From: "Michael S. Tsirkin" <mst@redhat.com>
To: Eugenio Perez Martin <eperezma@redhat.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
linux-kernel@vger.kernel.org, kvm list <kvm@vger.kernel.org>,
virtualization@lists.linux-foundation.org,
netdev@vger.kernel.org, Jason Wang <jasowang@redhat.com>
Subject: Re: [PATCH RFC v8 02/11] vhost: use batched get_vq_desc version
Date: Mon, 22 Jun 2020 11:55:04 -0400 [thread overview]
Message-ID: <20200622114622-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <CAJaqyWdwgy0fmReOgLfL4dAv-E+5k_7z3d9M+vHqt0aO2SmOFg@mail.gmail.com>
On Fri, Jun 19, 2020 at 08:07:57PM +0200, Eugenio Perez Martin wrote:
> On Mon, Jun 15, 2020 at 2:28 PM Eugenio Perez Martin
> <eperezma@redhat.com> wrote:
> >
> > On Thu, Jun 11, 2020 at 5:22 PM Konrad Rzeszutek Wilk
> > <konrad.wilk@oracle.com> wrote:
> > >
> > > On Thu, Jun 11, 2020 at 07:34:19AM -0400, Michael S. Tsirkin wrote:
> > > > As testing shows no performance change, switch to that now.
> > >
> > > What kind of testing? 100GiB? Low latency?
> > >
> >
> > Hi Konrad.
> >
> > I tested this version of the patch:
> > https://lkml.org/lkml/2019/10/13/42
> >
> > It was tested for throughput with DPDK's testpmd (as described in
> > http://doc.dpdk.org/guides/howto/virtio_user_as_exceptional_path.html)
> > and kernel pktgen. No latency tests were performed by me. Maybe it is
> > interesting to perform a latency test or just a different set of tests
> > over a recent version.
> >
> > Thanks!
>
> I have repeated the tests with v9, and results are a little bit different:
> * If I test opening it with testpmd, I see no change between versions
OK that is testpmd on guest, right? And vhost-net on the host?
> * If I forward packets between two vhost-net interfaces in the guest
> using a linux bridge in the host:
And here I guess you mean virtio-net in the guest kernel?
> - netperf UDP_STREAM shows a performance increase of 1.8, almost
> doubling performance. This gets lower as frame size increase.
> - rests of the test goes noticeably worse: UDP_RR goes from ~6347
> transactions/sec to 5830
OK so it seems plausible that we still have a bug where an interrupt
is delayed. That is the main difference between pmd and virtio.
Let's try disabling event index, and see what happens - that's
the trickiest part of interrupts.
> - TCP_STREAM goes from ~10.7 gbps to ~7Gbps
> - TCP_RR from 6223.64 transactions/sec to 5739.44
next prev parent reply other threads:[~2020-06-22 15:55 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-11 11:34 [PATCH RFC v8 00/11] vhost: ring format independence Michael S. Tsirkin
2020-06-11 11:34 ` [PATCH RFC v8 01/11] vhost: option to fetch descriptors through an independent struct Michael S. Tsirkin
2020-06-11 11:34 ` [PATCH RFC v8 02/11] vhost: use batched get_vq_desc version Michael S. Tsirkin
2020-06-11 15:22 ` Konrad Rzeszutek Wilk
2020-06-15 12:28 ` Eugenio Perez Martin
2020-06-19 18:07 ` Eugenio Perez Martin
2020-06-19 18:25 ` Eugenio Perez Martin
2020-06-22 9:07 ` Jason Wang
2020-06-22 10:44 ` Eugenio Perez Martin
2020-06-22 15:55 ` Michael S. Tsirkin [this message]
2020-06-22 16:11 ` Eugenio Perez Martin
2020-06-22 16:29 ` Michael S. Tsirkin
2020-06-23 16:15 ` Eugenio Perez Martin
2020-07-01 10:43 ` Eugenio Perez Martin
2020-07-01 11:11 ` Michael S. Tsirkin
2020-07-01 12:56 ` Eugenio Perez Martin
2020-07-01 12:39 ` Jason Wang
2020-07-01 13:04 ` Eugenio Perez Martin
2020-07-01 14:09 ` Jason Wang
2020-07-09 16:46 ` Eugenio Perez Martin
2020-07-09 17:37 ` Michael S. Tsirkin
2020-07-10 3:56 ` Jason Wang
2020-07-10 5:39 ` Eugenio Perez Martin
2020-07-10 5:58 ` Michael S. Tsirkin
2020-07-16 17:16 ` Eugenio Perez Martin
2020-07-20 8:55 ` Jason Wang
2020-07-20 13:07 ` Eugenio Perez Martin
2020-07-20 9:27 ` Michael S. Tsirkin
2020-07-20 11:16 ` Eugenio Pérez
2020-07-20 11:45 ` Michael S. Tsirkin
2020-07-21 2:55 ` Jason Wang
2020-07-29 18:37 ` Eugenio Perez Martin
2020-07-10 6:44 ` Jason Wang
2020-06-17 3:19 ` Jason Wang
2020-06-19 17:56 ` Eugenio Perez Martin
2020-06-22 16:00 ` Michael S. Tsirkin
2020-06-23 2:51 ` Jason Wang
2020-06-23 7:00 ` Eugenio Perez Martin
2020-06-23 7:15 ` Jason Wang
2020-06-23 8:25 ` Michael S. Tsirkin
2020-06-23 15:54 ` Eugenio Perez Martin
2020-06-11 11:34 ` [PATCH RFC v8 03/11] vhost/net: pass net specific struct pointer Michael S. Tsirkin
2020-06-15 16:08 ` Eugenio Perez Martin
2020-06-11 11:34 ` [PATCH RFC v8 04/11] vhost: reorder functions Michael S. Tsirkin
2020-06-11 11:34 ` [PATCH RFC v8 05/11] vhost: format-independent API for used buffers Michael S. Tsirkin
2020-06-15 16:11 ` Eugenio Perez Martin
2020-06-11 11:34 ` [PATCH RFC v8 06/11] vhost/net: convert to new API: heads->bufs Michael S. Tsirkin
2020-06-11 11:34 ` [PATCH RFC v8 07/11] vhost/net: avoid iov length math Michael S. Tsirkin
2020-06-11 11:34 ` [PATCH RFC v8 08/11] vhost/test: convert to the buf API Michael S. Tsirkin
2020-06-11 11:34 ` [PATCH RFC v8 09/11] vhost/scsi: switch to buf APIs Michael S. Tsirkin
2020-06-11 11:34 ` [PATCH RFC v8 10/11] vhost/vsock: switch to the buf API Michael S. Tsirkin
2020-06-11 11:34 ` [PATCH RFC v8 11/11] vhost: drop head based APIs Michael S. Tsirkin
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=20200622114622-mutt-send-email-mst@kernel.org \
--to=mst@redhat.com \
--cc=eperezma@redhat.com \
--cc=jasowang@redhat.com \
--cc=konrad.wilk@oracle.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--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;
as well as URLs for NNTP newsgroup(s).