From: Or Gerlitz <ogerlitz@Voltaire.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Herbert Xu <herbert@gondor.apana.org.au>,
Jamie Lokier <jamie@shareable.org>,
Anthony Liguori <anthony@codemonkey.ws>,
qemu-devel@nongnu.org, Jan Kiszka <jan.kiszka@web.de>,
Mark McLoughlin <markmc@redhat.com>, Dor Laor <dlaor@redhat.com>,
netdev@vger.kernel.org
Subject: Re: [Qemu-devel] [PATCH] net: add raw backend - some performance measurements
Date: Tue, 21 Jul 2009 14:05:32 +0300 [thread overview]
Message-ID: <4A65A0FC.6090204@Voltaire.com> (raw)
In-Reply-To: <20090721102733.GB22155@redhat.com>
Michael S. Tsirkin wrote:
> With raw sockets, send will block or fail if the TX queue for device is
> full. With tap+bridge, the buffer in tap has to fill up instead, which
> is not the same. I'm not sure this is the issue here, but could be: the
> benchmark is UDP, isn't it?
Michael,
What/where is this tap buffer? we're talking on VM TX, so looking on tun_get_user I see a call to
skb_copy_datagram_from_iovec() to copy from the user buffer to an skb, then a call to netif_rx_ni() and that's it... As for your question, indeed udp, the VM runs netperf/UDP_STREAM
Or.
WARNING: multiple messages have this Message-ID (diff)
From: Or Gerlitz <ogerlitz@Voltaire.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>,
Herbert Xu <herbert@gondor.apana.org.au>,
netdev@vger.kernel.org, Dor Laor <dlaor@redhat.com>,
qemu-devel@nongnu.org, Jan Kiszka <jan.kiszka@web.de>
Subject: Re: [Qemu-devel] [PATCH] net: add raw backend - some performance measurements
Date: Tue, 21 Jul 2009 14:05:32 +0300 [thread overview]
Message-ID: <4A65A0FC.6090204@Voltaire.com> (raw)
In-Reply-To: <20090721102733.GB22155@redhat.com>
Michael S. Tsirkin wrote:
> With raw sockets, send will block or fail if the TX queue for device is
> full. With tap+bridge, the buffer in tap has to fill up instead, which
> is not the same. I'm not sure this is the issue here, but could be: the
> benchmark is UDP, isn't it?
Michael,
What/where is this tap buffer? we're talking on VM TX, so looking on tun_get_user I see a call to
skb_copy_datagram_from_iovec() to copy from the user buffer to an skb, then a call to netif_rx_ni() and that's it... As for your question, indeed udp, the VM runs netperf/UDP_STREAM
Or.
next prev parent reply other threads:[~2009-07-21 11:05 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-07-01 15:46 [Qemu-devel] [PATCH] net: add raw backend Or Gerlitz
2009-07-01 16:21 ` Jamie Lokier
2009-07-02 12:25 ` Or Gerlitz
2009-07-03 2:39 ` Jamie Lokier
2009-07-07 13:33 ` Or Gerlitz
2009-07-07 14:57 ` Jamie Lokier
2009-07-08 14:45 ` Or Gerlitz
2009-07-14 13:54 ` Or Gerlitz
2009-07-15 20:38 ` Jamie Lokier
2009-07-15 21:06 ` Jan Kiszka
2009-07-15 21:52 ` Jamie Lokier
2009-07-16 8:29 ` Or Gerlitz
2009-07-20 14:13 ` [Qemu-devel] [PATCH] net: add raw backend - some performance measurements Or Gerlitz
2009-07-20 15:53 ` Herbert Xu
2009-07-20 18:20 ` Michael S. Tsirkin
2009-07-21 1:46 ` Herbert Xu
2009-07-21 7:03 ` Or Gerlitz
2009-07-21 7:25 ` Herbert Xu
2009-07-21 7:25 ` Herbert Xu
2009-07-21 10:17 ` Or Gerlitz
2009-07-21 10:17 ` Or Gerlitz
2009-07-21 10:27 ` Michael S. Tsirkin
2009-07-21 10:27 ` Michael S. Tsirkin
2009-07-21 11:05 ` Or Gerlitz [this message]
2009-07-21 11:05 ` Or Gerlitz
2009-07-21 12:01 ` Michael S. Tsirkin
2009-07-21 12:01 ` Michael S. Tsirkin
2009-07-21 12:14 ` Herbert Xu
2009-07-21 12:14 ` Herbert Xu
2009-07-21 13:41 ` Or Gerlitz
2009-07-21 13:41 ` Or Gerlitz
[not found] ` <5b31733c0907011250i7afcdbcdnb844290de4ad64f2@mail.gmail.com>
2009-07-02 12:08 ` [Qemu-devel] [PATCH] net: add raw backend Or Gerlitz
2009-07-02 15:43 ` Michael S. Tsirkin
2009-07-07 14:45 ` Or Gerlitz
2009-07-07 14:49 ` Michael S. Tsirkin
2009-07-08 14:46 ` Or Gerlitz
2009-07-08 15:06 ` Or Gerlitz
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=4A65A0FC.6090204@Voltaire.com \
--to=ogerlitz@voltaire.com \
--cc=anthony@codemonkey.ws \
--cc=dlaor@redhat.com \
--cc=herbert@gondor.apana.org.au \
--cc=jamie@shareable.org \
--cc=jan.kiszka@web.de \
--cc=markmc@redhat.com \
--cc=mst@redhat.com \
--cc=netdev@vger.kernel.org \
--cc=qemu-devel@nongnu.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.