On Wed, May 22, 2019 at 03:25:32PM +0300, Shamir Rabinovitch wrote:
> On Mon, May 20, 2019 at 12:18:40PM +0300, Leon Romanovsky wrote:
> > On Mon, May 20, 2019 at 10:53:20AM +0300, Shamir Rabinovitch wrote:
> > > In shared object model ib_pd can belong to 1 or more ib_ucontext.
> > > Fix the nldev code so it could report multiple context ids.
> > >
> > > Signed-off-by: Shamir Rabinovitch <shamir.rabinovitch@xxxxxxxxxx>
> > > ---
>
> [...]
>
> > > + if (!rdma_is_kernel_res(res)) {
> > > + pd_context(pd, &pd_context_ids);
> > > + list_for_each_entry(ctx_id, &pd_context_ids, list) {
> > > + if (nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_CTXN,
> > > + ctx_id->id))
> >
> > Did it work? You are overwriting RDMA_NLDEV_ATTR_RES_CTXN entry in the
> > loop. You need to add RDMA_NLDEV_ATTR_RES_CTX and
> > RDMA_NLDEV_ATTR_RES_CTX_ENTRY to include/uapi/rdma_netlink.h and
> > open nested table here (inside of PD) with list of contexts.
> >
> > > + goto err;
> > > + }
>
> Hi Leon,
>
> Just to clarify the above nesting...
>
> Do you expect the below NL attribute nesting in case of shared pd dump?
>
> RDMA_NLDEV_ATTR_RES_CTX
> RDMA_NLDEV_ATTR_RES_CTX_ENTRY
> RDMA_NLDEV_ATTR_RES_CTXN #1
> RDMA_NLDEV_ATTR_RES_CTXN #2
> ...
> RDMA_NLDEV_ATTR_RES_CTXN #N
>
>
> I tried this and rdmatool reported:
>
> [root@qemu-fc29 iproute2]# rdma/rdma res show pd dev mlx4_0
> dev mlx4_0 pdn 0 local_dma_lkey 0x8000 users 4 comm [ib_core]
> dev mlx4_0 pdn 1 local_dma_lkey 0x8000 users 4 comm [ib_core]
> error: Operation not supported
>
> Is this the expected behaviour from unmodified latest rdmatool?
>
> Thanks
Leon,
I tried this nesting (which make more sense to me) and results are the
same as above.
RDMA_NLDEV_ATTR_RES_CTX
RDMA_NLDEV_ATTR_RES_CTX_ENTRY
RDMA_NLDEV_ATTR_RES_CTXN
RDMA_NLDEV_ATTR_RES_CTX_ENTRY
RDMA_NLDEV_ATTR_RES_CTXN
...
Which is the nesting you expect ?
Is it OK that we get the rdma tool "error: Operation not supported" ?
Thanks
