From: Jens Axboe <axboe@kernel.dk>
To: Christoph Hellwig <hch@infradead.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>,
Chris Wright <chrisw@sous-sol.org>,
Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>,
kvm@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 5/5] virtio-blk: implement ->make_request
Date: Thu, 6 Oct 2011 15:53:09 +0200 [thread overview]
Message-ID: <20111006135309.GA27941@kernel.dk> (raw)
In-Reply-To: <20111005195530.558142038@bombadil.infradead.org>
On Wed, Oct 05 2011, Christoph Hellwig wrote:
> Add an alternate I/O path that implements ->make_request for virtio-blk.
> This is required for high IOPs devices which get slowed down to 1/5th of
> the native speed by all the locking, memory allocation and other overhead
> in the request based I/O path.
We definitely have some performance fruit hanging rather low in that
path, but a factor 5 performance difference sounds insanely excessive. I
haven't looked at virtio_blk in detail, but I've done 500K+ IOPS on
request based drivers before (yes, on real hardware). So it could be
that virtio_blk is just doing things rather suboptimally in some places,
and that it would be possible to claim most of that speedup there too.
--
Jens Axboe
next prev parent reply other threads:[~2011-10-06 13:53 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-10-05 19:54 [PATCH 0/5] RFC: ->make_request support for virtio-blk Christoph Hellwig
2011-10-05 19:54 ` [PATCH 1/5] block: add bio_map_sg Christoph Hellwig
2011-10-05 22:51 ` Boaz Harrosh
2011-10-06 13:40 ` Christoph Hellwig
2011-10-05 19:54 ` [PATCH 2/5] virtio: support unlocked queue kick Christoph Hellwig
2011-10-06 8:42 ` Stefan Hajnoczi
[not found] ` <87r52qgaf3.fsf@rustcorp.com.au>
2011-10-06 13:19 ` Michael S. Tsirkin
2011-11-01 14:40 ` Michael S. Tsirkin
2011-11-02 3:19 ` Rusty Russell
2011-11-02 7:25 ` Christoph Hellwig
2011-11-03 4:01 ` Rusty Russell
2011-11-03 5:15 ` Rusty Russell
2011-11-03 6:45 ` Minchan Kim
2011-10-05 19:54 ` [PATCH 3/5] virtio-blk: remove the unused list of pending requests Christoph Hellwig
2011-10-05 19:54 ` [PATCH 4/5] virtio-blk: reimplement the serial attribute without using requests Christoph Hellwig
2011-10-05 19:54 ` [PATCH 5/5] virtio-blk: implement ->make_request Christoph Hellwig
2011-10-06 1:52 ` Rusty Russell
2011-10-06 13:42 ` Christoph Hellwig
2011-10-06 13:53 ` Jens Axboe [this message]
2011-10-05 20:31 ` [PATCH 0/5] RFC: ->make_request support for virtio-blk Vivek Goyal
2011-10-05 21:53 ` Christoph Hellwig
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=20111006135309.GA27941@kernel.dk \
--to=axboe@kernel.dk \
--cc=chrisw@sous-sol.org \
--cc=hch@infradead.org \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=rusty@rustcorp.com.au \
--cc=stefanha@linux.vnet.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).