One node from the test plan showed, number of seeks still is not enough. Increase it again twice more to fit new formula introduces in commit e008b95a28ef "ms/mm: use sc->priority for slab shrink targets".
https://jira.sw.ru/browse/PSBM-77015 Signed-off-by: Kirill Tkhai <ktk...@virtuozzo.com> --- mm/tcache.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/tcache.c b/mm/tcache.c index 8b893e6b520..b5157d9861d 100644 --- a/mm/tcache.c +++ b/mm/tcache.c @@ -1202,7 +1202,7 @@ static unsigned long tcache_shrink_scan(struct shrinker *shrink, struct shrinker tcache_shrinker = { .count_objects = tcache_shrink_count, .scan_objects = tcache_shrink_scan, - .seeks = 4, + .seeks = 8, .batch = TCACHE_SCAN_BATCH, .flags = SHRINKER_NUMA_AWARE, }; _______________________________________________ Devel mailing list Devel@openvz.org https://lists.openvz.org/mailman/listinfo/devel