linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Andrea Arcangeli <andrea@cpushare.com>
To: David Rientjes <rientjes@google.com>
Cc: linux-mm@kvack.org, Andrew Morton <akpm@osdl.org>
Subject: Re: [PATCH 10 of 24] stop useless vm trashing while we wait the TIF_MEMDIE task to exit
Date: Thu, 3 Jan 2008 02:08:31 +0100	[thread overview]
Message-ID: <20080103010831.GL30939@v2.random> (raw)
In-Reply-To: <alpine.DEB.0.9999.0709211208140.11391@chino.kir.corp.google.com>

On Fri, Sep 21, 2007 at 12:10:23PM -0700, David Rientjes wrote:
> On Wed, 22 Aug 2007, Andrea Arcangeli wrote:
> 
> > diff --git a/mm/vmscan.c b/mm/vmscan.c
> > --- a/mm/vmscan.c
> > +++ b/mm/vmscan.c
> > @@ -1028,6 +1028,8 @@ static unsigned long shrink_zone(int pri
> >  		nr_inactive = 0;
> >  
> >  	while (nr_active || nr_inactive) {
> > +		if (is_VM_OOM())
> > +			break;
> >  		if (nr_active) {
> >  			nr_to_scan = min(nr_active,
> >  					(unsigned long)sc->swap_cluster_max);
> 
> This will need to use the new OOM zone-locking interface.  shrink_zones() 
> accepts struct zone** as one of its formals so while traversing each zone 
> this would simply become a test of zone_is_oom_locked(*z).

yes I changed this with zone_is_oom_locked. same logic as before, to
spend the time in schedule_timeout while the system tries to solve the
oom condition instead of trashing the whole cpu caches over the lru.

> 
> > @@ -1138,6 +1140,17 @@ unsigned long try_to_free_pages(struct z
> >  	}
> >  
> >  	for (priority = DEF_PRIORITY; priority >= 0; priority--) {
> > +		if (is_VM_OOM()) {
> > +			if (!test_thread_flag(TIF_MEMDIE)) {
> > +				/* get out of the way */
> > +				schedule_timeout_interruptible(1);
> > +				/* don't waste cpu if we're still oom */
> > +				if (is_VM_OOM())
> > +					goto out;
> > +			} else
> > +				goto out;
> > +		}
> > +
> >  		sc.nr_scanned = 0;
> >  		if (!priority)
> >  			disable_swap_token();
> > 
> 
> Same as above, and it becomes trivial since try_to_free_pages() also 
> accepts a struct zone** formal.

yes, converted this too.

--
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>

  reply	other threads:[~2008-01-03  1:08 UTC|newest]

Thread overview: 113+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-08-22 12:48 [PATCH 00 of 24] OOM related fixes Andrea Arcangeli
2007-08-22 12:48 ` [PATCH 01 of 24] remove nr_scan_inactive/active Andrea Arcangeli
2007-09-12 11:44   ` Andrew Morton
2008-01-02 17:50     ` Andrea Arcangeli
2007-08-22 12:48 ` [PATCH 02 of 24] avoid oom deadlock in nfs_create_request Andrea Arcangeli
2007-09-12 23:54   ` Christoph Lameter
2007-08-22 12:48 ` [PATCH 03 of 24] prevent oom deadlocks during read/write operations Andrea Arcangeli
2007-09-12 11:56   ` Andrew Morton
2007-09-12  2:18     ` Nick Piggin
2008-01-03  0:53     ` Andrea Arcangeli
2007-08-22 12:48 ` [PATCH 04 of 24] serialize oom killer Andrea Arcangeli
2007-09-12 12:02   ` Andrew Morton
2007-09-12 12:04     ` Andrew Morton
2007-09-12 12:11       ` Andrea Arcangeli
2008-01-03  0:55     ` Andrea Arcangeli
2007-09-13  0:09   ` Christoph Lameter
2007-09-13 18:32     ` David Rientjes
2007-09-13 18:37       ` Christoph Lameter
2007-09-13 18:46         ` David Rientjes
2007-09-13 18:53           ` Christoph Lameter
2007-09-14  0:36             ` David Rientjes
2007-09-14  2:31               ` Christoph Lameter
2007-09-14  3:33                 ` David Rientjes
2007-09-18 16:44                   ` David Rientjes
2007-09-18 16:44                     ` [patch 1/4] oom: move prototypes to appropriate header file David Rientjes
2007-09-18 16:44                       ` [patch 2/4] oom: move constraints to enum David Rientjes
2007-09-18 16:44                         ` [patch 3/4] oom: save zonelist pointer for oom killer calls David Rientjes
2007-09-18 16:44                           ` [patch 4/4] oom: serialize out of memory calls David Rientjes
2007-09-18 19:54                             ` Christoph Lameter
2007-09-18 19:56                               ` David Rientjes
2007-09-18 20:01                                 ` Christoph Lameter
2007-09-18 20:06                                   ` David Rientjes
2007-09-18 20:23                                     ` [patch 5/4] oom: rename serialization helper functions David Rientjes
2007-09-18 20:26                                       ` Christoph Lameter
2007-09-18 20:39                                         ` [patch 5/4 v2] " David Rientjes
2007-09-18 20:59                                           ` Christoph Lameter
2007-09-18 19:57                           ` [patch 3/4] oom: save zonelist pointer for oom killer calls Christoph Lameter
2007-09-18 20:13                             ` David Rientjes
2007-09-18 20:16                               ` Christoph Lameter
2007-09-18 20:47                                 ` [patch 6/4] oom: pass null to kfree if zonelist is not cleared David Rientjes
2007-09-18 21:01                                   ` Christoph Lameter
2007-09-18 21:13                                     ` David Rientjes
2007-09-18 21:25                                       ` Christoph Lameter
2007-09-18 22:16                                         ` David Rientjes
2007-09-19 17:09                                           ` Paul Jackson
2007-09-19 18:21                                             ` David Rientjes
2007-09-18 19:55                         ` [patch 2/4] oom: move constraints to enum Christoph Lameter
2007-08-22 12:48 ` [PATCH 05 of 24] avoid selecting already killed tasks Andrea Arcangeli
2007-09-13  0:13   ` Christoph Lameter
2007-08-22 12:48 ` [PATCH 06 of 24] reduce the probability of an OOM livelock Andrea Arcangeli
2007-09-12 12:17   ` Andrew Morton
2008-01-03  1:03     ` Andrea Arcangeli
2007-08-22 12:48 ` [PATCH 07 of 24] balance_pgdat doesn't return the number of pages freed Andrea Arcangeli
2007-09-12 12:18   ` Andrew Morton
2007-09-13  0:26     ` Christoph Lameter
2007-08-22 12:48 ` [PATCH 08 of 24] don't depend on PF_EXITING tasks to go away Andrea Arcangeli
2007-09-12 12:20   ` Andrew Morton
2008-01-03  0:56     ` Andrea Arcangeli
2007-08-22 12:48 ` [PATCH 09 of 24] fallback killing more tasks if tif-memdie doesn't " Andrea Arcangeli
2007-09-12 12:30   ` Andrew Morton
2007-09-12 12:34     ` Andrew Morton
2008-01-03  1:06     ` Andrea Arcangeli
2007-08-22 12:48 ` [PATCH 10 of 24] stop useless vm trashing while we wait the TIF_MEMDIE task to exit Andrea Arcangeli
2007-09-12 12:42   ` Andrew Morton
2007-09-13  0:36     ` Christoph Lameter
2007-09-21 19:10   ` David Rientjes
2008-01-03  1:08     ` Andrea Arcangeli [this message]
2007-08-22 12:48 ` [PATCH 11 of 24] the oom schedule timeout isn't needed with the VM_is_OOM logic Andrea Arcangeli
2007-09-12 12:44   ` Andrew Morton
2007-08-22 12:48 ` [PATCH 12 of 24] show mem information only when a task is actually being killed Andrea Arcangeli
2007-09-12 12:49   ` Andrew Morton
2007-08-22 12:49 ` [PATCH 13 of 24] simplify oom heuristics Andrea Arcangeli
2007-09-12 12:52   ` Andrew Morton
2007-09-12 13:40     ` Andrea Arcangeli
2007-09-12 20:52       ` Andrew Morton
2007-08-22 12:49 ` [PATCH 14 of 24] oom select should only take rss into account Andrea Arcangeli
2007-09-13  0:43   ` Christoph Lameter
2007-08-22 12:49 ` [PATCH 15 of 24] limit reclaim if enough pages have been freed Andrea Arcangeli
2007-09-12 12:57   ` Andrew Morton
2008-01-03  1:12     ` Andrea Arcangeli
2007-09-12 12:58   ` Andrew Morton
2007-09-12 13:38     ` Andrea Arcangeli
2007-08-22 12:49 ` [PATCH 16 of 24] avoid some lock operation in vm fast path Andrea Arcangeli
2007-09-12 12:59   ` Andrew Morton
2007-09-13  0:49     ` Christoph Lameter
2007-09-13  1:16       ` Andrew Morton
2007-09-13  1:33         ` Christoph Lameter
2007-09-13  1:41           ` KAMEZAWA Hiroyuki
2007-09-13  1:44           ` Andrew Morton
2007-08-22 12:49 ` [PATCH 17 of 24] apply the anti deadlock features only to global oom Andrea Arcangeli
2007-09-12 13:02   ` Andrew Morton
2007-09-13  0:53     ` Christoph Lameter
2007-09-13  0:52   ` Christoph Lameter
2007-08-22 12:49 ` [PATCH 18 of 24] run panic the same way in both places Andrea Arcangeli
2007-09-13  0:54   ` Christoph Lameter
2007-08-22 12:49 ` [PATCH 19 of 24] cacheline align VM_is_OOM to prevent false sharing Andrea Arcangeli
2007-09-12 13:02   ` Andrew Morton
2007-09-12 13:36     ` Andrea Arcangeli
2007-09-13  0:55       ` Christoph Lameter
2007-08-22 12:49 ` [PATCH 20 of 24] extract deadlock helper function Andrea Arcangeli
2007-08-22 12:49 ` [PATCH 21 of 24] select process to kill for cpusets Andrea Arcangeli
2007-09-12 13:05   ` Andrew Morton
2007-09-13  0:59     ` Christoph Lameter
2007-09-13  5:13       ` David Rientjes
2007-09-13 17:55         ` Christoph Lameter
2007-08-22 12:49 ` [PATCH 22 of 24] extract select helper function Andrea Arcangeli
2007-08-22 12:49 ` [PATCH 23 of 24] serialize for cpusets Andrea Arcangeli
2007-09-12 13:10   ` Andrew Morton
2007-09-12 13:34     ` Andrea Arcangeli
2007-09-12 19:08     ` David Rientjes
2007-09-13  1:02     ` Christoph Lameter
2007-08-22 12:49 ` [PATCH 24 of 24] add oom_kill_asking_task flag Andrea Arcangeli
2007-09-12 13:11   ` Andrew Morton

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20080103010831.GL30939@v2.random \
    --to=andrea@cpushare.com \
    --cc=akpm@osdl.org \
    --cc=linux-mm@kvack.org \
    --cc=rientjes@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).