On Tue, Jun 14, 2011 at 11:14:40AM -0600, Eric Blake wrote:
On 06/14/2011 11:04 AM, Daniel P. Berrange wrote:
> On Tue, Jun 14, 2011 at 06:18:17PM +0200, Michal Privoznik wrote:
>> For virtio disks and interfaces, qemu allows users to enable or disable
>> ioeventfd feature. This means, qemu can execute domain code, while
>> another thread waits for I/O event. Basically, in some cases it is win,
>> in some loss. This feature is available via 'asyncio' attribute in disk
>> and interface <driver> element. It accepts 'on' and 'off'.
Leaving this
>> attribute out defaults to hypervisor decision.
>
> I think this is rather misleading. AFAIK, ioeventfd has absolutely
> nothing todo with async I/O.
>
> There is already a completely different QEMU command line flag
> which toggles between the two QEMU async I/O implementations,
> which is set via the 'io=' attribute.
>
> So I don't think we can seriously use 'asyncio' for this new
> feature.
Hmm, we're at a bit of a conflict here; DV didn't like the name
'ioeventfd' as being too hypervisor-specific, and danpb things 'asyncio'
is not quite the right term.
The rest of the patch is fine, but we need to settle on the attribute
name that everyone can live with.
go for 'ioeventfd' it's very specific but that's normal after all,
Daniel
--
Daniel Veillard | libxml Gnome XML XSLT toolkit
http://xmlsoft.org/
daniel(a)veillard.com | Rpmfind RPM search engine
http://rpmfind.net/
http://veillard.com/ | virtualization library
http://libvirt.org/