On Fri, Aug 02, 2013 at 06:06:28PM +0200, Andrea Arcangeli wrote: > Zone reclaim locked breaks zone_reclaim_mode=1. If more than one > thread allocates memory at the same time, it forces a premature > allocation into remote NUMA nodes even when there's plenty of clean > cache to reclaim in the local nodes. > > Signed-off-by: Andrea Arcangeli <aarcange@xxxxxxxxxx> > Reviewed-by: Rik van Riel <riel@xxxxxxxxxx> > Acked-by: Rafael Aquini <aquini@xxxxxxxxxx> > Acked-by: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx> I would have preferred if the changelog included the history of why this exists at all (prevents excessive reclaim from parallel allocation requests) and why it should not currently be a problem (SWAP_CLUSTER_MAX should be strictly obeyed limiting the excessive reclaim to SWAP_CLUSTER_MAX * nr_parallel_requests). Hopefully we'll remember to connect any bugs about excessive reclaim + zone_reclaim to this patch :) Acked-by: Mel Gorman <mgorman@xxxxxxx> -- Mel Gorman SUSE Labs -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>