From: Nick Piggin <npiggin@suse.de>
To: Rik van Riel <riel@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>,
Andrea Arcangeli <aarcange@redhat.com>,
Mel Gorman <mel@csn.ul.ie>,
linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-mm@kvack.org, Dave Chinner <david@fromorbit.com>,
Chris Mason <chris.mason@oracle.com>
Subject: Re: [RFC PATCH 0/6] Do not call ->writepage[s] from direct reclaim and use a_ops->writepages() where possible
Date: Wed, 16 Jun 2010 17:57:23 +1000 [thread overview]
Message-ID: <20100616075723.GT6138@laptop> (raw)
In-Reply-To: <4C17D0C5.9030203@redhat.com>
On Tue, Jun 15, 2010 at 03:13:09PM -0400, Rik van Riel wrote:
> On 06/15/2010 12:54 PM, Christoph Hellwig wrote:
> >On Tue, Jun 15, 2010 at 12:49:49PM -0400, Rik van Riel wrote:
> >>This is already in a filesystem. Why does ->writepage get
> >>called a second time? Shouldn't this have a gfp_mask
> >>without __GFP_FS set?
> >
> >Why would it? GFP_NOFS is not for all filesystem code, but only for
> >code where we can't re-enter the filesystem due to deadlock potential.
>
> Why? How about because you know the stack is not big enough
> to have the XFS call path on it twice? :)
>
> Isn't the whole purpose of this patch series to prevent writepage
> from being called by the VM, when invoked from a deep callstack
> like xfs writepage?
>
> That sounds a lot like simply wanting to not have GFP_FS...
buffered write path uses __GFP_FS by design because huge amounts
of (dirty) memory can be allocated in doing pagecache writes. If
would be nasty if that was not allowed to wait for filesystem
activity.
WARNING: multiple messages have this Message-ID (diff)
From: Nick Piggin <npiggin@suse.de>
To: Rik van Riel <riel@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>,
Andrea Arcangeli <aarcange@redhat.com>,
Mel Gorman <mel@csn.ul.ie>,
linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-mm@kvack.org, Dave Chinner <david@fromorbit.com>,
Chris Mason <chris.mason@oracle.com>
Subject: Re: [RFC PATCH 0/6] Do not call ->writepage[s] from direct reclaim and use a_ops->writepages() where possible
Date: Wed, 16 Jun 2010 17:57:23 +1000 [thread overview]
Message-ID: <20100616075723.GT6138@laptop> (raw)
In-Reply-To: <4C17D0C5.9030203@redhat.com>
On Tue, Jun 15, 2010 at 03:13:09PM -0400, Rik van Riel wrote:
> On 06/15/2010 12:54 PM, Christoph Hellwig wrote:
> >On Tue, Jun 15, 2010 at 12:49:49PM -0400, Rik van Riel wrote:
> >>This is already in a filesystem. Why does ->writepage get
> >>called a second time? Shouldn't this have a gfp_mask
> >>without __GFP_FS set?
> >
> >Why would it? GFP_NOFS is not for all filesystem code, but only for
> >code where we can't re-enter the filesystem due to deadlock potential.
>
> Why? How about because you know the stack is not big enough
> to have the XFS call path on it twice? :)
>
> Isn't the whole purpose of this patch series to prevent writepage
> from being called by the VM, when invoked from a deep callstack
> like xfs writepage?
>
> That sounds a lot like simply wanting to not have GFP_FS...
buffered write path uses __GFP_FS by design because huge amounts
of (dirty) memory can be allocated in doing pagecache writes. If
would be nasty if that was not allowed to wait for filesystem
activity.
--
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:[~2010-06-16 7:57 UTC|newest]
Thread overview: 134+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-06-08 9:02 [RFC PATCH 0/6] Do not call ->writepage[s] from direct reclaim and use a_ops->writepages() where possible Mel Gorman
2010-06-08 9:02 ` Mel Gorman
2010-06-08 9:02 ` [PATCH 1/6] tracing, vmscan: Add trace events for kswapd wakeup, sleeping and direct reclaim Mel Gorman
2010-06-08 9:02 ` Mel Gorman
2010-06-08 9:02 ` [PATCH 2/6] tracing, vmscan: Add trace events for LRU page isolation Mel Gorman
2010-06-08 9:02 ` Mel Gorman
2010-06-08 9:02 ` [PATCH 3/6] tracing, vmscan: Add trace event when a page is written Mel Gorman
2010-06-08 9:02 ` Mel Gorman
2010-06-08 9:02 ` [PATCH 4/6] tracing, vmscan: Add a postprocessing script for reclaim-related ftrace events Mel Gorman
2010-06-08 9:02 ` Mel Gorman
2010-06-08 9:02 ` [PATCH 5/6] vmscan: Write out ranges of pages contiguous to the inode where possible Mel Gorman
2010-06-08 9:02 ` Mel Gorman
2010-06-11 6:10 ` Andrew Morton
2010-06-11 6:10 ` Andrew Morton
2010-06-11 12:49 ` Mel Gorman
2010-06-11 12:49 ` Mel Gorman
2010-06-11 19:07 ` Andrew Morton
2010-06-11 19:07 ` Andrew Morton
2010-06-11 20:44 ` Mel Gorman
2010-06-11 20:44 ` Mel Gorman
2010-06-11 21:33 ` Andrew Morton
2010-06-11 21:33 ` Andrew Morton
2010-06-12 0:17 ` Mel Gorman
2010-06-12 0:17 ` Mel Gorman
2010-06-11 16:27 ` Christoph Hellwig
2010-06-11 16:27 ` Christoph Hellwig
2010-06-08 9:02 ` [PATCH 6/6] vmscan: Do not writeback pages in direct reclaim Mel Gorman
2010-06-08 9:02 ` Mel Gorman
2010-06-11 6:17 ` Andrew Morton
2010-06-11 6:17 ` Andrew Morton
2010-06-11 12:54 ` Mel Gorman
2010-06-11 12:54 ` Mel Gorman
2010-06-11 16:25 ` Christoph Hellwig
2010-06-11 16:25 ` Christoph Hellwig
2010-06-11 17:43 ` Andrew Morton
2010-06-11 17:43 ` Andrew Morton
2010-06-11 17:49 ` Christoph Hellwig
2010-06-11 17:49 ` Christoph Hellwig
2010-06-11 18:13 ` Mel Gorman
2010-06-11 18:13 ` Mel Gorman
2010-06-08 9:08 ` [RFC PATCH 0/6] Do not call ->writepage[s] from direct reclaim and use a_ops->writepages() where possible Christoph Hellwig
2010-06-08 9:08 ` Christoph Hellwig
2010-06-08 9:28 ` Mel Gorman
2010-06-08 9:28 ` Mel Gorman
2010-06-11 16:29 ` Christoph Hellwig
2010-06-11 16:29 ` Christoph Hellwig
2010-06-11 18:15 ` Mel Gorman
2010-06-11 18:15 ` Mel Gorman
2010-06-11 19:12 ` Chris Mason
2010-06-11 19:12 ` Chris Mason
2010-06-09 2:52 ` KAMEZAWA Hiroyuki
2010-06-09 2:52 ` KAMEZAWA Hiroyuki
2010-06-09 9:52 ` Mel Gorman
2010-06-09 9:52 ` Mel Gorman
2010-06-10 0:38 ` KAMEZAWA Hiroyuki
2010-06-10 0:38 ` KAMEZAWA Hiroyuki
2010-06-10 1:10 ` Mel Gorman
2010-06-10 1:10 ` Mel Gorman
2010-06-10 1:29 ` KAMEZAWA Hiroyuki
2010-06-10 1:29 ` KAMEZAWA Hiroyuki
2010-06-11 5:57 ` Andrew Morton
2010-06-11 5:57 ` Andrew Morton
2010-06-11 12:33 ` Mel Gorman
2010-06-11 12:33 ` Mel Gorman
2010-06-11 16:30 ` Christoph Hellwig
2010-06-11 16:30 ` Christoph Hellwig
2010-06-11 18:17 ` Mel Gorman
2010-06-11 18:17 ` Mel Gorman
2010-06-15 14:00 ` Andrea Arcangeli
2010-06-15 14:00 ` Andrea Arcangeli
2010-06-15 14:11 ` Christoph Hellwig
2010-06-15 14:11 ` Christoph Hellwig
2010-06-15 14:22 ` Andrea Arcangeli
2010-06-15 14:22 ` Andrea Arcangeli
2010-06-15 14:43 ` Christoph Hellwig
2010-06-15 14:43 ` Christoph Hellwig
2010-06-15 15:08 ` Andrea Arcangeli
2010-06-15 15:08 ` Andrea Arcangeli
2010-06-15 15:25 ` Christoph Hellwig
2010-06-15 15:25 ` Christoph Hellwig
2010-06-15 15:45 ` Andrea Arcangeli
2010-06-15 15:45 ` Andrea Arcangeli
2010-06-15 16:26 ` Christoph Hellwig
2010-06-15 16:26 ` Christoph Hellwig
2010-06-15 16:31 ` Andrea Arcangeli
2010-06-15 16:31 ` Andrea Arcangeli
2010-06-15 16:49 ` Rik van Riel
2010-06-15 16:49 ` Rik van Riel
2010-06-15 16:54 ` Christoph Hellwig
2010-06-15 16:54 ` Christoph Hellwig
2010-06-15 19:13 ` Rik van Riel
2010-06-15 19:13 ` Rik van Riel
2010-06-15 19:17 ` Christoph Hellwig
2010-06-15 19:17 ` Christoph Hellwig
2010-06-15 19:44 ` Chris Mason
2010-06-15 19:44 ` Chris Mason
2010-06-16 7:57 ` Nick Piggin [this message]
2010-06-16 7:57 ` Nick Piggin
2010-06-16 16:59 ` Rik van Riel
2010-06-16 16:59 ` Rik van Riel
2010-06-16 17:04 ` Andrea Arcangeli
2010-06-16 17:04 ` Andrea Arcangeli
2010-06-15 16:54 ` Nick Piggin
2010-06-15 16:54 ` Nick Piggin
2010-06-15 15:38 ` Mel Gorman
2010-06-15 15:38 ` Mel Gorman
2010-06-15 16:14 ` Andrea Arcangeli
2010-06-15 16:14 ` Andrea Arcangeli
2010-06-15 16:22 ` Christoph Hellwig
2010-06-15 16:22 ` Christoph Hellwig
2010-06-15 16:30 ` Mel Gorman
2010-06-15 16:30 ` Mel Gorman
2010-06-15 16:34 ` Mel Gorman
2010-06-15 16:34 ` Mel Gorman
2010-06-15 16:54 ` Andrea Arcangeli
2010-06-15 16:54 ` Andrea Arcangeli
2010-06-15 16:35 ` Christoph Hellwig
2010-06-15 16:35 ` Christoph Hellwig
2010-06-15 16:37 ` Andrea Arcangeli
2010-06-15 16:37 ` Andrea Arcangeli
2010-06-15 17:43 ` Christoph Hellwig
2010-06-15 17:43 ` Christoph Hellwig
2010-06-15 16:45 ` Christoph Hellwig
2010-06-15 16:45 ` Christoph Hellwig
2010-06-15 14:51 ` Mel Gorman
2010-06-15 14:51 ` Mel Gorman
2010-06-15 14:55 ` Rik van Riel
2010-06-15 14:55 ` Rik van Riel
2010-06-15 15:08 ` Nick Piggin
2010-06-15 15:08 ` Nick Piggin
2010-06-15 15:10 ` Mel Gorman
2010-06-15 15:10 ` Mel Gorman
2010-06-15 16:28 ` Andrea Arcangeli
2010-06-15 16:28 ` Andrea Arcangeli
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=20100616075723.GT6138@laptop \
--to=npiggin@suse.de \
--cc=aarcange@redhat.com \
--cc=chris.mason@oracle.com \
--cc=david@fromorbit.com \
--cc=hch@infradead.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mel@csn.ul.ie \
--cc=riel@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.