From: Alasdair G Kergon <agk@redhat.com>
To: Mikulas Patocka <mpatocka@redhat.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>,
Andrew Morton <akpm@linux-foundation.org>,
linux-kernel@vger.kernel.org, agk@redhat.com, mbroz@redhat.com,
chris@arachsys.com
Subject: Re: [PATCH] Memory management livelock
Date: Fri, 3 Oct 2008 15:50:29 +0100 [thread overview]
Message-ID: <20081003145029.GE18542@agk.fab.redhat.com> (raw)
In-Reply-To: <Pine.LNX.4.64.0810030941480.27239@hs20-bc2-1.build.redhat.com>
On Fri, Oct 03, 2008 at 09:50:17AM -0400, Mikulas Patocka wrote:
> > > LVM does (that is where the bug was discovered). Basically, it scans all
> > > the block devices with direct IO and if someone else does buffered IO on
> > > any device simultaneously, it locks up.
> > Scans all block devices with direct IO? Hmm, why, I wonder? Should
> > really consider using buffered (posix_fadvise to readahead/dropbehind).
> LVM must not allocate any memory when doing IO because it suspends the
> block device and memory allocation could trigger writeback on the
> suspended device and deadlock.
> So it preallocates heap and stack, mlockall()s itself and does direct IO.
True, but unrelated to the scanning, which LVM performs *prior* to
entering such a state.
We use direct IO while scanning because it's essential all nodes in a
cluster see the same updated version of the data after any node updated
it.
Alasdair
--
agk@redhat.com
next prev parent reply other threads:[~2008-10-03 14:50 UTC|newest]
Thread overview: 67+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20080911101616.GA24064@agk.fab.redhat.com>
2008-09-22 21:10 ` [PATCH] Memory management livelock Mikulas Patocka
2008-09-23 0:48 ` Andrew Morton
2008-09-23 22:34 ` Mikulas Patocka
2008-09-23 22:49 ` Andrew Morton
2008-09-23 23:11 ` Mikulas Patocka
2008-09-23 23:46 ` Andrew Morton
2008-09-24 18:50 ` Mikulas Patocka
2008-09-24 18:51 ` [PATCH 1/3] " Mikulas Patocka
2008-09-24 18:52 ` [PATCH 2/3] " Mikulas Patocka
2008-10-02 5:54 ` Andrew Morton
2008-10-05 22:11 ` RFC: one-bit mutexes (was: Re: [PATCH 2/3] Memory management livelock) Mikulas Patocka
2008-10-11 12:06 ` Nick Piggin
2008-10-20 20:14 ` Mikulas Patocka
2008-10-21 1:51 ` Nick Piggin
2008-10-05 22:14 ` [PATCH 1/3] bit mutexes Mikulas Patocka
2008-10-05 22:14 ` [PATCH 2/3] Fix fsync livelock Mikulas Patocka
2008-10-05 22:33 ` Arjan van de Ven
2008-10-05 23:02 ` Mikulas Patocka
2008-10-05 23:07 ` Arjan van de Ven
2008-10-05 23:18 ` Mikulas Patocka
2008-10-05 23:28 ` Arjan van de Ven
2008-10-06 0:01 ` Mikulas Patocka
2008-10-06 0:30 ` Arjan van de Ven
2008-10-06 3:30 ` Mikulas Patocka
2008-10-06 4:20 ` Arjan van de Ven
2008-10-06 13:00 ` Mikulas Patocka
2008-10-06 13:50 ` Arjan van de Ven
2008-10-06 20:44 ` Mikulas Patocka
2008-10-08 10:56 ` Pavel Machek
2008-10-06 2:51 ` Dave Chinner
2008-10-05 22:16 ` [PATCH 3/3] Fix fsync-vs-write misbehavior Mikulas Patocka
2008-10-09 1:12 ` [PATCH] documentation: explain memory barriers Randy Dunlap
2008-10-09 1:17 ` Chris Snook
2008-10-09 1:31 ` Andrew Morton
2008-10-09 5:51 ` Chris Snook
2008-10-09 9:58 ` Ben Hutchings
2008-10-09 21:27 ` Nick Piggin
2008-10-09 17:29 ` Nick Piggin
2008-10-09 1:50 ` Valdis.Kletnieks
2008-10-09 17:35 ` Nick Piggin
2008-10-09 6:52 ` Valdis.Kletnieks
2008-09-24 18:53 ` [PATCH 3/3] Memory management livelock Mikulas Patocka
2008-10-03 2:32 ` [PATCH] " Nick Piggin
2008-10-03 2:40 ` Andrew Morton
2008-10-03 2:59 ` Nick Piggin
2008-10-03 3:14 ` Andrew Morton
2008-10-03 3:47 ` Nick Piggin
2008-10-03 3:56 ` Andrew Morton
2008-10-03 4:07 ` Nick Piggin
2008-10-03 4:17 ` Andrew Morton
2008-10-03 4:29 ` Nick Piggin
2008-10-03 11:43 ` Mikulas Patocka
2008-10-03 12:27 ` Nick Piggin
2008-10-03 13:53 ` Mikulas Patocka
2008-10-03 2:54 ` Nick Piggin
2008-10-03 11:26 ` Mikulas Patocka
2008-10-03 12:31 ` Nick Piggin
2008-10-03 13:50 ` Mikulas Patocka
2008-10-03 14:50 ` Alasdair G Kergon [this message]
2008-10-03 14:36 ` Alasdair G Kergon
2008-10-03 15:52 ` application syncing options (was Re: [PATCH] Memory management livelock) david
2008-10-06 0:04 ` Mikulas Patocka
2008-10-06 0:19 ` david
2008-10-06 3:42 ` Mikulas Patocka
2008-10-07 3:37 ` david
2008-10-07 15:44 ` Mikulas Patocka
2008-10-07 17:16 ` david
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=20081003145029.GE18542@agk.fab.redhat.com \
--to=agk@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=chris@arachsys.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mbroz@redhat.com \
--cc=mpatocka@redhat.com \
--cc=nickpiggin@yahoo.com.au \
/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.