From: Dave Chinner <david@fromorbit.com>
To: Christoph Hellwig <hch@lst.de>
Cc: linux-fsdevel@vger.kernel.org, xfs@oss.sgi.com,
linux-ext4@vger.kernel.org, ocfs2-devel@oss.oracle.com
Subject: Re: [PATCH 2/3] xfs: don't use ioends for direct write completions
Date: Mon, 8 Feb 2016 12:00:26 +1100 [thread overview]
Message-ID: <20160208010026.GL31407@dastard> (raw)
In-Reply-To: <1454524816-11392-3-git-send-email-hch@lst.de>
On Wed, Feb 03, 2016 at 07:40:15PM +0100, Christoph Hellwig wrote:
> We only need to communicate two bits of information to the direct I/O
> completion handler:
>
> (1) do we need to convert any unwritten extents in the range
> (2) do we need to check if we need to update the inode size based
> on the range passed to the completion handler
>
> We can use the private data passed to the get_block handler and the
> completion handler as a simple bitmask to communicate this information
> instead of the current complicated infrastructure reusing the ioends
> from the buffer I/O path, and thus avoiding a memory allocation and
> a context switch for any non-trivial direct write. As a nice side
> effect we also decouple the direct I/O path implementation from that
> of the buffered I/O path.
>
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> Reviewed-by: Brian Foster <bfoster@redhat.com>
This change is now dependent on the preceeding direct IO API
changes. Do I a) take the DIO API change through the XFS tree, or
b) use the older version of the patch that didn't have this
dependency and let somebody else deal with the API change and merge
issues?
I'm happy to take the DIO API change through the XFS tree, if that's
the fastest/easiest way to get the necessary DIO subsystem fixes
into the mainline tree for XFS. As such, the for-next tree that I'm
building right now will include the DIO API change patch....
-Dave.
--
Dave Chinner
david@fromorbit.com
WARNING: multiple messages have this Message-ID (diff)
From: Dave Chinner <david@fromorbit.com>
To: Christoph Hellwig <hch@lst.de>
Cc: linux-fsdevel@vger.kernel.org, linux-ext4@vger.kernel.org,
ocfs2-devel@oss.oracle.com, xfs@oss.sgi.com
Subject: Re: [PATCH 2/3] xfs: don't use ioends for direct write completions
Date: Mon, 8 Feb 2016 12:00:26 +1100 [thread overview]
Message-ID: <20160208010026.GL31407@dastard> (raw)
In-Reply-To: <1454524816-11392-3-git-send-email-hch@lst.de>
On Wed, Feb 03, 2016 at 07:40:15PM +0100, Christoph Hellwig wrote:
> We only need to communicate two bits of information to the direct I/O
> completion handler:
>
> (1) do we need to convert any unwritten extents in the range
> (2) do we need to check if we need to update the inode size based
> on the range passed to the completion handler
>
> We can use the private data passed to the get_block handler and the
> completion handler as a simple bitmask to communicate this information
> instead of the current complicated infrastructure reusing the ioends
> from the buffer I/O path, and thus avoiding a memory allocation and
> a context switch for any non-trivial direct write. As a nice side
> effect we also decouple the direct I/O path implementation from that
> of the buffered I/O path.
>
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> Reviewed-by: Brian Foster <bfoster@redhat.com>
This change is now dependent on the preceeding direct IO API
changes. Do I a) take the DIO API change through the XFS tree, or
b) use the older version of the patch that didn't have this
dependency and let somebody else deal with the API change and merge
issues?
I'm happy to take the DIO API change through the XFS tree, if that's
the fastest/easiest way to get the necessary DIO subsystem fixes
into the mainline tree for XFS. As such, the for-next tree that I'm
building right now will include the DIO API change patch....
-Dave.
--
Dave Chinner
david@fromorbit.com
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
WARNING: multiple messages have this Message-ID (diff)
From: Dave Chinner <david@fromorbit.com>
To: Christoph Hellwig <hch@lst.de>
Cc: linux-fsdevel@vger.kernel.org, xfs@oss.sgi.com,
linux-ext4@vger.kernel.org, ocfs2-devel@oss.oracle.com
Subject: [Ocfs2-devel] [PATCH 2/3] xfs: don't use ioends for direct write completions
Date: Mon, 8 Feb 2016 12:00:26 +1100 [thread overview]
Message-ID: <20160208010026.GL31407@dastard> (raw)
In-Reply-To: <1454524816-11392-3-git-send-email-hch@lst.de>
On Wed, Feb 03, 2016 at 07:40:15PM +0100, Christoph Hellwig wrote:
> We only need to communicate two bits of information to the direct I/O
> completion handler:
>
> (1) do we need to convert any unwritten extents in the range
> (2) do we need to check if we need to update the inode size based
> on the range passed to the completion handler
>
> We can use the private data passed to the get_block handler and the
> completion handler as a simple bitmask to communicate this information
> instead of the current complicated infrastructure reusing the ioends
> from the buffer I/O path, and thus avoiding a memory allocation and
> a context switch for any non-trivial direct write. As a nice side
> effect we also decouple the direct I/O path implementation from that
> of the buffered I/O path.
>
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> Reviewed-by: Brian Foster <bfoster@redhat.com>
This change is now dependent on the preceeding direct IO API
changes. Do I a) take the DIO API change through the XFS tree, or
b) use the older version of the patch that didn't have this
dependency and let somebody else deal with the API change and merge
issues?
I'm happy to take the DIO API change through the XFS tree, if that's
the fastest/easiest way to get the necessary DIO subsystem fixes
into the mainline tree for XFS. As such, the for-next tree that I'm
building right now will include the DIO API change patch....
-Dave.
--
Dave Chinner
david at fromorbit.com
next prev parent reply other threads:[~2016-02-08 1:00 UTC|newest]
Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-02-03 18:40 vfs/xfs: directio updates to ease COW handling V2 Christoph Hellwig
2016-02-03 18:40 ` [Ocfs2-devel] " Christoph Hellwig
2016-02-03 18:40 ` Christoph Hellwig
2016-02-03 18:40 ` [PATCH 1/3] direct-io: always call ->end_io if non-NULL Christoph Hellwig
2016-02-03 18:40 ` [Ocfs2-devel] " Christoph Hellwig
2016-02-03 18:40 ` Christoph Hellwig
2016-02-03 19:55 ` Darrick J. Wong
2016-02-03 19:55 ` [Ocfs2-devel] " Darrick J. Wong
2016-02-03 19:55 ` Darrick J. Wong
2016-02-04 7:14 ` Christoph Hellwig
2016-02-04 7:14 ` [Ocfs2-devel] " Christoph Hellwig
2016-02-04 7:14 ` Christoph Hellwig
2016-02-04 8:17 ` Darrick J. Wong
2016-02-04 8:17 ` [Ocfs2-devel] " Darrick J. Wong
2016-02-04 8:17 ` Darrick J. Wong
2016-02-03 18:40 ` [PATCH 2/3] xfs: don't use ioends for direct write completions Christoph Hellwig
2016-02-03 18:40 ` [Ocfs2-devel] " Christoph Hellwig
2016-02-03 18:40 ` Christoph Hellwig
2016-02-05 21:57 ` Darrick J. Wong
2016-02-05 21:57 ` [Ocfs2-devel] " Darrick J. Wong
2016-02-05 21:57 ` Darrick J. Wong
2016-02-05 22:36 ` Dave Chinner
2016-02-05 22:36 ` [Ocfs2-devel] " Dave Chinner
2016-02-05 22:36 ` Dave Chinner
2016-02-08 1:00 ` Dave Chinner [this message]
2016-02-08 1:00 ` [Ocfs2-devel] " Dave Chinner
2016-02-08 1:00 ` Dave Chinner
2016-02-08 6:17 ` Dave Chinner
2016-02-08 6:17 ` [Ocfs2-devel] " Dave Chinner
2016-02-08 6:17 ` Dave Chinner
2016-02-08 7:31 ` Christoph Hellwig
2016-02-08 7:31 ` [Ocfs2-devel] " Christoph Hellwig
2016-02-08 7:31 ` Christoph Hellwig
2016-02-08 9:16 ` Dave Chinner
2016-02-08 9:16 ` [Ocfs2-devel] " Dave Chinner
2016-02-08 9:16 ` Dave Chinner
2016-02-08 9:22 ` Christoph Hellwig
2016-02-08 9:22 ` [Ocfs2-devel] " Christoph Hellwig
2016-02-08 9:22 ` Christoph Hellwig
2016-02-03 18:40 ` [PATCH 3/3] xfs: fold xfs_vm_do_dio into xfs_vm_direct_IO Christoph Hellwig
2016-02-03 18:40 ` [Ocfs2-devel] " Christoph Hellwig
2016-02-03 18:40 ` Christoph Hellwig
2016-02-03 18:40 ` Christoph Hellwig
2016-02-03 19:43 ` vfs/xfs: directio updates to ease COW handling V2 Jeff Moyer
2016-02-03 19:43 ` [Ocfs2-devel] " Jeff Moyer
2016-02-03 19:43 ` Jeff Moyer
2016-02-03 20:01 ` Darrick J. Wong
2016-02-03 20:01 ` [Ocfs2-devel] " Darrick J. Wong
2016-02-03 20:01 ` Darrick J. Wong
2016-02-03 21:53 ` Jeff Moyer
2016-02-03 21:53 ` [Ocfs2-devel] " Jeff Moyer
2016-02-03 21:53 ` Jeff Moyer
-- strict thread matches above, loose matches on Subject: below --
2016-02-02 20:17 VFS/XFS: directio updates to ease COW handling Christoph Hellwig
2016-02-02 20:17 ` [PATCH 2/3] xfs: don't use ioends for direct write completions Christoph Hellwig
2016-02-02 20:17 ` Christoph Hellwig
2016-02-03 13:52 ` Brian Foster
2016-02-03 13:52 ` Brian Foster
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=20160208010026.GL31407@dastard \
--to=david@fromorbit.com \
--cc=hch@lst.de \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=ocfs2-devel@oss.oracle.com \
--cc=xfs@oss.sgi.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.