From: Dave Kleikamp <dave.kleikamp@oracle.com>
To: Zach Brown <zab@zabbo.net>
Cc: Jeff Moyer <jmoyer@redhat.com>,
"Maxim V. Patlasov" <mpatlasov@parallels.com>,
"linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [RFC PATCH v2 15/21] loop: use aio to perform io on the underlying file
Date: Fri, 20 Apr 2012 10:57:12 -0500 [thread overview]
Message-ID: <4F918758.8020001@oracle.com> (raw)
In-Reply-To: <4F91862F.2030502@zabbo.net>
On 04/20/2012 10:52 AM, Zach Brown wrote:
> On 04/20/2012 11:20 AM, Jeff Moyer wrote:
>> Dave Kleikamp<dave.kleikamp@oracle.com> writes:
>>
>>> On 04/20/2012 09:48 AM, Maxim V. Patlasov wrote:
>>>> On 03/30/2012 07:43 PM, Dave Kleikamp wrote:
>>>>> From: Zach Brown<zab@zabbo.net>
>>>>>
>>>>> This uses the new kernel aio interface to process loopback IO by
>>>>> submitting concurrent direct aio. Previously loop's IO was serialized
>>>>> by synchronous processing in a thread.
>>>>>
>>>>
>>>> The patch ignores REQ_FLUSH bit of bi_rw. Is it simply overlook?
>>>
>>> Good question. Since the loop device is sending only direct IO requests,
>>> it shouldn't be necessary to explicitly flush page cache, but REQ_FLUSH
>>
>> REQ_FLUSH isn't about the page cache, it's about flushing the volatile
>> disk write cache. You need to handle that.
>
> I guess O_DIRECT doesn't routinely issue flushes simply because it's too
> expensive? Apps that care about consistent IO and O_DIRECT are expected
> to not have writeback caching enabled? 'cause there's no way they're
> issuing syncs themselves.
If we weren't using aio, we might be okay, but we don't know that any
prior asynchronous request has completed.
>
> So yeah, I'd agree that the loop code should be reworked a bit so that
> both the filebacked and aio methods call vfs_sync() when they see
> REQ_FLUSH.
It's an easy fix. I don't anticipate that it will hurt performance too
badly.
>
> Bleh.
>
> - z
> (Sorry, no real time to dig into this now. Lots more time in two months!)
next prev parent reply other threads:[~2012-04-20 15:57 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-03-30 15:43 [RFC PATCH v2 00/21] loop: Issue O_DIRECT aio using bio_vec Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 01/21] iov_iter: move into its own file Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 02/21] iov_iter: add copy_to_user support Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 03/21] fuse: convert fuse to use iov_iter_copy_[to|from]_user Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 04/21] iov_iter: hide iovec details behind ops function pointers Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 05/21] iov_iter: add bvec support Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 06/21] iov_iter: add a shorten call Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 07/21] iov_iter: let callers extract iovecs and bio_vecs Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 08/21] dio: create a dio_aligned() helper function Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 09/21] dio: Convert direct_IO to use iov_iter Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 10/21] dio: add bio_vec support to __blockdev_direct_IO() Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 11/21] fs: pull iov_iter use higher up the stack Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 12/21] aio: add aio_kernel_() interface Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 13/21] aio: add aio support for iov_iter arguments Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 14/21] bio: add bvec_length(), like iov_length() Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 15/21] loop: use aio to perform io on the underlying file Dave Kleikamp
2012-04-20 14:48 ` Maxim V. Patlasov
2012-04-20 15:09 ` Dave Kleikamp
2012-04-20 15:20 ` Jeff Moyer
2012-04-20 15:52 ` Zach Brown
2012-04-20 15:57 ` Dave Kleikamp [this message]
2012-04-20 16:14 ` Maxim V. Patlasov
2012-04-20 17:19 ` Dave Kleikamp
2012-04-20 17:37 ` Maxim V. Patlasov
2012-04-20 16:35 ` Jeff Moyer
2012-04-20 17:48 ` Zach Brown
2012-04-20 16:14 ` Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 16/21] ext3: add support for .read_iter and .write_iter Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 17/21] ocfs2: add support for read_iter, write_iter, and direct_IO_bvec Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 18/21] ext4: add support for read_iter and write_iter Dave Kleikamp
2012-04-02 18:42 ` Ted Ts'o
2012-04-02 22:45 ` Dave Kleikamp
2012-04-03 0:11 ` Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 19/21] nfs: add support for read_iter, write_iter Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 20/21] btrfs: add support for read_iter and write_iter Dave Kleikamp
2012-03-30 15:43 ` [RFC PATCH v2 21/21] fs: add read_iter and write_iter to more file systems Dave Kleikamp
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=4F918758.8020001@oracle.com \
--to=dave.kleikamp@oracle.com \
--cc=jmoyer@redhat.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mpatlasov@parallels.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;
as well as URLs for NNTP newsgroup(s).