On Thu, Apr 28, 2016 at 11:25:35PM +0000, Weiny, Ira wrote: > Mellanox how hard will it be to change your drivers to that model? > I'm not even sure how the detection of Link Layer works any more. Hmm, I vaugely remember looking into this and thinking the mlx drivers already did this? It may even be that IB and rocee can do APM and could perhaps be part of the same struct ib_device without breaking the world. I actually have no idea. > But all that is different from the qib/hfi case where we have 1 port > with 2 protocols on it. If we are going to add PSM into the core > then I think it is _semantically_ appropriate for users to be able > to query for the protocols supported on a port and get back more > than 1. That doesn't make sense, the issue here is that we have a variety of verbs 'flavours'. PSM is not a verbs flavour. *If* PSM gets a kAPI (nobody is talking about doing this?) *and* gains multiple incompatible flavours (such as OPA and IB?) then it will need a unique set of cap tests and restrictions on which physical ports can be used together. Jason -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
