LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
* [PATCH 0/5] Fragmentation avoidance improvements
@ 2018-10-31 16:06 Mel Gorman
  2018-10-31 16:06 ` [PATCH 1/5] mm, page_alloc: Spread allocations across zones before introducing fragmentation Mel Gorman
                   ` (4 more replies)
  0 siblings, 5 replies; 8+ messages in thread
From: Mel Gorman @ 2018-10-31 16:06 UTC (permalink / raw)
  To: Linux-MM
  Cc: Andrew Morton, Vlastimil Babka, David Rientjes, Andrea Arcangeli,
	Zi Yan, LKML, Mel Gorman

Warning: This is a long intro with long changelogs and this is not a
	trivial area to either analyse or fix. TLDR -- 95% reduction in
	fragmentation events, patches 1-3 should be relatively ok. Patch
	4 and 5 need scrutiny but they are also independent or dropped.

It has been noted before that fragmentation avoidance (aka
anti-fragmentation) is far from perfect. Given a long enough time or an
adverse enough workload, memory still gets fragmented and the long-term
success of high-order allocations degrades. This series defines an adverse
workload, a definition of external fragmentation events (including serious)
ones and a series that reduces the level of those fragmentation events.

This series is *not* directly related to the recent __GFP_THISNODE
discussion and has no impact on the trivial test cases that were discussed
there. This series was also evaluated without the candidate fixes from
that discussion. The series does have consequences for high-order and
THP allocations though that are important to consider so the same people
are cc'd. It's also far from a complete solution but side-issues such as
compaction, usability and other factors would require different series. It's
also extremely important to note that this is analysed in the context of
one adverse workload. While other patterns of fragmentation are possible
(and workloads that are mostly slab allocations have a completely different
solution space), they would need test cases to be properly considered.

The details of the workload and the consequences are described in more
detail in the changelogs. However, from patch 1, this is a high-level
summary of the adverse workload. The exact details are found in the
mmtests implementation.

The broad details of the workload are as follows;

1. Create an XFS filesystem (not specified in the configuration but done
   as part of the testing for this patch)
2. Start 4 fio threads that write a number of 64K files inefficiently.
   Inefficiently means that files are created on first access and not
   created in advance (fio parameterr create_on_open=1) and fallocate
   is not used (fallocate=none). With multiple IO issuers this creates
   a mix of slab and page cache allocations over time. The total size
   of the files is 150% physical memory so that the slabs and page cache
   pages get mixed
3. Warm up a number of fio read-only threads accessing the same files
   created in step 2. This part runs for the same length of time it
   took to create the files. It'll fault back in old data and further
   interleave slab and page cache allocations. As it's now low on
   memory due to step 2, fragmentation occurs as pageblocks get
   stolen.
4. While step 3 is still running, start a process that tries to allocate
   75% of memory as huge pages with a number of threads. The number of
   threads is based on a (NR_CPUS_SOCKET - NR_FIO_THREADS)/4 to avoid THP
   threads contending with fio, any other threads or forcing cross-NUMA
   scheduling. Note that the test has not been used on a machine with less
   than 8 cores. The benchmark records whether huge pages were allocated
   and what the fault latency was in microseconds
5. Measure the number of events potentially causing external fragmentation,
   the fault latency and the huge page allocation success rate.
6. Cleanup

Overall the series reduces external fragmentation causing events by over 95%
on 1 and 2 socket machines, which in turn impacts high-order allocation
success rates over the long term. There are differences in latencies and
high-order allocation success rates. Latencies are a mixed bag as they
are vulnerable to exact system state and whether allocations succeeded so
they are treated as a secondary metric.

Patch 1 uses lower zones if they are populated and have free memory
	instead of fragmenting a higher zone. It's special cased to
	handle a Normal->DMA32 fallback with the reasons explained
	in the changelog.

Patch 2+3 boosts watermarks temporarily when an external fragmentation
	event occurs. kswapd wakes to reclaim a small amount of old memory
	and then wakes kcompactd on completion to recover the system
	slightly. This introduces some overhead in the slowpath. The level
	of boosting can be tuned or disabled depending on the tolerance
	for fragmentation vs allocation latency.

Patch 4 is more heavy handed. In the event of a movable allocation
	request that can stall, it'll wake kswapd as in patch 3.  However,
	if the expected fragmentation event is serious then the request
	will stall briefly on pfmemalloc_wait until kswapd completes
	light reclaim work and retry the allocation without stalling.
	This can avoid the fragmentation event entirely in some cases.
	The definition of a serious fragmentation event can be tuned
	or disabled.

Patch 5 is the hardest to prove it's a real benefit. In the event
	that fragmentation was unavoidable, it'll queue a pageblock for
	kcompactd to clean. It's a fixed-length queue that is neither
	guaranteed to have a slot available or successfully clean a
	pageblock.

Patches 4 and 5 can be treated independently or dropped. The bulk of
the improvement in fragmentation avoidance is from patches 1-3 (94-97%
reduction in fragmentation events for an adverse workload on both a
1-socket and 2-socket machine).

 Documentation/sysctl/vm.txt       |  42 +++++++
 include/linux/compaction.h        |   4 +
 include/linux/migrate.h           |   7 +-
 include/linux/mm.h                |   2 +
 include/linux/mmzone.h            |  18 ++-
 include/trace/events/compaction.h |  62 +++++++++++
 kernel/sysctl.c                   |  18 +++
 mm/compaction.c                   | 148 +++++++++++++++++++++++--
 mm/internal.h                     |  14 ++-
 mm/migrate.c                      |   6 +-
 mm/page_alloc.c                   | 228 ++++++++++++++++++++++++++++++++++----
 mm/vmscan.c                       | 123 ++++++++++++++++++--
 12 files changed, 621 insertions(+), 51 deletions(-)

-- 
2.16.4


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [PATCH 1/5] mm, page_alloc: Spread allocations across zones before introducing fragmentation
  2018-10-31 16:06 [PATCH 0/5] Fragmentation avoidance improvements Mel Gorman
@ 2018-10-31 16:06 ` Mel Gorman
  2018-10-31 16:06 ` [PATCH 2/5] mm: Move zone watermark accesses behind an accessor Mel Gorman
                   ` (3 subsequent siblings)
  4 siblings, 0 replies; 8+ messages in thread
From: Mel Gorman @ 2018-10-31 16:06 UTC (permalink / raw)
  To: Linux-MM
  Cc: Andrew Morton, Vlastimil Babka, David Rientjes, Andrea Arcangeli,
	Zi Yan, LKML, Mel Gorman

The page allocator zone lists are iterated based on the watermarks
of each zone which does not take anti-fragmentation into account. On
x86, node 0 may have multiple zones while other nodes have one zone. A
consequence is that tasks running on node 0 may fragment ZONE_NORMAL even
though ZONE_DMA32 has plenty of free memory. This patch special cases
the allocator fast path such that it'll try an allocation from a lower
local zone before fragmenting a higher zone. In this case, stealing of
pageblocks or orders larger than a pageblock are still allowed in the
fast path as they are uninteresting from a fragmentation point of view.

This was evaluated using a benchmark designed to fragment memory
before attempting THPs.  It's implemented in mmtests as the following
configurations

configs/config-global-dhp__workload_thpfioscale
configs/config-global-dhp__workload_thpfioscale-defrag
configs/config-global-dhp__workload_thpfioscale-madvhugepage

e.g. from mmtests
./run-mmtests.sh --run-monitor --config configs/config-global-dhp__workload_thpfioscale test-run-1

The broad details of the workload are as follows;

1. Create an XFS filesystem (not specified in the configuration but done
   as part of the testing for this patch)
2. Start 4 fio threads that write a number of 64K files inefficiently.
   Inefficiently means that files are created on first access and not
   created in advance (fio parameterr create_on_open=1) and fallocate
   is not used (fallocate=none). With multiple IO issuers this creates
   a mix of slab and page cache allocations over time. The total size
   of the files is 150% physical memory so that the slabs and page cache
   pages get mixed
3. Warm up a number of fio read-only threads accessing the same files
   created in step 2. This part runs for the same length of time it
   took to create the files. It'll fault back in old data and further
   interleave slab and page cache allocations. As it's now low on
   memory due to step 2, fragmentation occurs as pageblocks get
   stolen.
4. While step 3 is still running, start a process that tries to allocate
   75% of memory as huge pages with a number of threads. The number of
   threads is based on a (NR_CPUS_SOCKET - NR_FIO_THREADS)/4 to avoid THP
   threads contending with fio, any other threads or forcing cross-NUMA
   scheduling. Note that the test has not been used on a machine with less
   than 8 cores. The benchmark records whether huge pages were allocated
   and what the fault latency was in microseconds
5. Measure the number of events potentially causing external fragmentation,
   the fault latency and the huge page allocation success rate.
6. Cleanup

Note that due to the use of IO and page cache that this benchmark is not
suitable for running on large machines where the time to fragment memory
may be excessive. Also note that while this is one mix that generates
fragmentation that it's not the only mix that generates fragmentation.
Differences in workload that are more slab-intensive or whether SLUB is
used with high-order pages may yield different results.

When the page allocator fragments memory, it records the event using the
mm_page_alloc_extfrag event. If the fallback_order is smaller than a
pageblock order (order-9 on 64-bit x86) then it's considered an event
that may cause external fragmentation issues in the future. Hence, the
primary metric here is the number of external fragmentation events that
occur with order < 9. The secondary metric is allocation latency and huge
page allocation success rates but note that differences in latencies and
what the success rate also can affect the number of external fragmentation
event which is why it's a secondary metric.

1-socket Skylake machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 1 THP allocating thread
--------------------------------------

4.19 extfrag events < order 0:	71227
4.19+patch:                     36456 (49% reduction)

thpfioscale Fault Latencies
                                       4.19.0                 4.19.0
                                      vanilla           lowzone-v1r1
Amean     fault-base-1      605.84 (   0.00%)      599.92 *   0.98%*
Amean     fault-huge-1      296.00 (   0.00%)      179.84 *  39.24%*

                                  4.19.0                 4.19.0
                                 vanilla           lowzone-v1r1
Percentage huge-1        0.44 (   0.00%)        1.08 ( 146.15%)

Fault latencies are reduced. While allocation success rates are not much
higher, this configuration does not make any heavy effort to allocate
THP and fio is heavily active at the time and filling memory.  However,
a 49% reduction of serious fragmentation events reduces the changes of
external fragmentation being a problem in the future.

1-socket Skylake machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------

4.19 extfrag events < order 0:  40761
4.19+patch:                     36085 (11.47% reduction)

thpfioscale Fault Latencies
                                       4.19.0                 4.19.0
                                      vanilla           lowzone-v1r1
Amean     fault-base-1     1938.77 (   0.00%)     1938.47 (   0.02%)
Amean     fault-huge-1      774.80 (   0.00%)      749.40 *   3.28%*

thpfioscale Percentage Faults Huge
                                  4.19.0                 4.19.0
                                 vanilla           lowzone-v1r1
Percentage huge-1       83.59 (   0.00%)       83.79 (   0.24%)

Nothing dramatic. Fragmentation events are still reduced but the differences
in fault latencies and allocation success rates are similar.

2-socket Haswell machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 5 THP allocating threads
----------------------------------------------------------------

4.19 extfrag events < order 0:  882868
4.19+patch:                     476937 (46% reduction)

thpfioscale Fault Latencies
                                       4.19.0                 4.19.0
                                      vanilla           lowzone-v1r1
Amean     fault-base-5     1505.76 (   0.00%)     1602.01 (  -6.39%)
Amean     fault-huge-5      687.00 (   0.00%)        0.00 * 100.00%*

                                  4.19.0                 4.19.0
                                 vanilla           lowzone-v1r1
Percentage huge-5        0.07 (   0.00%)        0.00 (   0.00%)

The reduction of external fragmentation events is expected. The
latencies are off because the huge page allocations generally
failed and the patch does not have a direct impact on success
rates.

2-socket Haswell machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------

4.19 extfrag events < order 0: 803099
4.19+patch:                    654671 (23% reduction)

thpfioscale Fault Latencies
                                       4.19.0                 4.19.0
                                      vanilla           lowzone-v1r1
Amean     fault-base-5     5389.23 (   0.00%)     6678.61 * -23.93%*
Amean     fault-huge-5     5039.32 (   0.00%)     2796.35 *  44.51%*

thpfioscale Percentage Faults Huge
                                  4.19.0                 4.19.0
                                 vanilla           lowzone-v1r1
Percentage huge-5       30.69 (   0.00%)       57.92 (  88.71%)

In this case, there was both a reduction in the external fragmentation
causing events and the huge page allocation success rates were increased
substantially from 30.69% of attempts to 57.92%.

Overall, the patch significantly reduces the number of external
fragmentation causing events so the success of THP over long
periods of time would be improved for this adverse workload.

Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
---
 mm/internal.h   |  13 +++++---
 mm/page_alloc.c | 101 ++++++++++++++++++++++++++++++++++++++++++++++++++------
 2 files changed, 99 insertions(+), 15 deletions(-)

diff --git a/mm/internal.h b/mm/internal.h
index 87256ae1bef8..0dd659cf2a7e 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -480,10 +480,15 @@ unsigned long reclaim_clean_pages_from_list(struct zone *zone,
 #define ALLOC_OOM		ALLOC_NO_WATERMARKS
 #endif
 
-#define ALLOC_HARDER		0x10 /* try to alloc harder */
-#define ALLOC_HIGH		0x20 /* __GFP_HIGH set */
-#define ALLOC_CPUSET		0x40 /* check for correct cpuset */
-#define ALLOC_CMA		0x80 /* allow allocations from CMA areas */
+#define ALLOC_HARDER		 0x10 /* try to alloc harder */
+#define ALLOC_HIGH		 0x20 /* __GFP_HIGH set */
+#define ALLOC_CPUSET		 0x40 /* check for correct cpuset */
+#define ALLOC_CMA		 0x80 /* allow allocations from CMA areas */
+#ifdef CONFIG_ZONE_DMA32
+#define ALLOC_NOFRAGMENT	0x100 /* avoid mixing pageblock types */
+#else
+#define ALLOC_NOFRAGMENT	  0x0
+#endif
 
 enum ttu_flags;
 struct tlbflush_unmap_batch;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index e2ef1c17942f..db5d61868c96 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2364,20 +2364,30 @@ static bool unreserve_highatomic_pageblock(const struct alloc_context *ac,
  * condition simpler.
  */
 static __always_inline bool
-__rmqueue_fallback(struct zone *zone, int order, int start_migratetype)
+__rmqueue_fallback(struct zone *zone, int order, int start_migratetype,
+						unsigned int alloc_flags)
 {
 	struct free_area *area;
 	int current_order;
+	int min_order = order;
 	struct page *page;
 	int fallback_mt;
 	bool can_steal;
 
+	/*
+	 * Do not steal pages from freelists belonging to other pageblocks
+	 * i.e. orders < pageblock_order. In the event there is on local
+	 * zone free, the allocation will retry later.
+	 */
+	if (alloc_flags & ALLOC_NOFRAGMENT)
+		min_order = pageblock_order;
+
 	/*
 	 * Find the largest available free page in the other list. This roughly
 	 * approximates finding the pageblock with the most free pages, which
 	 * would be too costly to do exactly.
 	 */
-	for (current_order = MAX_ORDER - 1; current_order >= order;
+	for (current_order = MAX_ORDER - 1; current_order >= min_order;
 				--current_order) {
 		area = &(zone->free_area[current_order]);
 		fallback_mt = find_suitable_fallback(area, current_order,
@@ -2436,7 +2446,8 @@ __rmqueue_fallback(struct zone *zone, int order, int start_migratetype)
  * Call me with the zone->lock already held.
  */
 static __always_inline struct page *
-__rmqueue(struct zone *zone, unsigned int order, int migratetype)
+__rmqueue(struct zone *zone, unsigned int order, int migratetype,
+						unsigned int alloc_flags)
 {
 	struct page *page;
 
@@ -2446,7 +2457,8 @@ __rmqueue(struct zone *zone, unsigned int order, int migratetype)
 		if (migratetype == MIGRATE_MOVABLE)
 			page = __rmqueue_cma_fallback(zone, order);
 
-		if (!page && __rmqueue_fallback(zone, order, migratetype))
+		if (!page && __rmqueue_fallback(zone, order, migratetype,
+								alloc_flags))
 			goto retry;
 	}
 
@@ -2461,13 +2473,14 @@ __rmqueue(struct zone *zone, unsigned int order, int migratetype)
  */
 static int rmqueue_bulk(struct zone *zone, unsigned int order,
 			unsigned long count, struct list_head *list,
-			int migratetype)
+			int migratetype, unsigned int alloc_flags)
 {
 	int i, alloced = 0;
 
 	spin_lock(&zone->lock);
 	for (i = 0; i < count; ++i) {
-		struct page *page = __rmqueue(zone, order, migratetype);
+		struct page *page = __rmqueue(zone, order, migratetype,
+								alloc_flags);
 		if (unlikely(page == NULL))
 			break;
 
@@ -2923,6 +2936,7 @@ static inline void zone_statistics(struct zone *preferred_zone, struct zone *z)
 
 /* Remove page from the per-cpu list, caller must protect the list */
 static struct page *__rmqueue_pcplist(struct zone *zone, int migratetype,
+			unsigned int alloc_flags,
 			struct per_cpu_pages *pcp,
 			struct list_head *list)
 {
@@ -2932,7 +2946,7 @@ static struct page *__rmqueue_pcplist(struct zone *zone, int migratetype,
 		if (list_empty(list)) {
 			pcp->count += rmqueue_bulk(zone, 0,
 					pcp->batch, list,
-					migratetype);
+					migratetype, alloc_flags);
 			if (unlikely(list_empty(list)))
 				return NULL;
 		}
@@ -2948,7 +2962,8 @@ static struct page *__rmqueue_pcplist(struct zone *zone, int migratetype,
 /* Lock and remove page from the per-cpu list */
 static struct page *rmqueue_pcplist(struct zone *preferred_zone,
 			struct zone *zone, unsigned int order,
-			gfp_t gfp_flags, int migratetype)
+			gfp_t gfp_flags, int migratetype,
+			unsigned int alloc_flags)
 {
 	struct per_cpu_pages *pcp;
 	struct list_head *list;
@@ -2958,7 +2973,7 @@ static struct page *rmqueue_pcplist(struct zone *preferred_zone,
 	local_irq_save(flags);
 	pcp = &this_cpu_ptr(zone->pageset)->pcp;
 	list = &pcp->lists[migratetype];
-	page = __rmqueue_pcplist(zone,  migratetype, pcp, list);
+	page = __rmqueue_pcplist(zone,  migratetype, alloc_flags, pcp, list);
 	if (page) {
 		__count_zid_vm_events(PGALLOC, page_zonenum(page), 1 << order);
 		zone_statistics(preferred_zone, zone);
@@ -2981,7 +2996,7 @@ struct page *rmqueue(struct zone *preferred_zone,
 
 	if (likely(order == 0)) {
 		page = rmqueue_pcplist(preferred_zone, zone, order,
-				gfp_flags, migratetype);
+				gfp_flags, migratetype, alloc_flags);
 		goto out;
 	}
 
@@ -3000,7 +3015,7 @@ struct page *rmqueue(struct zone *preferred_zone,
 				trace_mm_page_alloc_zone_locked(page, order, migratetype);
 		}
 		if (!page)
-			page = __rmqueue(zone, order, migratetype);
+			page = __rmqueue(zone, order, migratetype, alloc_flags);
 	} while (page && check_new_pages(page, order));
 	spin_unlock(&zone->lock);
 	if (!page)
@@ -3242,6 +3257,36 @@ static bool zone_allows_reclaim(struct zone *local_zone, struct zone *zone)
 }
 #endif	/* CONFIG_NUMA */
 
+#ifdef CONFIG_ZONE_DMA32
+/*
+ * The restriction on ZONE_DMA32 as being a suitable zone to use to avoid
+ * fragmentation is subtle. If the preferred zone was HIGHMEM then
+ * premature use of a lower zone may cause lowmem pressure problems that
+ * are wose than fragmentation. If the next zone is ZONE_DMA then it is
+ * probably too small. It only makes sense to spread allocations to avoid
+ * fragmentation between the Normal and DMA32 zones.
+ */
+static inline unsigned int alloc_flags_nofragment(struct zone *zone)
+{
+	if (zone_idx(zone) != ZONE_NORMAL)
+		return 0;
+
+	/*
+	 * If ZONE_DMA32 exists, assume it is the one after ZONE_NORMAL and
+	 * the pointer is within zone->zone_pgdat->node_zones[].
+	 */
+	if (!populated_zone(--zone))
+		return 0;
+
+	return ALLOC_NOFRAGMENT;
+}
+#else
+static inline unsigned int alloc_flags_nofragment(struct zone *zone)
+{
+	return 0;
+}
+#endif
+
 /*
  * get_page_from_freelist goes through the zonelist trying to allocate
  * a page.
@@ -3253,11 +3298,14 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 	struct zoneref *z = ac->preferred_zoneref;
 	struct zone *zone;
 	struct pglist_data *last_pgdat_dirty_limit = NULL;
+	bool no_fallback;
 
+retry:
 	/*
 	 * Scan zonelist, looking for a zone with enough free.
 	 * See also __cpuset_node_allowed() comment in kernel/cpuset.c.
 	 */
+	no_fallback = alloc_flags & ALLOC_NOFRAGMENT;
 	for_next_zone_zonelist_nodemask(zone, z, ac->zonelist, ac->high_zoneidx,
 								ac->nodemask) {
 		struct page *page;
@@ -3296,6 +3344,22 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 			}
 		}
 
+		if (no_fallback) {
+			int local_nid;
+
+			/*
+			 * If moving to a remote node, retry but allow
+			 * fragmenting fallbacks. Locality is more important
+			 * than fragmentation avoidance.
+			 *
+			 */
+			local_nid = zone_to_nid(ac->preferred_zoneref->zone);
+			if (zone_to_nid(zone) != local_nid) {
+				alloc_flags &= ~ALLOC_NOFRAGMENT;
+				goto retry;
+			}
+		}
+
 		mark = zone->watermark[alloc_flags & ALLOC_WMARK_MASK];
 		if (!zone_watermark_fast(zone, order, mark,
 				       ac_classzone_idx(ac), alloc_flags)) {
@@ -3363,6 +3427,15 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 		}
 	}
 
+	/*
+	 * It's possible on a UMA machine to get through all zones that are
+	 * fragmented. If avoiding fragmentation, reset and try again
+	 */
+	if (no_fallback) {
+		alloc_flags &= ~ALLOC_NOFRAGMENT;
+		goto retry;
+	}
+
 	return NULL;
 }
 
@@ -4366,6 +4439,12 @@ __alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order, int preferred_nid,
 
 	finalise_ac(gfp_mask, &ac);
 
+	/*
+	 * Forbid the first pass from falling back to types that fragment
+	 * memory until all local zones are considered.
+	 */
+	alloc_flags |= alloc_flags_nofragment(ac.preferred_zoneref->zone);
+
 	/* First allocation attempt */
 	page = get_page_from_freelist(alloc_mask, order, alloc_flags, &ac);
 	if (likely(page))
-- 
2.16.4


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [PATCH 2/5] mm: Move zone watermark accesses behind an accessor
  2018-10-31 16:06 [PATCH 0/5] Fragmentation avoidance improvements Mel Gorman
  2018-10-31 16:06 ` [PATCH 1/5] mm, page_alloc: Spread allocations across zones before introducing fragmentation Mel Gorman
@ 2018-10-31 16:06 ` Mel Gorman
  2018-10-31 16:06 ` [PATCH 3/5] mm: Reclaim small amounts of memory when an external fragmentation event occurs Mel Gorman
                   ` (2 subsequent siblings)
  4 siblings, 0 replies; 8+ messages in thread
From: Mel Gorman @ 2018-10-31 16:06 UTC (permalink / raw)
  To: Linux-MM
  Cc: Andrew Morton, Vlastimil Babka, David Rientjes, Andrea Arcangeli,
	Zi Yan, LKML, Mel Gorman

This is a preparation patch only, no functional change.

Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
---
 include/linux/mmzone.h |  9 +++++----
 mm/compaction.c        |  2 +-
 mm/page_alloc.c        | 12 ++++++------
 3 files changed, 12 insertions(+), 11 deletions(-)

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index d4b0c79d2924..854d6c188888 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -267,9 +267,10 @@ enum zone_watermarks {
 	NR_WMARK
 };
 
-#define min_wmark_pages(z) (z->watermark[WMARK_MIN])
-#define low_wmark_pages(z) (z->watermark[WMARK_LOW])
-#define high_wmark_pages(z) (z->watermark[WMARK_HIGH])
+#define min_wmark_pages(z) (z->_watermark[WMARK_MIN])
+#define low_wmark_pages(z) (z->_watermark[WMARK_LOW])
+#define high_wmark_pages(z) (z->_watermark[WMARK_HIGH])
+#define wmark_pages(z, i) (z->_watermark[i])
 
 struct per_cpu_pages {
 	int count;		/* number of pages in the list */
@@ -360,7 +361,7 @@ struct zone {
 	/* Read-mostly fields */
 
 	/* zone watermarks, access with *_wmark_pages(zone) macros */
-	unsigned long watermark[NR_WMARK];
+	unsigned long _watermark[NR_WMARK];
 
 	unsigned long nr_reserved_highatomic;
 
diff --git a/mm/compaction.c b/mm/compaction.c
index faca45ebe62d..aa9473a64915 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -1430,7 +1430,7 @@ static enum compact_result __compaction_suitable(struct zone *zone, int order,
 	if (is_via_compact_memory(order))
 		return COMPACT_CONTINUE;
 
-	watermark = zone->watermark[alloc_flags & ALLOC_WMARK_MASK];
+	watermark = wmark_pages(zone, alloc_flags & ALLOC_WMARK_MASK);
 	/*
 	 * If watermarks for high-order allocation are already met, there
 	 * should be no need for compaction at all.
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index db5d61868c96..a51887765abc 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3360,7 +3360,7 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 			}
 		}
 
-		mark = zone->watermark[alloc_flags & ALLOC_WMARK_MASK];
+		mark = wmark_pages(zone, alloc_flags & ALLOC_WMARK_MASK);
 		if (!zone_watermark_fast(zone, order, mark,
 				       ac_classzone_idx(ac), alloc_flags)) {
 			int ret;
@@ -4787,7 +4787,7 @@ long si_mem_available(void)
 		pages[lru] = global_node_page_state(NR_LRU_BASE + lru);
 
 	for_each_zone(zone)
-		wmark_low += zone->watermark[WMARK_LOW];
+		wmark_low += low_wmark_pages(zone);
 
 	/*
 	 * Estimate the amount of memory available for userspace allocations,
@@ -7323,13 +7323,13 @@ static void __setup_per_zone_wmarks(void)
 
 			min_pages = zone->managed_pages / 1024;
 			min_pages = clamp(min_pages, SWAP_CLUSTER_MAX, 128UL);
-			zone->watermark[WMARK_MIN] = min_pages;
+			zone->_watermark[WMARK_MIN] = min_pages;
 		} else {
 			/*
 			 * If it's a lowmem zone, reserve a number of pages
 			 * proportionate to the zone's size.
 			 */
-			zone->watermark[WMARK_MIN] = tmp;
+			zone->_watermark[WMARK_MIN] = tmp;
 		}
 
 		/*
@@ -7341,8 +7341,8 @@ static void __setup_per_zone_wmarks(void)
 			    mult_frac(zone->managed_pages,
 				      watermark_scale_factor, 10000));
 
-		zone->watermark[WMARK_LOW]  = min_wmark_pages(zone) + tmp;
-		zone->watermark[WMARK_HIGH] = min_wmark_pages(zone) + tmp * 2;
+		zone->_watermark[WMARK_LOW]  = min_wmark_pages(zone) + tmp;
+		zone->_watermark[WMARK_HIGH] = min_wmark_pages(zone) + tmp * 2;
 
 		spin_unlock_irqrestore(&zone->lock, flags);
 	}
-- 
2.16.4


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [PATCH 3/5] mm: Reclaim small amounts of memory when an external fragmentation event occurs
  2018-10-31 16:06 [PATCH 0/5] Fragmentation avoidance improvements Mel Gorman
  2018-10-31 16:06 ` [PATCH 1/5] mm, page_alloc: Spread allocations across zones before introducing fragmentation Mel Gorman
  2018-10-31 16:06 ` [PATCH 2/5] mm: Move zone watermark accesses behind an accessor Mel Gorman
@ 2018-10-31 16:06 ` Mel Gorman
  2018-10-31 16:56   ` Mel Gorman
  2018-10-31 16:06 ` [PATCH 4/5] mm: Stall movable allocations until kswapd progresses during serious external fragmentation event Mel Gorman
  2018-10-31 16:06 ` [PATCH 5/5] mm: Target compaction on pageblocks that were recently fragmented Mel Gorman
  4 siblings, 1 reply; 8+ messages in thread
From: Mel Gorman @ 2018-10-31 16:06 UTC (permalink / raw)
  To: Linux-MM
  Cc: Andrew Morton, Vlastimil Babka, David Rientjes, Andrea Arcangeli,
	Zi Yan, LKML, Mel Gorman

An external fragmentation event was previously described as

    When the page allocator fragments memory, it records the event using
    the mm_page_alloc_extfrag event. If the fallback_order is smaller
    than a pageblock order (order-9 on 64-bit x86) then it's considered
    an event that will cause external fragmentation issues in the future.

The kernel reduces the probability of such events by increasing the
watermark sizes by calling set_recommended_min_free_kbytes early in the
lifetime of the system. This works reasonably well in general but if there
is enough sparsely populated pageblocks then the problem can still occur
as enough memory is free overall and kswapd stays asleep.

This patch introduces a watermark_boost_factor sysctl that allows a zone
watermark to be temporarily boosted when an external fragmentation causing
events occurs. The boosting will stall allocations below the boosted low
watermark and kswapd is woken unconditionally to reclaim an amount of
memory relative to the size of the high watermark and the
watermark_boost_factor until the boost is cleared. When kswapd finishes,
it wakes kcompactd at the pageblock order to clean some of the pageblocks
that may have been affected by the fragmentation event. kswapd avoids
any writeback or swap from reclaim context during this operation to avoid
excessive system disruption in the name of fragmentation avoidance. Care
is taken so that kswapd will do normal reclaim work if the system is
really low on memory.

This was evaluated using the same workloads as "mm, page_alloc: Spread
allocations across zones before introducing fragmentation".

1-socket Skylake machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 1 THP allocating thread
--------------------------------------

4.19 extfrag events < order 0:  71227
4.19+patch1:                    36456 (49% reduction)
4.19+patch1-3:                   4510 (94% reduction)

                                       4.19.0                 4.19.0
                                 lowzone-v1r1             boost-v1r5
Amean     fault-base-1      599.92 (   0.00%)      630.44 *  -5.09%*
Amean     fault-huge-1      179.84 (   0.00%)      179.22 (   0.35%)

                                  4.19.0                 4.19.0
                            lowzone-v1r1             boost-v1r5
Percentage huge-1        1.08 (   0.00%)        2.89 ( 168.75%)

Note that external fragmentation causing events are massively reduced
by this path whether in comparison to the previous kernel or the vanilla
kernel. There is some jitter in the fault latencies and they are a bit
more variable but the slight increase in THP allocation success rates
would account for some of that.

1-socket Skylake machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------

4.19 extfrag events < order 0:  40761
4.19+patch1:                    36085 (11% reduction)
4.19+patch1-3:                   1887 (95% reduction)

thpfioscale Fault Latencies
                                       4.19.0                 4.19.0
                                 lowzone-v1r1             boost-v1r5
Amean     fault-base-1     1938.47 (   0.00%)     1863.70 *   3.86%*
Amean     fault-huge-1      749.40 (   0.00%)      776.07 *  -3.56%*

thpfioscale Percentage Faults Huge
                                  4.19.0                 4.19.0
                            lowzone-v1r1             boost-v1r5
Percentage huge-1       83.79 (   0.00%)       86.92 (   3.73%)

As before, massive reduction in external fragmentation events, some
jitter on latencies and a slight increase in THP allocation success
rates.

2-socket Haswell machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 5 THP allocating threads
----------------------------------------------------------------

4.19 extfrag events < order 0:  882868
4.19+patch1:                    476937 (46% reduction)
4.19+patch1-3:                   29044 (97% reduction)

                                       4.19.0                 4.19.0
                                 lowzone-v1r1             boost-v1r5
Amean     fault-base-5     1602.01 (   0.00%)     1595.28 (   0.42%)
Amean     fault-huge-5        0.00 (   0.00%)      435.67 * -99.00%*

                                  4.19.0                 4.19.0
                            lowzone-v1r1             boost-v1r5
Percentage huge-5        0.00 (   0.00%)        0.15 ( 100.00%)

This is an illustration of why latencies are not the primary metric.
There is a 97% reduction in fragmentation causing events but the
huge page latencies are much higher because they went from never
succeeding to a small success.

2-socket Haswell machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------

4.19 extfrag events < order 0: 803099
4.19+patch1:                   654671 (23% reduction)
4.19+patch1-3:                  24352 (97% reduction)

thpfioscale Fault Latencies
                                       4.19.0                 4.19.0
                                 lowzone-v1r1             boost-v1r5
Amean     fault-base-5     6678.61 (   0.00%)     5935.74 (  11.12%)
Amean     fault-huge-5     2796.35 (   0.00%)     2611.69 (   6.60%)

                                  4.19.0                 4.19.0
                            lowzone-v1r1             boost-v1r5
Percentage huge-5       57.92 (   0.00%)       66.18 (  14.26%)

There is a large reduction in fragmentation events and is reflected
by a higher THP allocation success rate without a negative impact
on fault latencies.

Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
---
 Documentation/sysctl/vm.txt |  19 +++++++
 include/linux/mm.h          |   1 +
 include/linux/mmzone.h      |  11 ++--
 kernel/sysctl.c             |   8 +++
 mm/page_alloc.c             |  50 +++++++++++++++++-
 mm/vmscan.c                 | 123 ++++++++++++++++++++++++++++++++++++++++----
 6 files changed, 197 insertions(+), 15 deletions(-)

diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
index 7d73882e2c27..2244520d7913 100644
--- a/Documentation/sysctl/vm.txt
+++ b/Documentation/sysctl/vm.txt
@@ -63,6 +63,7 @@ files can be found in mm/swap.c.
 - swappiness
 - user_reserve_kbytes
 - vfs_cache_pressure
+- watermark_boost_factor
 - watermark_scale_factor
 - zone_reclaim_mode
 
@@ -856,6 +857,24 @@ ten times more freeable objects than there are.
 
 =============================================================
 
+watermark_boost_factor:
+
+This factor controls the level of reclaim when memory is being fragmented.
+It defines the percentage of the low watermark of a zone that will be
+reclaimed if pages of different mobility are being mixed within pageblocks.
+The intent is so that compaction has less work to do and increase the
+success rate of future high-order allocations such as SLUB allocations,
+THP and hugetlbfs pages.
+
+To make it sensible with respect to the matermark_scale_factor parameter,
+the unit is in fractions of 10,000. The default value of 15000 means
+that 150% of the high watermark will be reclaimed in the event of a
+pageblock being mixed due to fragmentation. If this value is smaller
+than a pageblock then a pageblocks worth of pages will be reclaimed (e.g.
+2MB on 64-bit x86). A boost factor of 0 will disable the feature.
+
+=============================================================
+
 watermark_scale_factor:
 
 This factor controls the aggressiveness of kswapd. It defines the
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 0416a7204be3..036bba4b84af 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -2174,6 +2174,7 @@ extern void zone_pcp_reset(struct zone *zone);
 
 /* page_alloc.c */
 extern int min_free_kbytes;
+extern int watermark_boost_factor;
 extern int watermark_scale_factor;
 
 /* nommu.c */
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 854d6c188888..30595df513c4 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -267,10 +267,10 @@ enum zone_watermarks {
 	NR_WMARK
 };
 
-#define min_wmark_pages(z) (z->_watermark[WMARK_MIN])
-#define low_wmark_pages(z) (z->_watermark[WMARK_LOW])
-#define high_wmark_pages(z) (z->_watermark[WMARK_HIGH])
-#define wmark_pages(z, i) (z->_watermark[i])
+#define min_wmark_pages(z) (z->_watermark[WMARK_MIN] + z->watermark_boost)
+#define low_wmark_pages(z) (z->_watermark[WMARK_LOW] + z->watermark_boost)
+#define high_wmark_pages(z) (z->_watermark[WMARK_HIGH] + z->watermark_boost)
+#define wmark_pages(z, i) (z->_watermark[i] + z->watermark_boost)
 
 struct per_cpu_pages {
 	int count;		/* number of pages in the list */
@@ -362,6 +362,7 @@ struct zone {
 
 	/* zone watermarks, access with *_wmark_pages(zone) macros */
 	unsigned long _watermark[NR_WMARK];
+	unsigned long watermark_boost;
 
 	unsigned long nr_reserved_highatomic;
 
@@ -886,6 +887,8 @@ static inline int is_highmem(struct zone *zone)
 struct ctl_table;
 int min_free_kbytes_sysctl_handler(struct ctl_table *, int,
 					void __user *, size_t *, loff_t *);
+int watermark_boost_factor_sysctl_handler(struct ctl_table *, int,
+					void __user *, size_t *, loff_t *);
 int watermark_scale_factor_sysctl_handler(struct ctl_table *, int,
 					void __user *, size_t *, loff_t *);
 extern int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES];
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index cc02050fd0c4..6886c7928bb4 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1450,6 +1450,14 @@ static struct ctl_table vm_table[] = {
 		.proc_handler	= min_free_kbytes_sysctl_handler,
 		.extra1		= &zero,
 	},
+	{
+		.procname	= "watermark_boost_factor",
+		.data		= &watermark_boost_factor,
+		.maxlen		= sizeof(watermark_boost_factor),
+		.mode		= 0644,
+		.proc_handler	= watermark_boost_factor_sysctl_handler,
+		.extra1		= &zero,
+	},
 	{
 		.procname	= "watermark_scale_factor",
 		.data		= &watermark_scale_factor,
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index a51887765abc..f799c5510789 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -263,6 +263,7 @@ compound_page_dtor * const compound_page_dtors[] = {
 
 int min_free_kbytes = 1024;
 int user_min_free_kbytes = -1;
+int watermark_boost_factor __read_mostly = 15000;
 int watermark_scale_factor = 10;
 
 static unsigned long nr_kernel_pages __meminitdata;
@@ -2118,6 +2119,21 @@ static bool can_steal_fallback(unsigned int order, int start_mt)
 	return false;
 }
 
+static inline void boost_watermark(struct zone *zone)
+{
+	unsigned long max_boost;
+
+	if (!watermark_boost_factor)
+		return;
+
+	max_boost = mult_frac(wmark_pages(zone, WMARK_HIGH),
+			watermark_boost_factor, 10000);
+	max_boost = max(pageblock_nr_pages, max_boost);
+
+	zone->watermark_boost = min(zone->watermark_boost + pageblock_nr_pages,
+		max_boost);
+}
+
 /*
  * This function implements actual steal behaviour. If order is large enough,
  * we can steal whole pageblock. If not, we first move freepages in this
@@ -2149,6 +2165,14 @@ static void steal_suitable_fallback(struct zone *zone, struct page *page,
 		goto single_page;
 	}
 
+	/*
+	 * Boost watermarks to increase reclaim pressure to reduce the
+	 * likelihood of future fallbacks. Wake kswapd now as the node
+	 * may be balanced overall and kswapd will not wake naturally.
+	 */
+	boost_watermark(zone);
+	wakeup_kswapd(zone, 0, 0, zone_idx(zone));
+
 	/* We are not allowed to try stealing from the whole block */
 	if (!whole_block)
 		goto single_page;
@@ -3266,11 +3290,19 @@ static bool zone_allows_reclaim(struct zone *local_zone, struct zone *zone)
  * probably too small. It only makes sense to spread allocations to avoid
  * fragmentation between the Normal and DMA32 zones.
  */
-static inline unsigned int alloc_flags_nofragment(struct zone *zone)
+static inline unsigned int alloc_flags_nofragment(struct zone *zone,
+							gfp_t gfp_mask)
 {
 	if (zone_idx(zone) != ZONE_NORMAL)
 		return 0;
 
+	/*
+	 * A fragmenting fallback will try waking kswapd. ALLOC_NOFRAGMENT
+	 * may break that so such callers can introduce fragmentation.
+	 */
+	if (!(gfp_mask & __GFP_KSWAPD_RECLAIM))
+		return 0;
+
 	/*
 	 * If ZONE_DMA32 exists, assume it is the one after ZONE_NORMAL and
 	 * the pointer is within zone->zone_pgdat->node_zones[].
@@ -4443,7 +4475,8 @@ __alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order, int preferred_nid,
 	 * Forbid the first pass from falling back to types that fragment
 	 * memory until all local zones are considered.
 	 */
-	alloc_flags |= alloc_flags_nofragment(ac.preferred_zoneref->zone);
+	alloc_flags |= alloc_flags_nofragment(ac.preferred_zoneref->zone,
+								gfp_mask);
 
 	/* First allocation attempt */
 	page = get_page_from_freelist(alloc_mask, order, alloc_flags, &ac);
@@ -7343,6 +7376,7 @@ static void __setup_per_zone_wmarks(void)
 
 		zone->_watermark[WMARK_LOW]  = min_wmark_pages(zone) + tmp;
 		zone->_watermark[WMARK_HIGH] = min_wmark_pages(zone) + tmp * 2;
+		zone->watermark_boost = 0;
 
 		spin_unlock_irqrestore(&zone->lock, flags);
 	}
@@ -7443,6 +7477,18 @@ int min_free_kbytes_sysctl_handler(struct ctl_table *table, int write,
 	return 0;
 }
 
+int watermark_boost_factor_sysctl_handler(struct ctl_table *table, int write,
+	void __user *buffer, size_t *length, loff_t *ppos)
+{
+	int rc;
+
+	rc = proc_dointvec_minmax(table, write, buffer, length, ppos);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
 int watermark_scale_factor_sysctl_handler(struct ctl_table *table, int write,
 	void __user *buffer, size_t *length, loff_t *ppos)
 {
diff --git a/mm/vmscan.c b/mm/vmscan.c
index c5ef7240cbcb..7a8161258f0d 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -3360,6 +3360,30 @@ static void age_active_anon(struct pglist_data *pgdat,
 	} while (memcg);
 }
 
+static bool pgdat_watermark_boosted(pg_data_t *pgdat, int classzone_idx)
+{
+	int i;
+	struct zone *zone;
+
+	/*
+	 * Check for watermark boosts top-down as the higher zones
+	 * are more likely to be boosted. Both watermarks and boosts
+	 * should not be checked at the time time as reclaim would
+	 * start prematurely when there is no boosting and a lower
+	 * zone is balanced.
+	 */
+	for (i = classzone_idx; i >= 0; i--) {
+		zone = pgdat->node_zones + i;
+		if (!managed_zone(zone))
+			continue;
+
+		if (zone->watermark_boost)
+			return true;
+	}
+
+	return false;
+}
+
 /*
  * Returns true if there is an eligible zone balanced for the request order
  * and classzone_idx
@@ -3370,9 +3394,12 @@ static bool pgdat_balanced(pg_data_t *pgdat, int order, int classzone_idx)
 	unsigned long mark = -1;
 	struct zone *zone;
 
+	/*
+	 * Check watermarks bottom-up as lower zones are more likely to
+	 * meet watermarks.
+	 */
 	for (i = 0; i <= classzone_idx; i++) {
 		zone = pgdat->node_zones + i;
-
 		if (!managed_zone(zone))
 			continue;
 
@@ -3497,23 +3524,42 @@ static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx)
 	int i;
 	unsigned long nr_soft_reclaimed;
 	unsigned long nr_soft_scanned;
+	unsigned long nr_boost_reclaim;
+	unsigned long zone_boosts[MAX_NR_ZONES] = { 0, };
+	bool boosted;
 	struct zone *zone;
 	struct scan_control sc = {
 		.gfp_mask = GFP_KERNEL,
 		.order = order,
-		.priority = DEF_PRIORITY,
-		.may_writepage = !laptop_mode,
 		.may_unmap = 1,
-		.may_swap = 1,
 	};
 
 	__fs_reclaim_acquire();
 
 	count_vm_event(PAGEOUTRUN);
 
+	/*
+	 * Account for the reclaim boost. Note that the zone boost is left in
+	 * place so that parallel allocations that are near the watermark will
+	 * stall or direct reclaim until kswapd is finished.
+	 */
+	nr_boost_reclaim = 0;
+	for (i = 0; i <= classzone_idx; i++) {
+		zone = pgdat->node_zones + i;
+		if (!managed_zone(zone))
+			continue;
+
+		nr_boost_reclaim += zone->watermark_boost;
+		zone_boosts[i] = zone->watermark_boost;
+	}
+	boosted = nr_boost_reclaim;
+
+restart:
+	sc.priority = DEF_PRIORITY;
 	do {
 		unsigned long nr_reclaimed = sc.nr_reclaimed;
 		bool raise_priority = true;
+		bool balanced;
 		bool ret;
 
 		sc.reclaim_idx = classzone_idx;
@@ -3540,13 +3586,39 @@ static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx)
 		}
 
 		/*
-		 * Only reclaim if there are no eligible zones. Note that
-		 * sc.reclaim_idx is not used as buffer_heads_over_limit may
-		 * have adjusted it.
+		 * If the pgdat is imbalanced then ignore boosting and preserve
+		 * the watermarks for a later time and restart. Note that the
+		 * zone watermarks will be still reset at the end of balancing
+		 * on the grounds that the normal reclaim should be enough to
+		 * re-evaluate if boosting is required when kswapd next wakes.
+		 */
+		balanced = pgdat_balanced(pgdat, sc.order, classzone_idx);
+		if (!balanced && nr_boost_reclaim) {
+			nr_boost_reclaim = 0;
+			goto restart;
+		}
+
+		/*
+		 * If boosting is not active then only reclaim if there are no
+		 * eligible zones. Note that sc.reclaim_idx is not used as
+		 * buffer_heads_over_limit may have adjusted it.
 		 */
-		if (pgdat_balanced(pgdat, sc.order, classzone_idx))
+		if (!nr_boost_reclaim && balanced)
 			goto out;
 
+		/* Limit the priority of boosting to avoid reclaim writeback */
+		if (nr_boost_reclaim && sc.priority == DEF_PRIORITY - 2)
+			raise_priority = false;
+
+		/*
+		 * Do not writeback or swap pages for boosted reclaim. The
+		 * intent is to relieve pressure not issue sub-optimal IO
+		 * from reclaim context. If no pages are reclaimed, the
+		 * reclaim will be aborted.
+		 */
+		sc.may_writepage = !laptop_mode && !nr_boost_reclaim;
+		sc.may_swap = !nr_boost_reclaim;
+
 		/*
 		 * Do some background aging of the anon list, to give
 		 * pages a chance to be referenced before reclaiming. All
@@ -3598,6 +3670,16 @@ static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx)
 		 * progress in reclaiming pages
 		 */
 		nr_reclaimed = sc.nr_reclaimed - nr_reclaimed;
+		nr_boost_reclaim -= min(nr_boost_reclaim, nr_reclaimed);
+
+		/*
+		 * If reclaim made no progress for a boost, stop reclaim as
+		 * IO cannot be queued and it could be an infinite loop in
+		 * extreme circumstances.
+		 */
+		if (nr_boost_reclaim && !nr_reclaimed)
+			break;
+
 		if (raise_priority || !nr_reclaimed)
 			sc.priority--;
 	} while (sc.priority >= 1);
@@ -3606,6 +3688,28 @@ static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx)
 		pgdat->kswapd_failures++;
 
 out:
+	/* If reclaim was boosted, account for the reclaim done in this pass */
+	if (boosted) {
+		unsigned long flags;
+
+		for (i = 0; i <= classzone_idx; i++) {
+			if (!zone_boosts[i])
+				continue;
+
+			/* Increments are under the zone lock */
+			zone = pgdat->node_zones + i;
+			spin_lock_irqsave(&zone->lock, flags);
+			zone->watermark_boost -= min(zone->watermark_boost, zone_boosts[i]);
+			spin_unlock_irqrestore(&zone->lock, flags);
+		}
+
+		/*
+		 * As there is now likely space, wakeup kcompact to defragment
+		 * pageblocks.
+		 */
+		wakeup_kcompactd(pgdat, pageblock_order, classzone_idx);
+	}
+
 	snapshot_refaults(NULL, pgdat);
 	__fs_reclaim_release();
 	/*
@@ -3833,7 +3937,8 @@ void wakeup_kswapd(struct zone *zone, gfp_t gfp_flags, int order,
 
 	/* Hopeless node, leave it to direct reclaim if possible */
 	if (pgdat->kswapd_failures >= MAX_RECLAIM_RETRIES ||
-	    pgdat_balanced(pgdat, order, classzone_idx)) {
+	    (pgdat_balanced(pgdat, order, classzone_idx) &&
+	     !pgdat_watermark_boosted(pgdat, classzone_idx))) {
 		/*
 		 * There may be plenty of free memory available, but it's too
 		 * fragmented for high-order allocations.  Wake up kcompactd
-- 
2.16.4


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [PATCH 4/5] mm: Stall movable allocations until kswapd progresses during serious external fragmentation event
  2018-10-31 16:06 [PATCH 0/5] Fragmentation avoidance improvements Mel Gorman
                   ` (2 preceding siblings ...)
  2018-10-31 16:06 ` [PATCH 3/5] mm: Reclaim small amounts of memory when an external fragmentation event occurs Mel Gorman
@ 2018-10-31 16:06 ` Mel Gorman
  2018-10-31 16:06 ` [PATCH 5/5] mm: Target compaction on pageblocks that were recently fragmented Mel Gorman
  4 siblings, 0 replies; 8+ messages in thread
From: Mel Gorman @ 2018-10-31 16:06 UTC (permalink / raw)
  To: Linux-MM
  Cc: Andrew Morton, Vlastimil Babka, David Rientjes, Andrea Arcangeli,
	Zi Yan, LKML, Mel Gorman

An external fragmentation causing events as already been described. A
serious external fragmentation causing event is described as one that steals
a contiguous range of pages of an order lower than fragment_stall_order
(PAGE_ALLOC_COSTLY_ORDER by default). If fragmentation would steal a
block smaller than this, this patch causes a movable allocation request
that is allowed to sleep to until kswapd makes progress. As kswapd has
just been woken due to a boosted watermark, it's expected to return quickly.

This stall is not guaranteed to avoid serious fragmentation causing events.
If memory pressure is high enough, the pages freed by kswapd may still
be used or they may not be in pageblocks that contain only movable
pages. Furthermore an allocation request that cannot stall (e.g. atomic
allocations) or if for unmovable/reclaimable pages will still proceed
without stalling.

1-socket Skylake machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 1 THP allocating thread
--------------------------------------

4.19 extfrag events < order 0:  71227
4.19+patch1:                    36456 (49% reduction)
4.19+patch1-3:                   4510 (94% reduction)
4.19+patch1-4:                    548 (99% reduction)

Fragmentation events reduced further. The latency and allocation rates
were similar so are not included for brevity.

1-socket Skylake machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------

4.19 extfrag events < order 0:  40761
4.19+patch1:                    36085 (11% reduction)
4.19+patch1-3:                   1887 (95% reduction)
4.19+patch1-4:                    394 (99% reduction)

thpfioscale Fault Latencies
                                       4.19.0                 4.19.0
                                   boost-v1r5             stall-v1r6
Amean     fault-base-1     1863.70 (   0.00%)     3943.28 *-111.58%*
Amean     fault-huge-1      776.07 (   0.00%)     2739.80 *-253.03%*

                                  4.19.0                 4.19.0
                              boost-v1r5             stall-v1r6
Percentage huge-1       86.92 (   0.00%)       98.55 (  13.39%)

Similar to the first case, the reduction in fragmentation events
is notable. However, on this occasion the latencies are much higher
but the allocation success rate is also way higher at 98% success
rate. This is a case where the increased success rate causing pressure
elsewhere but the reduced external framentation events means that
compaction is more effective. This is a classic trade-off on whether
allocation success rate is higher but if problematic, the behaviour
can be tuned.

2-socket Haswell machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 5 THP allocating threads
----------------------------------------------------------------

4.19 extfrag events < order 0:  882868
4.19+patch1:                    476937 (46% reduction)
4.19+patch1-3:                   29044 (97% reduction)
4.19+patch1-4:                   29290 (97% reduction)

There is little impact on fragmentation causing events but the
latency and allocation rates were similar.

2-socket Haswell machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------

4.19 extfrag events < order 0: 803099
4.19+patch1:                   654671 (23% reduction)
4.19+patch1-3:                  24352 (97% reduction)
4.19+patch1-4:                  16698 (98% reduction)

thpfioscale Fault Latencies
                                       4.19.0                 4.19.0
                                   boost-v1r5             stall-v1r6
Amean     fault-base-5     5935.74 (   0.00%)     8649.60 * -45.72%*
Amean     fault-huge-5     2611.69 (   0.00%)     2799.82 (  -7.20%)

                                  4.19.0                 4.19.0
                              boost-v1r5             stall-v1r6
Percentage huge-5       66.18 (   0.00%)       77.80 (  17.56%)

Similar to the 1-socket case, the fragmentation events are reduced
but the higher THP allocation success rates also impact the latencies
as compaction goes to work.

This patch does reduce fragmentation rates overall but it's not free as
some allocataions can stall for short periods of time. While it's within
acceptable limits for the adverse test case, there may be other workloads
that cannot tolerate the stalls. Either it can be tuned to disable the
feature or more ideally, the test case is made available for analysis
to see if the stall behaviour can be reduced while still limiting the
fragmentation events. On the flip-side, it has been checked that setting
the fragment_stall_order to 9 eliminated fragmentation events entirely
on the 1-socket machine and by 99.71% on the 2-socket machine.

Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
---
 Documentation/sysctl/vm.txt | 23 +++++++++++++++
 include/linux/mm.h          |  1 +
 include/linux/mmzone.h      |  2 ++
 kernel/sysctl.c             | 10 +++++++
 mm/internal.h               |  1 +
 mm/page_alloc.c             | 68 +++++++++++++++++++++++++++++++++++++++------
 6 files changed, 97 insertions(+), 8 deletions(-)

diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
index 2244520d7913..f7d3fcb9d4ce 100644
--- a/Documentation/sysctl/vm.txt
+++ b/Documentation/sysctl/vm.txt
@@ -31,6 +31,7 @@ files can be found in mm/swap.c.
 - dirty_writeback_centisecs
 - drop_caches
 - extfrag_threshold
+- fragment_stall_order
 - hugetlb_shm_group
 - laptop_mode
 - legacy_va_layout
@@ -275,6 +276,28 @@ any throttling.
 
 ==============================================================
 
+fragment_stall_order
+
+External fragmentation control is managed on a pageblock level where the
+page allocator tries to avoid mixing pages of different mobility within page
+blocks (e.g. order 9 on 64-bit x86). If external fragmentation is perfectly
+controlled then a THP allocation will often succeed up to the number of
+movable pageblocks in the system as reported by /proc/pagetypeinfo.
+
+When memory is low, the system may have to mix pageblocks and will wake
+kswapd to try control future fragmentation. fragment_stall_order controls if
+the allocating task will stall if possible until kswapd makes some progress
+in preference to fragmenting the system. This incurs a small stall penalty
+in exchange for future success at allocating huge pages. If the stalls
+are undesirable and high-order allocations are irrelevant then this can
+be disabled by writing 0 to the tunable. Writing the pageblock order will
+strongly (but not perfectly) control external fragmentation.
+
+The default will stall for fragmenting allocations smaller than the
+PAGE_ALLOC_COSTLY_ORDER (defined as order-3 at the time of writing).
+
+==============================================================
+
 hugetlb_shm_group
 
 hugetlb_shm_group contains group id that is allowed to create SysV
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 036bba4b84af..a1a2e2833986 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -2176,6 +2176,7 @@ extern void zone_pcp_reset(struct zone *zone);
 extern int min_free_kbytes;
 extern int watermark_boost_factor;
 extern int watermark_scale_factor;
+extern int fragment_stall_order;
 
 /* nommu.c */
 extern atomic_long_t mmap_pages_allocated;
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 30595df513c4..66e71a8ac8a6 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -891,6 +891,8 @@ int watermark_boost_factor_sysctl_handler(struct ctl_table *, int,
 					void __user *, size_t *, loff_t *);
 int watermark_scale_factor_sysctl_handler(struct ctl_table *, int,
 					void __user *, size_t *, loff_t *);
+int fragment_stall_order_sysctl_handler(struct ctl_table *, int,
+					void __user *, size_t *, loff_t *);
 extern int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES];
 int lowmem_reserve_ratio_sysctl_handler(struct ctl_table *, int,
 					void __user *, size_t *, loff_t *);
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 6886c7928bb4..d26f3d9a6400 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -125,6 +125,7 @@ static int zero;
 static int __maybe_unused one = 1;
 static int __maybe_unused two = 2;
 static int __maybe_unused four = 4;
+static int __maybe_unused max_order = MAX_ORDER;
 static unsigned long one_ul = 1;
 static int one_hundred = 100;
 static int one_thousand = 1000;
@@ -1467,6 +1468,15 @@ static struct ctl_table vm_table[] = {
 		.extra1		= &one,
 		.extra2		= &one_thousand,
 	},
+	{
+		.procname	= "fragment_stall_order",
+		.data		= &fragment_stall_order,
+		.maxlen		= sizeof(fragment_stall_order),
+		.mode		= 0644,
+		.proc_handler	= fragment_stall_order_sysctl_handler,
+		.extra1		= &zero,
+		.extra2		= &max_order,
+	},
 	{
 		.procname	= "percpu_pagelist_fraction",
 		.data		= &percpu_pagelist_fraction,
diff --git a/mm/internal.h b/mm/internal.h
index 0dd659cf2a7e..4f159a3b5c4f 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -489,6 +489,7 @@ unsigned long reclaim_clean_pages_from_list(struct zone *zone,
 #else
 #define ALLOC_NOFRAGMENT	  0x0
 #endif
+#define ALLOC_FRAGMENT_STALL	0x200 /* stall if fragmenting heavily */
 
 enum ttu_flags;
 struct tlbflush_unmap_batch;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index f799c5510789..63de66b893d3 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -265,6 +265,7 @@ int min_free_kbytes = 1024;
 int user_min_free_kbytes = -1;
 int watermark_boost_factor __read_mostly = 15000;
 int watermark_scale_factor = 10;
+int fragment_stall_order __read_mostly = (PAGE_ALLOC_COSTLY_ORDER + 1);
 
 static unsigned long nr_kernel_pages __meminitdata;
 static unsigned long nr_all_pages __meminitdata;
@@ -2134,6 +2135,21 @@ static inline void boost_watermark(struct zone *zone)
 		max_boost);
 }
 
+static void stall_fragmentation(pg_data_t *pgdat)
+{
+	DEFINE_WAIT(wait);
+	long remaining = 0;
+
+	if (current->flags & PF_MEMALLOC)
+		return;
+
+	prepare_to_wait(&pgdat->pfmemalloc_wait, &wait, TASK_INTERRUPTIBLE);
+	if (waitqueue_active(&pgdat->kswapd_wait))
+		wake_up_interruptible(&pgdat->kswapd_wait);
+	remaining = schedule_timeout(HZ/10);
+	finish_wait(&pgdat->pfmemalloc_wait, &wait);
+}
+
 /*
  * This function implements actual steal behaviour. If order is large enough,
  * we can steal whole pageblock. If not, we first move freepages in this
@@ -2142,8 +2158,9 @@ static inline void boost_watermark(struct zone *zone)
  * of pages are free or compatible, we can change migratetype of the pageblock
  * itself, so pages freed in the future will be put on the correct free list.
  */
-static void steal_suitable_fallback(struct zone *zone, struct page *page,
-					int start_type, bool whole_block)
+static bool steal_suitable_fallback(struct zone *zone, struct page *page,
+					int start_type, bool whole_block,
+					unsigned int alloc_flags)
 {
 	unsigned int current_order = page_order(page);
 	struct free_area *area;
@@ -2173,6 +2190,11 @@ static void steal_suitable_fallback(struct zone *zone, struct page *page,
 	boost_watermark(zone);
 	wakeup_kswapd(zone, 0, 0, zone_idx(zone));
 
+	if ((alloc_flags & ALLOC_FRAGMENT_STALL) &&
+	    current_order < fragment_stall_order) {
+		return false;
+	}
+
 	/* We are not allowed to try stealing from the whole block */
 	if (!whole_block)
 		goto single_page;
@@ -2213,11 +2235,12 @@ static void steal_suitable_fallback(struct zone *zone, struct page *page,
 			page_group_by_mobility_disabled)
 		set_pageblock_migratetype(page, start_type);
 
-	return;
+	return true;
 
 single_page:
 	area = &zone->free_area[current_order];
 	list_move(&page->lru, &area->free_list[start_type]);
+	return true;
 }
 
 /*
@@ -2456,13 +2479,14 @@ __rmqueue_fallback(struct zone *zone, int order, int start_migratetype,
 	page = list_first_entry(&area->free_list[fallback_mt],
 							struct page, lru);
 
-	steal_suitable_fallback(zone, page, start_migratetype, can_steal);
+	if (!steal_suitable_fallback(zone, page, start_migratetype, can_steal,
+								alloc_flags))
+		return false;
 
 	trace_mm_page_alloc_extfrag(page, order, current_order,
 		start_migratetype, fallback_mt);
 
 	return true;
-
 }
 
 /*
@@ -3331,6 +3355,7 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 	struct zone *zone;
 	struct pglist_data *last_pgdat_dirty_limit = NULL;
 	bool no_fallback;
+	bool fragment_stall;
 
 retry:
 	/*
@@ -3338,6 +3363,8 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 	 * See also __cpuset_node_allowed() comment in kernel/cpuset.c.
 	 */
 	no_fallback = alloc_flags & ALLOC_NOFRAGMENT;
+	fragment_stall = alloc_flags & ALLOC_FRAGMENT_STALL;
+
 	for_next_zone_zonelist_nodemask(zone, z, ac->zonelist, ac->high_zoneidx,
 								ac->nodemask) {
 		struct page *page;
@@ -3376,18 +3403,21 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 			}
 		}
 
-		if (no_fallback) {
+		if (no_fallback || fragment_stall) {
+			pg_data_t *pgdat = zone->zone_pgdat;
 			int local_nid;
 
 			/*
 			 * If moving to a remote node, retry but allow
 			 * fragmenting fallbacks. Locality is more important
 			 * than fragmentation avoidance.
-			 *
 			 */
+			if (fragment_stall)
+				stall_fragmentation(pgdat);
 			local_nid = zone_to_nid(ac->preferred_zoneref->zone);
 			if (zone_to_nid(zone) != local_nid) {
 				alloc_flags &= ~ALLOC_NOFRAGMENT;
+				alloc_flags &= ~ALLOC_FRAGMENT_STALL;
 				goto retry;
 			}
 		}
@@ -3463,8 +3493,9 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 	 * It's possible on a UMA machine to get through all zones that are
 	 * fragmented. If avoiding fragmentation, reset and try again
 	 */
-	if (no_fallback) {
+	if (no_fallback || fragment_stall) {
 		alloc_flags &= ~ALLOC_NOFRAGMENT;
+		alloc_flags &= ~ALLOC_FRAGMENT_STALL;
 		goto retry;
 	}
 
@@ -4192,6 +4223,14 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
 	 */
 	alloc_flags = gfp_to_alloc_flags(gfp_mask);
 
+	/*
+	 * Consider stalling on heavy for movable allocations in preference to
+	 * fragmenting unmovable/reclaimable pageblocks.
+	 */
+	if ((gfp_mask & (__GFP_MOVABLE|__GFP_DIRECT_RECLAIM)) ==
+			(__GFP_MOVABLE|__GFP_DIRECT_RECLAIM))
+		alloc_flags |= ALLOC_FRAGMENT_STALL;
+
 	/*
 	 * We need to recalculate the starting point for the zonelist iterator
 	 * because we might have used different nodemask in the fast path, or
@@ -4213,6 +4252,7 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
 	page = get_page_from_freelist(gfp_mask, order, alloc_flags, ac);
 	if (page)
 		goto got_pg;
+	alloc_flags &= ~ALLOC_FRAGMENT_STALL;
 
 	/*
 	 * For costly allocations, try direct compaction first, as it's likely
@@ -7489,6 +7529,18 @@ int watermark_boost_factor_sysctl_handler(struct ctl_table *table, int write,
 	return 0;
 }
 
+int fragment_stall_order_sysctl_handler(struct ctl_table *table, int write,
+	void __user *buffer, size_t *length, loff_t *ppos)
+{
+	int rc;
+
+	rc = proc_dointvec_minmax(table, write, buffer, length, ppos);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
 int watermark_scale_factor_sysctl_handler(struct ctl_table *table, int write,
 	void __user *buffer, size_t *length, loff_t *ppos)
 {
-- 
2.16.4


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [PATCH 5/5] mm: Target compaction on pageblocks that were recently fragmented
  2018-10-31 16:06 [PATCH 0/5] Fragmentation avoidance improvements Mel Gorman
                   ` (3 preceding siblings ...)
  2018-10-31 16:06 ` [PATCH 4/5] mm: Stall movable allocations until kswapd progresses during serious external fragmentation event Mel Gorman
@ 2018-10-31 16:06 ` Mel Gorman
  4 siblings, 0 replies; 8+ messages in thread
From: Mel Gorman @ 2018-10-31 16:06 UTC (permalink / raw)
  To: Linux-MM
  Cc: Andrew Morton, Vlastimil Babka, David Rientjes, Andrea Arcangeli,
	Zi Yan, LKML, Mel Gorman

Despite the earlier patches, external fragmentation events are still
inevitable as not all callers can stall or are appropriate to stall
(e.g. unmovable allocations that kswapd reclaim will not necessarily
help). In the event there is a mixed pageblock, it's desirable to move all
movable pages from that block so that unmovable/unreclaimable allocations
do not further pollute the address space.

This patch queues such pageblocks for early compaction and relies on
kswapd to wake kcompactd when some pages are reclaimed. Waking kcompactd
after kswapd makes progress is so that the compaction is more likely to
have a suitable migration destination.

This patch may be controversial as there are multiple other design
decisions that can be made. We could refuse to change pageblock ownership
in some cases but great care would need to be taken to avoid premature
OOMs or a livelock. Similarly, we could tag pageblocks as mixed and
search for them but that would increase scanning costs. Finally, there
is a corner case that a mixed pageblock that is after the point where a
free scanner can operate may fail to clean the pageblock but addressing
that would require a fundamental alteration to how compaction works.

Unlike the previous series, this one is harder to prove that it is a benefit
because it ideally require a very long-lived workload that is fragmenting
to show if it's really effective. The timing of such an allocation stream
would be critical and detecting the change would be difficult can be
within the noise. Hence, the potential benefit of this patch is more
conceptual than quantitive even though there are some positive results.

1-socket Skylake machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 1 THP allocating thread
--------------------------------------

4.19 extfrag events < order 0:  71227
4.19+patch1:                    36456 (49% reduction)
4.19+patch1-3:                   4510 (94% reduction)
4.19+patch1-4:                    548 (99% reduction)
4.19+patch1-5:                    422 (99% reduction)

                                       4.19.0                 4.19.0
                                   stall-v1r6         proactive-v1r6
Amean     fault-base-1      839.48 (   0.00%)      860.89 *  -2.55%*
Amean     fault-huge-1      172.74 (   0.00%)      159.49 (   7.67%)

                                  4.19.0                 4.19.0
                              stall-v1r6         proactive-v1r6
Percentage huge-1        1.04 (   0.00%)        2.29 ( 119.35%)

While there is an improvement in the reduction of fragmentation events
and allocation success rates, the differences are marginal enough that
it may not be significant.

1-socket Skylake machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------

4.19 extfrag events < order 0:  40761
4.19+patch1:                    36085 (11% reduction)
4.19+patch1-3:                   1887 (95% reduction)
4.19+patch1-4:                    394 (99% reduction)
4.19+patch1-5:                    440 (99% reduction)

thpfioscale Fault Latencies
                                       4.19.0                 4.19.0
                                   stall-v1r6         proactive-v1r6
Amean     fault-base-1     3943.28 (   0.00%)     2704.46 *  31.42%*
Amean     fault-huge-1     2739.80 (   0.00%)     2552.13 *   6.85%*

thpfioscale Percentage Faults Huge
                                  4.19.0                 4.19.0
                              stall-v1r6         proactive-v1r6
Percentage huge-1       98.55 (   0.00%)       98.76 (   0.20%)

Slight increase in fragmentation events albeit very small. The latency
is much improved as well as a slight increase in allocation success
rates but this may be a co-incidence of the system state.

2-socket Haswell machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 5 THP allocating threads
----------------------------------------------------------------

4.19 extfrag events < order 0:  882868
4.19+patch1:                    476937 (46% reduction)
4.19+patch1-3:                   29044 (97% reduction)
4.19+patch1-4:                   29290 (97% reduction)
4.19+patch1-5:                   30791 (97% reduction)

thpfioscale Fault Latencies
                                       4.19.0                 4.19.0
                                   stall-v1r6         proactive-v1r6
Amean     fault-base-5     1773.24 (   0.00%)     1519.89 *  14.29%*
Amean     fault-huge-5    17791.20 (   0.00%)      536.44 (  96.98%)

                                  4.19.0                 4.19.0
                              stall-v1r6         proactive-v1r6
Percentage huge-5        0.17 (   0.00%)        0.98 ( 490.00%)

Again, the fragmentation causing events is slightly increased although
this is likely within the noise. The latency is massively improved but
the success rate is only marginally improved. Given the low success rate,
it may be a co-incidence of the exact system state during the test but
the fact it happened on both 1 and 2 socket machines is encouraging.

2-socket Haswell machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------

4.19 extfrag events < order 0: 803099
4.19+patch1:                   654671 (23% reduction)
4.19+patch1-3:                  24352 (97% reduction)
4.19+patch1-4:                  16698 (98% reduction)
4.19+patch1-5:                  32623 (96% reduction)

thpfioscale Fault Latencies
                                       4.19.0                 4.19.0
                                   stall-v1r6         proactive-v1r6
Amean     fault-base-5     8649.60 (   0.00%)    13074.71 * -51.16%*
Amean     fault-huge-5     2799.82 (   0.00%)     3410.02 * -21.79%*

thpfioscale Percentage Faults Huge
                                  4.19.0                 4.19.0
                              stall-v1r6         proactive-v1r6
Percentage huge-5       77.80 (   0.00%)       83.30 (   7.06%)

This shows an increase in both fragmentation events and latency. However
it is somewhat balanced by the higher allocation success rates which in
themselves can increase fragmentation pressure.

This is less an obvious universal win. It does control fragmentation
better to some extent in that pageblocks can be found faster in some
cases but the nature of the workload makes it less clear-cut.

Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
---
 include/linux/compaction.h        |   4 ++
 include/linux/migrate.h           |   7 +-
 include/linux/mmzone.h            |   4 ++
 include/trace/events/compaction.h |  62 ++++++++++++++++
 mm/compaction.c                   | 146 +++++++++++++++++++++++++++++++++++---
 mm/migrate.c                      |   6 +-
 mm/page_alloc.c                   |   7 ++
 7 files changed, 225 insertions(+), 11 deletions(-)

diff --git a/include/linux/compaction.h b/include/linux/compaction.h
index 68250a57aace..1fc1ad055f66 100644
--- a/include/linux/compaction.h
+++ b/include/linux/compaction.h
@@ -177,6 +177,7 @@ bool compaction_zonelist_suitable(struct alloc_context *ac, int order,
 extern int kcompactd_run(int nid);
 extern void kcompactd_stop(int nid);
 extern void wakeup_kcompactd(pg_data_t *pgdat, int order, int classzone_idx);
+extern void kcompactd_queue_migration(struct zone *zone, struct page *page);
 
 #else
 static inline void reset_isolation_suitable(pg_data_t *pgdat)
@@ -225,6 +226,9 @@ static inline void wakeup_kcompactd(pg_data_t *pgdat, int order, int classzone_i
 {
 }
 
+static inline void kcompactd_queue_migration(struct zone *zone, struct page *page)
+{
+}
 #endif /* CONFIG_COMPACTION */
 
 #if defined(CONFIG_COMPACTION) && defined(CONFIG_SYSFS) && defined(CONFIG_NUMA)
diff --git a/include/linux/migrate.h b/include/linux/migrate.h
index f2b4abbca55e..f12cee38c0f0 100644
--- a/include/linux/migrate.h
+++ b/include/linux/migrate.h
@@ -61,7 +61,7 @@ static inline struct page *new_page_nodemask(struct page *page,
 
 #ifdef CONFIG_MIGRATION
 
-extern void putback_movable_pages(struct list_head *l);
+extern unsigned int putback_movable_pages(struct list_head *l);
 extern int migrate_page(struct address_space *mapping,
 			struct page *newpage, struct page *page,
 			enum migrate_mode mode);
@@ -82,7 +82,10 @@ extern int migrate_page_move_mapping(struct address_space *mapping,
 		int extra_count);
 #else
 
-static inline void putback_movable_pages(struct list_head *l) {}
+static inline unsigned int putback_movable_pages(struct list_head *l)
+{
+	return 0;
+}
 static inline int migrate_pages(struct list_head *l, new_page_t new,
 		free_page_t free, unsigned long private, enum migrate_mode mode,
 		int reason)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 66e71a8ac8a6..0a905add8112 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -495,6 +495,10 @@ struct zone {
 	unsigned int		compact_considered;
 	unsigned int		compact_defer_shift;
 	int			compact_order_failed;
+
+#define COMPACT_QUEUE_LENGTH 16
+	unsigned long		compact_queue[COMPACT_QUEUE_LENGTH];
+	int			nr_compact;
 #endif
 
 #if defined CONFIG_COMPACTION || defined CONFIG_CMA
diff --git a/include/trace/events/compaction.h b/include/trace/events/compaction.h
index 6074eff3d766..6b5b61177d8c 100644
--- a/include/trace/events/compaction.h
+++ b/include/trace/events/compaction.h
@@ -353,6 +353,68 @@ DEFINE_EVENT(kcompactd_wake_template, mm_compaction_kcompactd_wake,
 	TP_ARGS(nid, order, classzone_idx)
 );
 
+TRACE_EVENT(mm_compaction_wakeup_kcompactd_queue,
+
+	TP_PROTO(
+		int nid,
+		enum zone_type zoneid,
+		unsigned long pfn,
+		int nr_queued),
+
+	TP_ARGS(nid, pfn, zoneid, nr_queued),
+
+	TP_STRUCT__entry(
+		__field(int, nid)
+		__field(enum zone_type, zoneid)
+		__field(unsigned long, pfn)
+		__field(int, nr_queued)
+	),
+
+	TP_fast_assign(
+		__entry->nid = nid;
+		__entry->zoneid = zoneid;
+		__entry->pfn = pfn;
+		__entry->nr_queued = nr_queued;
+	),
+
+	TP_printk("nid=%d zoneid=%-8s pfn=%lu nr_queued=%d",
+		__entry->nid,
+		__print_symbolic(__entry->zoneid, ZONE_TYPE),
+		__entry->pfn,
+		__entry->nr_queued)
+);
+
+TRACE_EVENT(mm_compaction_kcompactd_migrated,
+
+	TP_PROTO(
+		int nid,
+		enum zone_type zoneid,
+		int nr_migrated,
+		int nr_failed),
+
+	TP_ARGS(nid, zoneid, nr_migrated, nr_failed),
+
+	TP_STRUCT__entry(
+		__field(int, nid)
+		__field(enum zone_type, zoneid)
+		__field(int, nr_migrated)
+		__field(int, nr_failed)
+	),
+
+	TP_fast_assign(
+		__entry->nid = nid;
+		__entry->zoneid = zoneid,
+		__entry->nr_migrated = nr_migrated;
+		__entry->nr_failed = nr_failed;
+	),
+
+	TP_printk("nid=%d zoneid=%-8s nr_migrated=%d nr_failed=%d",
+		__entry->nid,
+		__print_symbolic(__entry->zoneid, ZONE_TYPE),
+		__entry->nr_migrated,
+		__entry->nr_failed)
+);
+
 #endif /* _TRACE_COMPACTION_H */
 
 /* This part must be outside protection */
diff --git a/mm/compaction.c b/mm/compaction.c
index aa9473a64915..853538e568d9 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -1914,6 +1914,12 @@ void compaction_unregister_node(struct node *node)
 
 static inline bool kcompactd_work_requested(pg_data_t *pgdat)
 {
+	int zoneid;
+
+	for (zoneid = 0; zoneid < MAX_NR_ZONES; zoneid++)
+		if (pgdat->node_zones[zoneid].nr_compact)
+			return true;
+
 	return pgdat->kcompactd_max_order > 0 || kthread_should_stop();
 }
 
@@ -1937,6 +1943,93 @@ static bool kcompactd_node_suitable(pg_data_t *pgdat)
 	return false;
 }
 
+static void kcompactd_migrate_block(struct compact_control *cc,
+	unsigned long pfn)
+{
+	unsigned long end = min(pfn + pageblock_nr_pages, zone_end_pfn(cc->zone));
+	unsigned long total_migrated = 0, total_failed = 0;
+
+	cc->migrate_pfn = pfn;
+	while (pfn && pfn < end) {
+		int err;
+		unsigned long nr_migrated, nr_failed = 0;
+
+		pfn = isolate_migratepages_range(cc, pfn, end);
+		if (!pfn)
+			break;
+
+		nr_migrated = cc->nr_migratepages;
+		err = migrate_pages(&cc->migratepages, compaction_alloc,
+				compaction_free, (unsigned long)cc,
+				cc->mode, MR_COMPACTION);
+		if (err) {
+			nr_failed = putback_movable_pages(&cc->migratepages);
+			nr_migrated -= nr_failed;
+		}
+		cc->nr_migratepages = 0;
+		total_migrated += nr_migrated;
+		total_failed += nr_failed;
+	}
+
+	trace_mm_compaction_kcompactd_migrated(zone_to_nid(cc->zone),
+		zone_idx(cc->zone), total_migrated, total_failed);
+	return;
+}
+
+static void kcompactd_init_cc(struct compact_control *cc, struct zone *zone)
+{
+	cc->nr_freepages = 0;
+	cc->nr_migratepages = 0;
+	cc->total_migrate_scanned = 0;
+	cc->total_free_scanned = 0;
+	cc->zone = zone;
+	INIT_LIST_HEAD(&cc->freepages);
+	INIT_LIST_HEAD(&cc->migratepages);
+}
+
+static void kcompactd_do_queue(pg_data_t *pgdat)
+{
+	/*
+	 * With no special task, compact all zones so that a page of requested
+	 * order is allocatable.
+	 */
+	int zoneid;
+	struct zone *zone;
+	struct compact_control cc = {
+		.order = 0,
+		.total_migrate_scanned = 0,
+		.total_free_scanned = 0,
+		.classzone_idx = 0,
+		.mode = MIGRATE_SYNC,
+		.ignore_skip_hint = true,
+		.gfp_mask = GFP_KERNEL,
+	};
+	trace_mm_compaction_kcompactd_wake(pgdat->node_id, 0, -1);
+
+	migrate_prep();
+	for (zoneid = 0; zoneid < MAX_NR_ZONES; zoneid++) {
+		unsigned long pfn = ULONG_MAX;
+		int limit;
+
+		zone = &pgdat->node_zones[zoneid];
+		if (!populated_zone(zone))
+			continue;
+
+		kcompactd_init_cc(&cc, zone);
+		cc.free_pfn = pageblock_start_pfn(zone_end_pfn(zone) - 1);
+		limit = zone->nr_compact;
+		while (zone->nr_compact && limit--) {
+			unsigned long flags;
+
+			spin_lock_irqsave(&zone->lock, flags);
+			if (zone->nr_compact)
+				pfn = zone->compact_queue[--zone->nr_compact];
+			spin_unlock_irqrestore(&zone->lock, flags);
+			kcompactd_migrate_block(&cc, pfn);
+		}
+	}
+}
+
 static void kcompactd_do_work(pg_data_t *pgdat)
 {
 	/*
@@ -1956,7 +2049,6 @@ static void kcompactd_do_work(pg_data_t *pgdat)
 	};
 	trace_mm_compaction_kcompactd_wake(pgdat->node_id, cc.order,
 							cc.classzone_idx);
-	count_compact_event(KCOMPACTD_WAKE);
 
 	for (zoneid = 0; zoneid <= cc.classzone_idx; zoneid++) {
 		int status;
@@ -1972,13 +2064,7 @@ static void kcompactd_do_work(pg_data_t *pgdat)
 							COMPACT_CONTINUE)
 			continue;
 
-		cc.nr_freepages = 0;
-		cc.nr_migratepages = 0;
-		cc.total_migrate_scanned = 0;
-		cc.total_free_scanned = 0;
-		cc.zone = zone;
-		INIT_LIST_HEAD(&cc.freepages);
-		INIT_LIST_HEAD(&cc.migratepages);
+		kcompactd_init_cc(&cc, zone);
 
 		if (kthread_should_stop())
 			return;
@@ -2024,6 +2110,19 @@ static void kcompactd_do_work(pg_data_t *pgdat)
 
 void wakeup_kcompactd(pg_data_t *pgdat, int order, int classzone_idx)
 {
+	int i;
+
+	/* Wake kcompact if there are compaction queue entries */
+	for (i = 0; i < MAX_NR_ZONES; i++) {
+		struct zone *zone = &pgdat->node_zones[i];
+
+		if (!managed_zone(zone))
+			continue;
+
+		if (zone->nr_compact)
+			goto wake;
+	}
+
 	if (!order)
 		return;
 
@@ -2043,6 +2142,7 @@ void wakeup_kcompactd(pg_data_t *pgdat, int order, int classzone_idx)
 	if (!kcompactd_node_suitable(pgdat))
 		return;
 
+wake:
 	trace_mm_compaction_wakeup_kcompactd(pgdat->node_id, order,
 							classzone_idx);
 	wake_up_interruptible(&pgdat->kcompactd_wait);
@@ -2072,12 +2172,42 @@ static int kcompactd(void *p)
 		wait_event_freezable(pgdat->kcompactd_wait,
 				kcompactd_work_requested(pgdat));
 
+		count_compact_event(KCOMPACTD_WAKE);
+		kcompactd_do_queue(pgdat);
 		kcompactd_do_work(pgdat);
 	}
 
 	return 0;
 }
 
+/*
+ * Queue a pageblock to have all movable pages migrated from. Note that
+ * kcompactd is not woken at this point. This assumes that kswapd has
+ * been woken to reclaim pages above the boosted watermark. kcompactd
+ * will be woken when kswapd has made progress.
+ */
+void kcompactd_queue_migration(struct zone *zone, struct page *page)
+{
+	unsigned long pfn = page_to_pfn(page) & ~(pageblock_nr_pages - 1);
+	int nr_queued = -1;
+
+	/* Do not overflow the queue */
+	if (zone->nr_compact == COMPACT_QUEUE_LENGTH)
+		goto trace;
+
+	/* Only queue a pageblock once */
+	for (nr_queued = 0; nr_queued < zone->nr_compact; nr_queued++) {
+		if (zone->compact_queue[nr_queued] == pfn)
+			return;
+	}
+
+	zone->compact_queue[zone->nr_compact++] = pfn;
+
+trace:
+	trace_mm_compaction_wakeup_kcompactd_queue(zone_to_nid(zone),
+		zone_idx(zone), pfn, nr_queued);
+}
+
 /*
  * This kcompactd start function will be called by init and node-hot-add.
  * On node-hot-add, kcompactd will moved to proper cpus if cpus are hot-added.
diff --git a/mm/migrate.c b/mm/migrate.c
index 84381b55b2bd..b8ce5b56a2a9 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -164,12 +164,14 @@ void putback_movable_page(struct page *page)
  * built from lru, balloon, hugetlbfs page. See isolate_migratepages_range()
  * and isolate_huge_page().
  */
-void putback_movable_pages(struct list_head *l)
+unsigned int putback_movable_pages(struct list_head *l)
 {
 	struct page *page;
 	struct page *page2;
+	unsigned int nr_putback = 0;
 
 	list_for_each_entry_safe(page, page2, l, lru) {
+		nr_putback++;
 		if (unlikely(PageHuge(page))) {
 			putback_active_hugepage(page);
 			continue;
@@ -195,6 +197,8 @@ void putback_movable_pages(struct list_head *l)
 			putback_lru_page(page);
 		}
 	}
+
+	return nr_putback;
 }
 
 /*
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 63de66b893d3..77bcc35903e0 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2190,6 +2190,9 @@ static bool steal_suitable_fallback(struct zone *zone, struct page *page,
 	boost_watermark(zone);
 	wakeup_kswapd(zone, 0, 0, zone_idx(zone));
 
+	if (start_type == MIGRATE_MOVABLE || old_block_type == MIGRATE_MOVABLE)
+		kcompactd_queue_migration(zone, page);
+
 	if ((alloc_flags & ALLOC_FRAGMENT_STALL) &&
 	    current_order < fragment_stall_order) {
 		return false;
@@ -6359,7 +6362,11 @@ static void pgdat_init_split_queue(struct pglist_data *pgdat) {}
 #ifdef CONFIG_COMPACTION
 static void pgdat_init_kcompactd(struct pglist_data *pgdat)
 {
+	int i;
+
 	init_waitqueue_head(&pgdat->kcompactd_wait);
+	for (i = 0; i < MAX_NR_ZONES; i++)
+		pgdat->node_zones[i].nr_compact = 0;
 }
 #else
 static void pgdat_init_kcompactd(struct pglist_data *pgdat) {}
-- 
2.16.4


^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH 3/5] mm: Reclaim small amounts of memory when an external fragmentation event occurs
  2018-10-31 16:06 ` [PATCH 3/5] mm: Reclaim small amounts of memory when an external fragmentation event occurs Mel Gorman
@ 2018-10-31 16:56   ` Mel Gorman
  0 siblings, 0 replies; 8+ messages in thread
From: Mel Gorman @ 2018-10-31 16:56 UTC (permalink / raw)
  To: Linux-MM
  Cc: Andrew Morton, Vlastimil Babka, David Rientjes, Andrea Arcangeli,
	Zi Yan, LKML

On Wed, Oct 31, 2018 at 04:06:43PM +0000, Mel Gorman wrote:
> An external fragmentation event was previously described as
> 
>     When the page allocator fragments memory, it records the event using
>     the mm_page_alloc_extfrag event. If the fallback_order is smaller
>     than a pageblock order (order-9 on 64-bit x86) then it's considered
>     an event that will cause external fragmentation issues in the future.
> 

This had a build error reported by the 0-day bot. It's trivially fixed
with

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 77bcc35903e0..e36c279dfade 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3317,8 +3317,8 @@ static bool zone_allows_reclaim(struct zone *local_zone, struct zone *zone)
  * probably too small. It only makes sense to spread allocations to avoid
  * fragmentation between the Normal and DMA32 zones.
  */
-static inline unsigned int alloc_flags_nofragment(struct zone *zone,
-							gfp_t gfp_mask)
+static inline unsigned int
+alloc_flags_nofragment(struct zone *zone, gfp_t gfp_mask)
 {
 	if (zone_idx(zone) != ZONE_NORMAL)
 		return 0;
@@ -3340,7 +3340,8 @@ static inline unsigned int alloc_flags_nofragment(struct zone *zone,
 	return ALLOC_NOFRAGMENT;
 }
 #else
-static inline unsigned int alloc_flags_nofragment(struct zone *zone)
+static inline unsigned int
+alloc_flags_nofragment(struct zone *zone, gfp_t gfp_mask)
 {
 	return 0;
 }

-- 
Mel Gorman
SUSE Labs

^ permalink raw reply	[flat|nested] 8+ messages in thread

* [PATCH 4/5] mm: Stall movable allocations until kswapd progresses during serious external fragmentation event
  2018-11-07 18:38 [PATCH 0/5] Fragmentation avoidance improvements v2 Mel Gorman
@ 2018-11-07 18:38 ` Mel Gorman
  0 siblings, 0 replies; 8+ messages in thread
From: Mel Gorman @ 2018-11-07 18:38 UTC (permalink / raw)
  To: Linux-MM
  Cc: Andrew Morton, Vlastimil Babka, David Rientjes, Andrea Arcangeli,
	Zi Yan, LKML, Mel Gorman

An event that potentially causes external fragmentation problems has
already been described but there are degrees of severity.  A "serious"
event is defined as one that steals a contiguous range of pages of an order
lower than fragment_stall_order (PAGE_ALLOC_COSTLY_ORDER by default). If a
movable allocation request that is allowed to sleep needs to steal a small
block then it schedules until kswapd makes progress or a timeout passes.
The watermarks are also boosted slightly faster so that kswapd makes
greater effort to reclaim enough pages to avoid the fragmentation event.

This stall is not guaranteed to avoid serious fragmentation events.
If memory pressure is high enough, the pages freed by kswapd may be
reallocated or the free pages may not be in pageblocks that contain
only movable pages. Furthermore an allocation request that cannot stall
(e.g. atomic allocations) or unmovable/reclaimable allocations will still
proceed without stalling.

The worst-case scenario for stalling is a combination of both high memory
pressure where kswapd is having trouble keeping free pages over the
pfmemalloc_reserve and movable allocations are fragmenting memory. In this
case, an allocation request may sleep for longer. There are both vmstats
to identify stalls are happening and a tracepoint to quantify what the
stall durations are. Note that the granularity of the stall detection is
a jiffy so the delay accounting is not precise.

1-socket Skylake machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 1 THP allocating thread
--------------------------------------

4.20-rc1 extfrag events < order 9:  1023463
4.20-rc1+patch:                      358574 (65% reduction)
4.20-rc1+patch1-3:                    19274 (98% reduction)
4.20-rc1+patch1-4:                     1351 (99.9% reduction)

                                   4.20.0-rc1             4.20.0-rc1
                                   boost-v2r4             stall-v2r6
Amean     fault-base-1      659.85 (   0.00%)      648.66 *   1.70%*
Amean     fault-huge-1      172.19 (   0.00%)      167.79 (   2.56%)

thpfioscale Percentage Faults Huge
                              4.20.0-rc1             4.20.0-rc1
                              boost-v2r4             stall-v2r6
Percentage huge-1        1.68 (   0.00%)        1.16 ( -30.69%)

Fragmentation events are now reduced to negligible levels.

The latencies and allocation success rates are roughly similar.  Over the
course of 16 minutes, there were 100 stalls due to fragmentation avoidance
with a total stall time of 0.4 seconds.

1-socket Skylake machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------

4.20-rc1 extfrag events < order 9:  342549
4.20-rc1+patch:                     337890 ( 1% reduction)
4.20-rc1+patch1-3:                   12801 (96% reduction)
4.20-rc1+patch1-4:                    1117 (99.7% reduction)

                                   4.20.0-rc1             4.20.0-rc1
                                   boost-v2r4             stall-v2r6
Amean     fault-base-1     1578.91 (   0.00%)    43404.60 (-2649.02%)
Amean     fault-huge-1     1090.23 (   0.00%)     1424.32 * -30.64%*

                              4.20.0-rc1             4.20.0-rc1
                              boost-v2r4             stall-v2r6
Percentage huge-1       82.59 (   0.00%)       99.92 (  20.97%)

The fragmentation events were reduced but the latencies went a bit crazy.
The "problem" is that the allocation success rates were very high and
forward progress was being made. This put the system under further pressure
and while compactions were succeeding, the latencies were high in cases
where compaction failed. The THP allocation vm stats are illustrative in this case

                         4.20.0-rc1  4.20.0-rc1
                         boost-v2r4  stall-v2r6
THP fault alloc                4974        6016
THP fault fallback             1048           5
THP collapse alloc               65          56
THP collapse fail                 4           4
THP split                         0        3719
THP split failed                  0         224

Note the THP fault alloc stats where they almost all succeeded relative
to the baseline. While the latencies are much higher, it is the case that
the application specifically requested THP while the system was under
heavy memory pressure.

There were 314 stalls over the course of 16 minutes for a total stall
time of roughly 11 seconds. The distribution of stalls is as follows

    205 4000
      1 8000
      1 20000
      1 32000
      1 36000
      6 40000
      1 56000
     98 100000

This is showing that 98 of the stalls waited until the timeout expired
at 25 jiffies which 100000 microseconds on this particular configuration.
If this is considered problematic, the timeout can be reduced to tradeoff
fault times against fragmentation avoidance.

2-socket Haswell machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 5 THP allocating threads
----------------------------------------------------------------

4.20-rc1 extfrag events < order 9:  209820
4.20-rc1+patch:                     185923 (11% reduction)
4.20-rc1+patch1-3:                   11240 (95% reduction)
4.20-rc1+patch1-4:                   13241 (93% reduction)

                                   4.20.0-rc1             4.20.0-rc1
                                   boost-v2r4             stall-v2r6
Amean     fault-base-5     1395.28 (   0.00%)     1508.94 *  -8.15%*
Amean     fault-huge-5      539.69 (   0.00%)      614.88 * -13.93%*

                              4.20.0-rc1             4.20.0-rc1
                              boost-v2r4             stall-v2r6
Percentage huge-5        0.53 (   0.00%)        3.38 ( 534.38%

There is a slight increase in fragmentation events but given that it's
already heavily reduced, there are elements of luck. There is a small
increase in latencies which is partially offset by a slight increase in
THP allocation success rates. There were 65 stalls over the course of 87
minutes with stall time of a total of roughly 0.4 milliseconds.

2-socket Haswell machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------

4.20-rc1 extfrag events < order 9: 167464
4.20-rc1+patch:                    130081 (22% reduction)
4.20-rc1+patch1-3:                  12057 (92% reduction)
4.20-rc1+patch1-4:                  11060 (93% reduction)

thpfioscale Fault Latencies
                                   4.20.0-rc1             4.20.0-rc1
                                   boost-v2r4             stall-v2r6
Amean     fault-base-5     8691.83 (   0.00%)     9363.89 (  -7.73%)
Amean     fault-huge-5     2899.83 (   0.00%)     3638.29 * -25.47%*

                              4.20.0-rc1             4.20.0-rc1
                              boost-v2r4             stall-v2r6
Percentage huge-5       95.55 (   0.00%)       99.27 (   3.89%)

The fragmentation events are reduced and while there is some hit on
the latency, the success rate is near 100% while under heavy pressure.
There were 2486 stalls over the course of 85 minutes with a total stall
time of roughly 12 seconds.

This patch does reduce fragmentation rates overall but it's not free
as some allocataions can stall for short periods of time and there
are knock-on effects to latency when THP allocation success rates are
higher. While it's within acceptable limits for the adverse test case,
there may be other workloads that cannot tolerate the stalls. If this
occurs, it can be tuned to disable the feature or more ideally, the test
case is made available for analysis to see if the stall behaviour can be
reduced while still limiting the fragmentation events. On the flip-side,
it has been checked that setting the fragment_stall_order to 9 eliminated
fragmentation events entirely.

Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
---
 Documentation/sysctl/vm.txt   | 23 +++++++++++
 include/linux/mm.h            |  1 +
 include/linux/mmzone.h        |  2 +
 include/linux/vm_event_item.h |  1 +
 include/trace/events/kmem.h   | 21 ++++++++++
 kernel/sysctl.c               | 10 +++++
 mm/internal.h                 |  1 +
 mm/page_alloc.c               | 94 +++++++++++++++++++++++++++++++++++++------
 mm/vmstat.c                   |  1 +
 9 files changed, 142 insertions(+), 12 deletions(-)

diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
index 2244520d7913..f7d3fcb9d4ce 100644
--- a/Documentation/sysctl/vm.txt
+++ b/Documentation/sysctl/vm.txt
@@ -31,6 +31,7 @@ files can be found in mm/swap.c.
 - dirty_writeback_centisecs
 - drop_caches
 - extfrag_threshold
+- fragment_stall_order
 - hugetlb_shm_group
 - laptop_mode
 - legacy_va_layout
@@ -275,6 +276,28 @@ any throttling.
 
 ==============================================================
 
+fragment_stall_order
+
+External fragmentation control is managed on a pageblock level where the
+page allocator tries to avoid mixing pages of different mobility within page
+blocks (e.g. order 9 on 64-bit x86). If external fragmentation is perfectly
+controlled then a THP allocation will often succeed up to the number of
+movable pageblocks in the system as reported by /proc/pagetypeinfo.
+
+When memory is low, the system may have to mix pageblocks and will wake
+kswapd to try control future fragmentation. fragment_stall_order controls if
+the allocating task will stall if possible until kswapd makes some progress
+in preference to fragmenting the system. This incurs a small stall penalty
+in exchange for future success at allocating huge pages. If the stalls
+are undesirable and high-order allocations are irrelevant then this can
+be disabled by writing 0 to the tunable. Writing the pageblock order will
+strongly (but not perfectly) control external fragmentation.
+
+The default will stall for fragmenting allocations smaller than the
+PAGE_ALLOC_COSTLY_ORDER (defined as order-3 at the time of writing).
+
+==============================================================
+
 hugetlb_shm_group
 
 hugetlb_shm_group contains group id that is allowed to create SysV
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 81926daf6dfb..ef98eb3f8360 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -2196,6 +2196,7 @@ extern void zone_pcp_reset(struct zone *zone);
 extern int min_free_kbytes;
 extern int watermark_boost_factor;
 extern int watermark_scale_factor;
+extern int fragment_stall_order;
 
 /* nommu.c */
 extern atomic_long_t mmap_pages_allocated;
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index d352c1dab486..cffec484ac8a 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -890,6 +890,8 @@ int watermark_boost_factor_sysctl_handler(struct ctl_table *, int,
 					void __user *, size_t *, loff_t *);
 int watermark_scale_factor_sysctl_handler(struct ctl_table *, int,
 					void __user *, size_t *, loff_t *);
+int fragment_stall_order_sysctl_handler(struct ctl_table *, int,
+					void __user *, size_t *, loff_t *);
 extern int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES];
 int lowmem_reserve_ratio_sysctl_handler(struct ctl_table *, int,
 					void __user *, size_t *, loff_t *);
diff --git a/include/linux/vm_event_item.h b/include/linux/vm_event_item.h
index 47a3441cf4c4..7661abe5236e 100644
--- a/include/linux/vm_event_item.h
+++ b/include/linux/vm_event_item.h
@@ -43,6 +43,7 @@ enum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT,
 		PAGEOUTRUN, PGROTATED,
 		DROP_PAGECACHE, DROP_SLAB,
 		OOM_KILL,
+		FRAGMENTSTALL,
 #ifdef CONFIG_NUMA_BALANCING
 		NUMA_PTE_UPDATES,
 		NUMA_HUGE_PTE_UPDATES,
diff --git a/include/trace/events/kmem.h b/include/trace/events/kmem.h
index eb57e3037deb..caadd8681ac5 100644
--- a/include/trace/events/kmem.h
+++ b/include/trace/events/kmem.h
@@ -315,6 +315,27 @@ TRACE_EVENT(mm_page_alloc_extfrag,
 		__entry->change_ownership)
 );
 
+TRACE_EVENT(mm_fragmentation_stall,
+
+	TP_PROTO(int nid, unsigned long duration),
+
+	TP_ARGS(nid, duration),
+
+	TP_STRUCT__entry(
+		__field(	int,		nid		)
+		__field(	unsigned long,	duration	)
+	),
+
+	TP_fast_assign(
+		__entry->nid		= nid;
+		__entry->duration	= duration
+	),
+
+	TP_printk("nid=%d duration=%lu",
+		__entry->nid,
+		__entry->duration)
+);
+
 #endif /* _TRACE_KMEM_H */
 
 /* This part must be outside protection */
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 1825f712e73b..eb09c79ddbef 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -126,6 +126,7 @@ static int zero;
 static int __maybe_unused one = 1;
 static int __maybe_unused two = 2;
 static int __maybe_unused four = 4;
+static int __maybe_unused max_order = MAX_ORDER;
 static unsigned long one_ul = 1;
 static int one_hundred = 100;
 static int one_thousand = 1000;
@@ -1479,6 +1480,15 @@ static struct ctl_table vm_table[] = {
 		.extra1		= &one,
 		.extra2		= &one_thousand,
 	},
+	{
+		.procname	= "fragment_stall_order",
+		.data		= &fragment_stall_order,
+		.maxlen		= sizeof(fragment_stall_order),
+		.mode		= 0644,
+		.proc_handler	= fragment_stall_order_sysctl_handler,
+		.extra1		= &zero,
+		.extra2		= &max_order,
+	},
 	{
 		.procname	= "percpu_pagelist_fraction",
 		.data		= &percpu_pagelist_fraction,
diff --git a/mm/internal.h b/mm/internal.h
index 544355156c92..5506a4596d59 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -489,6 +489,7 @@ unsigned long reclaim_clean_pages_from_list(struct zone *zone,
 #else
 #define ALLOC_NOFRAGMENT	  0x0
 #endif
+#define ALLOC_FRAGMENT_STALL	0x200 /* stall if fragmenting heavily */
 
 enum ttu_flags;
 struct tlbflush_unmap_batch;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 4abac725a149..86a6e86c51bb 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -265,6 +265,7 @@ int min_free_kbytes = 1024;
 int user_min_free_kbytes = -1;
 int watermark_boost_factor __read_mostly = 15000;
 int watermark_scale_factor = 10;
+int fragment_stall_order __read_mostly = (PAGE_ALLOC_COSTLY_ORDER + 1);
 
 static unsigned long nr_kernel_pages __meminitdata;
 static unsigned long nr_all_pages __meminitdata;
@@ -2130,9 +2131,10 @@ static bool can_steal_fallback(unsigned int order, int start_mt)
 	return false;
 }
 
-static inline void boost_watermark(struct zone *zone)
+static inline void boost_watermark(struct zone *zone, bool fast_boost)
 {
 	unsigned long max_boost;
+	unsigned long nr;
 
 	if (!watermark_boost_factor)
 		return;
@@ -2140,9 +2142,36 @@ static inline void boost_watermark(struct zone *zone)
 	max_boost = mult_frac(wmark_pages(zone, WMARK_HIGH),
 			watermark_boost_factor, 10000);
 	max_boost = max(pageblock_nr_pages, max_boost);
+	nr = pageblock_nr_pages;
 
-	zone->watermark_boost = min(zone->watermark_boost + pageblock_nr_pages,
-		max_boost);
+	/* Scale relative to the MIGRATE_PCPTYPES similar to min_free_kbytes */
+	if (fast_boost)
+		nr += pageblock_nr_pages * (MIGRATE_PCPTYPES << 1);
+
+	zone->watermark_boost = min(zone->watermark_boost + nr, max_boost);
+}
+
+static void stall_fragmentation(struct zone *pzone)
+{
+	DEFINE_WAIT(wait);
+	long remaining = 0;
+	long timeout = HZ/10;
+	pg_data_t *pgdat = pzone->zone_pgdat;
+
+	if (current->flags & PF_MEMALLOC)
+		return;
+
+	boost_watermark(pzone, true);
+	prepare_to_wait(&pgdat->pfmemalloc_wait, &wait, TASK_INTERRUPTIBLE);
+	if (waitqueue_active(&pgdat->kswapd_wait))
+		wake_up_interruptible(&pgdat->kswapd_wait);
+	remaining = schedule_timeout(timeout);
+	finish_wait(&pgdat->pfmemalloc_wait, &wait);
+	if (remaining != timeout) {
+		trace_mm_fragmentation_stall(pgdat->node_id,
+			jiffies_to_usecs(timeout - remaining));
+		count_vm_event(FRAGMENTSTALL);
+	}
 }
 
 /*
@@ -2153,8 +2182,9 @@ static inline void boost_watermark(struct zone *zone)
  * of pages are free or compatible, we can change migratetype of the pageblock
  * itself, so pages freed in the future will be put on the correct free list.
  */
-static void steal_suitable_fallback(struct zone *zone, struct page *page,
-					int start_type, bool whole_block)
+static bool steal_suitable_fallback(struct zone *zone, struct page *page,
+					int start_type, bool whole_block,
+					unsigned int alloc_flags)
 {
 	unsigned int current_order = page_order(page);
 	struct free_area *area;
@@ -2181,9 +2211,14 @@ static void steal_suitable_fallback(struct zone *zone, struct page *page,
 	 * likelihood of future fallbacks. Wake kswapd now as the node
 	 * may be balanced overall and kswapd will not wake naturally.
 	 */
-	boost_watermark(zone);
+	boost_watermark(zone, false);
 	wakeup_kswapd(zone, 0, 0, zone_idx(zone));
 
+	if ((alloc_flags & ALLOC_FRAGMENT_STALL) &&
+	    current_order < fragment_stall_order) {
+		return false;
+	}
+
 	/* We are not allowed to try stealing from the whole block */
 	if (!whole_block)
 		goto single_page;
@@ -2224,11 +2259,12 @@ static void steal_suitable_fallback(struct zone *zone, struct page *page,
 			page_group_by_mobility_disabled)
 		set_pageblock_migratetype(page, start_type);
 
-	return;
+	return true;
 
 single_page:
 	area = &zone->free_area[current_order];
 	list_move(&page->lru, &area->free_list[start_type]);
+	return true;
 }
 
 /*
@@ -2467,13 +2503,14 @@ __rmqueue_fallback(struct zone *zone, int order, int start_migratetype,
 	page = list_first_entry(&area->free_list[fallback_mt],
 							struct page, lru);
 
-	steal_suitable_fallback(zone, page, start_migratetype, can_steal);
+	if (!steal_suitable_fallback(zone, page, start_migratetype, can_steal,
+								alloc_flags))
+		return false;
 
 	trace_mm_page_alloc_extfrag(page, order, current_order,
 		start_migratetype, fallback_mt);
 
 	return true;
-
 }
 
 /*
@@ -3340,9 +3377,12 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 						const struct alloc_context *ac)
 {
 	struct zoneref *z = ac->preferred_zoneref;
+	struct zone *pzone = z->zone;
 	struct zone *zone;
 	struct pglist_data *last_pgdat_dirty_limit = NULL;
 	bool no_fallback;
+	bool fragment_stall;
+	int wmark_idx = alloc_flags & ALLOC_WMARK_MASK;
 
 retry:
 	/*
@@ -3350,6 +3390,8 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 	 * See also __cpuset_node_allowed() comment in kernel/cpuset.c.
 	 */
 	no_fallback = alloc_flags & ALLOC_NOFRAGMENT;
+	fragment_stall = alloc_flags & ALLOC_FRAGMENT_STALL;
+
 	for_next_zone_zonelist_nodemask(zone, z, ac->zonelist, ac->high_zoneidx,
 								ac->nodemask) {
 		struct page *page;
@@ -3388,7 +3430,7 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 			}
 		}
 
-		if (no_fallback) {
+		if (no_fallback || fragment_stall) {
 			int local_nid;
 
 			/*
@@ -3396,9 +3438,12 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 			 * fragmenting fallbacks. Locality is more important
 			 * than fragmentation avoidance.
 			 */
-			local_nid = zone_to_nid(ac->preferred_zoneref->zone);
+			local_nid = zone_to_nid(pzone);
 			if (zone_to_nid(zone) != local_nid) {
+				if (fragment_stall)
+					stall_fragmentation(pzone);
 				alloc_flags &= ~ALLOC_NOFRAGMENT;
+				alloc_flags &= ~ALLOC_FRAGMENT_STALL;
 				goto retry;
 			}
 		}
@@ -3474,8 +3519,12 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 	 * It's possible on a UMA machine to get through all zones that are
 	 * fragmented. If avoiding fragmentation, reset and try again
 	 */
-	if (no_fallback) {
+	if (no_fallback || fragment_stall) {
+		if (fragment_stall)
+			stall_fragmentation(pzone);
+
 		alloc_flags &= ~ALLOC_NOFRAGMENT;
+		alloc_flags &= ~ALLOC_FRAGMENT_STALL;
 		goto retry;
 	}
 
@@ -4197,6 +4246,14 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
 	 */
 	alloc_flags = gfp_to_alloc_flags(gfp_mask);
 
+	/*
+	 * Consider stalling on heavy for movable allocations in preference to
+	 * fragmenting unmovable/reclaimable pageblocks.
+	 */
+	if ((gfp_mask & (__GFP_MOVABLE|__GFP_DIRECT_RECLAIM)) ==
+			(__GFP_MOVABLE|__GFP_DIRECT_RECLAIM))
+		alloc_flags |= ALLOC_FRAGMENT_STALL;
+
 	/*
 	 * We need to recalculate the starting point for the zonelist iterator
 	 * because we might have used different nodemask in the fast path, or
@@ -4218,6 +4275,7 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
 	page = get_page_from_freelist(gfp_mask, order, alloc_flags, ac);
 	if (page)
 		goto got_pg;
+	alloc_flags &= ~ALLOC_FRAGMENT_STALL;
 
 	/*
 	 * For costly allocations, try direct compaction first, as it's likely
@@ -7585,6 +7643,18 @@ int watermark_boost_factor_sysctl_handler(struct ctl_table *table, int write,
 	return 0;
 }
 
+int fragment_stall_order_sysctl_handler(struct ctl_table *table, int write,
+	void __user *buffer, size_t *length, loff_t *ppos)
+{
+	int rc;
+
+	rc = proc_dointvec_minmax(table, write, buffer, length, ppos);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
 int watermark_scale_factor_sysctl_handler(struct ctl_table *table, int write,
 	void __user *buffer, size_t *length, loff_t *ppos)
 {
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 6038ce593ce3..9bb78adf4445 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -1211,6 +1211,7 @@ const char * const vmstat_text[] = {
 	"drop_pagecache",
 	"drop_slab",
 	"oom_kill",
+	"fragment_stall",
 
 #ifdef CONFIG_NUMA_BALANCING
 	"numa_pte_updates",
-- 
2.16.4


^ permalink raw reply	[flat|nested] 8+ messages in thread

end of thread, other threads:[~2018-11-07 18:38 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-10-31 16:06 [PATCH 0/5] Fragmentation avoidance improvements Mel Gorman
2018-10-31 16:06 ` [PATCH 1/5] mm, page_alloc: Spread allocations across zones before introducing fragmentation Mel Gorman
2018-10-31 16:06 ` [PATCH 2/5] mm: Move zone watermark accesses behind an accessor Mel Gorman
2018-10-31 16:06 ` [PATCH 3/5] mm: Reclaim small amounts of memory when an external fragmentation event occurs Mel Gorman
2018-10-31 16:56   ` Mel Gorman
2018-10-31 16:06 ` [PATCH 4/5] mm: Stall movable allocations until kswapd progresses during serious external fragmentation event Mel Gorman
2018-10-31 16:06 ` [PATCH 5/5] mm: Target compaction on pageblocks that were recently fragmented Mel Gorman
2018-11-07 18:38 [PATCH 0/5] Fragmentation avoidance improvements v2 Mel Gorman
2018-11-07 18:38 ` [PATCH 4/5] mm: Stall movable allocations until kswapd progresses during serious external fragmentation event Mel Gorman

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).