The patch titled
Subject: mm, compaction: create compact_gap wrapper
has been removed from the -mm tree. Its filename was
mm-compaction-create-compact_gap-wrapper.patch
This patch was dropped because it was merged into mainline or a subsystem tree
------------------------------------------------------
From: Vlastimil Babka <vbabka@xxxxxxx>
Subject: mm, compaction: create compact_gap wrapper
Compaction uses a watermark gap of (2UL << order) pages at various places
and it's not immediately obvious why. Abstract it through a compact_gap()
wrapper to create a single place with a thorough explanation.
[vbabka@xxxxxxx: clarify the comment of compact_gap()]
Link: http://lkml.kernel.org/r/7b6aed1f-fdf8-2063-9ff4-bbe4de712d37@xxxxxxx
Link: http://lkml.kernel.org/r/20160810091226.6709-9-vbabka@xxxxxxx
Signed-off-by: Vlastimil Babka <vbabka@xxxxxxx>
Tested-by: Lorenzo Stoakes <lstoakes@xxxxxxxxx>
Acked-by: Michal Hocko <mhocko@xxxxxxxx>
Cc: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>
Cc: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx>
Cc: David Rientjes <rientjes@xxxxxxxxxx>
Cc: Rik van Riel <riel@xxxxxxxxxx>
Signed-off-by: Vlastimil Babka <vbabka@xxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/compaction.h | 23 +++++++++++++++++++++++
mm/compaction.c | 7 +++----
mm/vmscan.c | 6 +++---
3 files changed, 29 insertions(+), 7 deletions(-)
diff -puN include/linux/compaction.h~mm-compaction-create-compact_gap-wrapper include/linux/compaction.h
--- a/include/linux/compaction.h~mm-compaction-create-compact_gap-wrapper
+++ a/include/linux/compaction.h
@@ -58,6 +58,29 @@ enum compact_result {
struct alloc_context; /* in mm/internal.h */
+/*
+ * Number of free order-0 pages that should be available above given watermark
+ * to make sure compaction has reasonable chance of not running out of free
+ * pages that it needs to isolate as migration target during its work.
+ */
+static inline unsigned long compact_gap(unsigned int order)
+{
+ /*
+ * Although all the isolations for migration are temporary, compaction
+ * free scanner may have up to 1 << order pages on its list and then
+ * try to split an (order - 1) free page. At that point, a gap of
+ * 1 << order might not be enough, so it's safer to require twice that
+ * amount. Note that the number of pages on the list is also
+ * effectively limited by COMPACT_CLUSTER_MAX, as that's the maximum
+ * that the migrate scanner can have isolated on migrate list, and free
+ * scanner is only invoked when the number of isolated free pages is
+ * lower than that. But it's not worth to complicate the formula here
+ * as a bigger gap for higher orders than strictly necessary can also
+ * improve chances of compaction success.
+ */
+ return 2UL << order;
+}
+
#ifdef CONFIG_COMPACTION
extern int sysctl_compact_memory;
extern int sysctl_compaction_handler(struct ctl_table *table, int write,
diff -puN mm/compaction.c~mm-compaction-create-compact_gap-wrapper mm/compaction.c
--- a/mm/compaction.c~mm-compaction-create-compact_gap-wrapper
+++ a/mm/compaction.c
@@ -1391,11 +1391,10 @@ static enum compact_result __compaction_
return COMPACT_SUCCESS;
/*
- * Watermarks for order-0 must be met for compaction. Note the 2UL.
- * This is because during migration, copies of pages need to be
- * allocated and for a short time, the footprint is higher
+ * Watermarks for order-0 must be met for compaction to be able to
+ * isolate free pages for migration targets.
*/
- watermark = low_wmark_pages(zone) + (2UL << order);
+ watermark = low_wmark_pages(zone) + compact_gap(order);
if (!__zone_watermark_ok(zone, 0, watermark, classzone_idx,
alloc_flags, wmark_target))
return COMPACT_SKIPPED;
diff -puN mm/vmscan.c~mm-compaction-create-compact_gap-wrapper mm/vmscan.c
--- a/mm/vmscan.c~mm-compaction-create-compact_gap-wrapper
+++ a/mm/vmscan.c
@@ -2480,7 +2480,7 @@ static inline bool should_continue_recla
* If we have not reclaimed enough pages for compaction and the
* inactive lists are large enough, continue reclaiming
*/
- pages_for_compaction = (2UL << sc->order);
+ pages_for_compaction = compact_gap(sc->order);
inactive_lru_pages = node_page_state(pgdat, NR_INACTIVE_FILE);
if (get_nr_swap_pages() > 0)
inactive_lru_pages += node_page_state(pgdat, NR_INACTIVE_ANON);
@@ -2612,7 +2612,7 @@ static inline bool compaction_ready(stru
* there is a buffer of free pages available to give compaction
* a reasonable chance of completing and allocating the page
*/
- watermark = high_wmark_pages(zone) + (2UL << sc->order);
+ watermark = high_wmark_pages(zone) + compact_gap(sc->order);
watermark_ok = zone_watermark_ok_safe(zone, 0, watermark, sc->reclaim_idx);
/*
@@ -3169,7 +3169,7 @@ static bool kswapd_shrink_node(pg_data_t
* excessive reclaim. Assume that a process requested a high-order
* can direct reclaim/compact.
*/
- if (sc->order && sc->nr_reclaimed >= 2UL << sc->order)
+ if (sc->order && sc->nr_reclaimed >= compact_gap(sc->order))
sc->order = 0;
return sc->nr_scanned >= sc->nr_to_reclaim;
_
Patches currently in -mm which might be from vbabka@xxxxxxx are
fs-select-add-vmalloc-fallback-for-select2.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
