From mboxrd@z Thu Jan 1 00:00:00 1970 From: Roopa Prabhu Subject: Re: RFT: virtio_net: limit xmit polling Date: Sat, 25 Jun 2011 12:44:50 -0700 Message-ID: References: Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============5835043323597548260==" Cc: Krishna Kumar2 , habanero@linux.vnet.ibm.com, lguest@lists.ozlabs.org, Shirley Ma , kvm@vger.kernel.org, Carsten Otte , linux-s390@vger.kernel.org, Heiko Carstens , linux-kernel@vger.kernel.org, virtualization@lists.linux-foundation.org, steved@us.ibm.com, Christian Borntraeger , netdev@vger.kernel.org, Martin Schwidefsky , linux390@de.ibm.com To: Roopa Prabhu , Tom Lendacky , "Michael S. Tsirkin" Return-path: In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Mime-version: 1.0 Sender: virtualization-bounces@lists.linux-foundation.org Errors-To: virtualization-bounces@lists.linux-foundation.org List-Id: netdev.vger.kernel.org > This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. --===============5835043323597548260== Content-type: multipart/alternative; boundary="B_3391850690_10655769" > This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. --B_3391850690_10655769 Content-type: text/plain; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable Here are the results I am getting with a Cisco 10G VIC adapter. All tests are from the guest to an external host. virtio-net-limit-xmit-polling/base: TCP_STREAM: 8089Mbps TCP_MAERTS: 9334Mbps virtio-net-limit-xmit-polling/v0 TCP_STREAM: 8004Mbps TCP_MAERTS: 9338Mbps virtio-net-limit-xmit-polling/v1 TCP_STREAM: 8028Mbps TCP_MAERTS: 9339Mbps virtio-net-limit-xmit-polling/v2 TCP_STREAM: 8045Mbps TCP_MAERTS: 9337Mbps For the TCP_STREAM tests I don=B9t get consistent results. Every run gives me slightly different results. But its always between 7900Mbps to 8100Mbps. But I also see this with the base kernel so its not related to these patches.=20 Thanks, Roopa On 6/24/11 5:50 AM, "Roopa Prabhu" wrote: > Michael, =20 >=20 > I am testing this too. > I have finished one round of testing. But am running it again just to > confirm. > This time I will see if I can collect some exit stats too. Will post resu= lts > sometime this weekend. > I am just doing TCP_STREAM and TCP_MAERTS from guest to remote host. >=20 > Thanks, > Roopa >=20 >=20 > On 6/21/11 8:23 AM, "Tom Lendacky" wrote: >=20 >> On Sunday, June 19, 2011 05:27:00 AM Michael S. Tsirkin wrote: >>> > OK, different people seem to test different trees. In the hope to ge= t >>> > everyone on the same page, I created several variants of this patch s= o >>> > they can be compared. Whoever's interested, please check out the >>> > following, and tell me how these compare: >>=20 >> I'm in the process of testing these patches. Base and v0 are complete >> and v1 is near complete with v2 to follow. I'm testing with a variety >> of TCP_RR and TCP_STREAM/TCP_MAERTS tests involving local guest-to-guest >> tests and remote host-to-guest tests. I'll post the results in the next >> day or two when the tests finish. >>=20 >> Thanks, >> Tom >>=20 >>> > >>> > kernel: >>> > >>> > git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost.git >>> > >>> > virtio-net-limit-xmit-polling/base - this is net-next baseline to tes= t >>> > against virtio-net-limit-xmit-polling/v0 - fixes checks on out of cap= acity >>> > virtio-net-limit-xmit-polling/v1 - previous revision of the patch >>> > this does xmit,free,xmit,2*free,free >>> > virtio-net-limit-xmit-polling/v2 - new revision of the patch >>> > this does free,xmit,2*free,free >>> > >>> > There's also this on top: >>> > virtio-net-limit-xmit-polling/v3 -> don't delay avail index update >>> > I don't think it's important to test this one, yet >>> > >>> > Userspace to use: event index work is not yet merged upstream >>> > so the revision to use is still this: >>> > git://git.kernel.org/pub/scm/linux/kernel/git/mst/qemu-kvm.git >>> > virtio-net-event-idx-v3 >>=20 >=20 >=20 > _______________________________________________ > Virtualization mailing list > Virtualization@lists.linux-foundation.org > https://lists.linux-foundation.org/mailman/listinfo/virtualization --B_3391850690_10655769 Content-type: text/html; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable Re: RFT: virtio_net: limit xmit polling
Here are the results I am getting with a Cisco 10G VIC adapter.
All tests are from the guest to an external host.

virtio-net-limit-xmit-polling/base:
TCP_STREAM: 8089Mbps
TCP_MAERTS: 9334Mbps

virtio-net-limit-xmit-polling/v0
TCP_STREAM: 8004Mbps
TCP_MAERTS: 9338Mbps

virtio-net-limit-xmit-polling/v1
TCP_STREAM: 8028Mbps
TCP_MAERTS: 9339Mbps

virtio-net-limit-xmit-polling/v2
TCP_STREAM: 8045Mbps
TCP_MAERTS: 9337Mbps

For the TCP_STREAM tests I don’t get consistent results.
Every run gives me slightly different results. But its always between 7900M= bps to 8100Mbps.
But I also see this with the base kernel so its not related to these patche= s.

Thanks,
Roopa

On 6/24/11 5:50 AM, "Roopa Prabhu" <roprabhu@cisco.com> wrote:

<= SPAN STYLE=3D'font-size:11pt'>Michael,  

I am testing this too.
 I have finished one round of testing. But am running it again just to= confirm.
This time I will see if I can collect some exit stats too. Will post result= s sometime this weekend.
I am just doing TCP_STREAM and TCP_MAERTS from guest to remote host.

Thanks,
Roopa


On 6/21/11 8:23 AM, "Tom Lendacky" <tahm@linux.vnet.ibm.com> wrote:

<= SPAN STYLE=3D'font-size:11pt'>On Sunday, June 19, 2011 05:27:00 AM Michael S. = Tsirkin wrote:
> OK, different people seem to test different trees.  In the hope t= o get
> everyone on the same page, I created several variants of this patch so=
> they can be compared. Whoever's interested, please check out the
> following, and tell me how these compare:

I'm in the process of testing these patches.  Base and v0 are complete=
and v1 is near complete with v2 to follow.  I'm testing with a variety=
of TCP_RR and TCP_STREAM/TCP_MAERTS tests involving local guest-to-guest tests and remote host-to-guest tests.  I'll post the results in the ne= xt
day or two when the tests finish.

Thanks,
Tom

>
> kernel:
>
> git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost.git
>
> virtio-net-limit-xmit-polling/base - this is net-next baseline to test=
> against virtio-net-limit-xmit-polling/v0 - fixes checks on out of capa= city
> virtio-net-limit-xmit-polling/v1 - previous revision of the patch
>            &nbs= p;  this does xmit,free,xmit,2*free,free
> virtio-net-limit-xmit-polling/v2 - new revision of the patch
>            &nbs= p;  this does free,xmit,2*free,free
>
> There's also this on top:
> virtio-net-limit-xmit-polling/v3 -> don't delay avail index update<= BR> > I don't think it's important to test this one, yet
>
> Userspace to use: event index work is not yet merged upstream
> so the revision to use is still this:
> git://git.kernel.org/pub/scm/linux/kernel/git/mst/qemu-kvm.git
> virtio-net-event-idx-v3

=
___________= ____________________________________
Virtualization mailing list
Virtualization@lists.li= nux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/virtualization
--B_3391850690_10655769-- --===============5835043323597548260== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linux-foundation.org/mailman/listinfo/virtualization --===============5835043323597548260==--