lk/linux-kernel-test
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

[PATCH] Optimize off-node performance of zone reclaim

Browse Source 
Ensure that the performance of off node pages stays the same as before.
Off node pagefault tests showed an 18% drop in performance without this
patch.

- Increase the timeout to 30 seconds to reduce the overhead.

- Move all code possible out of the off node hot path for zone reclaim
  (Sorry Andrew, the struct initialization had to be sacrificed).
  The read_page_state() bit us there.

- Check first for the timeout before any other checks.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This commit is contained in:
Christoph Lameter
2006-02-01 03:05:25 -08:00
committed by Linus Torvalds
parent 537421be79
commit 8928862398
1 changed files with 15 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
mm/vmscan.c
View File

@@ -1589,24 +1589,20 @@ int zone_reclaim_mode __read_mostly;
/* /*
* Mininum time between zone reclaim scans * Mininum time between zone reclaim scans
*/ */
#define ZONE_RECLAIM_INTERVAL HZ/2 #define ZONE_RECLAIM_INTERVAL 30*HZ
/* /*
* Try to free up some pages from this zone through reclaim. * Try to free up some pages from this zone through reclaim.
*/ */
int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
{ {
int nr_pages = 1 << order; int nr_pages;
struct task_struct *p = current; struct task_struct *p = current;
struct reclaim_state reclaim_state; struct reclaim_state reclaim_state;
struct scan_control sc = { struct scan_control sc;
.gfp_mask = gfp_mask,
.may_writepage = 0, if (time_before(jiffies,
.may_swap = 0, zone->last_unsuccessful_zone_reclaim + ZONE_RECLAIM_INTERVAL))
.nr_mapped = read_page_state(nr_mapped), return 0;
.nr_scanned = 0,
.nr_reclaimed = 0,
.priority = 0
};
if (!(gfp_mask & __GFP_WAIT) || if (!(gfp_mask & __GFP_WAIT) ||
zone->zone_pgdat->node_id != numa_node_id() || zone->zone_pgdat->node_id != numa_node_id() ||
@@ -1614,12 +1610,17 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
atomic_read(&zone->reclaim_in_progress) > 0) atomic_read(&zone->reclaim_in_progress) > 0)
return 0; return 0;
if (time_before(jiffies, sc.may_writepage = 0;
zone->last_unsuccessful_zone_reclaim + ZONE_RECLAIM_INTERVAL)) sc.may_swap = 0;
return 0; sc.nr_scanned = 0;
sc.nr_reclaimed = 0;
sc.priority = 0;
sc.nr_mapped = read_page_state(nr_mapped);
sc.gfp_mask = gfp_mask;
disable_swap_token(); disable_swap_token();
nr_pages = 1 << order;
if (nr_pages > SWAP_CLUSTER_MAX) if (nr_pages > SWAP_CLUSTER_MAX)
sc.swap_cluster_max = nr_pages; sc.swap_cluster_max = nr_pages;
else else