Revision 21a446cf186570168b7281b154b1993968598aca authored by Trond Myklebust on 05 November 2018, 16:10:50 UTC, committed by Trond Myklebust on 12 November 2018, 21:39:13 UTC
If we exit the NFSv4 state manager due to a umount, then we can end up
leaving the NFS4CLNT_MANAGER_RUNNING flag set. If another mount causes
the nfs4_client to be rereferenced before it is destroyed, then we end
up never being able to recover state.
Fixes: 47c2199b6eb5 ("NFSv4.1: Ensure state manager thread dies on last ...")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Cc: stable@vger.kernel.org # v4.15+
1 parent ccda4af
| File | Mode | Size |
|---|---|---|
| Makefile | -rw-r--r-- | 1.2 KB |
| lockdep.c | -rw-r--r-- | 108.7 KB |
| lockdep_internals.h | -rw-r--r-- | 5.7 KB |
| lockdep_proc.c | -rw-r--r-- | 16.7 KB |
| lockdep_states.h | -rw-r--r-- | 207 bytes |
| locktorture.c | -rw-r--r-- | 28.3 KB |
| mcs_spinlock.h | -rw-r--r-- | 3.7 KB |
| mutex-debug.c | -rw-r--r-- | 2.8 KB |
| mutex-debug.h | -rw-r--r-- | 1.1 KB |
| mutex.c | -rw-r--r-- | 36.7 KB |
| mutex.h | -rw-r--r-- | 841 bytes |
| osq_lock.c | -rw-r--r-- | 5.6 KB |
| percpu-rwsem.c | -rw-r--r-- | 5.1 KB |
| qrwlock.c | -rw-r--r-- | 2.8 KB |
| qspinlock.c | -rw-r--r-- | 16.3 KB |
| qspinlock_paravirt.h | -rw-r--r-- | 16.1 KB |
| qspinlock_stat.h | -rw-r--r-- | 7.9 KB |
| rtmutex-debug.c | -rw-r--r-- | 4.6 KB |
| rtmutex-debug.h | -rw-r--r-- | 1.4 KB |
| rtmutex.c | -rw-r--r-- | 50.1 KB |
| rtmutex.h | -rw-r--r-- | 1.2 KB |
| rtmutex_common.h | -rw-r--r-- | 4.2 KB |
| rwsem-spinlock.c | -rw-r--r-- | 7.7 KB |
| rwsem-xadd.c | -rw-r--r-- | 19.9 KB |
| rwsem.c | -rw-r--r-- | 4.5 KB |
| rwsem.h | -rw-r--r-- | 4.3 KB |
| semaphore.c | -rw-r--r-- | 7.3 KB |
| spinlock.c | -rw-r--r-- | 9.5 KB |
| spinlock_debug.c | -rw-r--r-- | 5.5 KB |
| test-ww_mutex.c | -rw-r--r-- | 12.9 KB |
Computing file changes ...
