Re: [PATCH rdma-next v2 7/7] RDMA/nldev: Provide detailed QP information

[Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>]

[-- Attachment #1: Type: text/plain, Size: 1230 bytes --]

On Thu, Jan 11, 2018 at 09:17:56AM -0600, Steve Wise wrote:
> > > >
> > > > > I personally have no plans to fix CMA netlink code and for my
> opinion it
> > > > > should be removed, instead of beating that dead horse.
> > > >
> > > > What is so wrong with it you can't use it anyhow?
> > > >
> > >
> > > We certainly need to be able to get detailed cm_id information.  At
> least
> > > cm_id bound device, cm_id state, port num, qp-id associated with this
> cm_id,
> > > and src/dst ipaddrs/ports.  However, it could be fetched and added to qp
> > > attributes.  Ala iw_cxgb4's debugfs "qps" file:
> > >
> > > [root@stevo3 stevo]# cat /sys/kernel/debug/iw_cxgb4/0000\:04\:00.4/qps
> > > rc qp sq id 1136 rq id 1137 state 1 onchip 0 ep tid 12028 state 7
> > > 172.16.4.3:58725/58725->172.16.4.4:7174/7174
> > >
> > > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> >
> > It can be one option, another is to add new RDMA_RESTRACK_CMA object and
> > expose all relevant information with rdmatool, e.g. "rdma resoruce cma
> link
> > cxgb4/1".
> >
>
> I like the idea of RDMA_RESTRACK_CMA.  I think cm_id's are important objects
> like qps, cqs, pds, and mrs.

Now, it is clear why I want to kill current CMA netlink code :)

>
> Steve.
>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]