locking/mutex: Fix lockdep_assert_held() fail
authorPeter Zijlstra <peterz@infradead.org>
Tue, 17 Jan 2017 15:06:09 +0000 (16:06 +0100)
committerIngo Molnar <mingo@kernel.org>
Mon, 30 Jan 2017 10:42:59 +0000 (11:42 +0100)
commitb9c16a0e1f733c97e48798b2a9362c485bb3b731
treebe96b0e158eca9af36f084b6344900212f4c58ba
parent4009f4b3a9d8b74547269f293e6a920adf278996
locking/mutex: Fix lockdep_assert_held() fail

In commit:

  659cf9f5824a ("locking/ww_mutex: Optimize ww-mutexes by waking at most one waiter for backoff when acquiring the lock")

I replaced a comment with a lockdep_assert_held(). However it turns out
we hide that lock from lockdep for hysterical raisins, which results
in the assertion always firing.

Remove the old debug code as lockdep will easily spot the abuse it was
meant to catch, which will make the lock visible to lockdep and make
the assertion work as intended.

Reported-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Nicolai Haehnle <Nicolai.Haehnle@amd.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 659cf9f5824a ("locking/ww_mutex: Optimize ww-mutexes by waking at most one waiter for backoff when acquiring the lock")
Link: http://lkml.kernel.org/r/20170117150609.GB32474@worktop
Signed-off-by: Ingo Molnar <mingo@kernel.org>
kernel/locking/mutex-debug.h
kernel/locking/mutex.c
kernel/locking/mutex.h