From: Asias He <asias@redhat.com>
To: Rusty Russell <rusty@rustcorp.com.au>
Cc: mst@redhat.com, virtualization@lists.linux-foundation.org,
sjur.brandeland@stericsson.com
Subject: Re: [PATCH 4/6] virtio_ring: virtqueue_add_outbuf / virtqueue_add_inbuf.
Date: Thu, 7 Mar 2013 16:15:34 +0800 [thread overview]
Message-ID: <20130307081534.GA19603@hj.localdomain> (raw)
In-Reply-To: <87boaw9hgw.fsf@rustcorp.com.au>
On Thu, Mar 07, 2013 at 11:33:19AM +1100, Rusty Russell wrote:
> Asias He <asias@redhat.com> writes:
> > On Wed, Mar 06, 2013 at 04:23:24PM +1100, Rusty Russell wrote:
> >> These are specialized versions of virtqueue_add_buf(), which cover
> >> over 80% of cases and are far clearer.
> >>
> >> In particular, the scatterlists passed to these functions don't have
> >> to be clean (ie. we ignore end markers).
> >>
> >> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
> >
> > So, what is the plan for the following ideas discussed in the other
> > thread?
> >
> > '''
> > > Looking at code, it seems that most users really have a single sg, in
> > > low memory. So how about simply passing void * instead of sg? Whoever
> > > has multiple sgs can use the rich interface.
> >
> > Good point, let's do that:
> > 1) Make virtqueue_add_outbuf()/inbuf() take a void * and len.
> > 2) Transfer users across to use that.
> > 3) Make everyone else use clean scatterlists with virtqueue_add_sgs[].
> > 4) Remove virtqueue_add_bufs().
>
> Networking performance: there is still a performance penalty in using
> virtqueue_add_sgs(), and it can't use a simple void * and len.
>
> So I changed my mind. Again...
Ah, OK.
> Cheers,
> Rusty.
--
Asias
next prev parent reply other threads:[~2013-03-07 8:15 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-06 5:15 [PATCH 0/6] virtio_add_buf replacement Rusty Russell
2013-03-06 5:19 ` [PATCH 1/6] virtio_ring: virtqueue_add_sgs, to add multiple sgs Rusty Russell
2013-03-06 5:20 ` [PATCH 2/6] virtio_ring: don't count elements twice for add_buf path Rusty Russell
2013-03-06 5:22 ` [PATCH 3/6] virtio_ring: inline internal vring functions more aggressively Rusty Russell
2013-03-06 10:24 ` Michael S. Tsirkin
2013-03-06 23:02 ` Rusty Russell
2013-03-06 5:23 ` [PATCH 4/6] virtio_ring: virtqueue_add_outbuf / virtqueue_add_inbuf Rusty Russell
2013-03-06 8:37 ` Asias He
2013-03-07 0:33 ` Rusty Russell
2013-03-07 8:15 ` Asias He [this message]
2013-03-06 5:24 ` [PATCH 5/6] tools/virtio: make vringh_test use inbuf/outbuf Rusty Russell
2013-03-06 5:24 ` [PATCH 6/6] virtio_scsi: use virtqueue_add_inbuf() for virtscsi_kick_event Rusty Russell
2013-03-06 8:09 ` [PATCH 0/6] virtio_add_buf replacement Asias He
2013-03-07 0:35 ` Rusty Russell
2013-03-07 8:22 ` Asias He
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=20130307081534.GA19603@hj.localdomain \
--to=asias@redhat.com \
--cc=mst@redhat.com \
--cc=rusty@rustcorp.com.au \
--cc=sjur.brandeland@stericsson.com \
--cc=virtualization@lists.linux-foundation.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox