linux.git/mm/migrate.c, branch v4.5

mm: numa: quickly fail allocations for NUMA balancing on full nodes

2016-02-27T18:28:52+00:00

Commit 4167e9b2cf10 ("mm: remove GFP_THISNODE") removed the GFP_THISNODE
flag combination due to confusing semantics.  It noted that
alloc_misplaced_dst_page() was one such user after changes made by
commit e97ca8e5b864 ("mm: fix GFP_THISNODE callers and clarify").

Unfortunately when GFP_THISNODE was removed, users of
alloc_misplaced_dst_page() started waking kswapd and entering direct
reclaim because the wrong GFP flags are cleared.  The consequence is
that workloads that used to fit into memory now get reclaimed which is
addressed by this patch.

The problem can be demonstrated with "mutilate" that exercises memcached
which is software dedicated to memory object caching.  The configuration
uses 80% of memory and is run 3 times for varying numbers of clients.
The results on a 4-socket NUMA box are

mutilate
                            4.4.0                 4.4.0
                          vanilla           numaswap-v1
Hmean    1      8394.71 (  0.00%)     8395.32 (  0.01%)
Hmean    4     30024.62 (  0.00%)    34513.54 ( 14.95%)
Hmean    7     32821.08 (  0.00%)    70542.96 (114.93%)
Hmean    12    55229.67 (  0.00%)    93866.34 ( 69.96%)
Hmean    21    39438.96 (  0.00%)    85749.21 (117.42%)
Hmean    30    37796.10 (  0.00%)    50231.49 ( 32.90%)
Hmean    47    18070.91 (  0.00%)    38530.13 (113.22%)

The metric is queries/second with the more the better.  The results are
way outside of the noise and the reason for the improvement is obvious
from some of the vmstats

                                 4.4.0       4.4.0
                               vanillanumaswap-v1r1
Minor Faults                1929399272  2146148218
Major Faults                  19746529        3567
Swap Ins                      57307366        9913
Swap Outs                     50623229       17094
Allocation stalls                35909         443
DMA allocs                           0           0
DMA32 allocs                  72976349   170567396
Normal allocs               5306640898  5310651252
Movable allocs                       0           0
Direct pages scanned         404130893      799577
Kswapd pages scanned         160230174           0
Kswapd pages reclaimed        55928786           0
Direct pages reclaimed         1843936       41921
Page writes file                  2391           0
Page writes anon              50623229       17094

The vanilla kernel is swapping like crazy with large amounts of direct
reclaim and kswapd activity.  The figures are aggregate but it's known
that the bad activity is throughout the entire test.

Note that simple streaming anon/file memory consumers also see this
problem but it's not as obvious.  In those cases, kswapd is awake when
it should not be.

As there are at least two reclaim-related bugs out there, it's worth
spelling out the user-visible impact.  This patch only addresses bugs
related to excessive reclaim on NUMA hardware when the working set is
larger than a NUMA node.  There is a bug related to high kswapd CPU
usage but the reports are against laptops and other UMA hardware and is
not addressed by this patch.

Signed-off-by: Mel Gorman 
Cc: Vlastimil Babka 
Cc: Johannes Weiner 
Cc: David Rientjes 
Cc: 	[4.1+]
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

thp: introduce deferred_split_huge_page()

2016-01-16T01:56:32+00:00

Currently we don't split huge page on partial unmap.  It's not an ideal
situation.  It can lead to memory overhead.

Furtunately, we can detect partial unmap on page_remove_rmap().  But we
cannot call split_huge_page() from there due to locking context.

It's also counterproductive to do directly from munmap() codepath: in
many cases we will hit this from exit(2) and splitting the huge page
just to free it up in small pages is not what we really want.

The patch introduce deferred_split_huge_page() which put the huge page
into queue for splitting.  The splitting itself will happen when we get
memory pressure via shrinker interface.  The page will be dropped from
list on freeing through compound page destructor.

Signed-off-by: Kirill A. Shutemov 
Tested-by: Sasha Levin 
Tested-by: Aneesh Kumar K.V 
Acked-by: Vlastimil Babka 
Acked-by: Jerome Marchand 
Cc: Andrea Arcangeli 
Cc: Hugh Dickins 
Cc: Dave Hansen 
Cc: Mel Gorman 
Cc: Rik van Riel 
Cc: Naoya Horiguchi 
Cc: Steve Capper 
Cc: Johannes Weiner 
Cc: Michal Hocko 
Cc: Christoph Lameter 
Cc: David Rientjes 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

thp, mm: split_huge_page(): caller need to lock page

2016-01-16T01:56:32+00:00

We're going to use migration entries instead of compound_lock() to
stabilize page refcounts.  Setup and remove migration entries require
page to be locked.

Some of split_huge_page() callers already have the page locked.  Let's
require everybody to lock the page before calling split_huge_page().

Signed-off-by: Kirill A. Shutemov 
Tested-by: Sasha Levin 
Tested-by: Aneesh Kumar K.V 
Acked-by: Vlastimil Babka 
Acked-by: Jerome Marchand 
Cc: Andrea Arcangeli 
Cc: Hugh Dickins 
Cc: Dave Hansen 
Cc: Mel Gorman 
Cc: Rik van Riel 
Cc: Naoya Horiguchi 
Cc: Steve Capper 
Cc: Johannes Weiner 
Cc: Michal Hocko 
Cc: Christoph Lameter 
Cc: David Rientjes 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: rework mapcount accounting to enable 4k mapping of THPs

2016-01-16T01:56:32+00:00

We're going to allow mapping of individual 4k pages of THP compound.  It
means we need to track mapcount on per small page basis.

Straight-forward approach is to use ->_mapcount in all subpages to track
how many time this subpage is mapped with PMDs or PTEs combined.  But
this is rather expensive: mapping or unmapping of a THP page with PMD
would require HPAGE_PMD_NR atomic operations instead of single we have
now.

The idea is to store separately how many times the page was mapped as
whole -- compound_mapcount.  This frees up ->_mapcount in subpages to
track PTE mapcount.

We use the same approach as with compound page destructor and compound
order to store compound_mapcount: use space in first tail page,
->mapping this time.

Any time we map/unmap whole compound page (THP or hugetlb) -- we
increment/decrement compound_mapcount.  When we map part of compound
page with PTE we operate on ->_mapcount of the subpage.

page_mapcount() counts both: PTE and PMD mappings of the page.

Basically, we have mapcount for a subpage spread over two counters.  It
makes tricky to detect when last mapcount for a page goes away.

We introduced PageDoubleMap() for this.  When we split THP PMD for the
first time and there's other PMD mapping left we offset up ->_mapcount
in all subpages by one and set PG_double_map on the compound page.
These additional references go away with last compound_mapcount.

This approach provides a way to detect when last mapcount goes away on
per small page basis without introducing new overhead for most common
cases.

[akpm@linux-foundation.org: fix typo in comment]
[mhocko@suse.com: ignore partial THP when moving task]
Signed-off-by: Kirill A. Shutemov 
Tested-by: Aneesh Kumar K.V 
Acked-by: Jerome Marchand 
Cc: Sasha Levin 
Cc: Aneesh Kumar K.V 
Cc: Jerome Marchand 
Cc: Vlastimil Babka 
Cc: Andrea Arcangeli 
Cc: Hugh Dickins 
Cc: Dave Hansen 
Cc: Mel Gorman 
Cc: Rik van Riel 
Cc: Naoya Horiguchi 
Cc: Steve Capper 
Cc: Johannes Weiner 
Cc: Christoph Lameter 
Cc: David Rientjes 
Signed-off-by: Michal Hocko 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

rmap: add argument to charge compound page

2016-01-16T01:56:32+00:00

We're going to allow mapping of individual 4k pages of THP compound
page.  It means we cannot rely on PageTransHuge() check to decide if
map/unmap small page or THP.

The patch adds new argument to rmap functions to indicate whether we
want to operate on whole compound page or only the small page.

[n-horiguchi@ah.jp.nec.com: fix mapcount mismatch in hugepage migration]
Signed-off-by: Kirill A. Shutemov 
Tested-by: Sasha Levin 
Tested-by: Aneesh Kumar K.V 
Acked-by: Vlastimil Babka 
Acked-by: Jerome Marchand 
Cc: Andrea Arcangeli 
Cc: Hugh Dickins 
Cc: Dave Hansen 
Cc: Mel Gorman 
Cc: Rik van Riel 
Cc: Steve Capper 
Cc: Johannes Weiner 
Cc: Michal Hocko 
Cc: Christoph Lameter 
Cc: David Rientjes 
Signed-off-by: Naoya Horiguchi 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

page-flags: define PG_locked behavior on compound pages

2016-01-16T01:56:32+00:00

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: Kirill A. Shutemov 
Cc: Andrea Arcangeli 
Cc: Hugh Dickins 
Cc: Dave Hansen 
Cc: Mel Gorman 
Cc: Rik van Riel 
Cc: Vlastimil Babka 
Cc: Christoph Lameter 
Cc: Naoya Horiguchi 
Cc: Steve Capper 
Cc: "Aneesh Kumar K.V" 
Cc: Johannes Weiner 
Cc: Michal Hocko 
Cc: Jerome Marchand 
Cc: Jérôme Glisse 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm, page_alloc: rename __GFP_WAIT to __GFP_RECLAIM

2015-11-07T01:50:42+00:00

__GFP_WAIT was used to signal that the caller was in atomic context and
could not sleep.  Now it is possible to distinguish between true atomic
context and callers that are not willing to sleep.  The latter should
clear __GFP_DIRECT_RECLAIM so kswapd will still wake.  As clearing
__GFP_WAIT behaves differently, there is a risk that people will clear the
wrong flags.  This patch renames __GFP_WAIT to __GFP_RECLAIM to clearly
indicate what it does -- setting it allows all reclaim activity, clearing
them prevents it.

[akpm@linux-foundation.org: fix build]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Mel Gorman 
Acked-by: Michal Hocko 
Acked-by: Vlastimil Babka 
Acked-by: Johannes Weiner 
Cc: Christoph Lameter 
Acked-by: David Rientjes 
Cc: Vitaly Wool 
Cc: Rik van Riel 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm, page_alloc: distinguish between being unable to sleep, unwilling to sleep and avoiding waking kswapd

2015-11-07T01:50:42+00:00

__GFP_WAIT has been used to identify atomic context in callers that hold
spinlocks or are in interrupts.  They are expected to be high priority and
have access one of two watermarks lower than "min" which can be referred
to as the "atomic reserve".  __GFP_HIGH users get access to the first
lower watermark and can be called the "high priority reserve".

Over time, callers had a requirement to not block when fallback options
were available.  Some have abused __GFP_WAIT leading to a situation where
an optimisitic allocation with a fallback option can access atomic
reserves.

This patch uses __GFP_ATOMIC to identify callers that are truely atomic,
cannot sleep and have no alternative.  High priority users continue to use
__GFP_HIGH.  __GFP_DIRECT_RECLAIM identifies callers that can sleep and
are willing to enter direct reclaim.  __GFP_KSWAPD_RECLAIM to identify
callers that want to wake kswapd for background reclaim.  __GFP_WAIT is
redefined as a caller that is willing to enter direct reclaim and wake
kswapd for background reclaim.

This patch then converts a number of sites

o __GFP_ATOMIC is used by callers that are high priority and have memory
  pools for those requests. GFP_ATOMIC uses this flag.

o Callers that have a limited mempool to guarantee forward progress clear
  __GFP_DIRECT_RECLAIM but keep __GFP_KSWAPD_RECLAIM. bio allocations fall
  into this category where kswapd will still be woken but atomic reserves
  are not used as there is a one-entry mempool to guarantee progress.

o Callers that are checking if they are non-blocking should use the
  helper gfpflags_allow_blocking() where possible. This is because
  checking for __GFP_WAIT as was done historically now can trigger false
  positives. Some exceptions like dm-crypt.c exist where the code intent
  is clearer if __GFP_DIRECT_RECLAIM is used instead of the helper due to
  flag manipulations.

o Callers that built their own GFP flags instead of starting with GFP_KERNEL
  and friends now also need to specify __GFP_KSWAPD_RECLAIM.

The first key hazard to watch out for is callers that removed __GFP_WAIT
and was depending on access to atomic reserves for inconspicuous reasons.
In some cases it may be appropriate for them to use __GFP_HIGH.

The second key hazard is callers that assembled their own combination of
GFP flags instead of starting with something like GFP_KERNEL.  They may
now wish to specify __GFP_KSWAPD_RECLAIM.  It's almost certainly harmless
if it's missed in most cases as other activity will wake kswapd.

Signed-off-by: Mel Gorman 
Acked-by: Vlastimil Babka 
Acked-by: Michal Hocko 
Acked-by: Johannes Weiner 
Cc: Christoph Lameter 
Cc: David Rientjes 
Cc: Vitaly Wool 
Cc: Rik van Riel 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: migrate dirty page without clear_page_dirty_for_io etc

2015-11-06T03:34:48+00:00

clear_page_dirty_for_io() has accumulated writeback and memcg subtleties
since v2.6.16 first introduced page migration; and the set_page_dirty()
which completed its migration of PageDirty, later had to be moderated to
__set_page_dirty_nobuffers(); then PageSwapBacked had to skip that too.

No actual problems seen with this procedure recently, but if you look into
what the clear_page_dirty_for_io(page)+set_page_dirty(newpage) is actually
achieving, it turns out to be nothing more than moving the PageDirty flag,
and its NR_FILE_DIRTY stat from one zone to another.

It would be good to avoid a pile of irrelevant decrementations and
incrementations, and improper event counting, and unnecessary descent of
the radix_tree under tree_lock (to set the PAGECACHE_TAG_DIRTY which
radix_tree_replace_slot() left in place anyway).

Do the NR_FILE_DIRTY movement, like the other stats movements, while
interrupts still disabled in migrate_page_move_mapping(); and don't even
bother if the zone is the same.  Do the PageDirty movement there under
tree_lock too, where old page is frozen and newpage not yet visible:
bearing in mind that as soon as newpage becomes visible in radix_tree, an
un-page-locked set_page_dirty() might interfere (or perhaps that's just
not possible: anything doing so should already hold an additional
reference to the old page, preventing its migration; but play safe).

But we do still need to transfer PageDirty in migrate_page_copy(), for
those who don't go the mapping route through migrate_page_move_mapping().

Signed-off-by: Hugh Dickins 
Cc: Christoph Lameter 
Cc: "Kirill A. Shutemov" 
Cc: Rik van Riel 
Cc: Vlastimil Babka 
Cc: Davidlohr Bueso 
Cc: Oleg Nesterov 
Cc: Sasha Levin 
Cc: Dmitry Vyukov 
Cc: KOSAKI Motohiro 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: page migration avoid touching newpage until no going back

2015-11-06T03:34:48+00:00

We have had trouble in the past from the way in which page migration's
newpage is initialized in dribs and drabs - see commit 8bdd63809160 ("mm:
fix direct reclaim writeback regression") which proposed a cleanup.

We have no actual problem now, but I think the procedure would be clearer
(and alternative get_new_page pools safer to implement) if we assert that
newpage is not touched until we are sure that it's going to be used -
except for taking the trylock on it in __unmap_and_move().

So shift the early initializations from move_to_new_page() into
migrate_page_move_mapping(), mapping and NULL-mapping paths.  Similarly
migrate_huge_page_move_mapping(), but its NULL-mapping path can just be
deleted: you cannot reach hugetlbfs_migrate_page() with a NULL mapping.

Adjust stages 3 to 8 in the Documentation file accordingly.

Signed-off-by: Hugh Dickins 
Cc: Christoph Lameter 
Cc: "Kirill A. Shutemov" 
Cc: Rik van Riel 
Cc: Vlastimil Babka 
Cc: Davidlohr Bueso 
Cc: Oleg Nesterov 
Cc: Sasha Levin 
Cc: Dmitry Vyukov 
Cc: KOSAKI Motohiro 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds