Jason Gunthorpe <jguntho...@obsidianresearch.com> wrote:
> The commit comment states the goal is to have new apps see
> IBV_LINK_LAYER_UNSPECIFIED for all possible combinations of old stuff
> underneath.
> [...] The problem is with the userspace ABI out of libibverbs. Current
> libibverbs does not zero its ib_port_attr structure before returning.
> Build this with Or's patch applied, then run it against libibverbs 1.1.5
> struct ibv_port_attr attr;
> attr.link_layer = 100;
> ibv_query_port(..,&attr);
> assert(attr.link_layer == IBV_LINK_LAYER_UNSPECIFIED); // Fails!
impossible, see below why
> That is an ABI breaking issue that is not dealt with by Or's patch - couple
> choices:
> 1) Require and document callers using ibv_query_port to zero attr
> before hand when using link_layer
> 2) Crank the symbol version on ibv_query_port so dynamic linking fails
> 3) Inline the memset in ibverbs.h:
YES, #3 is what the patch does, and it follows your suggestion!!
+static inline int ___ibv_query_port(struct ibv_context *context,
+ uint8_t port_num,
+ struct ibv_port_attr *port_attr)
+{
+ port_attr->link_layer = IBV_LINK_LAYER_UNSPECIFIED;
+ port_attr->pad = 0;
+
+ return context->ops.query_port(context, port_num, port_attr);
+}
+
+#define ibv_query_port(context, port_num, port_attr) \
+ ___ibv_query_port(context, port_num, port_attr)
+
So we are okay now, correct?
Or.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
