A couple of the callers assume that ib_post_send() initializes
*bad_send_wr. It doesn't look like we can rely on the ->post_send()
function to initialize it. For example in the i40iw_post_recv()
function and there are some error paths there which don't set it.
Here are the static checker warnings for reference:
net/sunrpc/xprtrdma/verbs.c:1564 rpcrdma_post_recvs() error: uninitialized symbol 'bad_wr'.
net/sunrpc/xprtrdma/svc_rdma_rw.c:350 svc_rdma_post_chunk_ctxt() error: uninitialized symbol 'bad_wr'.
Signed-off-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 8caee90e6dee..0ebacbe5019e 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -3389,6 +3389,8 @@ static inline int ib_post_send(struct ib_qp *qp,
{
const struct ib_send_wr *dummy;
+ if (bad_send_wr)
+ *bad_send_wr = NULL;
return qp->device->post_send(qp, send_wr, bad_send_wr ? : &dummy);
}
--
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
