For fast flash disk, async IO could introduce overhead because of
context switch.  block-mq now supports IO poll, which improves
performance and latency a lot.  swapin is a good place to use this
technique, because the task is waiting for the swapin page to continue
execution.

In my virtual machine, directly read 4k data from a NVMe with iopoll is
about 60% better than that without poll.  With iopoll support in swapin
patch, my microbenchmark (a task does random memory write) is about
10%~25% faster.  CPU utilization increases a lot though, 2x and even 3x
CPU utilization.  This will depend on disk speed.

While iopoll in swapin isn't intended for all usage cases, it's a win
for latency sensistive workloads with high speed swap disk.  block layer
has knob to control poll in runtime.  If poll isn't enabled in block
layer, there should be no noticeable change in swapin.

I got a chance to run the same test in a NVMe with DRAM as the media.
In simple fio IO test, blkpoll boosts 50% performance in single thread
test and ~20% in 8 threads test.  So this is the base line.  In above
swap test, blkpoll boosts ~27% performance in single thread test.
blkpoll uses 2x CPU time though.

If we enable hybid polling, the performance gain has very slight drop
but CPU time is only 50% worse than that without blkpoll.  Also we can
adjust parameter of hybid poll, with it, the CPU time penality is
reduced further.  In 8 threads test, blkpoll doesn't help though.  The
performance is similar to that without blkpoll, but cpu utilization is
similar too.  There is lock contention in swap path.  The cpu time
spending on blkpoll isn't high.  So overall, blkpoll swapin isn't worse
than that without it.

The swapin readahead might read several pages in in the same time and
form a big IO request.  Since the IO will take longer time, it doesn't
make sense to do poll, so the patch only does iopoll for single page
swapin.

[akpm@linux-foundation.org: coding-style fixes]
Link: http://lkml.kernel.org/r/070c3c3e40b711e7b1390002c991e86a-b5408f0@7511894063d3764ff01ea8111f5a004d7dd700ed078797c204a24e620ddb965cSigned-off-by: NShaohua Li <shli@fb.com>
Cc: Tim Chen <tim.c.chen@intel.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Jens Axboe <axboe@fb.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The add_to_swap aims to allocate swap_space(ie, swap slot and swapcache)
so if it fails due to lack of space in case of THP or something(hdd swap
but tries THP swapout) *caller* rather than add_to_swap itself should
split the THP page and retry it with base page which is more natural.

Link: http://lkml.kernel.org/r/20170515112522.32457-4-ying.huang@intel.comSigned-off-by: NMinchan Kim <minchan@kernel.org>
Signed-off-by: N"Huang, Ying" <ying.huang@intel.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Ebru Akagunduz <ebru.akagunduz@gmail.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Shaohua Li <shli@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Now, get_swap_page takes struct page and allocates swap space according
to page size(ie, normal or THP) so it would be more cleaner to introduce
put_swap_page which is a counter function of get_swap_page.  Then, it
calls right swap slot free function depending on page's size.

[ying.huang@intel.com: minor cleanup and fix]
Link: http://lkml.kernel.org/r/20170515112522.32457-3-ying.huang@intel.comSigned-off-by: NMinchan Kim <minchan@kernel.org>
Signed-off-by: N"Huang, Ying" <ying.huang@intel.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Ebru Akagunduz <ebru.akagunduz@gmail.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Shaohua Li <shli@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Patch series "THP swap: Delay splitting THP during swapping out", v11.

This patchset is to optimize the performance of Transparent Huge Page
(THP) swap.

Recently, the performance of the storage devices improved so fast that
we cannot saturate the disk bandwidth with single logical CPU when do
page swap out even on a high-end server machine.  Because the
performance of the storage device improved faster than that of single
logical CPU.  And it seems that the trend will not change in the near
future.  On the other hand, the THP becomes more and more popular
because of increased memory size.  So it becomes necessary to optimize
THP swap performance.

The advantages of the THP swap support include:

 - Batch the swap operations for the THP to reduce lock
   acquiring/releasing, including allocating/freeing the swap space,
   adding/deleting to/from the swap cache, and writing/reading the swap
   space, etc. This will help improve the performance of the THP swap.

 - The THP swap space read/write will be 2M sequential IO. It is
   particularly helpful for the swap read, which are usually 4k random
   IO. This will improve the performance of the THP swap too.

 - It will help the memory fragmentation, especially when the THP is
   heavily used by the applications. The 2M continuous pages will be
   free up after THP swapping out.

 - It will improve the THP utilization on the system with the swap
   turned on. Because the speed for khugepaged to collapse the normal
   pages into the THP is quite slow. After the THP is split during the
   swapping out, it will take quite long time for the normal pages to
   collapse back into the THP after being swapped in. The high THP
   utilization helps the efficiency of the page based memory management
   too.

There are some concerns regarding THP swap in, mainly because possible
enlarged read/write IO size (for swap in/out) may put more overhead on
the storage device.  To deal with that, the THP swap in should be turned
on only when necessary.  For example, it can be selected via
"always/never/madvise" logic, to be turned on globally, turned off
globally, or turned on only for VMA with MADV_HUGEPAGE, etc.

This patchset is the first step for the THP swap support.  The plan is
to delay splitting THP step by step, finally avoid splitting THP during
the THP swapping out and swap out/in the THP as a whole.

As the first step, in this patchset, the splitting huge page is delayed
from almost the first step of swapping out to after allocating the swap
space for the THP and adding the THP into the swap cache.  This will
reduce lock acquiring/releasing for the locks used for the swap cache
management.

With the patchset, the swap out throughput improves 15.5% (from about
3.73GB/s to about 4.31GB/s) in the vm-scalability swap-w-seq test case
with 8 processes.  The test is done on a Xeon E5 v3 system.  The swap
device used is a RAM simulated PMEM (persistent memory) device.  To test
the sequential swapping out, the test case creates 8 processes, which
sequentially allocate and write to the anonymous pages until the RAM and
part of the swap device is used up.

This patch (of 5):

In this patch, splitting huge page is delayed from almost the first step
of swapping out to after allocating the swap space for the THP
(Transparent Huge Page) and adding the THP into the swap cache.  This
will batch the corresponding operation, thus improve THP swap out
throughput.

This is the first step for the THP swap optimization.  The plan is to
delay splitting the THP step by step and avoid splitting the THP
finally.

In this patch, one swap cluster is used to hold the contents of each THP
swapped out.  So, the size of the swap cluster is changed to that of the
THP (Transparent Huge Page) on x86_64 architecture (512).  For other
architectures which want such THP swap optimization,
ARCH_USES_THP_SWAP_CLUSTER needs to be selected in the Kconfig file for
the architecture.  In effect, this will enlarge swap cluster size by 2
times on x86_64.  Which may make it harder to find a free cluster when
the swap space becomes fragmented.  So that, this may reduce the
continuous swap space allocation and sequential write in theory.  The
performance test in 0day shows no regressions caused by this.

In the future of THP swap optimization, some information of the swapped
out THP (such as compound map count) will be recorded in the
swap_cluster_info data structure.

The mem cgroup swap accounting functions are enhanced to support charge
or uncharge a swap cluster backing a THP as a whole.

The swap cluster allocate/free functions are added to allocate/free a
swap cluster for a THP.  A fair simple algorithm is used for swap
cluster allocation, that is, only the first swap device in priority list
will be tried to allocate the swap cluster.  The function will fail if
the trying is not successful, and the caller will fallback to allocate a
single swap slot instead.  This works good enough for normal cases.  If
the difference of the number of the free swap clusters among multiple
swap devices is significant, it is possible that some THPs are split
earlier than necessary.  For example, this could be caused by big size
difference among multiple swap devices.

The swap cache functions is enhanced to support add/delete THP to/from
the swap cache as a set of (HPAGE_PMD_NR) sub-pages.  This may be
enhanced in the future with multi-order radix tree.  But because we will
split the THP soon during swapping out, that optimization doesn't make
much sense for this first step.

The THP splitting functions are enhanced to support to split THP in swap
cache during swapping out.  The page lock will be held during allocating
the swap cluster, adding the THP into the swap cache and splitting the
THP.  So in the code path other than swapping out, if the THP need to be
split, the PageSwapCache(THP) will be always false.

The swap cluster is only available for SSD, so the THP swap optimization
in this patchset has no effect for HDD.

[ying.huang@intel.com: fix two issues in THP optimize patch]
  Link: http://lkml.kernel.org/r/87k25ed8zo.fsf@yhuang-dev.intel.com
[hannes@cmpxchg.org: extensive cleanups and simplifications, reduce code size]
Link: http://lkml.kernel.org/r/20170515112522.32457-2-ying.huang@intel.comSigned-off-by: N"Huang, Ying" <ying.huang@intel.com>
Signed-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Suggested-by: Andrew Morton <akpm@linux-foundation.org> [for config option]
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> [for changes in huge_memory.c and huge_mm.h]
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Ebru Akagunduz <ebru.akagunduz@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Shaohua Li <shli@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Now vzalloc() is used in swap code to allocate various data structures,
such as swap cache, swap slots cache, cluster info, etc.  Because the
size may be too large on some system, so that normal kzalloc() may fail.
But using kzalloc() has some advantages, for example, less memory
fragmentation, less TLB pressure, etc.  So change the data structure
allocation in swap code to use kvzalloc() which will try kzalloc()
firstly, and fallback to vzalloc() if kzalloc() failed.

In general, although kmalloc() will reduce the number of high-order
pages in short term, vmalloc() will cause more pain for memory
fragmentation in the long term.  And the swap data structure allocation
that is changed in this patch is expected to be long term allocation.

From Dave Hansen:
 "for example, we have a two-page data structure. vmalloc() takes two
  effectively random order-0 pages, probably from two different 2M pages
  and pins them. That "kills" two 2M pages. kmalloc(), allocating two
  *contiguous* pages, will not cross a 2M boundary. That means it will
  only "kill" the possibility of a single 2M page. More 2M pages == less
  fragmentation.

The allocation in this patch occurs during swap on time, which is
usually done during system boot, so usually we have high opportunity to
allocate the contiguous pages successfully.

The allocation for swap_map[] in struct swap_info_struct is not changed,
because that is usually quite large and vmalloc_to_page() is used for
it.  That makes it a little harder to change.

Link: http://lkml.kernel.org/r/20170407064911.25447-1-ying.huang@intel.comSigned-off-by: NHuang Ying <ying.huang@intel.com>
Acked-by: NTim Chen <tim.c.chen@intel.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Acked-by: NRik van Riel <riel@redhat.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Shaohua Li <shli@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit cbab0e4e ("swap: avoid read_swap_cache_async() race to
deadlock while waiting on discard I/O completion") fixed a deadlock in
read_swap_cache_async().  Because at that time, in swap allocation path,
a swap entry may be set as SWAP_HAS_CACHE, then wait for discarding to
complete before the page for the swap entry is added to the swap cache.

But in commit 815c2c54 ("swap: make swap discard async"), the
discarding for swap become asynchronous, waiting for discarding to
complete will be done before the swap entry is set as SWAP_HAS_CACHE.
So the comments in code is incorrect now.  This patch fixes the
comments.

The cond_resched() added in the commit cbab0e4e is not necessary now
too.  But if we added some sleep in swap allocation path in the future,
there may be some hard to debug/reproduce deadlock bug.  So it is kept.

Link: http://lkml.kernel.org/r/20170317064635.12792-1-ying.huang@intel.comSigned-off-by: N"Huang, Ying" <ying.huang@intel.com>
Acked-by: NRafael Aquini <aquini@redhat.com>
Cc: Shaohua Li <shli@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Because during swap off, a swap entry may have swap_map[] ==
SWAP_HAS_CACHE (for example, just allocated).  If we return NULL in
__read_swap_cache_async(), the swap off will abort.  So when swap slot
cache is disabled, (for swap off), we will wait for page to be put into
swap cache in such race condition.  This should not be a problem for swap
slot cache, because swap slot cache should be drained after clearing
swap_slot_cache_enabled.

[ying.huang@intel.com: fix memory leak in __read_swap_cache_async()]
  Link: http://lkml.kernel.org/r/874lzt6znd.fsf@yhuang-dev.intel.com
Link: http://lkml.kernel.org/r/5e2c5f6abe8e6eb0797408897b1bba80938e9b9d.1484082593.git.tim.c.chen@linux.intel.comSigned-off-by: N"Huang, Ying" <ying.huang@intel.com>
Signed-off-by: NTim Chen <tim.c.chen@linux.intel.com>
Cc: Aaron Lu <aaron.lu@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net> escreveu:
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Shaohua Li <shli@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We add per cpu caches for swap slots that can be allocated and freed
quickly without the need to touch the swap info lock.

Two separate caches are maintained for swap slots allocated and swap
slots returned.  This is to allow the swap slots to be returned to the
global pool in a batch so they will have a chance to be coaelesced with
other slots in a cluster.  We do not reuse the slots that are returned
right away, as it may increase fragmentation of the slots.

The swap allocation cache is protected by a mutex as we may sleep when
searching for empty slots in cache.  The swap free cache is protected by
a spin lock as we cannot sleep in the free path.

We refill the swap slots cache when we run out of slots, and we disable
the swap slots cache and drain the slots if the global number of slots
fall below a low watermark threshold.  We re-enable the cache agian when
the slots available are above a high watermark.

[ying.huang@intel.com: use raw_cpu_ptr over this_cpu_ptr for swap slots access]
[tim.c.chen@linux.intel.com: add comments on locks in swap_slots.h]
  Link: http://lkml.kernel.org/r/20170118180327.GA24225@linux.intel.com
Link: http://lkml.kernel.org/r/35de301a4eaa8daa2977de6e987f2c154385eb66.1484082593.git.tim.c.chen@linux.intel.comSigned-off-by: NTim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: N"Huang, Ying" <ying.huang@intel.com>
Reviewed-by: NMichal Hocko <mhocko@suse.com>
Cc: Aaron Lu <aaron.lu@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net> escreveu:
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Shaohua Li <shli@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We can avoid needlessly allocating page for swap slots that are not used
by anyone.  No pages have to be read in for these slots.

Link: http://lkml.kernel.org/r/0784b3f20b9bd3aa5552219624cb78dc4ae710c9.1484082593.git.tim.c.chen@linux.intel.comSigned-off-by: NTim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: N"Huang, Ying" <ying.huang@intel.com>
Cc: Aaron Lu <aaron.lu@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net> escreveu:
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Shaohua Li <shli@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The patch is to improve the scalability of the swap out/in via using
fine grained locks for the swap cache.  In current kernel, one address
space will be used for each swap device.  And in the common
configuration, the number of the swap device is very small (one is
typical).  This causes the heavy lock contention on the radix tree of
the address space if multiple tasks swap out/in concurrently.

But in fact, there is no dependency between pages in the swap cache.  So
that, we can split the one shared address space for each swap device
into several address spaces to reduce the lock contention.  In the
patch, the shared address space is split into 64MB trunks.  64MB is
chosen to balance the memory space usage and effect of lock contention
reduction.

The size of struct address_space on x86_64 architecture is 408B, so with
the patch, 6528B more memory will be used for every 1GB swap space on
x86_64 architecture.

One address space is still shared for the swap entries in the same 64M
trunks.  To avoid lock contention for the first round of swap space
allocation, the order of the swap clusters in the initial free clusters
list is changed.  The swap space distance between the consecutive swap
clusters in the free cluster list is at least 64M.  After the first
round of allocation, the swap clusters are expected to be freed
randomly, so the lock contention should be reduced effectively.

Link: http://lkml.kernel.org/r/735bab895e64c930581ffb0a05b661e01da82bc5.1484082593.git.tim.c.chen@linux.intel.comSigned-off-by: N"Huang, Ying" <ying.huang@intel.com>
Signed-off-by: NTim Chen <tim.c.chen@linux.intel.com>
Cc: Aaron Lu <aaron.lu@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net> escreveu:
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Shaohua Li <shli@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This patch is to improve the performance of swap cache operations when
the type of the swap device is not 0.  Originally, the whole swap entry
value is used as the key of the swap cache, even though there is one
radix tree for each swap device.  If the type of the swap device is not
0, the height of the radix tree of the swap cache will be increased
unnecessary, especially on 64bit architecture.  For example, for a 1GB
swap device on the x86_64 architecture, the height of the radix tree of
the swap cache is 11.  But if the offset of the swap entry is used as
the key of the swap cache, the height of the radix tree of the swap
cache is 4.  The increased height causes unnecessary radix tree
descending and increased cache footprint.

This patch reduces the height of the radix tree of the swap cache via
using the offset of the swap entry instead of the whole swap entry value
as the key of the swap cache.  In 32 processes sequential swap out test
case on a Xeon E5 v3 system with RAM disk as swap, the lock contention
for the spinlock of the swap cache is reduced from 20.15% to 12.19%,
when the type of the swap device is 1.

Use the whole swap entry as key,

  perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.__add_to_swap_cache.add_to_swap_cache.add_to_swap.shrink_page_list: 10.37,
  perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__remove_mapping.shrink_page_list.shrink_inactive_list.shrink_node_memcg: 9.78,

Use the swap offset as key,

  perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.__add_to_swap_cache.add_to_swap_cache.add_to_swap.shrink_page_list: 6.25,
  perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.__remove_mapping.shrink_page_list.shrink_inactive_list.shrink_node_memcg: 5.94,

Link: http://lkml.kernel.org/r/1473270649-27229-1-git-send-email-ying.huang@intel.comSigned-off-by: N"Huang, Ying" <ying.huang@intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Aaron Lu <aaron.lu@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The global zero page is used to satisfy an anonymous read fault.  If
THP(Transparent HugePage) is enabled then the global huge zero page is
used.  The global huge zero page uses an atomic counter for reference
counting and is allocated/freed dynamically according to its counter
value.

CPU time spent on that counter will greatly increase if there are a lot
of processes doing anonymous read faults.  This patch proposes a way to
reduce the access to the global counter so that the CPU load can be
reduced accordingly.

To do this, a new flag of the mm_struct is introduced:
MMF_USED_HUGE_ZERO_PAGE.  With this flag, the process only need to touch
the global counter in two cases:

 1 The first time it uses the global huge zero page;
 2 The time when mm_user of its mm_struct reaches zero.

Note that right now, the huge zero page is eligible to be freed as soon
as its last use goes away.  With this patch, the page will not be
eligible to be freed until the exit of the last process from which it
was ever used.

And with the use of mm_user, the kthread is not eligible to use huge
zero page either.  Since no kthread is using huge zero page today, there
is no difference after applying this patch.  But if that is not desired,
I can change it to when mm_count reaches zero.

Case used for test on Haswell EP:

  usemem -n 72 --readonly -j 0x200000 100G

Which spawns 72 processes and each will mmap 100G anonymous space and
then do read only access to that space sequentially with a step of 2MB.

  CPU cycles from perf report for base commit:
      54.03%  usemem   [kernel.kallsyms]   [k] get_huge_zero_page
  CPU cycles from perf report for this commit:
       0.11%  usemem   [kernel.kallsyms]   [k] mm_get_huge_zero_page

Performance(throughput) of the workload for base commit: 1784430792
Performance(throughput) of the workload for this commit: 4726928591
164% increase.

Runtime of the workload for base commit: 707592 us
Runtime of the workload for this commit: 303970 us
50% drop.

Link: http://lkml.kernel.org/r/fe51a88f-446a-4622-1363-ad1282d71385@intel.comSigned-off-by: NAaron Lu <aaron.lu@intel.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Ebru Akagunduz <ebru.akagunduz@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

File pages use a set of radix tree tags (DIRTY, TOWRITE, WRITEBACK,
etc.) to accelerate finding the pages with a specific tag in the radix
tree during inode writeback.  But for anonymous pages in the swap cache,
there is no inode writeback.  So there is no need to find the pages with
some writeback tags in the radix tree.  It is not necessary to touch
radix tree writeback tags for pages in the swap cache.

Per Rik van Riel's suggestion, a new flag AS_NO_WRITEBACK_TAGS is
introduced for address spaces which don't need to update the writeback
tags.  The flag is set for swap caches.  It may be used for DAX file
systems, etc.

With this patch, the swap out bandwidth improved 22.3% (from ~1.2GB/s to
~1.48GBps) in the vm-scalability swap-w-seq test case with 8 processes.
The test is done on a Xeon E5 v3 system.  The swap device used is a RAM
simulated PMEM (persistent memory) device.  The improvement comes from
the reduced contention on the swap cache radix tree lock.  To test
sequential swapping out, the test case uses 8 processes, which
sequentially allocate and write to the anonymous pages until RAM and
part of the swap device is used up.

Details of comparison is as follow,

base             base+patch
---------------- --------------------------
         %stddev     %change         %stddev
             \          |                \
   2506952 ±  2%     +28.1%    3212076 ±  7%  vm-scalability.throughput
   1207402 ±  7%     +22.3%    1476578 ±  6%  vmstat.swap.so
     10.86 ± 12%     -23.4%       8.31 ± 16%  perf-profile.cycles-pp._raw_spin_lock_irq.__add_to_swap_cache.add_to_swap_cache.add_to_swap.shrink_page_list
     10.82 ± 13%     -33.1%       7.24 ± 14%  perf-profile.cycles-pp._raw_spin_lock_irqsave.__remove_mapping.shrink_page_list.shrink_inactive_list.shrink_zone_memcg
     10.36 ± 11%    -100.0%       0.00 ± -1%  perf-profile.cycles-pp._raw_spin_lock_irqsave.__test_set_page_writeback.bdev_write_page.__swap_writepage.swap_writepage
     10.52 ± 12%    -100.0%       0.00 ± -1%  perf-profile.cycles-pp._raw_spin_lock_irqsave.test_clear_page_writeback.end_page_writeback.page_endio.pmem_rw_page

Link: http://lkml.kernel.org/r/1472578089-5560-1-git-send-email-ying.huang@intel.comSigned-off-by: N"Huang, Ying" <ying.huang@intel.com>
Acked-by: NRik van Riel <riel@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Shaohua Li <shli@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Tejun Heo <tj@kernel.org>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

There are now a number of accounting oddities such as mapped file pages
being accounted for on the node while the total number of file pages are
accounted on the zone.  This can be coped with to some extent but it's
confusing so this patch moves the relevant file-based accounted.  Due to
throttling logic in the page allocator for reliable OOM detection, it is
still necessary to track dirty and writeback pages on a per-zone basis.

[mgorman@techsingularity.net: fix NR_ZONE_WRITE_PENDING accounting]
  Link: http://lkml.kernel.org/r/1468404004-5085-5-git-send-email-mgorman@techsingularity.net
Link: http://lkml.kernel.org/r/1467970510-21195-20-git-send-email-mgorman@techsingularity.netSigned-off-by: NMel Gorman <mgorman@techsingularity.net>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: Hillf Danton <hillf.zj@alibaba-inc.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Rik van Riel <riel@surriel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Christian Borntraeger reported a kernel panic after corrupt page counts,
and it turned out to be a regression introduced with commit aa88b68c
("thp: keep huge zero page pinned until tlb flush"), at least on s390.

put_huge_zero_page() was moved over from zap_huge_pmd() to
release_pages(), and it was replaced by tlb_remove_page().  However,
release_pages() might not always be triggered by (the arch-specific)
tlb_remove_page().

On s390 we call free_page_and_swap_cache() from tlb_remove_page(), and
not tlb_flush_mmu() -> free_pages_and_swap_cache() like the generic
version, because we don't use the MMU-gather logic.  Although both
functions have very similar names, they are doing very unsimilar things,
in particular free_page_xxx is just doing a put_page(), while
free_pages_xxx calls release_pages().

This of course results in very harmful put_page()s on the huge zero
page, on architectures where tlb_remove_page() is implemented in this
way.  It seems to affect only s390 and sh, but sh doesn't have THP
support, so the problem (currently) probably only exists on s390.

The following quick hack fixed the issue:

Link: http://lkml.kernel.org/r/20160602172141.75c006a9@thinkpadSigned-off-by: NGerald Schaefer <gerald.schaefer@de.ibm.com>
Reported-by: NChristian Borntraeger <borntraeger@de.ibm.com>
Tested-by: NChristian Borntraeger <borntraeger@de.ibm.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: <stable@vger.kernel.org>	[4.6.x]
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

v3.16 commit 07a42788 ("mm: shmem: avoid atomic operation during
shmem_getpage_gfp") rightly replaced one instance of SetPageSwapBacked
by __SetPageSwapBacked, pointing out that the newly allocated page is
not yet visible to other users (except speculative get_page_unless_zero-
ers, who may not update page flags before their further checks).

That was part of a series in which Mel was focused on tmpfs profiles:
but almost all SetPageSwapBacked uses can be so optimized, with the same
justification.

Remove ClearPageSwapBacked from __read_swap_cache_async() error path:
it's not an error to free a page with PG_swapbacked set.

Follow a convention of __SetPageLocked, __SetPageSwapBacked instead of
doing it differently in different places; but that's for tidiness - if
the ordering actually mattered, we should not be using the __variants.

There's probably scope for further __SetPageFlags in other places, but
SwapBacked is the one I'm interested in at the moment.
Signed-off-by: NHugh Dickins <hughd@google.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Andres Lagar-Cavilla <andreslc@google.com>
Cc: Yang Shi <yang.shi@linaro.org>
Cc: Ning Qu <quning@gmail.com>
Reviewed-by: NMel Gorman <mgorman@techsingularity.net>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time
ago with promise that one day it will be possible to implement page
cache with bigger chunks than PAGE_SIZE.

This promise never materialized.  And unlikely will.

We have many places where PAGE_CACHE_SIZE assumed to be equal to
PAGE_SIZE.  And it's constant source of confusion on whether
PAGE_CACHE_* or PAGE_* constant should be used in a particular case,
especially on the border between fs and mm.

Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much
breakage to be doable.

Let's stop pretending that pages in page cache are special.  They are
not.

The changes are pretty straight-forward:

 - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;

 - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;

 - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN};

 - page_cache_get() -> get_page();

 - page_cache_release() -> put_page();

This patch contains automated changes generated with coccinelle using
script below.  For some reason, coccinelle doesn't patch header files.
I've called spatch for them manually.

The only adjustment after coccinelle is revert of changes to
PAGE_CAHCE_ALIGN definition: we are going to drop it later.

There are few places in the code where coccinelle didn't reach.  I'll
fix them manually in a separate patch.  Comments and documentation also
will be addressed with the separate patch.

virtual patch

@@
expression E;
@@
- E << (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E

@@
expression E;
@@
- E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E

@@
@@
- PAGE_CACHE_SHIFT
+ PAGE_SHIFT

@@
@@
- PAGE_CACHE_SIZE
+ PAGE_SIZE

@@
@@
- PAGE_CACHE_MASK
+ PAGE_MASK

@@
expression E;
@@
- PAGE_CACHE_ALIGN(E)
+ PAGE_ALIGN(E)

@@
expression E;
@@
- page_cache_get(E)
+ get_page(E)

@@
expression E;
@@
- page_cache_release(E)
+ put_page(E)
Signed-off-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This patchset introduces swap accounting to cgroup2.

This patch (of 7):

In the legacy hierarchy we charge memsw, which is dubious, because:

 - memsw.limit must be >= memory.limit, so it is impossible to limit
   swap usage less than memory usage. Taking into account the fact that
   the primary limiting mechanism in the unified hierarchy is
   memory.high while memory.limit is either left unset or set to a very
   large value, moving memsw.limit knob to the unified hierarchy would
   effectively make it impossible to limit swap usage according to the
   user preference.

 - memsw.usage != memory.usage + swap.usage, because a page occupying
   both swap entry and a swap cache page is charged only once to memsw
   counter. As a result, it is possible to effectively eat up to
   memory.limit of memory pages *and* memsw.limit of swap entries, which
   looks unexpected.

That said, we should provide a different swap limiting mechanism for
cgroup2.

This patch adds mem_cgroup->swap counter, which charges the actual number
of swap entries used by a cgroup.  It is only charged in the unified
hierarchy, while the legacy hierarchy memsw logic is left intact.

The swap usage can be monitored using new memory.swap.current file and
limited using memory.swap.max.

Note, to charge swap resource properly in the unified hierarchy, we have
to make swap_entry_free uncharge swap only when ->usage reaches zero, not
just ->count, i.e.  when all references to a swap entry, including the one
taken by swap cache, are gone.  This is necessary, because otherwise
swap-in could result in uncharging swap even if the page is still in swap
cache and hence still occupies a swap entry.  At the same time, this
shouldn't break memsw counter logic, where a page is never charged twice
for using both memory and swap, because in case of legacy hierarchy we
uncharge swap on commit (see mem_cgroup_commit_charge).
Signed-off-by: NVladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Linux doesn't have an ability to free pages lazy while other OS already
have been supported that named by madvise(MADV_FREE).

The gain is clear that kernel can discard freed pages rather than
swapping out or OOM if memory pressure happens.

Without memory pressure, freed pages would be reused by userspace
without another additional overhead(ex, page fault + allocation +
zeroing).

Jason Evans said:

: Facebook has been using MAP_UNINITIALIZED
: (https://lkml.org/lkml/2012/1/18/308) in some of its applications for
: several years, but there are operational costs to maintaining this
: out-of-tree in our kernel and in jemalloc, and we are anxious to retire it
: in favor of MADV_FREE.  When we first enabled MAP_UNINITIALIZED it
: increased throughput for much of our workload by ~5%, and although the
: benefit has decreased using newer hardware and kernels, there is still
: enough benefit that we cannot reasonably retire it without a replacement.
:
: Aside from Facebook operations, there are numerous broadly used
: applications that would benefit from MADV_FREE.  The ones that immediately
: come to mind are redis, varnish, and MariaDB.  I don't have much insight
: into Android internals and development process, but I would hope to see
: MADV_FREE support eventually end up there as well to benefit applications
: linked with the integrated jemalloc.
:
: jemalloc will use MADV_FREE once it becomes available in the Linux kernel.
: In fact, jemalloc already uses MADV_FREE or equivalent everywhere it's
: available: *BSD, OS X, Windows, and Solaris -- every platform except Linux
: (and AIX, but I'm not sure it even compiles on AIX).  The lack of
: MADV_FREE on Linux forced me down a long series of increasingly
: sophisticated heuristics for madvise() volume reduction, and even so this
: remains a common performance issue for people using jemalloc on Linux.
: Please integrate MADV_FREE; many people will benefit substantially.

How it works:

When madvise syscall is called, VM clears dirty bit of ptes of the
range.  If memory pressure happens, VM checks dirty bit of page table
and if it found still "clean", it means it's a "lazyfree pages" so VM
could discard the page instead of swapping out.  Once there was store
operation for the page before VM peek a page to reclaim, dirty bit is
set so VM can swap out the page instead of discarding.

One thing we should notice is that basically, MADV_FREE relies on dirty
bit in page table entry to decide whether VM allows to discard the page
or not.  IOW, if page table entry includes marked dirty bit, VM
shouldn't discard the page.

However, as a example, if swap-in by read fault happens, page table
entry doesn't have dirty bit so MADV_FREE could discard the page
wrongly.

For avoiding the problem, MADV_FREE did more checks with PageDirty and
PageSwapCache.  It worked out because swapped-in page lives on swap
cache and since it is evicted from the swap cache, the page has PG_dirty
flag.  So both page flags check effectively prevent wrong discarding by
MADV_FREE.

However, a problem in above logic is that swapped-in page has PG_dirty
still after they are removed from swap cache so VM cannot consider the
page as freeable any more even if madvise_free is called in future.

Look at below example for detail.

    ptr = malloc();
    memset(ptr);
    ..
    ..
    .. heavy memory pressure so all of pages are swapped out
    ..
    ..
    var = *ptr; -> a page swapped-in and could be removed from
                   swapcache. Then, page table doesn't mark
                   dirty bit and page descriptor includes PG_dirty
    ..
    ..
    madvise_free(ptr); -> It doesn't clear PG_dirty of the page.
    ..
    ..
    ..
    .. heavy memory pressure again.
    .. In this time, VM cannot discard the page because the page
    .. has *PG_dirty*

To solve the problem, this patch clears PG_dirty if only the page is
owned exclusively by current process when madvise is called because
PG_dirty represents ptes's dirtiness in several processes so we could
clear it only if we own it exclusively.

Firstly, heavy users would be general allocators(ex, jemalloc, tcmalloc
and hope glibc supports it) and jemalloc/tcmalloc already have supported
the feature for other OS(ex, FreeBSD)

  barrios@blaptop:~/benchmark/ebizzy$ lscpu
  Architecture:          x86_64
  CPU op-mode(s):        32-bit, 64-bit
  Byte Order:            Little Endian
  CPU(s):                12
  On-line CPU(s) list:   0-11
  Thread(s) per core:    1
  Core(s) per socket:    1
  Socket(s):             12
  NUMA node(s):          1
  Vendor ID:             GenuineIntel
  CPU family:            6
  Model:                 2
  Stepping:              3
  CPU MHz:               3200.185
  BogoMIPS:              6400.53
  Virtualization:        VT-x
  Hypervisor vendor:     KVM
  Virtualization type:   full
  L1d cache:             32K
  L1i cache:             32K
  L2 cache:              4096K
  NUMA node0 CPU(s):     0-11
  ebizzy benchmark(./ebizzy -S 10 -n 512)

  Higher avg is better.

   vanilla-jemalloc             MADV_free-jemalloc

  1 thread
  records: 10                   records: 10
  avg:   2961.90                avg:  12069.70
  std:     71.96(2.43%)         std:    186.68(1.55%)
  max:   3070.00                max:  12385.00
  min:   2796.00                min:  11746.00

  2 thread
  records: 10                   records: 10
  avg:   5020.00                avg:  17827.00
  std:    264.87(5.28%)         std:    358.52(2.01%)
  max:   5244.00                max:  18760.00
  min:   4251.00                min:  17382.00

  4 thread
  records: 10                   records: 10
  avg:   8988.80                avg:  27930.80
  std:   1175.33(13.08%)        std:   3317.33(11.88%)
  max:   9508.00                max:  30879.00
  min:   5477.00                min:  21024.00

  8 thread
  records: 10                   records: 10
  avg:  13036.50                avg:  33739.40
  std:    170.67(1.31%)         std:   5146.22(15.25%)
  max:  13371.00                max:  40572.00
  min:  12785.00                min:  24088.00

  16 thread
  records: 10                   records: 10
  avg:  11092.40                avg:  31424.20
  std:    710.60(6.41%)         std:   3763.89(11.98%)
  max:  12446.00                max:  36635.00
  min:   9949.00                min:  25669.00

  32 thread
  records: 10                   records: 10
  avg:  11067.00                avg:  34495.80
  std:    971.06(8.77%)         std:   2721.36(7.89%)
  max:  12010.00                max:  38598.00
  min:   9002.00                min:  30636.00

In summary, MADV_FREE is about much faster than MADV_DONTNEED.

This patch (of 12):

Add core MADV_FREE implementation.

[akpm@linux-foundation.org: small cleanups]
Signed-off-by: NMinchan Kim <minchan@kernel.org>
Acked-by: NMichal Hocko <mhocko@suse.com>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: Mika Penttil <mika.penttila@nextfour.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Jason Evans <je@fb.com>
Cc: Daniel Micay <danielmicay@gmail.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Shaohua Li <shli@kernel.org>
Cc: <yalin.wang2010@gmail.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: "James E.J. Bottomley" <jejb@parisc-linux.org>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: "Shaohua Li" <shli@kernel.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Chen Gang <gang.chen.5i5j@gmail.com>
Cc: Chris Zankel <chris@zankel.net>
Cc: Darrick J. Wong <darrick.wong@oracle.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Helge Deller <deller@gmx.de>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Matt Turner <mattst88@gmail.com>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Roland Dreier <roland@kernel.org>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Shaohua Li <shli@kernel.org>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

lock_page() must operate on the whole compound page.  It doesn't make
much sense to lock part of compound page.  Change code to use head
page's PG_locked, if tail page is passed.

This patch also gets rid of custom helper functions --
__set_page_locked() and __clear_page_locked().  They are replaced with
helpers generated by __SETPAGEFLAG/__CLEARPAGEFLAG.  Tail pages to these
helper would trigger VM_BUG_ON().

SLUB uses PG_locked as a bit spin locked.  IIUC, tail pages should never
appear there.  VM_BUG_ON() is added to make sure that this assumption is
correct.

[akpm@linux-foundation.org: fix fs/cifs/file.c]
Signed-off-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Christoph Lameter <cl@linux.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Steve Capper <steve.capper@linaro.org>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Jérôme Glisse <jglisse@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

zswap_get_swap_cache_page and read_swap_cache_async have pretty much the
same code with only significant difference in return value and usage of
swap_readpage.

I a helper __read_swap_cache_async() with the common code.  Behavior
change: now zswap_get_swap_cache_page will use radix_tree_maybe_preload
instead radix_tree_preload.  Looks like, this wasn't changed only by the
reason of code duplication.
Signed-off-by: NDmitry Safonov <0x7f454c46@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@fb.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: David Herrmann <dh.herrmann@gmail.com>
Cc: Seth Jennings <sjennings@variantweb.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We converted some of the usages of ACCESS_ONCE to READ_ONCE in the mm/
tree since it doesn't work reliably on non-scalar types.

This patch removes the rest of the usages of ACCESS_ONCE, and use the new
READ_ONCE API for the read accesses.  This makes things cleaner, instead
of using separate/multiple sets of APIs.
Signed-off-by: NJason Low <jason.low2@hp.com>
Acked-by: NMichal Hocko <mhocko@suse.cz>
Acked-by: NDavidlohr Bueso <dave@stgolabs.net>
Acked-by: NRik van Riel <riel@redhat.com>
Reviewed-by: NChristian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Now that we never use the backing_dev_info pointer in struct address_space
we can simply remove it and save 4 to 8 bytes in every inode.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Acked-by: NRyusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Reviewed-by: NTejun Heo <tj@kernel.org>
Reviewed-by: NJan Kara <jack@suse.cz>
Signed-off-by: NJens Axboe <axboe@fb.com>

This bdi flag isn't too useful - we can determine that a vma is backed by
either swap or shmem trivially in the caller.

This also allows removing the backing_dev_info instaces for swap and shmem
in favor of noop_backing_dev_info.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NTejun Heo <tj@kernel.org>
Reviewed-by: NJan Kara <jack@suse.cz>
Signed-off-by: NJens Axboe <axboe@fb.com>

Now that the external page_cgroup data structure and its lookup is gone,
the only code remaining in there is swap slot accounting.

Rename it and move the conditional compilation into mm/Makefile.
Signed-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Acked-by: NMichal Hocko <mhocko@suse.cz>
Acked-by: NVladimir Davydov <vdavydov@parallels.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Acked-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Tejun Heo <tj@kernel.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

free_pages_and_swap_cache limits release_pages to PAGEVEC_SIZE chunks.
This is not a big deal for the normal release path but it completely kills
memcg uncharge batching which reduces res_counter spin_lock contention.
Dave has noticed this with his page fault scalability test case on a large
machine when the lock was basically dominating on all CPUs:

    80.18%    80.18%  [kernel]               [k] _raw_spin_lock
                  |
                  --- _raw_spin_lock
                     |
                     |--66.59%-- res_counter_uncharge_until
                     |          res_counter_uncharge
                     |          uncharge_batch
                     |          uncharge_list
                     |          mem_cgroup_uncharge_list
                     |          release_pages
                     |          free_pages_and_swap_cache
                     |          tlb_flush_mmu_free
                     |          |
                     |          |--90.12%-- unmap_single_vma
                     |          |          unmap_vmas
                     |          |          unmap_region
                     |          |          do_munmap
                     |          |          vm_munmap
                     |          |          sys_munmap
                     |          |          system_call_fastpath
                     |          |          __GI___munmap
                     |          |
                     |           --9.88%-- tlb_flush_mmu
                     |                     tlb_finish_mmu
                     |                     unmap_region
                     |                     do_munmap
                     |                     vm_munmap
                     |                     sys_munmap
                     |                     system_call_fastpath
                     |                     __GI___munmap

In his case the load was running in the root memcg and that part has been
handled by reverting 05b84301 ("mm: memcontrol: use root_mem_cgroup
res_counter") because this is a clear regression, but the problem remains
inside dedicated memcgs.

There is no reason to limit release_pages to PAGEVEC_SIZE batches other
than lru_lock held times.  This logic, however, can be moved inside the
function.  mem_cgroup_uncharge_list and free_hot_cold_page_list do not
hold any lock for the whole pages_to_free list so it is safe to call them
in a single run.

The release_pages() code was previously breaking the lru_lock each
PAGEVEC_SIZE pages (ie, 14 pages).  However this code has no usage of
pagevecs so switch to breaking the lock at least every SWAP_CLUSTER_MAX
(32) pages.  This means that the lock acquisition frequency is
approximately halved and the max hold times are approximately doubled.

The now unneeded batching is removed from free_pages_and_swap_cache().

Also update the grossly out-of-date release_pages documentation.
Signed-off-by: NMichal Hocko <mhocko@suse.cz>
Signed-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Reported-by: NDave Hansen <dave@sr71.net>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Cc: Greg Thelen <gthelen@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This is designed to avoid a few ifdefs in .c files but it's obnoxious
because it can cause unsuspecting "migrate_page" symbols to get turned into
"NULL".

Just nuke it and use the ifdefs.

Cc: Konstantin Khlebnikov <k.khlebnikov@samsung.com>
Cc: Rafael Aquini <aquini@redhat.com>
Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This patch (of 6):

The i_mmap_writable field counts existing writable mappings of an
address_space.  To allow drivers to prevent new writable mappings, make
this counter signed and prevent new writable mappings if it is negative.
This is modelled after i_writecount and DENYWRITE.

This will be required by the shmem-sealing infrastructure to prevent any
new writable mappings after the WRITE seal has been set.  In case there
exists a writable mapping, this operation will fail with EBUSY.

Note that we rely on the fact that iff you already own a writable mapping,
you can increase the counter without using the helpers.  This is the same
that we do for i_writecount.
Signed-off-by: NDavid Herrmann <dh.herrmann@gmail.com>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: Ryan Lortie <desrt@desrt.ca>
Cc: Lennart Poettering <lennart@poettering.net>
Cc: Daniel Mack <zonque@gmail.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The memcg uncharging code that is involved towards the end of a page's
lifetime - truncation, reclaim, swapout, migration - is impressively
complicated and fragile.

Because anonymous and file pages were always charged before they had their
page->mapping established, uncharges had to happen when the page type
could still be known from the context; as in unmap for anonymous, page
cache removal for file and shmem pages, and swap cache truncation for swap
pages.  However, these operations happen well before the page is actually
freed, and so a lot of synchronization is necessary:

- Charging, uncharging, page migration, and charge migration all need
  to take a per-page bit spinlock as they could race with uncharging.

- Swap cache truncation happens during both swap-in and swap-out, and
  possibly repeatedly before the page is actually freed.  This means
  that the memcg swapout code is called from many contexts that make
  no sense and it has to figure out the direction from page state to
  make sure memory and memory+swap are always correctly charged.

- On page migration, the old page might be unmapped but then reused,
  so memcg code has to prevent untimely uncharging in that case.
  Because this code - which should be a simple charge transfer - is so
  special-cased, it is not reusable for replace_page_cache().

But now that charged pages always have a page->mapping, introduce
mem_cgroup_uncharge(), which is called after the final put_page(), when we
know for sure that nobody is looking at the page anymore.

For page migration, introduce mem_cgroup_migrate(), which is called after
the migration is successful and the new page is fully rmapped.  Because
the old page is no longer uncharged after migration, prevent double
charges by decoupling the page's memcg association (PCG_USED and
pc->mem_cgroup) from the page holding an actual charge.  The new bits
PCG_MEM and PCG_MEMSW represent the respective charges and are transferred
to the new page during migration.

mem_cgroup_migrate() is suitable for replace_page_cache() as well,
which gets rid of mem_cgroup_replace_page_cache().  However, care
needs to be taken because both the source and the target page can
already be charged and on the LRU when fuse is splicing: grab the page
lock on the charge moving side to prevent changing pc->mem_cgroup of a
page under migration.  Also, the lruvecs of both pages change as we
uncharge the old and charge the new during migration, and putback may
race with us, so grab the lru lock and isolate the pages iff on LRU to
prevent races and ensure the pages are on the right lruvec afterward.

Swap accounting is massively simplified: because the page is no longer
uncharged as early as swap cache deletion, a new mem_cgroup_swapout() can
transfer the page's memory+swap charge (PCG_MEMSW) to the swap entry
before the final put_page() in page reclaim.

Finally, page_cgroup changes are now protected by whatever protection the
page itself offers: anonymous pages are charged under the page table lock,
whereas page cache insertions, swapin, and migration hold the page lock.
Uncharging happens under full exclusion with no outstanding references.
Charging and uncharging also ensure that the page is off-LRU, which
serializes against charge migration.  Remove the very costly page_cgroup
lock and set pc->flags non-atomically.

[mhocko@suse.cz: mem_cgroup_charge_statistics needs preempt_disable]
[vdavydov@parallels.com: fix flags definition]
Signed-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Tested-by: NJet Chen <jet.chen@intel.com>
Acked-by: NMichal Hocko <mhocko@suse.cz>
Tested-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NVladimir Davydov <vdavydov@parallels.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

cold is a bool, make it one.  Make the likely case the "if" part of the
block instead of the else as according to the optimisation manual this is
preferred.
Signed-off-by: NMel Gorman <mgorman@suse.de>
Acked-by: NRik van Riel <riel@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Jan Kara <jack@suse.cz>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This is a patch to improve swap readahead algorithm.  It's from Hugh and
I slightly changed it.

Hugh's original changelog:

swapin readahead does a blind readahead, whether or not the swapin is
sequential.  This may be ok on harddisk, because large reads have
relatively small costs, and if the readahead pages are unneeded they can
be reclaimed easily - though, what if their allocation forced reclaim of
useful pages? But on SSD devices large reads are more expensive than
small ones: if the readahead pages are unneeded, reading them in caused
significant overhead.

This patch adds very simplistic random read detection.  Stealing the
PageReadahead technique from Konstantin Khlebnikov's patch, avoiding the
vma/anon_vma sophistications of Shaohua Li's patch, swapin_nr_pages()
simply looks at readahead's current success rate, and narrows or widens
its readahead window accordingly.  There is little science to its
heuristic: it's about as stupid as can be whilst remaining effective.

The table below shows elapsed times (in centiseconds) when running a
single repetitive swapping load across a 1000MB mapping in 900MB ram
with 1GB swap (the harddisk tests had taken painfully too long when I
used mem=500M, but SSD shows similar results for that).

Vanilla is the 3.6-rc7 kernel on which I started; Shaohua denotes his
Sep 3 patch in mmotm and linux-next; HughOld denotes my Oct 1 patch
which Shaohua showed to be defective; HughNew this Nov 14 patch, with
page_cluster as usual at default of 3 (8-page reads); HughPC4 this same
patch with page_cluster 4 (16-page reads); HughPC0 with page_cluster 0
(1-page reads: no readahead).

HDD for swapping to harddisk, SSD for swapping to VertexII SSD.  Seq for
sequential access to the mapping, cycling five times around; Rand for
the same number of random touches.  Anon for a MAP_PRIVATE anon mapping;
Shmem for a MAP_SHARED anon mapping, equivalent to tmpfs.

One weakness of Shaohua's vma/anon_vma approach was that it did not
optimize Shmem: seen below.  Konstantin's approach was perhaps mistuned,
50% slower on Seq: did not compete and is not shown below.

HDD        Vanilla Shaohua HughOld HughNew HughPC4 HughPC0
Seq Anon     73921   76210   75611   76904   78191  121542
Seq Shmem    73601   73176   73855   72947   74543  118322
Rand Anon   895392  831243  871569  845197  846496  841680
Rand Shmem 1058375 1053486  827935  764955  764376  756489

SSD        Vanilla Shaohua HughOld HughNew HughPC4 HughPC0
Seq Anon     24634   24198   24673   25107   21614   70018
Seq Shmem    24959   24932   25052   25703   22030   69678
Rand Anon    43014   26146   28075   25989   26935   25901
Rand Shmem   45349   45215   28249   24268   24138   24332

These tests are, of course, two extremes of a very simple case: under
heavier mixed loads I've not yet observed any consistent improvement or
degradation, and wider testing would be welcome.

Shaohua Li:

Test shows Vanilla is slightly better in sequential workload than Hugh's
patch.  I observed with Hugh's patch sometimes the readahead size is
shrinked too fast (from 8 to 1 immediately) in sequential workload if
there is no hit.  And in such case, continuing doing readahead is good
actually.

I don't prepare a sophisticated algorithm for the sequential workload
because so far we can't guarantee sequential accessed pages are swap out
sequentially.  So I slightly change Hugh's heuristic - don't shrink
readahead size too fast.

Here is my test result (unit second, 3 runs average):
	Vanilla		Hugh		New
Seq	356		370		360
Random	4525		2447		2444

Attached graph is the swapin/swapout throughput I collected with 'vmstat
2'.  The first part is running a random workload (till around 1200 of
the x-axis) and the second part is running a sequential workload.
swapin and swapout throughput are almost identical in steady state in
both workloads.  These are expected behavior.  while in Vanilla, swapin
is much bigger than swapout especially in random workload (because wrong
readahead).

Original patches by: Shaohua Li and Konstantin Khlebnikov.

[fengguang.wu@intel.com: swapin_nr_pages() can be static]
Signed-off-by: NHugh Dickins <hughd@google.com>
Signed-off-by: NShaohua Li <shli@fusionio.com>
Signed-off-by: NFengguang Wu <fengguang.wu@intel.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Konstantin Khlebnikov <khlebnikov@openvz.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Most of the VM_BUG_ON assertions are performed on a page.  Usually, when
one of these assertions fails we'll get a BUG_ON with a call stack and
the registers.

I've recently noticed based on the requests to add a small piece of code
that dumps the page to various VM_BUG_ON sites that the page dump is
quite useful to people debugging issues in mm.

This patch adds a VM_BUG_ON_PAGE(cond, page) which beyond doing what
VM_BUG_ON() does, also dumps the page before executing the actual
BUG_ON.

[akpm@linux-foundation.org: fix up includes]
Signed-off-by: NSasha Levin <sasha.levin@oracle.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

With users of radix_tree_preload() run from interrupt (block/blk-ioc.c is
one such possible user), the following race can happen:

radix_tree_preload()
...
radix_tree_insert()
  radix_tree_node_alloc()
    if (rtp->nr) {
      ret = rtp->nodes[rtp->nr - 1];
<interrupt>
...
radix_tree_preload()
...
radix_tree_insert()
  radix_tree_node_alloc()
    if (rtp->nr) {
      ret = rtp->nodes[rtp->nr - 1];

And we give out one radix tree node twice.  That clearly results in radix
tree corruption with different results (usually OOPS) depending on which
two users of radix tree race.

We fix the problem by making radix_tree_node_alloc() always allocate fresh
radix tree nodes when in interrupt.  Using preloading when in interrupt
doesn't make sense since all the allocations have to be atomic anyway and
we cannot steal nodes from process-context users because some users rely
on radix_tree_insert() succeeding after radix_tree_preload().
in_interrupt() check is somewhat ugly but we cannot simply key off passed
gfp_mask as that is acquired from root_gfp_mask() and thus the same for
all preload users.

Another part of the fix is to avoid node preallocation in
radix_tree_preload() when passed gfp_mask doesn't allow waiting.  Again,
preallocation in such case doesn't make sense and when preallocation would
happen in interrupt we could possibly leak some allocated nodes.  However,
some users of radix_tree_preload() require following radix_tree_insert()
to succeed.  To avoid unexpected effects for these users,
radix_tree_preload() only warns if passed gfp mask doesn't allow waiting
and we provide a new function radix_tree_maybe_preload() for those users
which get different gfp mask from different call sites and which are
prepared to handle radix_tree_insert() failure.
Signed-off-by: NJan Kara <jack@suse.cz>
Cc: Jens Axboe <jaxboe@fusionio.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

read_swap_cache_async() can race against get_swap_page(), and stumble
across a SWAP_HAS_CACHE entry in the swap map whose page wasn't brought
into the swapcache yet.

This transient swap_map state is expected to be transitory, but the
actual placement of discard at scan_swap_map() inserts a wait for I/O
completion thus making the thread at read_swap_cache_async() to loop
around its -EEXIST case, while the other end at get_swap_page() is
scheduled away at scan_swap_map().  This can leave the system deadlocked
if the I/O completion happens to be waiting on the CPU waitqueue where
read_swap_cache_async() is busy looping and !CONFIG_PREEMPT.

This patch introduces a cond_resched() call to make the aforementioned
read_swap_cache_async() busy loop condition to bail out when necessary,
thus avoiding the subtle race window.
Signed-off-by: NRafael Aquini <aquini@redhat.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Acked-by: NKOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: Shaohua Li <shli@kernel.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In page reclaim, huge page is split.  split_huge_page() adds tail pages
to LRU list.  Since we are reclaiming a huge page, it's better we
reclaim all subpages of the huge page instead of just the head page.
This patch adds split tail pages to shrink page list so the tail pages
can be reclaimed soon.

Before this patch, run a swap workload:
  thp_fault_alloc 3492
  thp_fault_fallback 608
  thp_collapse_alloc 6
  thp_collapse_alloc_failed 0
  thp_split 916

With this patch:
  thp_fault_alloc 4085
  thp_fault_fallback 16
  thp_collapse_alloc 90
  thp_collapse_alloc_failed 0
  thp_split 1272

fallback allocation is reduced a lot.

[akpm@linux-foundation.org: fix CONFIG_SWAP=n build]
Signed-off-by: NShaohua Li <shli@fusionio.com>
Acked-by: NRik van Riel <riel@redhat.com>
Acked-by: NMinchan Kim <minchan@kernel.org>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Reviewed-by: NWanpeng Li <liwanp@linux.vnet.ibm.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

swap_writepage() is currently where frontswap hooks into the swap write
path to capture pages with the frontswap_store() function.  However, if
a frontswap backend wants to "resume" the writeback of a page to the
swap device, it can't call swap_writepage() as the page will simply
reenter the backend.

This patch separates swap_writepage() into a top and bottom half, the
bottom half named __swap_writepage() to allow a frontswap backend, like
zswap, to resume writeback beyond the frontswap_store() hook.

__add_to_swap_cache() is also made non-static so that the page for which
writeback is to be resumed can be added to the swap cache.
Signed-off-by: NSeth Jennings <sjenning@linux.vnet.ibm.com>
Signed-off-by: NBob Liu <bob.liu@oracle.com>
Acked-by: NMinchan Kim <minchan@kernel.org>
Reviewed-by: NDan Magenheimer <dan.magenheimer@oracle.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

swap_lock is heavily contended when I test swap to 3 fast SSD (even
slightly slower than swap to 2 such SSD).  The main contention comes
from swap_info_get().  This patch tries to fix the gap with adding a new
per-partition lock.

Global data like nr_swapfiles, total_swap_pages, least_priority and
swap_list are still protected by swap_lock.

nr_swap_pages is an atomic now, it can be changed without swap_lock.  In
theory, it's possible get_swap_page() finds no swap pages but actually
there are free swap pages.  But sounds not a big problem.

Accessing partition specific data (like scan_swap_map and so on) is only
protected by swap_info_struct.lock.

Changing swap_info_struct.flags need hold swap_lock and
swap_info_struct.lock, because scan_scan_map() will check it.  read the
flags is ok with either the locks hold.

If both swap_lock and swap_info_struct.lock must be hold, we always hold
the former first to avoid deadlock.

swap_entry_free() can change swap_list.  To delete that code, we add a
new highest_priority_index.  Whenever get_swap_page() is called, we
check it.  If it's valid, we use it.

It's a pity get_swap_page() still holds swap_lock().  But in practice,
swap_lock() isn't heavily contended in my test with this patch (or I can
say there are other much more heavier bottlenecks like TLB flush).  And
BTW, looks get_swap_page() doesn't really need the lock.  We never free
swap_info[] and we check SWAP_WRITEOK flag.  The only risk without the
lock is we could swapout to some low priority swap, but we can quickly
recover after several rounds of swap, so sounds not a big deal to me.
But I'd prefer to fix this if it's a real problem.

"swap: make each swap partition have one address_space" improved the
swapout speed from 1.7G/s to 2G/s.  This patch further improves the
speed to 2.3G/s, so around 15% improvement.  It's a multi-process test,
so TLB flush isn't the biggest bottleneck before the patches.

[arnd@arndb.de: fix it for nommu]
[hughd@google.com: add missing unlock]
[minchan@kernel.org: get rid of lockdep whinge on sys_swapon]
Signed-off-by: NShaohua Li <shli@fusionio.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Seth Jennings <sjenning@linux.vnet.ibm.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com>
Cc: Dan Magenheimer <dan.magenheimer@oracle.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NHugh Dickins <hughd@google.com>
Signed-off-by: NMinchan Kim <minchan@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

When I use several fast SSD to do swap, swapper_space.tree_lock is
heavily contended.  This makes each swap partition have one
address_space to reduce the lock contention.  There is an array of
address_space for swap.  The swap entry type is the index to the array.

In my test with 3 SSD, this increases the swapout throughput 20%.

[akpm@linux-foundation.org: revert unneeded change to  __add_to_swap_cache]
Signed-off-by: NShaohua Li <shli@fusionio.com>
Cc: Hugh Dickins <hughd@google.com>
Acked-by: NRik van Riel <riel@redhat.com>
Acked-by: NMinchan Kim <minchan@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Currently swapfiles are managed entirely by the core VM by using ->bmap to
allocate space and write to the blocks directly.  This effectively ensures
that the underlying blocks are allocated and avoids the need for the swap
subsystem to locate what physical blocks store offsets within a file.

If the swap subsystem is to use the filesystem information to locate the
blocks, it is critical that information such as block groups, block
bitmaps and the block descriptor table that map the swap file were
resident in memory.  This patch adds address_space_operations that the VM
can call when activating or deactivating swap backed by a file.

  int swap_activate(struct file *);
  int swap_deactivate(struct file *);

The ->swap_activate() method is used to communicate to the file that the
VM relies on it, and the address_space should take adequate measures such
as reserving space in the underlying device, reserving memory for mempools
and pinning information such as the block descriptor table in memory.  The
->swap_deactivate() method is called on sys_swapoff() if ->swap_activate()
returned success.

After a successful swapfile ->swap_activate, the swapfile is marked
SWP_FILE and swapper_space.a_ops will proxy to
sis->swap_file->f_mappings->a_ops using ->direct_io to write swapcache
pages and ->readpage to read.

It is perfectly possible that direct_IO be used to read the swap pages but
it is an unnecessary complication.  Similarly, it is possible that
->writepage be used instead of direct_io to write the pages but filesystem
developers have stated that calling writepage from the VM is undesirable
for a variety of reasons and using direct_IO opens up the possibility of
writing back batches of swap pages in the future.

[a.p.zijlstra@chello.nl: Original patch]
Signed-off-by: NMel Gorman <mgorman@suse.de>
Acked-by: NRik van Riel <riel@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: Eric B Munson <emunson@mgebm.net>
Cc: Eric Paris <eparis@redhat.com>
Cc: James Morris <jmorris@namei.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Mike Christie <michaelc@cs.wisc.edu>
Cc: Neil Brown <neilb@suse.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Cc: Trond Myklebust <Trond.Myklebust@netapp.com>
Cc: Xiaotian Feng <dfeng@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@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

With the patch:
Reads Queued:     734,315,    2,937MiB  Writes Queued:     300,188,    1,200MiB
Read Dispatches:  214,972,    2,937MiB  Write Dispatches:  215,176,    1,200MiB
Reads Requeued:         0               Writes Requeued:         0
Reads Completed:  214,971,    2,937MiB  Writes Completed:  215,177,    1,200MiB
Read Merges:      519,343,    2,077MiB  Write Merges:       73,325,  293,300KiB
IO unplugs:       337,130               Timer unplugs:      11,184

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: NChristian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
Acked-by: NRik van Riel <riel@redhat.com>
Acked-by: NJens Axboe <axboe@kernel.dk>
Reviewed-by: NMinchan Kim <minchan@kernel.org>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>