[Qemu-devel] Re: [PATCH 0/4] Block DMA helpers (v2)

[Anthony Liguori]

Avi Kivity wrote:
> Anthony Liguori wrote:
> > BTW, what are your plans for virtio?  
>
> To hope someone else does it.
>
> > I'd reckon that just trying to map the full SG list and failing 
> > gracefully if unable would be sufficient.  You could also convert the 
> > element to contain a QEMUSGList and then make use of the IO helpers 
> > in virtio-blk and virtio-net.  It just gets a little funky for 
> > simpler things like virtio-console and virtio-balloon.
>
> One way is to convert the generic transport layer to use two 
> QEMUSGLists (one for incoming, one for outgoing) and leave it at 
> that.  virtio-blk could use the block dma helpers, virtio-net could do 
> something else, and we'd add a qemu_sglist_to_buffer(sg, offset, len, 
> buf) and qemu_sglist_from_buffer(sg, offset, len, buf) to allow 
> low-bandwidth easy access to virtio data.

Yeah, that's probably the best way of doing it.

> This requires that the generic layer be able to tell where a request 
> ends; don't know if that's the case now.

Maybe create a sub QEMUSGList from another QEMUSGList.  The other option 
is passing an offset and size to anything that takes a QEMUSGList.  You 
could potentially get smart with how sub SG lists were managed by making 
them just store an internal offset/size.  Although things start to get 
overly complex at some stage.

Regards,

Anthony Liguori