int may_enter_fs;
enum page_references references = PAGEREF_RECLAIM_CLEAN;
bool dirty, writeback;
- int ret = SWAP_SUCCESS;
cond_resched();
* processes. Try to unmap it here.
*/
if (page_mapped(page)) {
- switch (ret = try_to_unmap(page,
- ttu_flags | TTU_BATCH_FLUSH)) {
- case SWAP_FAIL:
+ if (!try_to_unmap(page, ttu_flags | TTU_BATCH_FLUSH)) {
nr_unmap_fail++;
goto activate_locked;
- case SWAP_SUCCESS:
- ; /* try to free the page below */
}
}
* Both inactive lists should also be large enough that each inactive
* page has a chance to be referenced again before it is reclaimed.
*
+ * If that fails and refaulting is observed, the inactive list grows.
+ *
* The inactive_ratio is the target ratio of ACTIVE to INACTIVE pages
* on this LRU, maintained by the pageout code. A zone->inactive_ratio
* of 3 means 3:1 or 25% of the pages are kept on the inactive list.
* 10TB 320 32GB
*/
static bool inactive_list_is_low(struct lruvec *lruvec, bool file,
- struct scan_control *sc, bool trace)
+ struct mem_cgroup *memcg,
+ struct scan_control *sc, bool actual_reclaim)
{
- unsigned long inactive_ratio;
- unsigned long inactive, active;
- enum lru_list inactive_lru = file * LRU_FILE;
enum lru_list active_lru = file * LRU_FILE + LRU_ACTIVE;
+ struct pglist_data *pgdat = lruvec_pgdat(lruvec);
+ enum lru_list inactive_lru = file * LRU_FILE;
+ unsigned long inactive, active;
+ unsigned long inactive_ratio;
+ unsigned long refaults;
unsigned long gb;
/*
inactive = lruvec_lru_size(lruvec, inactive_lru, sc->reclaim_idx);
active = lruvec_lru_size(lruvec, active_lru, sc->reclaim_idx);
- gb = (inactive + active) >> (30 - PAGE_SHIFT);
- if (gb)
- inactive_ratio = int_sqrt(10 * gb);
+ if (memcg)
+ refaults = memcg_page_state(memcg, WORKINGSET_ACTIVATE);
else
- inactive_ratio = 1;
+ refaults = node_page_state(pgdat, WORKINGSET_ACTIVATE);
+
+ /*
+ * When refaults are being observed, it means a new workingset
+ * is being established. Disable active list protection to get
+ * rid of the stale workingset quickly.
+ */
+ if (file && actual_reclaim && lruvec->refaults != refaults) {
+ inactive_ratio = 0;
+ } else {
+ gb = (inactive + active) >> (30 - PAGE_SHIFT);
+ if (gb)
+ inactive_ratio = int_sqrt(10 * gb);
+ else
+ inactive_ratio = 1;
+ }
- if (trace)
- trace_mm_vmscan_inactive_list_is_low(lruvec_pgdat(lruvec)->node_id,
- sc->reclaim_idx,
- lruvec_lru_size(lruvec, inactive_lru, MAX_NR_ZONES), inactive,
- lruvec_lru_size(lruvec, active_lru, MAX_NR_ZONES), active,
- inactive_ratio, file);
+ if (actual_reclaim)
+ trace_mm_vmscan_inactive_list_is_low(pgdat->node_id, sc->reclaim_idx,
+ lruvec_lru_size(lruvec, inactive_lru, MAX_NR_ZONES), inactive,
+ lruvec_lru_size(lruvec, active_lru, MAX_NR_ZONES), active,
+ inactive_ratio, file);
return inactive * inactive_ratio < active;
}
static unsigned long shrink_list(enum lru_list lru, unsigned long nr_to_scan,
- struct lruvec *lruvec, struct scan_control *sc)
+ struct lruvec *lruvec, struct mem_cgroup *memcg,
+ struct scan_control *sc)
{
if (is_active_lru(lru)) {
- if (inactive_list_is_low(lruvec, is_file_lru(lru), sc, true))
+ if (inactive_list_is_low(lruvec, is_file_lru(lru),
+ memcg, sc, true))
shrink_active_list(nr_to_scan, lruvec, sc, lru);
return 0;
}
* lruvec even if it has plenty of old anonymous pages unless the
* system is under heavy pressure.
*/
- if (!inactive_list_is_low(lruvec, true, sc, false) &&
+ if (!inactive_list_is_low(lruvec, true, memcg, sc, false) &&
lruvec_lru_size(lruvec, LRU_INACTIVE_FILE, sc->reclaim_idx) >> sc->priority) {
scan_balance = SCAN_FILE;
goto out;
nr[lru] -= nr_to_scan;
nr_reclaimed += shrink_list(lru, nr_to_scan,
- lruvec, sc);
+ lruvec, memcg, sc);
}
}
* Even if we did not try to evict anon pages at all, we want to
* rebalance the anon lru active/inactive ratio.
*/
- if (inactive_list_is_low(lruvec, false, sc, true))
+ if (inactive_list_is_low(lruvec, false, memcg, sc, true))
shrink_active_list(SWAP_CLUSTER_MAX, lruvec,
sc, LRU_ACTIVE_ANON);
}
sc->memcg_low_skipped = 1;
continue;
}
- mem_cgroup_events(memcg, MEMCG_LOW, 1);
+ mem_cgroup_event(memcg, MEMCG_LOW);
}
reclaimed = sc->nr_reclaimed;
sc->gfp_mask = orig_mask;
}
+static void snapshot_refaults(struct mem_cgroup *root_memcg, pg_data_t *pgdat)
+{
+ struct mem_cgroup *memcg;
+
+ memcg = mem_cgroup_iter(root_memcg, NULL, NULL);
+ do {
+ unsigned long refaults;
+ struct lruvec *lruvec;
+
+ if (memcg)
+ refaults = memcg_page_state(memcg, WORKINGSET_ACTIVATE);
+ else
+ refaults = node_page_state(pgdat, WORKINGSET_ACTIVATE);
+
+ lruvec = mem_cgroup_lruvec(pgdat, memcg);
+ lruvec->refaults = refaults;
+ } while ((memcg = mem_cgroup_iter(root_memcg, memcg, NULL)));
+}
+
/*
* This is the main entry point to direct page reclaim.
*
struct scan_control *sc)
{
int initial_priority = sc->priority;
+ pg_data_t *last_pgdat;
+ struct zoneref *z;
+ struct zone *zone;
retry:
delayacct_freepages_start();
sc->may_writepage = 1;
} while (--sc->priority >= 0);
+ last_pgdat = NULL;
+ for_each_zone_zonelist_nodemask(zone, z, zonelist, sc->reclaim_idx,
+ sc->nodemask) {
+ if (zone->zone_pgdat == last_pgdat)
+ continue;
+ last_pgdat = zone->zone_pgdat;
+ snapshot_refaults(sc->target_mem_cgroup, zone->zone_pgdat);
+ }
+
delayacct_freepages_end();
if (sc->nr_reclaimed)
struct zonelist *zonelist;
unsigned long nr_reclaimed;
int nid;
+ unsigned int noreclaim_flag;
struct scan_control sc = {
.nr_to_reclaim = max(nr_pages, SWAP_CLUSTER_MAX),
.gfp_mask = (current_gfp_context(gfp_mask) & GFP_RECLAIM_MASK) |
sc.gfp_mask,
sc.reclaim_idx);
- current->flags |= PF_MEMALLOC;
+ noreclaim_flag = memalloc_noreclaim_save();
nr_reclaimed = do_try_to_free_pages(zonelist, &sc);
- current->flags &= ~PF_MEMALLOC;
+ memalloc_noreclaim_restore(noreclaim_flag);
trace_mm_vmscan_memcg_reclaim_end(nr_reclaimed);
do {
struct lruvec *lruvec = mem_cgroup_lruvec(pgdat, memcg);
- if (inactive_list_is_low(lruvec, false, sc, true))
+ if (inactive_list_is_low(lruvec, false, memcg, sc, true))
shrink_active_list(SWAP_CLUSTER_MAX, lruvec,
sc, LRU_ACTIVE_ANON);
pgdat->kswapd_failures++;
out:
+ snapshot_refaults(NULL, pgdat);
/*
* Return the order kswapd stopped reclaiming at as
* prepare_kswapd_sleep() takes it into account. If another caller
struct zonelist *zonelist = node_zonelist(numa_node_id(), sc.gfp_mask);
struct task_struct *p = current;
unsigned long nr_reclaimed;
+ unsigned int noreclaim_flag;
- p->flags |= PF_MEMALLOC;
+ noreclaim_flag = memalloc_noreclaim_save();
lockdep_set_current_reclaim_state(sc.gfp_mask);
reclaim_state.reclaimed_slab = 0;
p->reclaim_state = &reclaim_state;
p->reclaim_state = NULL;
lockdep_clear_current_reclaim_state();
- p->flags &= ~PF_MEMALLOC;
+ memalloc_noreclaim_restore(noreclaim_flag);
return nr_reclaimed;
}
struct task_struct *p = current;
struct reclaim_state reclaim_state;
int classzone_idx = gfp_zone(gfp_mask);
+ unsigned int noreclaim_flag;
struct scan_control sc = {
.nr_to_reclaim = max(nr_pages, SWAP_CLUSTER_MAX),
.gfp_mask = (gfp_mask = current_gfp_context(gfp_mask)),
* and we also need to be able to write out pages for RECLAIM_WRITE
* and RECLAIM_UNMAP.
*/
- p->flags |= PF_MEMALLOC | PF_SWAPWRITE;
+ noreclaim_flag = memalloc_noreclaim_save();
+ p->flags |= PF_SWAPWRITE;
lockdep_set_current_reclaim_state(gfp_mask);
reclaim_state.reclaimed_slab = 0;
p->reclaim_state = &reclaim_state;
}
p->reclaim_state = NULL;
- current->flags &= ~(PF_MEMALLOC | PF_SWAPWRITE);
+ current->flags &= ~PF_SWAPWRITE;
+ memalloc_noreclaim_restore(noreclaim_flag);
lockdep_clear_current_reclaim_state();
return sc.nr_reclaimed >= nr_pages;
}
projects / karo-tx-linux.git / blobdiff