lockdep: Fix wrong assumption in match_held_lock
authorPeter Zijlstra <a.p.zijlstra@chello.nl>
Fri, 5 Aug 2011 12:26:17 +0000 (14:26 +0200)
committerIngo Molnar <mingo@elte.hu>
Tue, 9 Aug 2011 09:57:35 +0000 (11:57 +0200)
commit80e0401e35410a69bfae05b454db8a7187edd6b8
treed43eb51e0ca861e18284a85ccb790b321ed08320
parent2f84dd70916ccadd25e94d28363182a978f569b6
lockdep: Fix wrong assumption in match_held_lock

match_held_lock() was assuming it was being called on a lock class
that had already seen usage.

This condition was true for bug-free code using lockdep_assert_held(),
since you're in fact holding the lock when calling it. However the
assumption fails the moment you assume the assertion can fail, which
is the whole point of having the assertion in the first place.

Anyway, now that there's more lockdep_is_held() users, notably
__rcu_dereference_check(), its much easier to trigger this since we
test for a number of locks and we only need to hold any one of them to
be good.

Reported-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1312547787.28695.2.camel@twins
Signed-off-by: Ingo Molnar <mingo@elte.hu>
kernel/lockdep.c