> > enum verbs_context_mask {
> > VERBS_CONTEXT_XRCD = 1 << 0,
> > - VERBS_CONTEXT_RESERVED = 1 << 1
> > + VERBS_CONTEXT_SRQ = 1 << 1,
> > + VERBS_CONTEXT_RESERVED = 1 << 2
> > };
>
> Why is _RESERVED being re-numbered here? That worries me..
>
> For that matter, what is it for?
I called it reserved, you can call it max. It's to let libibverbs guard
against vendor libraries built against future versions of the library, but run
against an older one, using something like this:
if (mask >= VERBS_CONTEXT_RESERVED)
abort...
> Frankly, I would ditch mask for verbs_context op members and other
> structures that are size based. What does it mean if the size includes
> get_srq_num, but VERBS_CONTEXT_SRQ is 0 and *get_srq_num is
> null/non-null?
These mask values indicate to libibverbs that it can cast from struct ibv_* to
struct verbs_*.
> > diff --git a/src/libibverbs.map b/src/libibverbs.map
> > index c190eb9..00f9062 100644
> > +++ b/src/libibverbs.map
> > @@ -103,4 +103,6 @@ IBVERBS_1.1 {
> >
> > ibv_cmd_open_xrcd;
> > ibv_cmd_close_xrcd;
> > + ibv_cmd_create_srq_ex;
> > +
> > } IBVERBS_1.0;
>
> Hum, so drivers that implement XRC will need to be paired with a new
> verbs... Bit disappointing, did we want to try and address that?
The extensions were added specifically to handle XRC.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
