All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ming Lei <ming.lei@canonical.com>
To: Jens Axboe <axboe@kernel.dk>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Cc: Christoph Hellwig <hch@lst.de>
Subject: Re: [PATCH v3 0/10] block: per-distpatch_queue flush machinery
Date: Sat, 13 Sep 2014 21:52:52 +0800	[thread overview]
Message-ID: <CACVXFVMyYWHHOhQQObkaUSnf+CUB5NhD-RTFJCgqciotj85D9w@mail.gmail.com> (raw)
In-Reply-To: <1410577674-17019-1-git-send-email-ming.lei@canonical.com>

On Sat, Sep 13, 2014 at 11:07 AM, Ming Lei <ming.lei@canonical.com> wrote:
> Hi,
>
> As recent discussion, especially suggested by Christoph, this patchset
> implements per-distpatch_queue flush machinery, so that:
>
>         - current init_request and exit_request callbacks can
>         cover flush request too, then the buggy copying way of
>         initializing flush request's pdu can be fixed
>
>         - flushing performance gets improved in case of multi hw-queue
>
> About 70% throughput improvement is observed in sync write
> over multi dispatch-queue virtio-blk, see details in commit log
> of patch 10/10.
>
> This patchset can be pulled from below tree too:
>
>         git://kernel.ubuntu.com/ming/linux.git v3.17-block-dev_v3
>
> V3:
>         - don't return failure code from blk_alloc_flush_queue() to
>         avoid freeing invalid buffer in case of allocation failure
>         - remove blk_init_flush() and blk_exit_flush()
>         - remove unnecessary WARN_ON() from blk_alloc_flush_queue()

Looks there are still two problems in V3:

- blk_free_flush_queue() should be called only for legacy queue after
the patch 10's conversion

- pdu copy should be removed from blk_mq_clone_flush_request()

I will post V4 once these two problems are fixed.

Thanks,
>
> V2:
>         - refactor blk_mq_init_hw_queues() and its pair, also it is a fix
>         on failure path, so that conversion to per-queue flush becomes simple.
>         - allocate/initialize flush queue in blk_mq_init_hw_queues()
>         - add sync write tests on virtio-blk which is backed by SSD image
>
> V1:
>         - commit log typo fix
>         - introduce blk_alloc_flush_queue() and its pair earlier, so
>         that patch 5 and 8 become easier for review
>
>  block/blk-core.c       |   12 ++--
>  block/blk-flush.c      |  129 +++++++++++++++++++++++++-------------
>  block/blk-mq.c         |  160 ++++++++++++++++++++++++++++++------------------
>  block/blk-mq.h         |    1 -
>  block/blk-sysfs.c      |    4 +-
>  block/blk.h            |   35 ++++++++++-
>  include/linux/blk-mq.h |    2 +
>  include/linux/blkdev.h |   10 +--
>  8 files changed, 230 insertions(+), 123 deletions(-)
>
>
>
> Thanks,
> --
> Ming Lei
>
>

      parent reply	other threads:[~2014-09-13 13:52 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-09-13  3:07 [PATCH v3 0/10] block: per-distpatch_queue flush machinery Ming Lei
2014-09-13  3:07 ` [PATCH v3 01/10] blk-mq: allocate flush_rq in blk_mq_init_flush() Ming Lei
2014-09-13  3:07 ` [PATCH v3 02/10] block: introduce blk_init_flush and its pair Ming Lei
2014-09-13  3:07 ` [PATCH v3 03/10] block: move flush initialization to blk_flush_init Ming Lei
2014-09-13  3:07 ` [PATCH v3 04/10] block: avoid to use q->flush_rq directly Ming Lei
2014-09-13  3:07 ` [PATCH v3 05/10] block: introduce blk_flush_queue to drive flush machinery Ming Lei
2014-09-13  3:07 ` [PATCH v3 06/10] block: remove blk_init_flush() and its pair Ming Lei
2014-09-13  3:07 ` [PATCH v3 07/10] block: flush: avoid to figure out flush queue unnecessarily Ming Lei
2014-09-13  3:07 ` [PATCH v3 08/10] block: introduce 'blk_mq_ctx' parameter to blk_get_flush_queue Ming Lei
2014-09-13  3:07 ` [PATCH v3 09/10] blk-mq: handle failure path for initializing hctx Ming Lei
2014-09-13  3:07 ` [PATCH v3 10/10] blk-mq: support per-distpatch_queue flush machinery Ming Lei
2014-09-13 13:52 ` Ming Lei [this message]

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=CACVXFVMyYWHHOhQQObkaUSnf+CUB5NhD-RTFJCgqciotj85D9w@mail.gmail.com \
    --to=ming.lei@canonical.com \
    --cc=axboe@kernel.dk \
    --cc=hch@lst.de \
    --cc=linux-kernel@vger.kernel.org \
    /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.