Re: BUG_ON(!newowner) in fixup_pi_state_owner()

[Thomas Gleixner]

On Wed, Nov 04 2020 at 12:17, Thomas Gleixner wrote:
> On Wed, Nov 04 2020 at 11:24, Thomas Gleixner wrote:
>> On Wed, Nov 04 2020 at 08:42, Mike Galbraith wrote:
>>> On Wed, 2020-11-04 at 01:56 +0100, Mike Galbraith wrote:
>>> --- a/kernel/futex.c
>>> +++ b/kernel/futex.c
>>> @@ -2383,7 +2383,18 @@ static int fixup_pi_state_owner(u32 __us
>>>  		 * Since we just failed the trylock; there must be an owner.
>>>  		 */
>>>  		newowner = rt_mutex_owner(&pi_state->pi_mutex);
>>> -		BUG_ON(!newowner);
>>> +
>>> +		/*
>>> +		 * Why? Because I know what I'm doing with these beasts?  Nope,
>>> +		 * but what the hell, a busy restart loop let f_boosted become
>>> +		 * owner, so go for it. Box still boots, works, no longer makes
>>> +		 * boom with fbomb_v2, and as an added bonus, didn't even blow
>>> +		 * futextests all up.  Maybe it'll help... or not, we'll see.
>>> +		 */
>>> +		if (unlikely(!newowner)) {
>>> +			err = -EAGAIN;
>>> +			goto handle_err;
>>
>> Yes, that cures it, but does not really explain why newowner is
>> NULL. Lemme stare more.
>
> Aside of that it's going to create inconsistent state in the worst
> case. There is something really fishy in the trace Gratian provided....

Ok. With some more info added to the trace I could convince myself, that
Gratians trace is not so fishy at all and that Mike's fix is actually
the right thing to do.

I'll post that with a proper comment and changelog. Mike, can I add your
Signed-off-by to the thing?