diff options
author | Mel Gorman <mgorman@suse.de> | 2012-11-27 10:31:44 +0000 |
---|---|---|
committer | Mel Gorman <mgorman@suse.de> | 2012-12-11 14:42:58 +0000 |
commit | 7548341b28956ccd35a63ab12f01d8541041aa70 (patch) | |
tree | efa2a33587ac785c6462fee04146c52258740e66 | |
parent | 220018d388b8ab1fca1c5f0c6474bab47ad2c9c0 (diff) | |
download | op-kernel-dev-7548341b28956ccd35a63ab12f01d8541041aa70.zip op-kernel-dev-7548341b28956ccd35a63ab12f01d8541041aa70.tar.gz |
mm: numa: Account for failed allocations and isolations as migration failures
Subject says it all. Allocation failures and a failure to isolate should
be accounted as a migration failure. This is partially another
difference between base page and transhuge page migration. A base page
migration makes multiple attempts for these conditions before it would
be accounted for as a failure.
Signed-off-by: Mel Gorman <mgorman@suse.de>
-rw-r--r-- | mm/migrate.c | 5 |
1 files changed, 4 insertions, 1 deletions
diff --git a/mm/migrate.c b/mm/migrate.c index 9341a50..26537c4 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1635,12 +1635,15 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm, new_page = alloc_pages_node(node, (GFP_TRANSHUGE | GFP_THISNODE) & ~__GFP_WAIT, HPAGE_PMD_ORDER); - if (!new_page) + if (!new_page) { + count_vm_events(PGMIGRATE_FAIL, HPAGE_PMD_NR); goto out_dropref; + } page_xchg_last_nid(new_page, page_last_nid(page)); isolated = numamigrate_isolate_page(pgdat, page); if (!isolated) { + count_vm_events(PGMIGRATE_FAIL, HPAGE_PMD_NR); put_page(new_page); goto out_keep_locked; } |