <wency@cn.fujitsu.com> Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Mel Gorman [Sat, 17 May 2014 13:17:50 +0000 (23:17 +1000)]
mm: page_alloc: do not cache reclaim distances
pgdat->reclaim_nodes tracks if a remote node is allowed to be reclaimed by
zone_reclaim due to its distance. As it is expected that zone_reclaim_mode
will be rarely enabled it is unreasonable for all machines to take a penalty.
Fortunately, the zone_reclaim_mode() path is already slow and it is the path
that takes the hit.
Signed-off-by: Mel Gorman <mgorman@suse.de> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Zhang Yanfei <zhangyanfei@cn.fujitsu.com> Acked-by: Michal Hocko <mhocko@suse.cz> Reviewed-by: Christoph Lameter <cl@linux.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Mel Gorman [Sat, 17 May 2014 13:17:49 +0000 (23:17 +1000)]
mm: disable zone_reclaim_mode by default
When it was introduced, zone_reclaim_mode made sense as NUMA distances
punished and workloads were generally partitioned to fit into a NUMA node.
NUMA machines are now common but few of the workloads are NUMA-aware and
it's routine to see major performance degradation due to zone_reclaim_mode
being enabled but relatively few can identify the problem.
Those that require zone_reclaim_mode are likely to be able to detect when
it needs to be enabled and tune appropriately so lets have a sensible
default for the bulk of users.
This patch (of 2):
zone_reclaim_mode causes processes to prefer reclaiming memory from local
node instead of spilling over to other nodes. This made sense initially when
NUMA machines were almost exclusively HPC and the workload was partitioned
into nodes. The NUMA penalties were sufficiently high to justify reclaiming
the memory. On current machines and workloads it is often the case that
zone_reclaim_mode destroys performance but not all users know how to detect
this. Favour the common case and disable it by default. Users that are
sophisticated enough to know they need zone_reclaim_mode will detect it.
Signed-off-by: Mel Gorman <mgorman@suse.de> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Zhang Yanfei <zhangyanfei@cn.fujitsu.com> Acked-by: Michal Hocko <mhocko@suse.cz> Reviewed-by: Christoph Lameter <cl@linux.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
WARNING: braces {} are not necessary for any arm of this statement
#282: FILE: mm/hugetlb.c:1650:
+ if (hstate_is_gigantic(h)) {
[...]
+ } else {
[...]
total: 0 errors, 1 warnings, 219 lines checked
./patches/hugetlb-add-support-for-gigantic-page-allocation-at-runtime.patch has style problems, please review.
If any of these errors are false positives, please report
them to the maintainer, see CHECKPATCH in MAINTAINERS.
Please run checkpatch prior to sending patches
Cc: Luiz Capitulino <lcapitulino@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Luiz Capitulino [Sat, 17 May 2014 13:17:49 +0000 (23:17 +1000)]
hugetlb: add support for gigantic page allocation at runtime
HugeTLB is limited to allocating hugepages whose size are less than
MAX_ORDER order. This is so because HugeTLB allocates hugepages via the
buddy allocator. Gigantic pages (that is, pages whose size is greater
than MAX_ORDER order) have to be allocated at boottime.
However, boottime allocation has at least two serious problems. First, it
doesn't support NUMA and second, gigantic pages allocated at boottime
can't be freed.
This commit solves both issues by adding support for allocating gigantic
pages during runtime. It works just like regular sized hugepages, meaning
that the interface in sysfs is the same, it supports NUMA, and gigantic
pages can be freed.
For example, on x86_64 gigantic pages are 1GB big. To allocate two 1G
gigantic pages on node 1, one can do:
The one problem with gigantic page allocation at runtime is that it can't
be serviced by the buddy allocator. To overcome that problem, this commit
scans all zones from a node looking for a large enough contiguous region.
When one is found, it's allocated by using CMA, that is, we call
alloc_contig_range() to do the actual allocation. For example, on x86_64
we scan all zones looking for a 1GB contiguous region. When one is found,
it's allocated by alloc_contig_range().
One expected issue with that approach is that such gigantic contiguous
regions tend to vanish as runtime goes by. The best way to avoid this for
now is to make gigantic page allocations very early during system boot,
say from a init script. Other possible optimization include using
compaction, which is supported by CMA but is not explicitly used by this
commit.
It's also important to note the following:
1. Gigantic pages allocated at boottime by the hugepages= command-line
option can be freed at runtime just fine
2. This commit adds support for gigantic pages only to x86_64. The
reason is that I don't have access to nor experience with other archs.
The code is arch indepedent though, so it should be simple to add
support to different archs
3. I didn't add support for hugepage overcommit, that is allocating
a gigantic page on demand when
/proc/sys/vm/nr_overcommit_hugepages > 0. The reason is that I don't
think it's reasonable to do the hard and long work required for
allocating a gigantic page at fault time. But it should be simple
to add this if wanted
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com> Reviewed-by: Davidlohr Bueso <davidlohr@hp.com> Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Reviewed-by: Zhang Yanfei <zhangyanfei@cn.fujitsu.com> Reviewed-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: Rik van Riel <riel@redhat.com> Cc: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Luiz Capitulino [Sat, 17 May 2014 13:17:49 +0000 (23:17 +1000)]
hugetlb: move helpers up in the file
Next commit will add new code which will want to call
for_each_node_mask_to_alloc() macro. Move it, its buddy
for_each_node_mask_to_free() and their dependencies up in the file so the
new code can use them. This is just code movement, no logic change.
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com> Reviewed-by: Andrea Arcangeli <aarcange@redhat.com> Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Reviewed-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com> Reviewed-by: Davidlohr Bueso <davidlohr@hp.com> Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Reviewed-by: Zhang Yanfei <zhangyanfei@cn.fujitsu.com> Cc: David Rientjes <rientjes@google.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: Rik van Riel <riel@redhat.com> Cc: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Luiz Capitulino [Sat, 17 May 2014 13:17:48 +0000 (23:17 +1000)]
hugetlb: prep_compound_gigantic_page(): drop __init marker
The HugeTLB subsystem uses the buddy allocator to allocate hugepages
during runtime. This means that hugepages allocation during runtime is
limited to MAX_ORDER order. For archs supporting gigantic pages (that is,
page sizes greater than MAX_ORDER), this in turn means that those pages
can't be allocated at runtime.
HugeTLB supports gigantic page allocation during boottime, via the boot
allocator. To this end the kernel provides the command-line options
hugepagesz= and hugepages=, which can be used to instruct the kernel to
allocate N gigantic pages during boot.
For example, x86_64 supports 2M and 1G hugepages, but only 2M hugepages
can be allocated and freed at runtime. If one wants to allocate 1G
gigantic pages, this has to be done at boot via the hugepagesz= and
hugepages= command-line options.
Now, gigantic page allocation at boottime has two serious problems:
1. Boottime allocation is not NUMA aware. On a NUMA machine the kernel
evenly distributes boottime allocated hugepages among nodes.
For example, suppose you have a four-node NUMA machine and want
to allocate four 1G gigantic pages at boottime. The kernel will
allocate one gigantic page per node.
On the other hand, we do have users who want to be able to specify
which NUMA node gigantic pages should allocated from. So that they
can place virtual machines on a specific NUMA node.
2. Gigantic pages allocated at boottime can't be freed
At this point it's important to observe that regular hugepages allocated
at runtime don't have those problems. This is so because HugeTLB
interface for runtime allocation in sysfs supports NUMA and runtime
allocated pages can be freed just fine via the buddy allocator.
This series adds support for allocating gigantic pages at runtime. It
does so by allocating gigantic pages via CMA instead of the buddy
allocator. Releasing gigantic pages is also supported via CMA. As this
series builds on top of the existing HugeTLB interface, it makes gigantic
page allocation and releasing just like regular sized hugepages. This
also means that NUMA support just works.
For example, to allocate two 1G gigantic pages on node 1, one can do:
Vladimir Davydov [Sat, 17 May 2014 13:17:47 +0000 (23:17 +1000)]
slab: document kmalloc_order
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Li Zhong [Sat, 17 May 2014 13:17:47 +0000 (23:17 +1000)]
memory-hotplug: update documentation to hide information about SECTIONS and remove end_phys_index
Seems we all agree that information about SECTION, e.g. section size,
sections per memory block should be kept as kernel internals, and not
exposed to userspace.
This patch updates Documentation/memory-hotplug.txt to refer to memory
blocks instead of memory sections where appropriate and added a paragraph
to explain that memory blocks are made of memory sections. The
documentation update is mostly provided by Nathan.
Also, as end_phys_index in code is actually not the end section id, but
the end memory block id, which should always be the same as phys_index.
So it is removed here.
Signed-off-by: Li Zhong <zhong@linux.vnet.ibm.com> Reviewed-by: Zhang Yanfei <zhangyanfei@cn.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Dave Hansen [Sat, 17 May 2014 13:17:47 +0000 (23:17 +1000)]
mm: pass VM_BUG_ON() reason to dump_page()
I recently added a patch to let folks pass a "reason" string dump_page()
which gets dumped out along with the page's data. This essentially saves
the bug-reader a trip in to the source to figure out why we BUG_ON()'d.
The new VM_BUG_ON_PAGE() passes in NULL for "reason". It seems like we
might as well pass the BUG_ON() condition if we have it. This will bloat
kernels a bit with ~160 new strings, but this is all under a debugging
option anyway.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.cz> Cc: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Johannes Weiner [Sat, 17 May 2014 13:17:46 +0000 (23:17 +1000)]
mm: memcontrol: remove hierarchy restrictions for swappiness and oom_control
Per-memcg swappiness and oom killing can currently not be tweaked on a
memcg that is part of a hierarchy, but not the root of that hierarchy.
Users have complained that they can't configure this when they turned on
hierarchy mode. In fact, with hierarchy mode becoming the default, this
restriction disables the tunables entirely.
But there is no good reason for this restriction. The settings for
swappiness and OOM killing are taken from whatever memcg whose limit
triggered reclaim and OOM invocation, regardless of its position in the
hierarchy tree.
Allow setting swappiness on any group. The knob on the root memcg already
reads the global VM swappiness, make it writable as well.
Allow disabling the OOM killer on any non-root memcg.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.cz> Cc: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
printk is meant to be used with an associated log level. There are some
instances of printk scattered around the mm code where the log level is
missing. Add a log level and adhere to suggestions by
scripts/checkpatch.pl by moving to the pr_* macros.
Also add the typical pr_fmt definition so that print statements can be
easily traced back to the modules where they occur, correlated one with
another, etc. This will require the removal of some (now redundant)
prefixes on a few print statements.
Signed-off-by: Mitchel Humpherys <mitchelh@codeaurora.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
thp: consolidate assert checks in __split_huge_page()
It doesn't make sense to have two assert checks for each invariant: one
for printing and one for BUG().
Let's trigger BUG() if we print error message.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Akinobu Mita [Sat, 17 May 2014 13:17:44 +0000 (23:17 +1000)]
arch/x86/kernel/pci-dma.c: fix dma_generic_alloc_coherent() when CONFIG_DMA_CMA is enabled
dma_generic_alloc_coherent() firstly attempts to allocate by
dma_alloc_from_contiguous() if CONFIG_DMA_CMA is enabled. But the memory
region allocated by it may not fit within the device's DMA mask. This
change makes it fall back to usual alloc_pages_node() allocation for such
cases.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Don Dutile <ddutile@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Andi Kleen <andi@firstfloor.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Akinobu Mita [Sat, 17 May 2014 13:17:44 +0000 (23:17 +1000)]
cma: add placement specifier for "cma=" kernel parameter
Currently, "cma=" kernel parameter is used to specify the size of CMA, but
we can't specify where it is located. We want to locate CMA below 4GB for
devices only supporting 32-bit addressing on 64-bit systems without iommu.
This enables to specify the placement of CMA by extending "cma=" kernel
parameter.
Examples:
1. locate 64MB CMA below 4GB by "cma=64M@0-4G"
2. locate 64MB CMA exact at 512MB by "cma=64M@512M"
Note that the DMA contiguous memory allocator on x86 assumes that
page_address() works for the pages to allocate. So this change requires
to limit end address of contiguous memory area upto max_pfn_mapped to
prevent from locating it on highmem area by the argument of
dma_contiguous_reserve().
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Don Dutile <ddutile@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Akinobu Mita [Sat, 17 May 2014 13:17:44 +0000 (23:17 +1000)]
memblock: introduce memblock_alloc_range()
This introduces memblock_alloc_range() which allocates memblock from the
specified range of physical address. I would like to use this function to
specify the location of CMA.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Don Dutile <ddutile@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Akinobu Mita [Sat, 17 May 2014 13:17:44 +0000 (23:17 +1000)]
intel-iommu: add missing include of dma-contiguous.h
This patch fixes build error on ia64, that is introduced by the patch
intel-iommu-integrate-dma-cma.patch in -mm tree, and this change should
be folded into it.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Reported-by: Fengguang Wu <fengguang.wu@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Akinobu Mita [Sat, 17 May 2014 13:17:43 +0000 (23:17 +1000)]
intel-iommu: integrate DMA CMA
This adds support for the DMA Contiguous Memory Allocator for intel-iommu.
This change enables dma_alloc_coherent() to allocate big contiguous
memory.
It is achieved in the same way as nommu_dma_ops currently does, i.e.
trying to allocate memory by dma_alloc_from_contiguous() and alloc_pages()
is used as a fallback.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Don Dutile <ddutile@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Akinobu Mita [Sat, 17 May 2014 13:17:43 +0000 (23:17 +1000)]
x86: enable DMA CMA with swiotlb
The DMA Contiguous Memory Allocator support on x86 is disabled when
swiotlb config option is enabled. So DMA CMA is always disabled on x86_64
because swiotlb is always enabled. This attempts to support for DMA CMA
with enabling swiotlb config option.
The contiguous memory allocator on x86 is integrated in the function
dma_generic_alloc_coherent() which is .alloc callback in nommu_dma_ops for
dma_alloc_coherent().
x86_swiotlb_alloc_coherent() which is .alloc callback in swiotlb_dma_ops
tries to allocate with dma_generic_alloc_coherent() firstly and then
swiotlb_alloc_coherent() is called as a fallback.
The main part of supporting DMA CMA with swiotlb is that changing
x86_swiotlb_free_coherent() which is .free callback in swiotlb_dma_ops for
dma_free_coherent() so that it can distinguish memory allocated by
dma_generic_alloc_coherent() from one allocated by
swiotlb_alloc_coherent() and release it with dma_generic_free_coherent()
which can handle contiguous memory. This change requires making
is_swiotlb_buffer() global function.
This also needs to change .free callback in the dma_map_ops for amd_gart
and sta2x11, because these dma_ops are also using
dma_generic_alloc_coherent().
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Acked-by: Marek Szyprowski <m.szyprowski@samsung.com> Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Don Dutile <ddutile@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Akinobu Mita [Sat, 17 May 2014 13:17:42 +0000 (23:17 +1000)]
arch/x86/kernel/pci-dma.c: avoid duplicated memset in dma_generic_alloc_coherent()
Fix a duplicated memset that was introduced by the patch
x86-make-dma_alloc_coherent-return-zeroed-memory-if-cma-is-enabled.patch
in -mm tree, and this change should be folded into it.
If dma_generic_alloc_coherent() is called with __GFP_ZERO, it does a
duplicated memset to the memory area allocated by alloc_pages_node() with
__GFP_ZERO. This change fixes that inefficiency by clearing __GFP_ZERO
bit in gfp flages before calling alloc_pages_node(). Note that
dma_generic_alloc_coherent() always returns zeroed memory.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Don Dutile <ddutile@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Andi Kleen <andi@firstfloor.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Akinobu Mita [Sat, 17 May 2014 13:17:42 +0000 (23:17 +1000)]
x86: make dma_alloc_coherent() return zeroed memory if CMA is enabled
This patchset enhances the DMA Contiguous Memory Allocator on x86.
Currently the DMA CMA is only supported with pci-nommu dma_map_ops and
furthermore it can't be enabled on x86_64. But I would like to allocate
big contiguous memory with dma_alloc_coherent() and tell it to the device
that requires it, regardless of which dma mapping implementation is
actually used in the system.
So this makes it work with swiotlb and intel-iommu dma_map_ops, too. And
this also extends "cma=" kernel parameter to specify placement constraint
by the physical address range of memory allocations. For example, CMA
allocates memory below 4GB by "cma=64M@0-4G", it is required for the
devices only supporting 32-bit addressing on 64-bit systems without iommu.
This patch (of 5):
Calling dma_alloc_coherent() with __GFP_ZERO must return zeroed memory.
But when the contiguous memory allocator (CMA) is enabled on x86 and the
memory region is allocated by dma_alloc_from_contiguous(), it doesn't
return zeroed memory. Because dma_generic_alloc_coherent() forgot to fill
the memory region with zero if it was allocated by
dma_alloc_from_contiguous()
Most implementations of dma_alloc_coherent() return zeroed memory
regardless of whether __GFP_ZERO is specified. So this fixes it by
unconditionally zeroing the allocated memory region.
Alternatively, we could fix dma_alloc_from_contiguous() to return zeroed
out memory and remove memset() from all caller of it. But we can't simply
remove the memset on arm because __dma_clear_buffer() is used there for
ensuring cache flushing and it is used in many places. Of course we can
do redundant memset in dma_alloc_from_contiguous(), but I think this patch
is less impact for fixing this problem.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Don Dutile <ddutile@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
For single threaded workloads, we can avoid flushing and iterating through
the entire list of tasks, making the whole function a lot faster,
requiring only a single atomic read for the mm_users.
Suleiman Souhlal [Sat, 17 May 2014 13:17:41 +0000 (23:17 +1000)]
mm: only force scan in reclaim when none of the LRUs are big enough.
Prior to this change, we would decide whether to force scan a LRU during
reclaim if that LRU itself was too small for the current priority.
However, this can lead to the file LRU getting force scanned even if there
are a lot of anonymous pages we can reclaim, leading to hot file pages
getting needlessly reclaimed.
To address this, we instead only force scan when none of the reclaimable
LRUs are big enough.
Gives huge improvements with zswap. For example, when doing -j20 kernel
build in a 500MB container with zswap enabled, runtime (in seconds) is
greatly reduced:
x without this change
+ with this change
N Min Max Median Avg Stddev
x 5 700.997 790.076 763.928 754.05 39.59493
+ 5 141.634 197.899 155.706 161.9 21.270224
Difference at 95.0% confidence
-592.15 +/- 46.3521
-78.5293% +/- 6.14709%
(Student's t, pooled s = 31.7819)
Should also give some improvements in regular (non-zswap) swap cases.
Yes, hughd found significant speedup using regular swap, with several
memcgs under pressure; and it should also be effective in the non-memcg
case, whenever one or another zone LRU is forced too small.
Signed-off-by: Suleiman Souhlal <suleiman@google.com> Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Suleiman Souhlal <suleiman@google.com> Cc: Mel Gorman <mgorman@suse.de> Acked-by: Rik van Riel <riel@redhat.com> Acked-by: Rafael Aquini <aquini@redhat.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: Yuanhan Liu <yuanhan.liu@linux.intel.com> Cc: Seth Jennings <sjennings@variantweb.net> Cc: Bob Liu <bob.liu@oracle.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Luigi Semenzato <semenzato@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Roman Pen [Sat, 17 May 2014 13:17:41 +0000 (23:17 +1000)]
fs/mpage.c: forgotten WRITE_SYNC in case of data integrity write
In case of wbc->sync_mode == WB_SYNC_ALL we need to do data integrity
write, thus mark request as WRITE_SYNC.
akpm: afaict this change will cause the data integrity write bios to be
placed onto the second queue in cfq_io_cq.cfqq[], which presumably results
in special treatment. The documentation for REQ_SYNC is horrid.
Signed-off-by: Roman Pen <r.peniaev@gmail.com> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Tejun Heo <tj@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
projects / karo-tx-linux.git / log