[PATCH] mm: Reduce the amount of work done when updating min_free_kbytes

[PATCH] mm: Reduce the amount of work done when updating min_free_kbytes

[Mel Gorman]

When min_free_kbytes is updated, some pageblocks are marked MIGRATE_RESERVE.
Ordinarily, this work is unnoticable as it happens early in boot but on
large machines with 1TB of memory, this has been reported to delay
boot times, probably due to the NUMA distances involved.

The bulk of the work is due to calling calling pageblock_is_reserved()
an unnecessary amount of times and accessing far more struct page
metadata than is necessary. This patch significantly reduces the
amount of work done by setup_zone_migrate_reserve() improving boot
times on 1TB machines.

Signed-off-by: Mel Gorman <mgorman@suse.de>
---
 mm/page_alloc.c |   35 +++++++++++++++++++----------------
 1 files changed, 19 insertions(+), 16 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 9dd443d..c95e4c9 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3401,25 +3401,28 @@ static void setup_zone_migrate_reserve(struct zone *zone)
 		if (page_to_nid(page) != zone_to_nid(zone))
 			continue;
 
-		/* Blocks with reserved pages will never free, skip them. */
-		block_end_pfn = min(pfn + pageblock_nr_pages, end_pfn);
-		if (pageblock_is_reserved(pfn, block_end_pfn))
-			continue;
-
 		block_migratetype = get_pageblock_migratetype(page);
 
-		/* If this block is reserved, account for it */
-		if (reserve > 0 && block_migratetype == MIGRATE_RESERVE) {
-			reserve--;
-			continue;
-		}
+		/* Only test what is necessary when the reserves are not met */
+		if (reserve > 0) {
+			/* Blocks with reserved pages will never free, skip them. */
+			block_end_pfn = min(pfn + pageblock_nr_pages, end_pfn);
+			if (pageblock_is_reserved(pfn, block_end_pfn))
+				continue;
 
-		/* Suitable for reserving if this block is movable */
-		if (reserve > 0 && block_migratetype == MIGRATE_MOVABLE) {
-			set_pageblock_migratetype(page, MIGRATE_RESERVE);
-			move_freepages_block(zone, page, MIGRATE_RESERVE);
-			reserve--;
-			continue;
+			/* If this block is reserved, account for it */
+			if (block_migratetype == MIGRATE_RESERVE) {
+				reserve--;
+				continue;
+			}
+
+			/* Suitable for reserving if this block is movable */
+			if (block_migratetype == MIGRATE_MOVABLE) {
+				set_pageblock_migratetype(page, MIGRATE_RESERVE);
+				move_freepages_block(zone, page, MIGRATE_RESERVE);
+				reserve--;
+				continue;
+			}
 		}
 
 		/*

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] mm: Reduce the amount of work done when updating min_free_kbytes

[Andrew Morton]

On Fri, 11 Nov 2011 16:21:19 +0000
Mel Gorman <mgorman@suse.de> wrote:

> When min_free_kbytes is updated, some pageblocks are marked MIGRATE_RESERVE.
> Ordinarily, this work is unnoticable as it happens early in boot but on
> large machines with 1TB of memory, this has been reported to delay
> boot times, probably due to the NUMA distances involved.
> 
> The bulk of the work is due to calling calling pageblock_is_reserved()
> an unnecessary amount of times and accessing far more struct page
> metadata than is necessary. This patch significantly reduces the
> amount of work done by setup_zone_migrate_reserve() improving boot
> times on 1TB machines.
> 

By how much? :)

(I mainly ask because I'm curious to know how long the kernel takes to
boot on a 1TB machine...)

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] mm: Reduce the amount of work done when updating min_free_kbytes

[Mel Gorman]

On Mon, Nov 14, 2011 at 03:21:00PM -0800, Andrew Morton wrote:
> On Fri, 11 Nov 2011 16:21:19 +0000
> Mel Gorman <mgorman@suse.de> wrote:
> 
> > When min_free_kbytes is updated, some pageblocks are marked MIGRATE_RESERVE.
> > Ordinarily, this work is unnoticable as it happens early in boot but on
> > large machines with 1TB of memory, this has been reported to delay
> > boot times, probably due to the NUMA distances involved.
> > 
> > The bulk of the work is due to calling calling pageblock_is_reserved()
> > an unnecessary amount of times and accessing far more struct page
> > metadata than is necessary. This patch significantly reduces the
> > amount of work done by setup_zone_migrate_reserve() improving boot
> > times on 1TB machines.
> > 
> 
> By how much? :)
> 
> (I mainly ask because I'm curious to know how long the kernel takes to
> boot on a 1TB machine...)
> 


Good question. I don't have access to the machine but based on the dmesg
they posted before and after, this patch reduced boot times by 27
seconds.

With only dmesg, I don't know how long it is taking to start services
and mount of the filesystem but assuming no major problems or timeouts
from drivers it looks like it is taking about 6 minutes to boot.

-- 
Mel Gorman
SUSE Labs

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>