mm: don't miss the last page because of round-off error
authorRoman Gushchin <guro@fb.com>
Fri, 26 Oct 2018 22:03:27 +0000 (15:03 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 26 Oct 2018 23:25:19 +0000 (16:25 -0700)
commit68600f623d69da428c6163275f97ca126e1a8ec5
tree3894896d27376c03b8242a8be5d20e70af2b28a1
parent591edfb10a949d635ed770c6e85ec5286206c07e
mm: don't miss the last page because of round-off error

I've noticed, that dying memory cgroups are often pinned in memory by a
single pagecache page.  Even under moderate memory pressure they sometimes
stayed in such state for a long time.  That looked strange.

My investigation showed that the problem is caused by applying the LRU
pressure balancing math:

  scan = div64_u64(scan * fraction[lru], denominator),

where

  denominator = fraction[anon] + fraction[file] + 1.

Because fraction[lru] is always less than denominator, if the initial scan
size is 1, the result is always 0.

This means the last page is not scanned and has
no chances to be reclaimed.

Fix this by rounding up the result of the division.

In practice this change significantly improves the speed of dying cgroups
reclaim.

[guro@fb.com: prevent double calculation of DIV64_U64_ROUND_UP() arguments]
Link: http://lkml.kernel.org/r/20180829213311.GA13501@castle
Link: http://lkml.kernel.org/r/20180827162621.30187-3-guro@fb.com
Signed-off-by: Roman Gushchin <guro@fb.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Rik van Riel <riel@surriel.com>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
include/linux/math64.h
mm/vmscan.c