mm, page_alloc: uninline the bad page part of check_new_page()
authorVlastimil Babka <vbabka@suse.cz>
Fri, 20 May 2016 00:14:41 +0000 (17:14 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 20 May 2016 02:12:14 +0000 (19:12 -0700)
commit4e6118016eb7986109ad61b00186579f384f956a
treef3c06e8c7d5c726bacb8dafb5fd959484850bc99
parente2769dbdc51f1baa1908ecf6c84d50f19577e1db
mm, page_alloc: uninline the bad page part of check_new_page()

Bad pages should be rare so the code handling them doesn't need to be
inline for performance reasons.  Put it to separate function which
returns void.  This also assumes that the initial page_expected_state()
result will match the result of the thorough check, i.e.  the page
doesn't become "good" in the meanwhile.  This matches the same
expectations already in place in free_pages_check().

!DEBUG_VM bloat-o-meter:

  add/remove: 1/0 grow/shrink: 0/1 up/down: 134/-274 (-140)
  function                                     old     new   delta
  check_new_page_bad                             -     134    +134
  get_page_from_freelist                      3468    3194    -274

Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/page_alloc.c