From: James Bottomley <jbottomley@parallels.com>
To: "sedat.dilek@gmail.com" <sedat.dilek@gmail.com>
Cc: Dave Kleikamp <dave.kleikamp@oracle.com>,
"linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
Zach Brown <zab@zabbo.net>,
Maxim Patlasov <MPatlasov@parallels.com>
Subject: Re: [PATCH V5 00/30] loop: Issue O_DIRECT aio using bio_vec
Date: Thu, 17 Jan 2013 23:49:37 +0000 [thread overview]
Message-ID: <1358466576.2394.44.camel@dabdike.int.hansenpartnership.com> (raw)
In-Reply-To: <CA+icZUWc_hwX6A3YjD3p66FG4Qbgraukm+0GJXUPzPOWb=twJQ@mail.gmail.com>
On Wed, 2013-01-16 at 19:25 +0100, Sedat Dilek wrote:
> On Wed, Jan 16, 2013 at 5:32 PM, James Bottomley
> <jbottomley@parallels.com> wrote:
> > On Wed, 2013-01-09 at 13:58 -0600, Dave Kleikamp wrote:
> >> This patchset was begun by Zach Brown and was originally submitted for
> >> review in October, 2009. Feedback was positive, and I have picked up
> >> where he left off, porting his patches to the latest mainline kernel
> >> and adding support more file systems.
> >>
> >> This patch series adds a kernel interface to fs/aio.c so that kernel code can
> >> issue concurrent asynchronous IO to file systems. It adds an aio command and
> >> file system methods which specify io memory with pages instead of userspace
> >> addresses.
> >>
> >> This series was written to reduce the current overhead loop imposes by
> >> performing synchronus buffered file system IO from a kernel thread. These
> >> patches turn loop into a light weight layer that translates bios into iocbs.
> >>
> >> It introduces new file ops, read_iter() and write_iter(), that replace the
> >> aio_read() and aio_write() operations. The iov_iter structure can now contain
> >> either a user-space iovec or a kernel-space bio_vec. Since it would be
> >> overly complicated to replace every instance of aio_read() and aio_write(),
> >> the old operations are not removed, but file systems implementing the new
> >> ones need not keep the old ones.
> >>
> >> Changes from V4 include:
> >> * moved iov-iter.c from mm/ to fs/
> >> * removed dio_aligned helper
> >> * insured that FUA write to loop device is committed to media
> >> * removed no-longer-used REQ_KERNEL define
> >>
> >> These patches apply to 3.8-rc2 and are also available at:
> >> git://github.com/kleikamp/linux-shaggy.git loop_2013_01_04
> >>
> >> My hopes are that this patchset is finally ready for linux-next.
> >
> > Just a note that we at parallels are anxiously awaiting this too. Our
> > rewrite of the linux loop device to be more efficient (and not double
> > cache) depends on this work. I should also note that we've been testing
> > these patches (albeit backported to a RHEL kernel as a current work
> > base) with no problems reported so far.
> >
>
> Hi James,
>
> do you have a location for me for testing this patchset from parallels?
You mean the RHEL kernels? Not really; they're just proof of concept
and not released yet. Our mainline effort won't really begin until we
know the fate of the DIO patches.
James
next prev parent reply other threads:[~2013-01-17 23:49 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-01-09 19:58 Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 01/30] iov_iter: move into its own file Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 02/30] iov_iter: iov_iter_copy_from_user() should use non-atomic copy Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 03/30] iov_iter: add copy_to_user support Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 04/30] fuse: convert fuse to use iov_iter_copy_[to|from]_user Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 05/30] iov_iter: hide iovec details behind ops function pointers Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 06/30] iov_iter: add bvec support Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 07/30] iov_iter: add a shorten call Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 08/30] iov_iter: let callers extract iovecs and bio_vecs Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 09/30] dio: Convert direct_IO to use iov_iter Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 10/30] dio: add bio_vec support to __blockdev_direct_IO() Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 11/30] fs: pull iov_iter use higher up the stack Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 12/30] aio: add aio_kernel_() interface Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 13/30] aio: add aio support for iov_iter arguments Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 14/30] bio: add bvec_length(), like iov_length() Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 15/30] loop: use aio to perform io on the underlying file Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 16/30] fs: create file_readable() and file_writable() functions Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 17/30] fs: use read_iter and write_iter rather than aio_read and aio_write Dave Kleikamp
2013-01-18 21:26 ` Jeff Moyer
2013-01-18 22:03 ` Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 19/30] ocfs2: add support for read_iter, write_iter, and direct_IO_bvec Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 20/30] ext4: add support for read_iter and write_iter Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 21/30] nfs: add support for read_iter, write_iter Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 22/30] nfs: simplify swap Dave Kleikamp
2013-01-09 20:01 ` Rik van Riel
2013-01-09 19:58 ` [PATCH V5 23/30] btrfs: add support for read_iter and write_iter Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 24/30] block_dev: add support for read_iter, write_iter Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 25/30] xfs: add support for read_iter and write_iter Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 26/30] gfs2: Convert aio_read/write ops to read/write_iter Dave Kleikamp
2013-01-10 10:10 ` Steven Whitehouse
2013-01-10 14:34 ` Dave Kleikamp
2013-01-11 16:24 ` Steven Whitehouse
2013-01-09 19:58 ` [PATCH V5 27/30] udf: convert file ops from aio_read/write " Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 28/30] afs: add support for read_iter and write_iter Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 29/30] ecrpytfs: Convert aio_read/write ops to read/write_iter Dave Kleikamp
2013-01-09 19:58 ` [PATCH V5 30/30] ubifs: convert file ops from aio_read/write " Dave Kleikamp
2013-01-10 15:46 ` [PATCH V5 00/30] loop: Issue O_DIRECT aio using bio_vec Sedat Dilek
2013-01-11 21:51 ` Dave Kleikamp
2013-01-12 10:52 ` Sedat Dilek
2013-01-12 12:54 ` Sedat Dilek
2013-01-16 16:32 ` James Bottomley
2013-01-16 18:25 ` Sedat Dilek
2013-01-17 23:49 ` James Bottomley [this message]
2013-01-18 17:16 ` Jeff Moyer
2013-01-18 17:56 ` Jeff Moyer
2013-01-18 17:58 ` Dave Kleikamp
2013-01-18 21:48 ` Jeff Moyer
2013-01-18 22:33 ` Dave Kleikamp
2013-01-22 16:14 ` Jeff Moyer
2013-01-20 22:26 ` Dave Chinner
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=1358466576.2394.44.camel@dabdike.int.hansenpartnership.com \
--to=jbottomley@parallels.com \
--cc=MPatlasov@parallels.com \
--cc=dave.kleikamp@oracle.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=sedat.dilek@gmail.com \
--cc=zab@zabbo.net \
/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
all inboxes | Powered by JetHome®