From: Dave Jones <davej@redhat.com>
To: David Rientjes <rientjes@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
KOSAKI Motohiro <kosaki.motohiro@gmail.com>,
bhutchings@solarflare.com,
Konstantin Khlebnikov <khlebnikov@openvz.org>,
Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>,
Hugh Dickins <hughd@google.com>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [patch for-3.7] mm, mempolicy: fix printing stack contents in numa_maps
Date: Wed, 17 Oct 2012 15:32:29 -0400 [thread overview]
Message-ID: <20121017193229.GC16805@redhat.com> (raw)
In-Reply-To: <alpine.DEB.2.00.1210171219010.28214@chino.kir.corp.google.com>
On Wed, Oct 17, 2012 at 12:21:10PM -0700, David Rientjes wrote:
> On Wed, 17 Oct 2012, Dave Jones wrote:
>
> > On Tue, Oct 16, 2012 at 10:24:32PM -0700, David Rientjes wrote:
> > > On Wed, 17 Oct 2012, Dave Jones wrote:
> > >
> > > > BUG: sleeping function called from invalid context at kernel/mutex.c:269
> > >
> > > Hmm, looks like we need to change the refcount semantics entirely. We'll
> > > need to make get_vma_policy() always take a reference and then drop it
> > > accordingly. This work sif get_vma_policy() can grab a reference while
> > > holding task_lock() for the task policy fallback case.
> > >
> > > Comments on this approach?
> >
> > Seems to be surviving my testing at least..
> >
>
> Sounds good. Is it possible to verify that policy_cache isn't getting
> larger than normal in /proc/slabinfo, i.e. when all processes with a
> task mempolicy or shared vma policy have exited, are there still a
> significant number of active objects?
Killing the fuzzer caused it to drop dramatically.
Before:
(15:29:59:davej@bitcrush:trinity[master])$ sudo cat /proc/slabinfo | grep policy
shared_policy_node 2931 2967 376 43 4 : tunables 0 0 0 : slabdata 69 69 0
numa_policy 2971 6545 464 35 4 : tunables 0 0 0 : slabdata 187 187 0
After:
(15:30:16:davej@bitcrush:trinity[master])$ sudo cat /proc/slabinfo | grep policy
shared_policy_node 0 215 376 43 4 : tunables 0 0 0 : slabdata 5 5 0
numa_policy 15 175 464 35 4 : tunables 0 0 0 : slabdata 5 5 0
--
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>
WARNING: multiple messages have this Message-ID (diff)
From: Dave Jones <davej@redhat.com>
To: David Rientjes <rientjes@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
KOSAKI Motohiro <kosaki.motohiro@gmail.com>,
bhutchings@solarflare.com,
Konstantin Khlebnikov <khlebnikov@openvz.org>,
Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>,
Hugh Dickins <hughd@google.com>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [patch for-3.7] mm, mempolicy: fix printing stack contents in numa_maps
Date: Wed, 17 Oct 2012 15:32:29 -0400 [thread overview]
Message-ID: <20121017193229.GC16805@redhat.com> (raw)
In-Reply-To: <alpine.DEB.2.00.1210171219010.28214@chino.kir.corp.google.com>
On Wed, Oct 17, 2012 at 12:21:10PM -0700, David Rientjes wrote:
> On Wed, 17 Oct 2012, Dave Jones wrote:
>
> > On Tue, Oct 16, 2012 at 10:24:32PM -0700, David Rientjes wrote:
> > > On Wed, 17 Oct 2012, Dave Jones wrote:
> > >
> > > > BUG: sleeping function called from invalid context at kernel/mutex.c:269
> > >
> > > Hmm, looks like we need to change the refcount semantics entirely. We'll
> > > need to make get_vma_policy() always take a reference and then drop it
> > > accordingly. This work sif get_vma_policy() can grab a reference while
> > > holding task_lock() for the task policy fallback case.
> > >
> > > Comments on this approach?
> >
> > Seems to be surviving my testing at least..
> >
>
> Sounds good. Is it possible to verify that policy_cache isn't getting
> larger than normal in /proc/slabinfo, i.e. when all processes with a
> task mempolicy or shared vma policy have exited, are there still a
> significant number of active objects?
Killing the fuzzer caused it to drop dramatically.
Before:
(15:29:59:davej@bitcrush:trinity[master])$ sudo cat /proc/slabinfo | grep policy
shared_policy_node 2931 2967 376 43 4 : tunables 0 0 0 : slabdata 69 69 0
numa_policy 2971 6545 464 35 4 : tunables 0 0 0 : slabdata 187 187 0
After:
(15:30:16:davej@bitcrush:trinity[master])$ sudo cat /proc/slabinfo | grep policy
shared_policy_node 0 215 376 43 4 : tunables 0 0 0 : slabdata 5 5 0
numa_policy 15 175 464 35 4 : tunables 0 0 0 : slabdata 5 5 0
next prev parent reply other threads:[~2012-10-17 19:32 UTC|newest]
Thread overview: 117+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-08 15:09 mpol_to_str revisited Dave Jones
2012-10-08 15:09 ` Dave Jones
2012-10-08 15:15 ` Dave Jones
2012-10-08 15:15 ` Dave Jones
2012-10-08 20:46 ` David Rientjes
2012-10-08 20:46 ` David Rientjes
2012-10-08 20:35 ` David Rientjes
2012-10-08 20:35 ` David Rientjes
2012-10-08 20:52 ` Dave Jones
2012-10-08 20:52 ` Dave Jones
2012-10-16 0:48 ` David Rientjes
2012-10-16 0:48 ` David Rientjes
2012-10-09 0:33 ` Ben Hutchings
2012-10-16 2:34 ` KOSAKI Motohiro
2012-10-16 2:34 ` KOSAKI Motohiro
2012-10-16 3:58 ` David Rientjes
2012-10-16 3:58 ` David Rientjes
2012-10-16 5:10 ` KOSAKI Motohiro
2012-10-16 5:10 ` KOSAKI Motohiro
2012-10-16 6:10 ` David Rientjes
2012-10-16 6:10 ` David Rientjes
2012-10-16 23:39 ` KOSAKI Motohiro
2012-10-16 23:39 ` KOSAKI Motohiro
2012-10-17 0:12 ` David Rientjes
2012-10-17 0:12 ` David Rientjes
2012-10-17 0:31 ` [patch for-3.7] mm, mempolicy: fix printing stack contents in numa_maps David Rientjes
2012-10-17 0:31 ` David Rientjes
2012-10-17 1:38 ` KOSAKI Motohiro
2012-10-17 1:38 ` KOSAKI Motohiro
2012-10-17 1:49 ` David Rientjes
2012-10-17 1:49 ` David Rientjes
2012-10-17 1:53 ` KOSAKI Motohiro
2012-10-17 1:53 ` KOSAKI Motohiro
2012-10-17 4:05 ` Dave Jones
2012-10-17 4:05 ` Dave Jones
2012-10-17 5:24 ` David Rientjes
2012-10-17 5:24 ` David Rientjes
2012-10-17 5:42 ` Kamezawa Hiroyuki
2012-10-17 5:42 ` Kamezawa Hiroyuki
2012-10-17 8:49 ` KOSAKI Motohiro
2012-10-17 8:49 ` KOSAKI Motohiro
2012-10-17 19:50 ` David Rientjes
2012-10-17 19:50 ` David Rientjes
2012-10-17 21:05 ` KOSAKI Motohiro
2012-10-17 21:05 ` KOSAKI Motohiro
2012-10-17 21:27 ` David Rientjes
2012-10-17 21:27 ` David Rientjes
2012-10-17 18:14 ` Dave Jones
2012-10-17 18:14 ` Dave Jones
2012-10-17 19:21 ` David Rientjes
2012-10-17 19:21 ` David Rientjes
2012-10-17 19:32 ` Dave Jones [this message]
2012-10-17 19:32 ` Dave Jones
2012-10-17 19:38 ` David Rientjes
2012-10-17 19:38 ` David Rientjes
2012-10-17 19:45 ` Dave Jones
2012-10-17 19:45 ` Dave Jones
2012-10-17 20:28 ` [patch for-3.7] mm, mempolicy: avoid taking mutex inside spinlock when reading numa_maps David Rientjes
2012-10-17 20:28 ` David Rientjes
2012-10-17 21:31 ` [patch for-3.7 v2] " David Rientjes
2012-10-17 21:31 ` David Rientjes
2012-10-18 4:06 ` Kamezawa Hiroyuki
2012-10-18 4:06 ` Kamezawa Hiroyuki
2012-10-18 4:14 ` Linus Torvalds
2012-10-18 4:14 ` Linus Torvalds
2012-10-18 4:41 ` Kamezawa Hiroyuki
2012-10-18 4:41 ` Kamezawa Hiroyuki
2012-10-18 4:34 ` Kamezawa Hiroyuki
2012-10-18 4:34 ` Kamezawa Hiroyuki
2012-10-18 20:03 ` David Rientjes
2012-10-18 20:03 ` David Rientjes
2012-10-19 8:35 ` [patch for-3.7 v3] mm, mempolicy: hold task->mempolicy refcount while " Kamezawa Hiroyuki
2012-10-19 8:35 ` Kamezawa Hiroyuki
2012-10-19 9:28 ` David Rientjes
2012-10-19 9:28 ` David Rientjes
2012-10-22 2:47 ` Kamezawa Hiroyuki
2012-10-22 2:47 ` Kamezawa Hiroyuki
2012-10-22 20:55 ` Andrew Morton
2012-10-22 20:55 ` Andrew Morton
2012-10-22 20:56 ` David Rientjes
2012-10-22 20:56 ` David Rientjes
2012-10-19 19:15 ` KOSAKI Motohiro
2012-10-19 19:15 ` KOSAKI Motohiro
2012-10-19 6:51 ` [patch for-3.7 v2] mm, mempolicy: avoid taking mutex inside spinlock when " KOSAKI Motohiro
2012-10-19 6:51 ` KOSAKI Motohiro
2012-10-18 4:35 ` David Rientjes
2012-10-18 4:35 ` David Rientjes
2012-10-24 23:30 ` [patch for-3.7] mm, mempolicy: fix printing stack contents in numa_maps Sasha Levin
2012-10-24 23:30 ` Sasha Levin
2012-10-24 23:34 ` David Rientjes
2012-10-24 23:34 ` David Rientjes
2012-10-24 23:37 ` Sasha Levin
2012-10-24 23:37 ` Sasha Levin
2012-10-25 0:08 ` David Rientjes
2012-10-25 0:08 ` David Rientjes
2012-10-25 0:54 ` KOSAKI Motohiro
2012-10-25 0:54 ` KOSAKI Motohiro
2012-10-25 1:15 ` David Rientjes
2012-10-25 1:15 ` David Rientjes
2012-10-25 12:19 ` Peter Zijlstra
2012-10-25 12:19 ` Peter Zijlstra
2012-10-25 14:39 ` Peter Zijlstra
2012-10-25 14:39 ` Peter Zijlstra
2012-10-25 17:23 ` Sasha Levin
2012-10-25 17:23 ` Sasha Levin
2012-10-25 20:22 ` David Rientjes
2012-10-25 20:22 ` David Rientjes
2012-10-25 23:09 ` Linus Torvalds
2012-10-25 23:09 ` Linus Torvalds
2012-10-26 8:48 ` Peter Zijlstra
2012-10-26 8:48 ` Peter Zijlstra
2012-10-31 18:29 ` Sasha Levin
2012-10-31 18:29 ` Sasha Levin
2012-11-21 0:59 ` Sasha Levin
2012-11-21 0:59 ` Sasha Levin
2012-10-17 1:33 ` mpol_to_str revisited KOSAKI Motohiro
2012-10-17 1:33 ` KOSAKI Motohiro
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=20121017193229.GC16805@redhat.com \
--to=davej@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=bhutchings@solarflare.com \
--cc=hughd@google.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=khlebnikov@openvz.org \
--cc=kosaki.motohiro@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=n-horiguchi@ah.jp.nec.com \
--cc=rientjes@google.com \
--cc=torvalds@linux-foundation.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.