From: Zi Yan <z...@nvidia.com> commit d20bdd571ee5c9966191568527ecdb1bd4b52368 upstream.
In isolate_migratepages_block, if we have too many isolated pages and nr_migratepages is not zero, we should try to migrate what we have without wasting time on isolating. In theory it's possible that multiple parallel compactions will cause too_many_isolated() to become true even if each has isolated less than COMPACT_CLUSTER_MAX, and loop forever in the while loop. Bailing immediately prevents that. [vba...@suse.cz: changelog addition] Fixes: 1da2f328fa64 (“mm,thp,compaction,cma: allow THP migration for CMA allocations”) Suggested-by: Vlastimil Babka <vba...@suse.cz> Signed-off-by: Zi Yan <z...@nvidia.com> Signed-off-by: Andrew Morton <a...@linux-foundation.org> Cc: <sta...@vger.kernel.org> Cc: Mel Gorman <mgor...@techsingularity.net> Cc: Michal Hocko <mho...@kernel.org> Cc: Rik van Riel <r...@surriel.com> Cc: Yang Shi <shy828...@gmail.com> Link: https://lkml.kernel.org/r/20201030183809.3616803-2-zi....@sent.com Signed-off-by: Linus Torvalds <torva...@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gre...@linuxfoundation.org> --- mm/compaction.c | 4 ++++ 1 file changed, 4 insertions(+) --- a/mm/compaction.c +++ b/mm/compaction.c @@ -818,6 +818,10 @@ isolate_migratepages_block(struct compac * delay for some time until fewer pages are isolated */ while (unlikely(too_many_isolated(pgdat))) { + /* stop isolation if there are still pages not migrated */ + if (cc->nr_migratepages) + return 0; + /* async migration should just abort */ if (cc->mode == MIGRATE_ASYNC) return 0;
