Re: [PATCH] kvm:virtio-net: Run TX from the I/O thread

[Avi Kivity <avi@redhat.com>]

(copying Anthony this time)

Mark McLoughlin wrote:
> Hi Alex,
>
> On Wed, 2009-01-21 at 16:08 -0700, Alex Williamson wrote:
>   
>> This is an attempt to improve the latency of virtio-net while not hurting
>> throughput.  I wanted to try moving packet TX into a different thread
>> so we can quickly return to the guest after it kicks us to send packets
>> out.  I also switched the order of when the tx_timer comes into play, so
>> we can get an inital burst of packets out, then wait for the timer to
>> fire and notify us if there's more to do.  Here's what it does for me
>> (average of 5 runs each, testing to a remote system on a 1Gb network):
>>
>> netperf TCP_STREAM: 939.22Mb/s -> 935.24Mb/s =  99.58%
>> netperf TCP_RR:      2028.72/s -> 3927.99/s  = 193.62%
>> tbench:              92.99MB/s -> 99.97MB/s  = 107.51%
>>
>> I'd be interested to hear if it helps or hurts anyone else.  Thanks,
>>     
>
> Avi and I went back and forth on this one in great detail before:
>
>   http://www.mail-archive.com/kvm@vger.kernel.org/msg06431.html
>
> Avi's arguments make a lot of sense, but looking over those patches
> again now, I still think that applying them would be a better approach
> than what we have right now.
>   

I can go with that.  Anthony, should I wait for a qemu iothread so it
can go upstream directly?

-- 
error compiling committee.c: too many arguments to function