Currently in addition to a shrink_control struct shrink_slab() takes two arguments, nr_pages_scanned and lru_pages, which are used for balancing slab reclaim versus page reclaim - roughly speaking, shrink_slab() will try to scan nr_pages_scanned/lru_pages fraction of all slab objects. However, shrink_slab() is not always called after page cache reclaim. For example, drop_slab() uses shrink_slab() to drop as many slab objects as possible and thus has to pass phony values 1000/1000 to it, which do not make sense for nr_pages_scanned/lru_pages. Moreover, as soon as kmemcg reclaim is introduced, we will have to make up phony values for nr_pages_scanned and lru_pages again when doing kmem-only reclaim for a memory cgroup, which is possible if the cgroup has its kmem limit less than the total memory limit.
Signed-off-by: Vladimir Davydov <[email protected]> Cc: Johannes Weiner <[email protected]> Cc: Michal Hocko <[email protected]> Cc: Andrew Morton <[email protected]> Cc: Mel Gorman <[email protected]> Cc: Rik van Riel <[email protected]> --- include/linux/mm.h | 3 +-- include/trace/events/vmscan.h | 20 ++++++++++---------- mm/vmscan.c | 26 +++++++++++++------------- 3 files changed, 24 insertions(+), 25 deletions(-) diff --git a/include/linux/mm.h b/include/linux/mm.h index 1cedd00..71c7f50 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1926,8 +1926,7 @@ int drop_caches_sysctl_handler(struct ctl_table *, int, #endif unsigned long shrink_slab(struct shrink_control *shrink, - unsigned long nr_pages_scanned, - unsigned long lru_pages); + unsigned long fraction, unsigned long denominator); #ifndef CONFIG_MMU #define randomize_va_space 0 diff --git a/include/trace/events/vmscan.h b/include/trace/events/vmscan.h index 132a985..6bed4ab 100644 --- a/include/trace/events/vmscan.h +++ b/include/trace/events/vmscan.h @@ -181,11 +181,11 @@ DEFINE_EVENT(mm_vmscan_direct_reclaim_end_template, mm_vmscan_memcg_softlimit_re TRACE_EVENT(mm_shrink_slab_start, TP_PROTO(struct shrinker *shr, struct shrink_control *sc, - long nr_objects_to_shrink, unsigned long pgs_scanned, - unsigned long lru_pgs, unsigned long cache_items, + long nr_objects_to_shrink, unsigned long frac, + unsigned long denom, unsigned long cache_items, unsigned long long delta, unsigned long total_scan), - TP_ARGS(shr, sc, nr_objects_to_shrink, pgs_scanned, lru_pgs, + TP_ARGS(shr, sc, nr_objects_to_shrink, frac, denom, cache_items, delta, total_scan), TP_STRUCT__entry( @@ -193,8 +193,8 @@ TRACE_EVENT(mm_shrink_slab_start, __field(void *, shrink) __field(long, nr_objects_to_shrink) __field(gfp_t, gfp_flags) - __field(unsigned long, pgs_scanned) - __field(unsigned long, lru_pgs) + __field(unsigned long, frac) + __field(unsigned long, denom) __field(unsigned long, cache_items) __field(unsigned long long, delta) __field(unsigned long, total_scan) @@ -205,20 +205,20 @@ TRACE_EVENT(mm_shrink_slab_start, __entry->shrink = shr->scan_objects; __entry->nr_objects_to_shrink = nr_objects_to_shrink; __entry->gfp_flags = sc->gfp_mask; - __entry->pgs_scanned = pgs_scanned; - __entry->lru_pgs = lru_pgs; + __entry->frac = frac; + __entry->denom = denom; __entry->cache_items = cache_items; __entry->delta = delta; __entry->total_scan = total_scan; ), - TP_printk("%pF %p: objects to shrink %ld gfp_flags %s pgs_scanned %ld lru_pgs %ld cache items %ld delta %lld total_scan %ld", + TP_printk("%pF %p: objects to shrink %ld gfp_flags %s frac %ld denom %ld cache items %ld delta %lld total_scan %ld", __entry->shrink, __entry->shr, __entry->nr_objects_to_shrink, show_gfp_flags(__entry->gfp_flags), - __entry->pgs_scanned, - __entry->lru_pgs, + __entry->frac, + __entry->denom, __entry->cache_items, __entry->delta, __entry->total_scan) diff --git a/mm/vmscan.c b/mm/vmscan.c index eea668d..6946997 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -219,7 +219,7 @@ EXPORT_SYMBOL(unregister_shrinker); static unsigned long shrink_slab_node(struct shrink_control *shrinkctl, struct shrinker *shrinker, - unsigned long nr_pages_scanned, unsigned long lru_pages) + unsigned long fraction, unsigned long denominator) { unsigned long freed = 0; unsigned long long delta; @@ -243,9 +243,9 @@ shrink_slab_node(struct shrink_control *shrinkctl, struct shrinker *shrinker, nr = atomic_long_xchg(&shrinker->nr_deferred[nid], 0); total_scan = nr; - delta = (4 * nr_pages_scanned) / shrinker->seeks; + delta = (4 * fraction) / shrinker->seeks; delta *= max_pass; - do_div(delta, lru_pages + 1); + do_div(delta, denominator + 1); total_scan += delta; if (total_scan < 0) { printk(KERN_ERR @@ -278,7 +278,7 @@ shrink_slab_node(struct shrink_control *shrinkctl, struct shrinker *shrinker, total_scan = max_pass * 2; trace_mm_shrink_slab_start(shrinker, shrinkctl, nr, - nr_pages_scanned, lru_pages, + fraction, denominator, max_pass, delta, total_scan); while (total_scan >= batch_size) { @@ -322,23 +322,23 @@ shrink_slab_node(struct shrink_control *shrinkctl, struct shrinker *shrinker, * If the vm encountered mapped pages on the LRU it increase the pressure on * slab to avoid swapping. * - * We do weird things to avoid (scanned*seeks*entries) overflowing 32 bits. + * We do weird things to avoid (fraction*seeks*entries) overflowing 32 bits. * - * `lru_pages' represents the number of on-LRU pages in all the zones which - * are eligible for the caller's allocation attempt. It is used for balancing - * slab reclaim versus page reclaim. + * `fraction' and `denominator' are used for balancing slab reclaim versus page + * reclaim. To scan slab objects proportionally to page cache, pass the number + * of pages scanned and the total number of on-LRU pages in all the zones which + * are eligible for the caller's allocation attempt respectively. * * Returns the number of slab objects which we shrunk. */ unsigned long shrink_slab(struct shrink_control *shrinkctl, - unsigned long nr_pages_scanned, - unsigned long lru_pages) + unsigned long fraction, unsigned long denominator) { struct shrinker *shrinker; unsigned long freed = 0; - if (nr_pages_scanned == 0) - nr_pages_scanned = SWAP_CLUSTER_MAX; + if (fraction == 0) + fraction = SWAP_CLUSTER_MAX; if (!down_read_trylock(&shrinker_rwsem)) { /* @@ -361,7 +361,7 @@ unsigned long shrink_slab(struct shrink_control *shrinkctl, break; freed += shrink_slab_node(shrinkctl, shrinker, - nr_pages_scanned, lru_pages); + fraction, denominator); } } -- 1.7.10.4 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
