From: Greg Thelen <gthelen@google.com>
To: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Vivek Goyal <vgoyal@redhat.com>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
balbir@linux.vnet.ibm.com, Andrea Righi <arighi@develer.com>,
Peter Zijlstra <peterz@infradead.org>,
Trond Myklebust <trond.myklebust@fys.uio.no>,
Suleiman Souhlal <suleiman@google.com>,
"Kirill A. Shutemov" <kirill@shutemov.name>,
Andrew Morton <akpm@linux-foundation.org>,
containers@lists.linux-foundation.org,
linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH -mmotm 1/5] memcg: disable irq at page cgroup lock
Date: Wed, 14 Apr 2010 23:38:30 -0700 [thread overview]
Message-ID: <i2k49b004811004142338je898a741q78a12c902c52724a@mail.gmail.com> (raw)
In-Reply-To: <20100415152104.62593f37.nishimura@mxp.nes.nec.co.jp>
On Wed, Apr 14, 2010 at 11:21 PM, Daisuke Nishimura
<nishimura@mxp.nes.nec.co.jp> wrote:
> On Wed, 14 Apr 2010 21:48:25 -0700, Greg Thelen <gthelen@google.com> wrote:
>> On Wed, Apr 14, 2010 at 7:40 PM, Daisuke Nishimura
>> <nishimura@mxp.nes.nec.co.jp> wrote:
>> > On Wed, 14 Apr 2010 13:14:07 -0700, Greg Thelen <gthelen@google.com> wrote:
>> >> Vivek Goyal <vgoyal@redhat.com> writes:
>> >>
>> >> > On Tue, Apr 13, 2010 at 11:55:12PM -0700, Greg Thelen wrote:
>> >> >> On Thu, Mar 18, 2010 at 8:00 PM, KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> wrote:
>> >> >> > On Fri, 19 Mar 2010 08:10:39 +0530
>> >> >> > Balbir Singh <balbir@linux.vnet.ibm.com> wrote:
>> >> >> >
>> >> >> >> * KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> [2010-03-19 10:23:32]:
>> >> >> >>
>> >> >> >> > On Thu, 18 Mar 2010 21:58:55 +0530
>> >> >> >> > Balbir Singh <balbir@linux.vnet.ibm.com> wrote:
>> >> >> >> >
>> >> >> >> > > * KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> [2010-03-18 13:35:27]:
>> >> >> >> >
>> >> >> >> > > > Then, no probelm. It's ok to add mem_cgroup_udpate_stat() indpendent from
>> >> >> >> > > > mem_cgroup_update_file_mapped(). The look may be messy but it's not your
>> >> >> >> > > > fault. But please write "why add new function" to patch description.
>> >> >> >> > > >
>> >> >> >> > > > I'm sorry for wasting your time.
>> >> >> >> > >
>> >> >> >> > > Do we need to go down this route? We could check the stat and do the
>> >> >> >> > > correct thing. In case of FILE_MAPPED, always grab page_cgroup_lock
>> >> >> >> > > and for others potentially look at trylock. It is OK for different
>> >> >> >> > > stats to be protected via different locks.
>> >> >> >> > >
>> >> >> >> >
>> >> >> >> > I _don't_ want to see a mixture of spinlock and trylock in a function.
>> >> >> >> >
>> >> >> >>
>> >> >> >> A well documented well written function can help. The other thing is to
>> >> >> >> of-course solve this correctly by introducing different locking around
>> >> >> >> the statistics. Are you suggesting the later?
>> >> >> >>
>> >> >> >
>> >> >> > No. As I wrote.
>> >> >> > - don't modify codes around FILE_MAPPED in this series.
>> >> >> > - add a new functions for new statistics
>> >> >> > Then,
>> >> >> > - think about clean up later, after we confirm all things work as expected.
>> >> >>
>> >> >> I have ported Andrea Righi's memcg dirty page accounting patches to latest
>> >> >> mmtom-2010-04-05-16-09. In doing so I have to address this locking issue. Does
>> >> >> the following look good? I will (of course) submit the entire patch for review,
>> >> >> but I wanted make sure I was aiming in the right direction.
>> >> >>
>> >> >> void mem_cgroup_update_page_stat(struct page *page,
>> >> >> enum mem_cgroup_write_page_stat_item idx, bool charge)
>> >> >> {
>> >> >> static int seq;
>> >> >> struct page_cgroup *pc;
>> >> >>
>> >> >> if (mem_cgroup_disabled())
>> >> >> return;
>> >> >> pc = lookup_page_cgroup(page);
>> >> >> if (!pc || mem_cgroup_is_root(pc->mem_cgroup))
>> >> >> return;
>> >> >>
>> >> >> /*
>> >> >> * This routine does not disable irq when updating stats. So it is
>> >> >> * possible that a stat update from within interrupt routine, could
>> >> >> * deadlock. Use trylock_page_cgroup() to avoid such deadlock. This
>> >> >> * makes the memcg counters fuzzy. More complicated, or lower
>> >> >> * performing locking solutions avoid this fuzziness, but are not
>> >> >> * currently needed.
>> >> >> */
>> >> >> if (irqs_disabled()) {
>> >> > ^^^^^^^^^
>> >> > Or may be in_interrupt()?
>> >>
>> >> Good catch. I will replace irqs_disabled() with in_interrupt().
>> >>
>> > I think you should check both. __remove_from_page_cache(), which will update
>> > DIRTY, is called with irq disabled(iow, under mapping->tree_lock) but not in
>> > interrupt context.
>>
>> The only reason to use trylock in this case is to prevent deadlock
>> when running in a context that may have preempted or interrupted a
>> routine that already holds the bit locked. In the
>> __remove_from_page_cache() irqs are disabled, but that does not imply
>> that a routine holding the spinlock has been preempted. When the bit
>> is locked, preemption is disabled. The only way to interrupt a holder
>> of the bit for an interrupt to occur (I/O, timer, etc). So I think
>> that in_interrupt() is sufficient. Am I missing something?
>>
> IIUC, it's would be enough to prevent deadlock where one CPU tries to acquire
> the same page cgroup lock. But there is still some possibility where 2 CPUs
> can cause dead lock each other(please see the commit e767e056).
> IOW, my point is "don't call lock_page_cgroup() under mapping->tree_lock".
I see your point. Thank you for explaining.
--
Greg
--
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>
next prev parent reply other threads:[~2010-04-15 6:39 UTC|newest]
Thread overview: 69+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-03-14 23:26 [PATCH -mmotm 0/5] memcg: per cgroup dirty limit (v7) Andrea Righi
2010-03-14 23:26 ` [PATCH -mmotm 1/5] memcg: disable irq at page cgroup lock Andrea Righi
2010-03-15 0:06 ` KAMEZAWA Hiroyuki
2010-03-15 10:00 ` Andrea Righi
2010-03-17 7:04 ` Balbir Singh
2010-03-17 11:58 ` Balbir Singh
2010-03-17 23:54 ` KAMEZAWA Hiroyuki
2010-03-18 0:45 ` KAMEZAWA Hiroyuki
2010-03-18 2:16 ` Daisuke Nishimura
2010-03-18 2:58 ` KAMEZAWA Hiroyuki
2010-03-18 5:12 ` Balbir Singh
2010-03-18 4:19 ` Balbir Singh
2010-03-18 4:21 ` KAMEZAWA Hiroyuki
2010-03-18 6:25 ` Balbir Singh
2010-03-18 4:35 ` KAMEZAWA Hiroyuki
2010-03-18 16:28 ` Balbir Singh
2010-03-19 1:23 ` KAMEZAWA Hiroyuki
2010-03-19 2:40 ` Balbir Singh
2010-03-19 3:00 ` KAMEZAWA Hiroyuki
[not found] ` <xr93hbnepmj6.fsf@ninji.mtv.corp.google.com>
2010-04-14 6:55 ` Greg Thelen
2010-04-14 9:29 ` KAMEZAWA Hiroyuki
2010-04-14 14:04 ` Vivek Goyal
2010-04-14 19:31 ` Greg Thelen
2010-04-15 0:14 ` KAMEZAWA Hiroyuki
2010-04-14 16:22 ` Greg Thelen
2010-04-15 0:22 ` KAMEZAWA Hiroyuki
2010-04-14 14:05 ` Vivek Goyal
2010-04-14 20:14 ` Greg Thelen
2010-04-15 2:40 ` Daisuke Nishimura
2010-04-15 4:48 ` Greg Thelen
2010-04-15 6:21 ` Daisuke Nishimura
2010-04-15 6:38 ` Greg Thelen [this message]
2010-04-15 6:54 ` KAMEZAWA Hiroyuki
2010-04-23 20:17 ` Greg Thelen
2010-04-23 20:54 ` Peter Zijlstra
2010-04-24 15:53 ` Greg Thelen
2010-04-23 20:57 ` Peter Zijlstra
2010-04-24 2:22 ` KAMEZAWA Hiroyuki
2010-04-23 21:19 ` Peter Zijlstra
2010-04-24 2:19 ` KAMEZAWA Hiroyuki
2010-04-14 14:44 ` Balbir Singh
2010-03-14 23:26 ` [PATCH -mmotm 2/5] memcg: dirty memory documentation Andrea Righi
2010-03-16 7:41 ` Daisuke Nishimura
2010-03-17 17:48 ` Greg Thelen
2010-03-17 19:02 ` Balbir Singh
2010-03-17 22:43 ` Andrea Righi
2010-03-14 23:26 ` [PATCH -mmotm 3/5] page_cgroup: introduce file cache flags Andrea Righi
2010-03-14 23:26 ` [PATCH -mmotm 4/5] memcg: dirty pages accounting and limiting infrastructure Andrea Righi
2010-03-15 2:26 ` KAMEZAWA Hiroyuki
2010-03-16 2:32 ` Daisuke Nishimura
2010-03-16 14:11 ` Vivek Goyal
2010-03-16 15:09 ` Daisuke Nishimura
2010-03-17 22:37 ` Andrea Righi
2010-03-17 22:52 ` Andrea Righi
2010-03-18 6:48 ` Greg Thelen
2010-03-14 23:26 ` [PATCH -mmotm 5/5] memcg: dirty pages instrumentation Andrea Righi
2010-03-15 2:31 ` KAMEZAWA Hiroyuki
2010-03-15 2:36 ` [PATCH -mmotm 0/5] memcg: per cgroup dirty limit (v7) KAMEZAWA Hiroyuki
2010-03-15 10:02 ` Andrea Righi
2010-03-15 17:12 ` Vivek Goyal
2010-03-15 17:19 ` Vivek Goyal
2010-03-17 11:54 ` Balbir Singh
2010-03-17 13:34 ` Vivek Goyal
2010-03-17 18:53 ` Balbir Singh
2010-03-17 19:15 ` Vivek Goyal
2010-03-17 19:17 ` Balbir Singh
2010-03-17 19:48 ` Vivek Goyal
2010-03-17 6:44 ` Balbir Singh
-- strict thread matches above, loose matches on Subject: below --
2010-03-09 23:00 [PATCH -mmotm 0/5] memcg: per cgroup dirty limit (v6) Andrea Righi
2010-03-09 23:00 ` [PATCH -mmotm 1/5] memcg: disable irq at page cgroup lock Andrea Righi
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=i2k49b004811004142338je898a741q78a12c902c52724a@mail.gmail.com \
--to=gthelen@google.com \
--cc=akpm@linux-foundation.org \
--cc=arighi@develer.com \
--cc=balbir@linux.vnet.ibm.com \
--cc=containers@lists.linux-foundation.org \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=kirill@shutemov.name \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=nishimura@mxp.nes.nec.co.jp \
--cc=peterz@infradead.org \
--cc=suleiman@google.com \
--cc=trond.myklebust@fys.uio.no \
--cc=vgoyal@redhat.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).