On Wed, Feb 16, 2022 at 11:42 PM Doug Ledford <dledford@xxxxxxxxxx> wrote: > > On Wed, Feb 16, 2022 at 9:08 AM Yongji Xie <xieyongji@xxxxxxxxxxxxx> wrote: >> >> On Wed, Feb 16, 2022 at 8:06 PM Leon Romanovsky <leon@xxxxxxxxxx> wrote: >> > >> > On Wed, Feb 16, 2022 at 06:03:29PM +0800, Junji Wei wrote: >> > > >> > > > On Feb 16, 2022, at 5:48 PM, Leon Romanovsky <leon@xxxxxxxxxx> wrote: >> > > > >> > > > On Wed, Feb 16, 2022 at 04:00:53PM +0800, Junji Wei wrote: >> > > > >> > > > <...> >> > > > >> > > >>> >> > > >>> What is the use case for this virtio-rdma? Especially in context of RXE. >> > > >> >> > > >> Hmm... yes, we didn’t find one. In passthrough case we can use RXE directly. >> > > > >> > > > It doesn't sound like a good sales pitch. >> > > >> > > Maybe I misunderstanded what you mean. We mean we didn’t find a user case >> > > for virtio-rdma with passthrough net device. Do you want to know the user >> > > case for our virtio-rdma(RoCE) proposal? >> > >> > Yes, please. >> > >> >> I think one point is: when running RDMA accelerated applications in >> VM, the virtio-rdma solution should get better performance than RXE >> since it has a shorter data path (guest app -> host dpdk, bypass guest >> kernel). > > > What's the security model? Native RDMA has security on a per QP basis. DPDK requires root/CAP_NET_RAW access and allows anything. How are you ensuring that rogue apps can't do bad things with this? > Should this be already protected at the virtio level? I think virtio-net would meet the same problem if we have a malicious guest. And DPDK is under our control, I think we can do some validation if the app violates the RDMA semantics. Thanks, Yongji
