Re: [patch V5 16/26] signal: Replace resched_timer logic

[Thomas Gleixner <tglx@linutronix.de>]

On Tue, Oct 29 2024 at 17:22, Thomas Gleixner wrote:
> On Tue, Oct 29 2024 at 16:56, Frederic Weisbecker wrote:
>>> @@ -568,10 +568,10 @@ static void collect_signal(int sig, stru
>>>  		list_del_init(&first->list);
>>>  		copy_siginfo(info, &first->info);
>>>  
>>> -		*resched_timer = (first->flags & SIGQUEUE_PREALLOC) &&
>>> -				 (info->si_code == SI_TIMER);
>>> -
>>> -		__sigqueue_free(first);
>>> +		if (unlikely((first->flags & SIGQUEUE_PREALLOC) && (info->si_code == SI_TIMER)))
>>> +			*timer_sigq = first;
>>> +		else
>>> +			__sigqueue_free(first);
>>
>> So this isn't calling __sigqueue_free() unconditionally anymore. What if
>> the timer has been freed already, what is going to free the sigqueue?
>
> __sigqueue_free() does not free timers marked with SIGQUEUE_PREALLOC.
>
> sigqueue_free() takes care of that, which is invoked from
> posixtimer_free_timer(). It clears SIGQUEUE_PREALLOC and if it is queued
> it lets it pending and delivery will free it.
>
> That's not any different from before this change.
>
> Though thinking more about it. As we drop the signal in that case
> anyway, we could remove it from pending in sigqueue_free() directly. Let
> me look into that.

Hrm. That requires to do partially what collect_signal() does to keep
the sigset correct. I'm not sure it's worth the trouble.

With the subsequent changes which embed the sigqueue into the timer
itself, the current flow is simple as it has just one place where it
drops the reference count which was acquired when queueing the timer,
i.e. in the delivery path.

Thanks,

        tglx