printk: Fix rq->lock vs logbuf_lock unlock lock inversion
Bu, Yitian [Mon, 18 Feb 2013 12:53:37 +0000 (12:53 +0000)]
commit dbda92d16f8655044e082930e4e9d244b87fde77 upstream.

commit 07354eb1a74d1 ("locking printk: Annotate logbuf_lock as raw")
reintroduced a lock inversion problem which was fixed in commit
0b5e1c5255 ("printk: Release console_sem after logbuf_lock"). This
happened probably when fixing up patch rejects.

Restore the ordering and unlock logbuf_lock before releasing
console_sem.

Signed-off-by: ybu <ybu@qti.qualcomm.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/E807E903FE6CBE4D95E420FBFCC273B827413C@nasanexd01h.na.qualcomm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

kernel/printk.c

index 8212c1a..d37d45c 100644 (file)
@@ -1369,9 +1369,9 @@ static int console_trylock_for_printk(unsigned int cpu)
                }
        }
        logbuf_cpu = UINT_MAX;
+       raw_spin_unlock(&logbuf_lock);
        if (wake)
                up(&console_sem);
-       raw_spin_unlock(&logbuf_lock);
        return retval;
 }