On Tue, Feb 27, 2018 at 1:05 PM, Jason Gunthorpe <jgg@xxxxxxxx> wrote: > On Fri, Feb 23, 2018 at 06:59:08AM +0000, Parav Pandit wrote: > >> I am adding following APIs. >> struct ib_gid_attr* ib_get_cached_gid_ref(device, port, index); >> ib_put_cached_gid_ref(); >> >> So ib_get_gid_context_by_index() API makes sense to call after calling ib_get_cached_gid_ref(). >> This ensure that GID won't get deleted while ib_get_gid_context_by_index() is invoked. > > Considering this need in the drivers I wonder if we shouldn't be > hoisting the context pointer into the ib_gid_attr. > > Selvin, is it OK to wait? Your original patch has the same basic > racing issue Parav describes, but it looks like it just malfunctions > not crashes? With the current stack, scheme used by bnxt_re to get the hw_gid index works in most of the cases. So i am okay to wait for Parav's changes. > > 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
