RDMA/rxe: Do not NULL deref on debugging failure path
authorJason Gunthorpe <jgg@nvidia.com>
Tue, 22 Nov 2022 19:37:39 +0000 (15:37 -0400)
committerJason Gunthorpe <jgg@nvidia.com>
Tue, 22 Nov 2022 19:53:02 +0000 (15:53 -0400)
Correct the mistake, mr is obviously NULL in this code path.

Fixes: 2778b72b1df0 ("RDMA/rxe: Replace pr_xxx by rxe_dbg_xxx in rxe_mr.c")
Link: https://lore.kernel.org/r/Y3eeJW0AdyJYhYyQ@kili
Reported-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
drivers/infiniband/sw/rxe/rxe_mr.c

index b1423000e4bcdad218cc947306b3dfdc2c47fa36..b7c9ff1ddf0e14a15ca273838a23916c38c25f56 100644 (file)
@@ -519,7 +519,7 @@ int rxe_invalidate_mr(struct rxe_qp *qp, u32 key)
 
        mr = rxe_pool_get_index(&rxe->mr_pool, key >> 8);
        if (!mr) {
-               rxe_dbg_mr(mr, "No MR for key %#x\n", key);
+               rxe_dbg_qp(qp, "No MR for key %#x\n", key);
                ret = -EINVAL;
                goto err;
        }