On Fri, Jun 05, 2020 at 06:03:38PM +0800, Jason Wang wrote: > > On 2020/6/5 上午12:49, Michael S. Tsirkin wrote: > > > > 2. Second argument to translate_desc is a GPA, isn't it? > > > No, it's IOVA, this function will be called only when IOTLB is enabled. > > > > > > Thanks > > Right IOVA. Point stands how does it make sense to cast > > a userspace pointer to an IOVA? I guess it's just > > because it's talking to qemu actually, so it's abusing > > the notation a bit ... > > > > Yes, but the issues are: > > 1) VHOST_SET_VRING_ADDR is used for iotlb and !iotlb > 2) so did the memory accessors > > Unless we differ separate IOTLB datapath out, there's probably not easy to > have another notation. > > Thanks
With the batching/format independence rework, it might be possible to separate that down the road. We'll see. -- MST