lockdep: Fix a race between /proc/lock_stat and module unload
authorPeter Zijlstra <peterz@infradead.org>
Tue, 2 Jun 2015 10:50:13 +0000 (12:50 +0200)
committerIngo Molnar <mingo@kernel.org>
Sun, 7 Jun 2015 13:46:30 +0000 (15:46 +0200)
commitcee34d88cabd1ba5fc93e09b5b12232bc9338c7c
tree278d17b06e20285f8c2822d724ecc41a69b27e9d
parent37ef1647b7f73d4ff4c7993984599b6c4f26443a
lockdep: Fix a race between /proc/lock_stat and module unload

The lock_class iteration of /proc/lock_stat is not serialized against
the lockdep_free_key_range() call from module unload.

Therefore it can happen that we find a class of which ->name/->key are
no longer valid.

There is a further bug in zap_class() that left ->name dangling. Cure
this. Use RCU_INIT_POINTER() because NULL.

Since lockdep_free_key_range() is rcu_sched serialized, we can read
both ->name and ->key under rcu_read_lock_sched() (preempt-disable)
and be assured that if we observe a !NULL value it stays safe to use
for as long as we hold that lock.

If we observe both NULL, skip the entry.

Reported-by: Jerome Marchand <jmarchan@redhat.com>
Tested-by: Jerome Marchand <jmarchan@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20150602105013.GS3644@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
kernel/locking/lockdep.c
kernel/locking/lockdep_proc.c