Re: [Qemu-devel] [RFC v4 00/58] Memory API

[Anthony Liguori]

On 07/20/2011 01:10 AM, Sasha Levin wrote:
> On Tue, 2011-07-19 at 21:53 -0500, Anthony Liguori wrote:
> > QEMU does use it and it's quite important.  Coalesced MMIO is really
> > about write caching MMIO exits.  It only works with devices that have
> > registers where writing has no side effects.  Moreover, it only really
> > works well when there are lots and lots of writes to these registers
> > simultaneously.
> >
> > Couple that with the fact that the buffer is a fixed size and it's
> > really not flexible enough to be useful for a wide variety of devices.
> >
> > But for VGA planar mode writes, it works wonders.  It would be terrible
> > to totally lose it.  That said, I'm not at all convinced it's useful for
> > much other than VGA planar mode.
>
> Why was the coalesced approach taken in the first place? When I tried
> using it for VGA in /tools/kvm it just seemed to me like a builtin
> virtio-memory transport.

I'm not really sure what you mean by that.  Coalesced I/O is old.  It 
predates the ioeventfd.

I think using a pipe/socket makes a bit more sense.

Regards,

Anthony Liguori

> Thats why I think planar VGA would be fine if we deprecate coalesced
> mmio in favor of either socket ioeventfds or a new virtio-memory device.