Jason Gunthorpe 4aa1615268 RDMA/core: Fix ordering of CQ pool destruction
rxe will hold a refcount on the IB device as long as CQ objects exist,
this causes destruction of a rxe device to hang if the CQ pool has any
cached CQs since they are being destroyed after the refcount must go to
zero.

Treat the CQ pool like a client and create/destroy it before/after all
other clients. No users of CQ pool can exist past a client remove call.

Link: https://lore.kernel.org/r/e8a240aa-9e9b-3dca-062f-9130b787f29b@acm.org
Fixes: c7ff819aefea ("RDMA/core: Introduce shared CQ pool API")
Tested-by: Bart Van Assche <bvanassche@acm.org>
Tested-by: Yi Zhang <yi.zhang@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2020-09-14 15:20:18 -03:00
..
2020-08-28 21:17:56 +02:00
2020-09-04 13:19:19 -07:00
2020-09-13 09:02:59 -07:00
2020-08-02 18:26:51 +03:00
2020-08-15 08:18:22 -07:00
2020-09-12 12:43:58 -07:00
2020-09-04 12:12:39 -07:00
2020-09-13 09:02:59 -07:00
2020-09-11 09:49:23 +10:00
2020-08-26 10:44:15 -07:00
2020-08-29 12:37:00 -07:00
2020-08-05 19:24:27 -07:00
2020-09-03 08:22:16 -07:00
2020-08-28 10:15:33 -07:00
2020-09-13 08:52:21 -07:00
2020-09-04 12:41:55 +02:00
2020-08-14 16:00:09 -07:00
2020-09-11 11:25:55 -07:00
2020-09-08 11:42:58 -07:00
2020-09-11 11:35:55 -07:00
2020-09-08 11:42:58 -07:00
2020-08-26 10:58:20 -07:00
2020-09-08 17:50:58 +02:00
2020-09-06 09:59:27 -07:00