]> git.proxmox.com Git - mirror_ubuntu-kernels.git/commit
mm: vmscan: remove double slab pressure by inc'ing sc->nr_scanned
authorYang Shi <yang.shi@linux.alibaba.com>
Fri, 12 Jul 2019 03:59:27 +0000 (20:59 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 12 Jul 2019 18:05:46 +0000 (11:05 -0700)
commitaf5d440365894b5ca51f29866c1a01496dce52c4
tree690efae5a91613b4e4b487660eba6532e23296a4
parent23a5c8cb7a91939cb08bc2dc880a7aa882bc6241
mm: vmscan: remove double slab pressure by inc'ing sc->nr_scanned

Commit 9092c71bb724 ("mm: use sc->priority for slab shrink targets") has
broken up the relationship between sc->nr_scanned and slab pressure.
The sc->nr_scanned can't double slab pressure anymore.  So, it sounds no
sense to still keep sc->nr_scanned inc'ed.  Actually, it would prevent
from adding pressure on slab shrink since excessive sc->nr_scanned would
prevent from scan->priority raise.

The bonnie test doesn't show this would change the behavior of slab
shrinkers.

w/ w/o
  /sec    %CP      /sec      %CP
Sequential delete:  3960.6    94.6    3997.6     96.2
Random delete:  2518      63.8    2561.6     64.6

The slight increase of "/sec" without the patch would be caused by the
slight increase of CPU usage.

Link: http://lkml.kernel.org/r/1559025859-72759-1-git-send-email-yang.shi@linux.alibaba.com
Signed-off-by: Yang Shi <yang.shi@linux.alibaba.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/vmscan.c