mm: vmscan: clarify how swappiness, highest priority, memcg interact
authorJohannes Weiner <hannes@cmpxchg.org>
Sat, 23 Feb 2013 00:32:14 +0000 (16:32 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Sun, 24 Feb 2013 01:50:09 +0000 (17:50 -0800)
commit10316b313cbde4b778c3d1b4b2fe2adbcbe84a48
tree3e91c097ac6409604acae2695fb2984ebda55615
parentd778df51c09264076fe0208c099ef7d428f21790
mm: vmscan: clarify how swappiness, highest priority, memcg interact

A swappiness of 0 has a slightly different meaning for global reclaim
(may swap if file cache really low) and memory cgroup reclaim (never
swap, ever).

In addition, global reclaim at highest priority will scan all LRU lists
equal to their size and ignore other balancing heuristics.  UNLESS
swappiness forbids swapping, then the lists are balanced based on recent
reclaim effectiveness.  UNLESS file cache is running low, then anonymous
pages are force-scanned.

This (total mess of a) behaviour is implicit and not obvious from the
way the code is organized.  At least make it apparent in the code flow
and document the conditions.  It will be it easier to come up with sane
semantics later.

Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Rik van Riel <riel@redhat.com>
Reviewed-by: Satoru Moriya <satoru.moriya@hds.com>
Reviewed-by: Michal Hocko <mhocko@suse.cz>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Hugh Dickins <hughd@google.com>
Cc: Simon Jeons <simon.jeons@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/vmscan.c