From: Christoph Hellwig <hch@infradead.org>
To: Mikulas Patocka <mpatocka@redhat.com>
Cc: axboe@kernel.dk, dm-devel <dm-devel@redhat.com>,
Alasdair G Kergon <agk@redhat.com>,
linux-kernel@vger.kernel.org
Subject: Re: Queue upcall locking (was: [RFC][PATCH] fix dm_any_congested() to properly sync up with suspend code path)
Date: Mon, 10 Nov 2008 08:54:01 -0500 [thread overview]
Message-ID: <20081110135401.GA11606@infradead.org> (raw)
In-Reply-To: <Pine.LNX.4.64.0811100746180.24696@hs20-bc2-1.build.redhat.com>
On Mon, Nov 10, 2008 at 08:11:51AM -0500, Mikulas Patocka wrote:
> For upstream Linux developers: you are holding a spinlock and calling
> bdi*_congested functions that can take indefinite amount of time (there
> are even users reporting having 50 disks in one logical volume or so). I
> think it would be good to move these calls out of spinlocks.
Umm, they shouldn't block that long, as that completely defeats their
purpose. These functions are mostly used to avoid throwing more I/O at
a congested device if pdflush could do more useful things instead. But
if it blocks in those functions anyway we wouldn't have to bother using
them. Do you have more details about the uses cases when this happens
and where the routines spend so much time?
WARNING: multiple messages have this Message-ID (diff)
From: Christoph Hellwig <hch@infradead.org>
To: Mikulas Patocka <mpatocka@redhat.com>
Cc: Chandra Seetharaman <sekharan@us.ibm.com>,
Alasdair G Kergon <agk@redhat.com>,
dm-devel <dm-devel@redhat.com>,
linux-kernel@vger.kernel.org, axboe@kernel.dk
Subject: Re: Queue upcall locking (was: [dm-devel] [RFC][PATCH] fix dm_any_congested() to properly sync up with suspend code path)
Date: Mon, 10 Nov 2008 08:54:01 -0500 [thread overview]
Message-ID: <20081110135401.GA11606@infradead.org> (raw)
In-Reply-To: <Pine.LNX.4.64.0811100746180.24696@hs20-bc2-1.build.redhat.com>
On Mon, Nov 10, 2008 at 08:11:51AM -0500, Mikulas Patocka wrote:
> For upstream Linux developers: you are holding a spinlock and calling
> bdi*_congested functions that can take indefinite amount of time (there
> are even users reporting having 50 disks in one logical volume or so). I
> think it would be good to move these calls out of spinlocks.
Umm, they shouldn't block that long, as that completely defeats their
purpose. These functions are mostly used to avoid throwing more I/O at
a congested device if pdflush could do more useful things instead. But
if it blocks in those functions anyway we wouldn't have to bother using
them. Do you have more details about the uses cases when this happens
and where the routines spend so much time?
next prev parent reply other threads:[~2008-11-10 13:54 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-11-06 4:00 [RFC][PATCH] fix dm_any_congested() to properly sync up with suspend code path Chandra Seetharaman
2008-11-10 13:11 ` Queue upcall locking (was: [RFC][PATCH] fix dm_any_congested() to properly sync up with suspend code path) Mikulas Patocka
2008-11-10 13:11 ` Queue upcall locking (was: [dm-devel] " Mikulas Patocka
2008-11-10 13:54 ` Christoph Hellwig [this message]
2008-11-10 13:54 ` Christoph Hellwig
2008-11-10 14:19 ` Queue upcall locking (was: " Mikulas Patocka
2008-11-10 14:19 ` Queue upcall locking (was: [dm-devel] " Mikulas Patocka
2008-11-10 14:26 ` Peter Zijlstra
2008-11-10 14:32 ` Queue upcall locking (was: " Mikulas Patocka
2008-11-10 14:32 ` Queue upcall locking (was: [dm-devel] " Mikulas Patocka
2008-11-10 14:46 ` Peter Zijlstra
2008-11-10 14:55 ` Queue upcall locking (was: " Mikulas Patocka
2008-11-10 14:55 ` Queue upcall locking (was: [dm-devel] " Mikulas Patocka
2008-11-10 14:35 ` Queue upcall locking (was: " Alasdair G Kergon
2008-11-10 14:35 ` Queue upcall locking (was: [dm-devel] " Alasdair G Kergon
2008-11-11 18:18 ` Queue upcall locking (was: " Christoph Hellwig
2008-11-11 18:18 ` Queue upcall locking (was: [dm-devel] " Christoph Hellwig
2008-11-11 22:08 ` Queue upcall locking (was: " Mikulas Patocka
2008-11-11 22:08 ` Queue upcall locking (was: [dm-devel] " Mikulas Patocka
2008-11-10 14:27 ` Queue upcall locking (was: " Alasdair G Kergon
2008-11-10 14:27 ` Queue upcall locking (was: [dm-devel] " Alasdair G Kergon
2008-11-10 14:40 ` Queue upcall locking (was: " Mikulas Patocka
2008-11-10 14:40 ` Queue upcall locking (was: [dm-devel] " Mikulas Patocka
2008-11-11 18:22 ` Christoph Hellwig
2008-11-11 18:21 ` Christoph Hellwig
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=20081110135401.GA11606@infradead.org \
--to=hch@infradead.org \
--cc=agk@redhat.com \
--cc=axboe@kernel.dk \
--cc=dm-devel@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mpatocka@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 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.