qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Zhi Yong Wu <zwu.kernel@gmail.com>
To: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Cc: Kevin Wolf <kwolf@redhat.com>,
	Anthony Liguori <aliguori@us.ibm.com>,
	Marcelo Tosatti <mtosatti@redhat.com>,
	qemu-devel@nongnu.org, Zhi Yong Wu <wuzhy@cn.ibm.com>,
	Avi Kivity <avi@redhat.com>, Christoph Hellwig <hch@lst.de>
Subject: Re: [Qemu-devel] [PATCH 0/6] block: do request processing in a coroutine
Date: Tue, 11 Oct 2011 14:49:04 +0800	[thread overview]
Message-ID: <CAEH94Lgt2OgiEb3zmcpdBLeu=2qpXmrBAs60SceOKdFcLQrDhw@mail.gmail.com> (raw)
In-Reply-To: <1317831427-477-1-git-send-email-stefanha@linux.vnet.ibm.com>

On Thu, Oct 6, 2011 at 12:17 AM, Stefan Hajnoczi
<stefanha@linux.vnet.ibm.com> wrote:
> Block layer features like dirty block tracing, I/O throttling, and live block
> copy are forced to duplicate code due to the three different interfaces:
> synchronous, asynchronous, and coroutines.
>
> Since there are bdrv_read(), bdrv_aio_readv(), and bdrv_co_readv() interfaces
> for read (and similar for write), per-request processing needs to be duplicated
> for each of these execution contexts.  For example, dirty block tracking code
> is duplicated across these three interfaces.
>
> This patch series unifies request processing so that there is only one code
> path.  I see this as a prerequisite to the live block copy (image streaming)
> code I am working on, so I'm pushing it now.
>
> The short-term win from this series is that it becomes easy to add live block
> copy and other features.  We now have a single code path where the perf-request
> processing is done.
>
> The longer-term win will be dropping the BlockDriver .bdrv_read(),
> .bdrv_write(), .bdrv_aio_readv(), and .bdrv_aio_writev() interfaces.  By doing
> that we can bring all BlockDrivers onto a common interface, namely
> .bdrv_co_readv() and .bdrv_co_writev().  It will also allow us to drop most of
> the sync and aio emulation code.
>
> A consequence of this patch series is that every I/O request goes through at
> least one coroutine.  There is no longer a direct .bdrv_read(), .bdrv_write(),
> .bdrv_aio_readv(), or .bdrv_aio_writev() call - we're trying to phase out those
> interfaces.  I have not noticed performance degradation in correctness tests
> but we need to confirm that there has not been a performance regression.
>
> Stefan Hajnoczi (6):
>  block: directly invoke .bdrv_aio_*() in bdrv_co_io_em()
>  block: split out bdrv_co_do_readv() and bdrv_co_do_writev()
>  block: switch bdrv_read()/bdrv_write() to coroutines
>  block: switch bdrv_aio_readv() to coroutines
>  block: mark blocks dirty on coroutine write completion
>  block: switch bdrv_aio_writev() to coroutines
>
>  block.c |  273 +++++++++++++++++++++++++++++++--------------------------------
>  1 files changed, 134 insertions(+), 139 deletions(-)
OK. When i am available, i will play with it.
>
> --
> 1.7.6.3
>
>
>



-- 
Regards,

Zhi Yong Wu

  parent reply	other threads:[~2011-10-11  6:49 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-10-05 16:17 [Qemu-devel] [PATCH 0/6] block: do request processing in a coroutine Stefan Hajnoczi
2011-10-05 16:17 ` [Qemu-devel] [PATCH 1/6] block: directly invoke .bdrv_aio_*() in bdrv_co_io_em() Stefan Hajnoczi
2011-10-05 16:17 ` [Qemu-devel] [PATCH 2/6] block: split out bdrv_co_do_readv() and bdrv_co_do_writev() Stefan Hajnoczi
2011-10-05 16:17 ` [Qemu-devel] [PATCH 3/6] block: switch bdrv_read()/bdrv_write() to coroutines Stefan Hajnoczi
2011-10-11  6:44   ` Zhi Yong Wu
2011-10-12  9:03     ` Stefan Hajnoczi
2011-10-12  9:11       ` Zhi Yong Wu
2011-10-12 12:53   ` Kevin Wolf
2011-10-05 16:17 ` [Qemu-devel] [PATCH 4/6] block: switch bdrv_aio_readv() " Stefan Hajnoczi
2011-10-12 13:07   ` Kevin Wolf
2011-10-05 16:17 ` [Qemu-devel] [PATCH 5/6] block: mark blocks dirty on coroutine write completion Stefan Hajnoczi
2011-10-05 16:17 ` [Qemu-devel] [PATCH 6/6] block: switch bdrv_aio_writev() to coroutines Stefan Hajnoczi
2011-10-11  6:46   ` Zhi Yong Wu
2011-10-11  6:49 ` Zhi Yong Wu [this message]
2011-10-12 13:21 ` [Qemu-devel] [PATCH 0/6] block: do request processing in a coroutine Kevin Wolf

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='CAEH94Lgt2OgiEb3zmcpdBLeu=2qpXmrBAs60SceOKdFcLQrDhw@mail.gmail.com' \
    --to=zwu.kernel@gmail.com \
    --cc=aliguori@us.ibm.com \
    --cc=avi@redhat.com \
    --cc=hch@lst.de \
    --cc=kwolf@redhat.com \
    --cc=mtosatti@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@linux.vnet.ibm.com \
    --cc=wuzhy@cn.ibm.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 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).