Trond Myklebust d72eb7187d NFSv4: Fix a potential sleep while atomic in nfs4_do_reclaim()
[ Upstream commit c77e22834ae9a11891cb613bd9a551be1b94f2bc ]

John Hubbard reports seeing the following stack trace:

nfs4_do_reclaim
   rcu_read_lock /* we are now in_atomic() and must not sleep */
       nfs4_purge_state_owners
           nfs4_free_state_owner
               nfs4_destroy_seqid_counter
                   rpc_destroy_wait_queue
                       cancel_delayed_work_sync
                           __cancel_work_timer
                               __flush_work
                                   start_flush_work
                                       might_sleep:
                                        (kernel/workqueue.c:2975: BUG)

The solution is to separate out the freeing of the state owners
from nfs4_purge_state_owners(), and perform that outside the atomic
context.

Reported-by: John Hubbard <jhubbard@nvidia.com>
Fixes: 0aaaf5c424c7f ("NFS: Cache state owners after files are closed")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
2019-09-06 10:18:07 +02:00
..
2017-12-20 10:04:56 +01:00
2019-08-06 18:28:27 +02:00
2019-06-11 12:23:53 +02:00
2018-12-17 21:55:12 +01:00
2017-11-30 08:37:20 +00:00
2019-03-23 08:44:21 +01:00
2018-12-21 14:09:50 +01:00
2015-11-10 12:07:22 -08:00
2015-11-16 23:54:45 -08:00
2015-11-16 23:54:45 -08:00
2017-06-14 13:16:24 +02:00
2018-02-16 20:09:43 +01:00
2017-06-14 13:16:24 +02:00
2018-09-09 20:04:36 +02:00