On Thu, 9 Jul 2015 16:07:47 +0300
"Michael S. Tsirkin" [off-list ref] wrote:

On Thu, Jul 09, 2015 at 02:57:33PM +0200, Paolo Bonzini wrote:

quoted

On 09/07/2015 11:48, Laurent Vivier wrote:

quoted

On 09/07/2015 09:49, Thomas Huth wrote:

quoted

The option for supporting cross-endianness legacy guests in
the vhost and tun code should only be available on systems
that support cross-endian guests.

I'm sure I misunderstand something, but what happens if we use QEMU with
TCG instead of KVM, i.e. a big endian powerpc kernel guest on x86_64
little endian host ?

TCG does not yet support irqfd/ioeventfd, so it cannot be used with vhost.

Paolo

vhost does not require irqfd anymore.  I think ioeventfd actually works
fine though I didn't try, it would be easy to support.

That's an interesting issue, thanks for pointing this out, Laurent! So
do we now rather want to leave everything as it currently is, in case
somebody wants to use vhost-net with a cross-endian TCG guest one day?

Or do we assume that either
a) TCG is so slow anyway that nobody wants to accelerate it with vhost
or
b) TCG vhost likely won't happen that soon so we hope that everybody
will already be using virtio 1.0 at that point in time (with a fixed
endianness) 
?
... then I think we should go on and include this patch.

 Thomas