On Mon, 8 Apr 2019 18:07:48 +0200 Damjan Marion <[email protected]> wrote:
> > On 8 Apr 2019, at 17:35, Stephen Hemminger <[email protected]> > > wrote: > > > > On Sat, 30 Mar 2019 01:01:29 +0100 > > Damjan Marion <[email protected]> wrote: > > > >>> On 30 Mar 2019, at 00:58, Sirshak Das <[email protected]> wrote: > >>> > >>> Apologies, DPDK build failure is not valid. wipe-release fixed it for me. > >>> One help still required would be a sample config for using the rdma-core > >>> instead of DPDK would like see the performance difference. > >> > >> rdma-core is work in progress, today it functionally work but you cannot > >> expect great performance. > >> We are going to add direct descriptor ring access for mlx5 similar to what > >> dpdk does. > >> > >> Stay tuned…. > >> > > > > The rdma-core stuff is likely to be a problem on Azure. The Mellanox device > > is always hidden as > > a secondary device behind a synthetic virtual device based on VMBus. There > > are two DPDK different > > ways this is used. One is with vdev_netvsc/failsafe/tap and the other is > > with netvsc PMD. > > In either case, the virtual device expects to see the MLX device show up as > > a VF if enabled. > > > > So unless you want to write native VPP drivers for VMBus as well, I don't > > see how having native > > rdma-core driver will help. It will only get confused. > > Why not :) Is there any spec? > There is not a spec of VMBus network code, you would have to look at examples in DPDK, Linux and FreeBSD to understand it. I also can not guarantee that it won't change in the future. For example, Linux supports the scatter mode of WS2019 but FreeBSD and DPDK do not.
-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#12733): https://lists.fd.io/g/vpp-dev/message/12733 Mute This Topic: https://lists.fd.io/mt/30795618/21656 Group Owner: [email protected] Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [[email protected]] -=-=-=-=-=-=-=-=-=-=-=-
