Re: [kvm-devel] [PATCH 1/5] Support more than 3.5GB with virtio (v3)

[Anthony L. <ali...@us...>]

Avi Kivity wrote:
> Anthony Liguori wrote:
>> We're pretty sloppy in virtio right now about phys_ram_base 
>> assumptions.  This
>> patch is an incremental step between what we have today and a full 
>> blown DMA
>> API.  I backported the DMA API but the performance impact was not 
>> acceptable
>> to me. There's only a slight performance impact with this particular 
>> patch.
>>
>> Since we're no longer assuming guest physical memory is contiguous, 
>> we need
>> a more complex way to validate the memory regions than just checking 
>> if it's
>> within ram_size.
>>   
>
> Applied patches 1-2.  Since patch 4 is under contention on qemu-devel, 
> and 3 and 5 depend on it, I'd like to get the can_receive semantic 
> change accepted first.

I'll send it upstream, but I think it's much less of a divergence than 
the current virtio_net_poll hacks.

Regards,

Anthony Liguori