+ mm-page_alloc-fair-zone-allocator-policy-v2.patch added to -mm tree

+ mm-page_alloc-fair-zone-allocator-policy-v2.patch added to -mm tree

[akpm]

Subject: + mm-page_alloc-fair-zone-allocator-policy-v2.patch added to -mm tree
To: hannes@cmpxchg.org,aarcange@redhat.com,mgorman@suse.de,paul.bollee@gmail.com,riel@redhat.com,zcalusic@bitsync.net
From: akpm@linux-foundation.org
Date: Wed, 07 Aug 2013 14:11:55 -0700


The patch titled
     Subject: mm-page_alloc-fair-zone-allocator-policy-v2
has been added to the -mm tree.  Its filename is
     mm-page_alloc-fair-zone-allocator-policy-v2.patch

This patch should soon appear at
    http://ozlabs.org/~akpm/mmots/broken-out/mm-page_alloc-fair-zone-allocator-policy-v2.patch
and later at
    http://ozlabs.org/~akpm/mmotm/broken-out/mm-page_alloc-fair-zone-allocator-policy-v2.patch

Before you just go and hit "reply", please:
   a) Consider who else should be cc'ed
   b) Prefer to cc a suitable mailing list as well
   c) Ideally: find the original patch on the mailing list and do a
      reply-to-all to that, adding suitable additional cc's

*** Remember to use Documentation/SubmitChecklist when testing your code ***

The -mm tree is included into linux-next and is updated
there every 3-4 working days

------------------------------------------------------
From: Johannes Weiner <hannes@cmpxchg.org>
Subject: mm-page_alloc-fair-zone-allocator-policy-v2

Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Tested-by: Zlatko Calusic <zcalusic@bitsync.net>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Paul Bolle <paul.bollee@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 include/linux/mmzone.h |    2 -
 mm/page_alloc.c        |   52 ++++++++++++++++++++++++++++++---------
 2 files changed, 42 insertions(+), 12 deletions(-)

diff -puN include/linux/mmzone.h~mm-page_alloc-fair-zone-allocator-policy-v2 include/linux/mmzone.h
--- a/include/linux/mmzone.h~mm-page_alloc-fair-zone-allocator-policy-v2
+++ a/include/linux/mmzone.h
@@ -352,7 +352,7 @@ struct zone {
 	 * free areas of different sizes
 	 */
 	spinlock_t		lock;
-	atomic_t		alloc_batch;
+	int			alloc_batch;
 	int                     all_unreclaimable; /* All pages pinned */
 #if defined CONFIG_COMPACTION || defined CONFIG_CMA
 	/* Set to true when the PG_migrate_skip bits should be cleared */
diff -puN mm/page_alloc.c~mm-page_alloc-fair-zone-allocator-policy-v2 mm/page_alloc.c
--- a/mm/page_alloc.c~mm-page_alloc-fair-zone-allocator-policy-v2
+++ a/mm/page_alloc.c
@@ -1817,6 +1817,11 @@ static void zlc_clear_zones_full(struct
 	bitmap_zero(zlc->fullzones, MAX_ZONES_PER_ZONELIST);
 }
 
+static bool zone_local(struct zone *local_zone, struct zone *zone)
+{
+	return node_distance(local_zone->node, zone->node) == LOCAL_DISTANCE;
+}
+
 static bool zone_allows_reclaim(struct zone *local_zone, struct zone *zone)
 {
 	return node_isset(local_zone->node, zone->zone_pgdat->reclaim_nodes);
@@ -1854,6 +1859,11 @@ static void zlc_clear_zones_full(struct
 {
 }
 
+static bool zone_local(struct zone *local_zone, struct zone *zone)
+{
+	return true;
+}
+
 static bool zone_allows_reclaim(struct zone *local_zone, struct zone *zone)
 {
 	return true;
@@ -1905,9 +1915,21 @@ zonelist_scan:
 		 * zone size to ensure fair page aging.  The zone a
 		 * page was allocated in should have no effect on the
 		 * time the page has in memory before being reclaimed.
+		 *
+		 * When zone_reclaim_mode is enabled, try to stay in
+		 * local zones in the fastpath.  If that fails, the
+		 * slowpath is entered, which will do another pass
+		 * starting with the local zones, but ultimately fall
+		 * back to remote zones that do not partake in the
+		 * fairness round-robin cycle of this zonelist.
 		 */
-		if (atomic_read(&zone->alloc_batch) <= 0)
-			continue;
+		if (alloc_flags & ALLOC_WMARK_LOW) {
+			if (zone->alloc_batch <= 0)
+				continue;
+			if (zone_reclaim_mode &&
+			    !zone_local(preferred_zone, zone))
+				continue;
+		}
 		/*
 		 * When allocating a page cache page for writing, we
 		 * want to get it from a zone that is within its dirty
@@ -2015,7 +2037,7 @@ this_zone_full:
 	}
 
 	if (page) {
-		atomic_sub(1U << order, &zone->alloc_batch);
+		zone->alloc_batch -= 1U << order;
 		/*
 		 * page->pfmemalloc is set when ALLOC_NO_WATERMARKS was
 		 * necessary to allocate the page. The expectation is
@@ -2359,16 +2381,24 @@ __alloc_pages_high_priority(gfp_t gfp_ma
 static void prepare_slowpath(gfp_t gfp_mask, unsigned int order,
 			     struct zonelist *zonelist,
 			     enum zone_type high_zoneidx,
-			     enum zone_type classzone_idx)
+			     struct zone *preferred_zone)
 {
 	struct zoneref *z;
 	struct zone *zone;
 
 	for_each_zone_zonelist(zone, z, zonelist, high_zoneidx) {
-		atomic_set(&zone->alloc_batch,
-			   high_wmark_pages(zone) - low_wmark_pages(zone));
 		if (!(gfp_mask & __GFP_NO_KSWAPD))
-			wakeup_kswapd(zone, order, classzone_idx);
+			wakeup_kswapd(zone, order, zone_idx(preferred_zone));
+		/*
+		 * Only reset the batches of zones that were actually
+		 * considered in the fast path, we don't want to
+		 * thrash fairness information for zones that are not
+		 * actually part of this zonelist's round-robin cycle.
+		 */
+		if (zone_reclaim_mode && !zone_local(preferred_zone, zone))
+			continue;
+		zone->alloc_batch = high_wmark_pages(zone) -
+			low_wmark_pages(zone);
 	}
 }
 
@@ -2466,7 +2496,7 @@ __alloc_pages_slowpath(gfp_t gfp_mask, u
 
 restart:
 	prepare_slowpath(gfp_mask, order, zonelist,
-			 high_zoneidx, zone_idx(preferred_zone));
+			 high_zoneidx, preferred_zone);
 
 	/*
 	 * OK, we're below the kswapd watermark and have kicked background
@@ -4768,7 +4798,7 @@ static void __paginginit free_area_init_
 		zone->zone_pgdat = pgdat;
 
 		/* For bootup, initialized properly in watermark setup */
-		atomic_set(&zone->alloc_batch, zone->managed_pages);
+		zone->alloc_batch = zone->managed_pages;
 
 		zone_pcp_init(zone);
 		lruvec_init(&zone->lruvec);
@@ -5541,8 +5571,8 @@ static void __setup_per_zone_wmarks(void
 		zone->watermark[WMARK_LOW]  = min_wmark_pages(zone) + (tmp >> 2);
 		zone->watermark[WMARK_HIGH] = min_wmark_pages(zone) + (tmp >> 1);
 
-		atomic_set(&zone->alloc_batch,
-			   high_wmark_pages(zone) - low_wmark_pages(zone));
+		zone->alloc_batch = high_wmark_pages(zone) -
+			low_wmark_pages(zone);
 
 		setup_zone_migrate_reserve(zone);
 		spin_unlock_irqrestore(&zone->lock, flags);
_

Patches currently in -mm which might be from hannes@cmpxchg.org are

memcg-dont-initialize-kmem-cache-destroying-work-for-root-caches.patch
mm-kill-one-if-loop-in-__free_pages_bootmem.patch
mm-vmscan-fix-numa-reclaim-balance-problem-in-kswapd.patch
mm-page_alloc-rearrange-watermark-checking-in-get_page_from_freelist.patch
mm-page_alloc-fair-zone-allocator-policy.patch
mm-page_alloc-fair-zone-allocator-policy-v2.patch
mm-revert-page-writebackc-subtract-min_free_kbytes-from-dirtyable-memory.patch
memcg-remove-redundant-code-in-mem_cgroup_force_empty_write.patch
memcg-vmscan-integrate-soft-reclaim-tighter-with-zone-shrinking-code.patch
memcg-get-rid-of-soft-limit-tree-infrastructure.patch
vmscan-memcg-do-softlimit-reclaim-also-for-targeted-reclaim.patch
memcg-enhance-memcg-iterator-to-support-predicates.patch
memcg-track-children-in-soft-limit-excess-to-improve-soft-limit.patch
memcg-vmscan-do-not-attempt-soft-limit-reclaim-if-it-would-not-scan-anything.patch
memcg-track-all-children-over-limit-in-the-root.patch
memcg-vmscan-do-not-fall-into-reclaim-all-pass-too-quickly.patch
arch-mm-remove-obsolete-init-oom-protection.patch
arch-mm-do-not-invoke-oom-killer-on-kernel-fault-oom.patch
arch-mm-pass-userspace-fault-flag-to-generic-fault-handler.patch
x86-finish-user-fault-error-path-with-fatal-signal.patch
mm-memcg-enable-memcg-oom-killer-only-for-user-faults.patch
mm-memcg-rework-and-document-oom-waiting-and-wakeup.patch
mm-memcg-do-not-trap-chargers-with-full-callstack-on-oom.patch
mm-page_alloc-add-unlikely-macro-to-help-compiler-optimization.patch
mm-move-pgtable-related-functions-to-right-place.patch
swap-clean-up-ifdef-in-page_mapping.patch
mm-vmalloc-remove-useless-variable-in-vmap_block.patch
mm-vmalloc-use-well-defined-find_last_bit-func.patch
swap-add-a-simple-detector-for-inappropriate-swapin-readahead-fix.patch
debugging-keep-track-of-page-owners-fix-2-fix-fix-fix.patch