[Qemu-devel] Re: Re: [RFC][PATCH] performance improvement for windows guests, running on top of virtio block device

["Michael S. Tsirkin" <mst@redhat.com>]

On Mon, Jan 11, 2010 at 08:37:10AM -0600, Anthony Liguori wrote:
> On 01/11/2010 08:29 AM, Avi Kivity wrote:
>> On 01/11/2010 03:49 PM, Anthony Liguori wrote:
>>>> So instead of disabling notify while requests are active we might want
>>>> to only disable it while we are inside virtio_blk_handle_output.
>>>> Something like the following minimally tested patch:
>>>
>>>
>>> I'd suggest that we get even more aggressive and install an idle  
>>> bottom half that checks the queue for newly submitted requests.  If  
>>> we keep getting requests submitted before a new one completes, we'll  
>>> never take an I/O exit.
>>>
>>
>> That has the downside of bouncing a cache line on unrelated exits.
>
> The read and write sides of the ring are widely separated in physical  
> memory specifically to avoid cache line bouncing.
>
>>   It probably doesn't matter with qemu as it is now, since it will  
>> bounce qemu_mutex, but it will hurt with large guests (especially if  
>> they have many rings).
>>
>> IMO we should get things to work well without riding on unrelated  
>> exits, especially as we're trying to reduce those exits.
>
> A block I/O request can potentially be very, very long lived.  By  
> serializing requests like this, there's a high likelihood that it's  
> going to kill performance with anything capable of processing multiple  
> requests.
>
> OTOH, if we aggressively poll the ring when we have an opportunity to,  
> there's very little down side to that and it addresses the serialization  
> problem.
>
>>> The same approach is probably a good idea for virtio-net.
>>
>> With vhost-net you don't see exits.
>
> The point is, when we've disabled notification, we should poll on the  
> ring for additional requests instead of waiting for one to complete  
> before looking at another one.
>
> Even with vhost-net, this logic is still applicable although potentially  
> harder to achieve.
>
> Regards,
>
> Anthony Liguori
>

vhost net does this already: it has a mode where it poll when skbs have
left send queue: for tap this is when they have crossed the bridge, for packet
socket this is when they have been transmitted.

-- 
MST