Re: [PATCH 5/5] ioeventfd: Introduce KVM_IOEVENTFD_FLAG_SOCKET

[Avi Kivity <avi@redhat.com>]

On 07/14/2011 03:37 PM, Pekka Enberg wrote:
> Hi Avi,
>
> On Thu, Jul 14, 2011 at 2:54 PM, Avi Kivity<avi@redhat.com>  wrote:
> >>  I don't think it needs to be faster for *significant number* of users
> >>  but yes, I completely agree that we need to make sure KVM gains more
> >>  than the costs are.
> >
> >  Significant, for me, means it's measured in a percentage, not as digits on
> >  various limbs.  2% is a significant amount of users.  5 is not.
>
> Just to make my position clear: I think it's enough that there's one
> user that benefits as long as complexity isn't too high and I think
> x86/Voyager is a pretty good example of that. So while you can argue
> *for* complexity if there are enough users, when there's only few
> users, it's really about whether a feature adds significant complexity
> or not.

Everything adds complexity.  And I think x86/voyager was a mistake.

> >>  We want to use 8250 emulation instead of virtio-serial because it's
> >>  more compatible with kernel debugging mechanisms. Also, it makes
> >>  debugging virtio code much easier when we don't need to use virtio to
> >>  deliver console output while debugging it. We want to make it fast so
> >>  that we don't need to switch over to another console type after early
> >>  boot.
> >>
> >>  What's unreasonable about that?
> >
> >  Does virtio debugging really need super-fast serial?  Does it need serial at
> >  all?
>
> Text mode guests should be super-fast, not virtio debugging. We want
> to use 8250 instead of virtio serial or virtio console (which we
> support, btw) because it's more compatible with Linux. Debugging
> virtio with 8250 has turned out to be useful in the past.

Use virtio-console when you're in production (it will be much much 
faster than socket-mmio 8250), and 8250 when debugging.

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