Doug / smarc-fsl-linux-kernel

Download as
Browse Code ยป

Commit bbdb396a60b2ebf7de3b717991e5d3e28c8b7bbd

Authored by Martin Bligh
Committed by Linus Torvalds
1 parent 3bb1a852ab
Exists in master and in 7 other branches imx_3.0.35_4.1.0, imx_3.10.17_1.0.1_ga, imx_3.10.53_1.1.0_ga, imx_3.14.28_1.0.0_ga, smarc-imx_3.10.53_1.1.0_ga, smarc-imx_3.14.28_1.0.0_ga, smarc-l5.0.0_1.0.0-ga

[PATCH] Use min of two prio settings in calculating distress for reclaim 

If try_to_free_pages / balance_pgdat are called with a gfp_mask specifying
GFP_IO and/or GFP_FS, they will reclaim the requisite number of pages, and the
reset prev_priority to DEF_PRIORITY (or to some other high (ie: unurgent)
value).

However, another reclaimer without those gfp_mask flags set (say, GFP_NOIO)
may still be struggling to reclaim pages.  The concurrent overwrite of
zone->prev_priority will cause this GFP_NOIO thread to unexpectedly cease
deactivating mapped pages, thus causing reclaim difficulties.

Fix this is to key the distress calculation not off zone->prev_priority, but
also take into account the local caller's priority by using
min(zone->prev_priority, sc->priority)

Signed-off-by: Martin J. Bligh <mbligh@google.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Showing 1 changed file with 4 additions and 4 deletions Side-by-side Diff

... ... @@ -760,7 +760,7 @@
760 760 * But we had to alter page->flags anyway.
761 761 */
762 762 static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
763   - struct scan_control *sc)
  763 + struct scan_control *sc, int priority)
764 764 {
765 765 unsigned long pgmoved;
766 766 int pgdeactivate = 0;
... ... @@ -784,7 +784,7 @@
784 784 * `distress' is a measure of how much trouble we're having
785 785 * reclaiming pages. 0 -> no problems. 100 -> great trouble.
786 786 */
787   - distress = 100 >> zone->prev_priority;
  787 + distress = 100 >> min(zone->prev_priority, priority);
788 788  
789 789 /*
790 790 * The point of this algorithm is to decide when to start
... ... @@ -936,7 +936,7 @@
936 936 nr_to_scan = min(nr_active,
937 937 (unsigned long)sc->swap_cluster_max);
938 938 nr_active -= nr_to_scan;
939   - shrink_active_list(nr_to_scan, zone, sc);
  939 + shrink_active_list(nr_to_scan, zone, sc, priority);
940 940 }
941 941  
942 942 if (nr_inactive) {
... ... @@ -1384,7 +1384,7 @@
1384 1384 if (zone->nr_scan_active >= nr_pages || pass > 3) {
1385 1385 zone->nr_scan_active = 0;
1386 1386 nr_to_scan = min(nr_pages, zone->nr_active);
1387   - shrink_active_list(nr_to_scan, zone, sc);
  1387 + shrink_active_list(nr_to_scan, zone, sc, prio);
1388 1388 }
1389 1389 }
1390 1390