| Message ID | 20210908052928.17375-3-rpearsonhpe@gmail.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | RDMA/rxe: Various bug fixes. | expand |
diff --git a/drivers/infiniband/sw/rxe/rxe_mcast.c b/drivers/infiniband/sw/rxe/rxe_mcast.c index 0ea9a5aa4ec0..1c1d1b53312d 100644 --- a/drivers/infiniband/sw/rxe/rxe_mcast.c +++ b/drivers/infiniband/sw/rxe/rxe_mcast.c @@ -85,7 +85,7 @@ int rxe_mcast_add_grp_elem(struct rxe_dev *rxe, struct rxe_qp *qp, goto out; } - elem = rxe_alloc(&rxe->mc_elem_pool); + elem = rxe_alloc_locked(&rxe->mc_elem_pool); if (!elem) { err = -ENOMEM; goto out;
rxe_mcast_add_grp_elem() in rxe_mcast.c calls rxe_alloc() while holding spinlocks which in turn calls kzalloc(size, GFP_KERNEL) which is incorrect. This patch replaces rxe_alloc() by rxe_alloc_locked() which uses GFP_ATOMIC. This bug was caused by the below mentioned commit and failing to handle the need for the atomic allocate. Fixes: 4276fd0dddc9 ("Remove RXE_POOL_ATOMIC") Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> --- v2: rebase on version 5.14 drivers/infiniband/sw/rxe/rxe_mcast.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)