lkml.org 
[lkml]   [2005]   [Nov]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
DateSun, 27 Nov 2005 15:21:08 -0800
FromAndrew Morton <>
SubjectRe: 2.6.14 kswapd eating too much CPU
Marcelo Tosatti <marcelo.tosatti@cyclades.com> wrote:
>
> It does seem to scan SLABs intensively:
>

It might be worth trying the below. Mainly for the debugging check.


From: Andrea Arcangeli <andrea@suse.de>

Greg Edwards found some deadlock in the icache shrinker.

I believe the major bug is that the VM is currently potentially setting nr
= LONG_MAX before shrinking the icache (and the icache shrinker never
returns -1, which means the api doesn't currently require shrinkers to
return -1 when they're finished).

The below is the most obviously safe way I could address this problem
(still untested).

This is not necessairly the way we want to fix it in mainline, but it at
least shows what I believe to be the major cuplrit in the code (i.e. nr
growing insane ;).

Signed-off-by: Andrea Arcangeli <andrea@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
mm/vmscan.c | 18 +++++++++++++++---
1 files changed, 15 insertions(+), 3 deletions(-)
diff -puN mm/vmscan.c~shrinker-nr-=-long_max-means-deadlock-for-icache mm/vmscan.c
--- devel/mm/vmscan.c~shrinker-nr-=-long_max-means-deadlock-for-icache 2005-11-19 02:53:18.000000000 -0800
+++ devel-akpm/mm/vmscan.c 2005-11-19 03:00:25.000000000 -0800
@@ -201,13 +201,25 @@ static int shrink_slab(unsigned long sca
list_for_each_entry(shrinker, &shrinker_list, list) {
unsigned long long delta;
unsigned long total_scan;
+ unsigned long max_pass = (*shrinker->shrinker)(0, gfp_mask);

delta = (4 * scanned) / shrinker->seeks;
- delta *= (*shrinker->shrinker)(0, gfp_mask);
+ delta *= max_pass;
do_div(delta, lru_pages + 1);
shrinker->nr += delta;
- if (shrinker->nr < 0)
- shrinker->nr = LONG_MAX; /* It wrapped! */
+ if (shrinker->nr < 0) {
+ printk(KERN_ERR "%s: nr=%ld\n",
+ __FUNCTION__, shrinker->nr);
+ shrinker->nr = max_pass;
+ }
+
+ /*
+ * Avoid risking looping forever due to too large nr value:
+ * never try to free more than twice the estimate number of
+ * freeable entries.
+ */
+ if (shrinker->nr > max_pass * 2)
+ shrinker->nr = max_pass * 2;

total_scan = shrinker->nr;
shrinker->nr = 0;
_
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-11-28 00:23    [from the cache]
©2003-2010