From: Avi Kivity <avi@redhat.com>
To: Mark McLoughlin <markmc@redhat.com>
Cc: "David S. Ahern" <daahern@cisco.com>,
Avi Kivity <avi@qumranet.com>,
kvm@vger.kernel.org
Subject: Re: [PATCH 5/6] kvm: qemu: virtio-net: handle all tx in I/O thread without timer
Date: Thu, 06 Nov 2008 19:43:10 +0200 [thread overview]
Message-ID: <49132CAE.10102@redhat.com> (raw)
In-Reply-To: <1225990976.10879.34.camel@blaa>
Mark McLoughlin wrote:
> My machine has four CPUs, with two 6Mb L2 caches - each cache is shared
> between two of the CPUs, so I set things up as follows:
>
> pcpu#3 - netserver, I/O thread, vcpu#0
> pcup#4 - vcpu#1, virtio_net irq, netperf
>
> which (hopefully) ensures that we're only doing one copy using RAM and
> the rest are using the L1/L2 caches.
>
This pinning can drastically alter the results in other ways. If the
vcpu doing the transmit and the iothread can run in parallel, each
guest->host kick will result in a vmexit, IPI, and iothread execution,
with much less blocking of interrupts.
When the iothread and vcpu are collocated, the scheduler will ensure
natural batching and you will only vmexit when the ring is full (or
empty, in the case of the iothread), or if you spent quite a long time
processing it.
--
I have a truly marvellous patch that fixes the bug which this
signature is too narrow to contain.
next prev parent reply other threads:[~2008-11-06 17:43 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-10-30 17:51 [PATCH 0/6] Kill off the virtio_net tx mitigation timer Mark McLoughlin
2008-10-30 17:51 ` [PATCH 1/6] kvm: qemu: virtio: remove unused variable Mark McLoughlin
2008-10-30 17:51 ` [PATCH 2/6] kvm: qemu: dup the qemu_eventfd() return Mark McLoughlin
2008-10-30 17:51 ` [PATCH 3/6] kvm: qemu: add qemu_eventfd_write() and qemu_eventfd_read() Mark McLoughlin
2008-10-30 17:51 ` [PATCH 4/6] kvm: qemu: aggregate reads from eventfd Mark McLoughlin
2008-10-30 17:51 ` [PATCH 5/6] kvm: qemu: virtio-net: handle all tx in I/O thread without timer Mark McLoughlin
2008-10-30 17:51 ` [PATCH 6/6] kvm: qemu: virtio-net: drop mutex during tx tapfd write Mark McLoughlin
2008-11-04 11:43 ` Avi Kivity
2008-10-30 19:24 ` [PATCH 5/6] kvm: qemu: virtio-net: handle all tx in I/O thread without timer Anthony Liguori
2008-10-31 9:16 ` Mark McLoughlin
2008-11-03 15:07 ` Mark McLoughlin
2008-11-02 9:56 ` Avi Kivity
2008-11-04 15:23 ` David S. Ahern
2008-11-06 17:02 ` Mark McLoughlin
2008-11-06 17:13 ` David S. Ahern
2008-11-06 17:43 ` Avi Kivity [this message]
2008-10-30 19:20 ` [PATCH 0/6] Kill off the virtio_net tx mitigation timer Anthony Liguori
2008-11-02 9:48 ` Avi Kivity
2008-11-03 12:23 ` Mark McLoughlin
2008-11-03 12:40 ` Avi Kivity
2008-11-03 15:04 ` Mark McLoughlin
2008-11-03 15:19 ` Avi Kivity
2008-11-06 16:46 ` Mark McLoughlin
2008-11-06 17:38 ` Avi Kivity
2008-11-06 17:45 ` Mark McLoughlin
2008-11-09 11:29 ` Avi Kivity
2008-11-02 9:57 ` Avi Kivity
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=49132CAE.10102@redhat.com \
--to=avi@redhat.com \
--cc=avi@qumranet.com \
--cc=daahern@cisco.com \
--cc=kvm@vger.kernel.org \
--cc=markmc@redhat.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