Thread (25 messages) 25 messages, 7 authors, 2012-07-12

Re: [PATCH -mm v3] mm: have order > 0 compaction start off where it left

From: Minchan Kim <minchan@kernel.org>
Date: 2012-07-12 02:26:23
Also in: lkml

Hi Rik,

On Wed, Jul 11, 2012 at 04:18:00PM -0400, Rik van Riel wrote:
This patch makes the comment for cc->wrapped longer, explaining
what is really going on. It also incorporates the comment fix
pointed out by Minchan.

Additionally, Minchan found that, when no pages get isolated,
high_pte could be a value that is much lower than desired,
s/high_pte/high_pfn
which might potentially cause compaction to skip a range of
pages.

Only assign zone->compact_cache_free_pfn if we actually
isolated free pages for compaction.

Split out the calculation to get the start of the last page
block in a zone into its own, commented function.

Signed-off-by: Rik van Riel <redacted>
Acked-by: Minchan Kim <minchan@kernel.org>
quoted hunk ↗ jump to hunk
---
 include/linux/mmzone.h |    2 +-
 mm/compaction.c        |   30 ++++++++++++++++++++++--------
 mm/internal.h          |    6 +++++-
 3 files changed, 28 insertions(+), 10 deletions(-)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index e629594..e957fa1 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -370,7 +370,7 @@ struct zone {
 	spinlock_t		lock;
 	int                     all_unreclaimable; /* All pages pinned */
 #if defined CONFIG_COMPACTION || defined CONFIG_CMA
-	/* pfn where the last order > 0 compaction isolated free pages */
+	/* pfn where the last incremental compaction isolated free pages */
 	unsigned long		compact_cached_free_pfn;
 #endif
 #ifdef CONFIG_MEMORY_HOTPLUG
diff --git a/mm/compaction.c b/mm/compaction.c
index 2668b77..3812c3e 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -472,10 +472,11 @@ static void isolate_freepages(struct zone *zone,
 		 * looking for free pages, the search will restart here as
 		 * page migration may have returned some pages to the allocator
 		 */
-		if (isolated)
+		if (isolated) {
 			high_pfn = max(high_pfn, pfn);
-		if (cc->order > 0)
-			zone->compact_cached_free_pfn = high_pfn;
+			if (cc->order > 0)
+				zone->compact_cached_free_pfn = high_pfn;
+		}
 	}
 
 	/* split_free_page does not map the pages */
@@ -569,6 +570,21 @@ static isolate_migrate_t isolate_migratepages(struct zone *zone,
 	return ISOLATE_SUCCESS;
 }
 
+/*
+ * Returns the start pfn of the laste page block in a zone.
s/laste/last/



--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help