Message ID | 20170829173444.4289-2-roland@kernel.org (mailing list archive) |
---|---|
State | Accepted |
Headers | show |
On 8/29/2017 1:34 PM, Roland Dreier wrote: > From: Roland Dreier <roland@purestorage.com> > > For RoCE, ib_init_ah_from_wc() can follow the path > > ib_init_ah_from_wc() -> > rdma_addr_find_l2_eth_by_grh() -> > rdma_resolve_ip() > > and rdma_resolve_ip() will sleep in kzalloc() and wait_for_completion(). > > However, developers will not see any warnings if they use ib_init_ah_from_wc() > in an atomic context and test only on IB, because the function doesn't > sleep in that case. > > Add a might_sleep() so that lockdep will catch bugs no matter what hardware is > used to test. > > Signed-off-by: Roland Dreier <roland@purestorage.com> Thanks, will pick this up today too.
diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index b456e3ca1876..84868f76f8b9 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -478,6 +478,8 @@ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num, union ib_gid dgid; union ib_gid sgid; + might_sleep(); + memset(ah_attr, 0, sizeof *ah_attr); ah_attr->type = rdma_ah_find_type(device, port_num); if (rdma_cap_eth_ah(device, port_num)) {