Re: [TIP][RFC 4/7] futex: finish_futex_lock_pi()

[Thomas Gleixner <tglx@linutronix.de>]

On Mon, 2 Mar 2009, Darren Hart wrote:
> +	} else {
> +		/* dvhart FIXME: can't we just BUG_ON in this case?

No. There is no reason to crash the kernel if this happens. All what
happens is that a userspace application becomes a bit unhappy.

I did not put a WARN_ON there as the stack trace is known, but we
could do a WARN to trigger the kerneloops detector.

> +		 * Paranoia check. If we did not take the lock in the trylock
> +		 * above, then we should not be the owner of the rtmutex,
> +		 * neither the real nor the pending one:
> +		 */
> +		if (rt_mutex_owner(&q->pi_state->pi_mutex) == current)
> +			printk(KERN_ERR "finish_futex_lock_pi: "
> +					"ret = %d pi-mutex: %p "
> +					"pi-state %p\n", ret,
> +					q->pi_state->pi_mutex.owner,
> +					q->pi_state->owner);
> +	}

Thanks,

	tglx