Re: [PATCH v3] eventpoll: Fix priority inversion problem

[Xi Ruoyao <xry111@xry111.site>]

[-- Attachment #1: Type: text/plain, Size: 2307 bytes --]

On Thu, 2025-07-10 at 10:32 +0200, Nam Cao wrote:
> On Thu, Jul 10, 2025 at 02:54:06PM +0800, Xi Ruoyao wrote:
> > On Thu, 2025-07-10 at 08:21 +0200, Nam Cao wrote:
> > > I am curious if Gnome is using some epoll options which are unused on my
> > > system.
> > 
> > > I presume you can still access dmesg despite the freeze. Do you mind
> > > running the below patch, let me know what's in your dmesg? It may help
> > > identifying that code path.
> > 
> > Attached the system journal (dmesg was truncated due to too many lines).
> > I guess the relevant part should be between line 6947 ("New session 2 of
> > user xry111") and line 8022 ("start operation timed out. Terminating").
> 
> Thanks! I have an idea..
> 
> Looking at the boot log you sent, I noticed some time gap immediately after
> EPOLL_CTL_DEL.
> 
> So I looked at EPOLL_CTL_DEL again, and noticed something that could
> explain your timed out issue:
> 
>   1. EPOLL_CTL_DEL may need to temporarily remove the entire event list.
> 
>   2. While the above is happening, another task may do epoll_wait(). It sees
>      nothing in the event list, and goes to sleep.
> 
>   3. EPOLL_CTL_DEL is now finished and puts the items back into the event
>      list. However, the task from (2.) is not woken up, therefore it keep
>      sleeping despite there are events available.
> 
> If this is really what causing you problem, the below patch should fix it:
> 
> 
> diff --git a/fs/eventpoll.c b/fs/eventpoll.c
> index 895256cd2786..a8fb8ec51751 100644
> --- a/fs/eventpoll.c
> +++ b/fs/eventpoll.c
> @@ -813,8 +813,13 @@ static bool __ep_remove(struct eventpoll *ep, struct epitem *epi, bool force)
>  				put_back_last = n;
>  			__llist_add(n, &put_back);
>  		}
> -		if (put_back_last)
> +		if (put_back_last) {
>  			llist_add_batch(put_back.first, put_back_last, &ep->rdllist);
> +
> +			/* borrow the memory barrier from llist_add_batch() */
> +			if (waitqueue_active(&ep->wq))
> +				wake_up(&ep->wq);
> +		}
>  	}
>  
>  	wakeup_source_unregister(ep_wakeup_source(epi));

It didn't work :(.

I've attached a new log.  This time to reduce noise I didn't make any
keyboard/mouse input until "start operation timed out."

-- 
Xi Ruoyao <xry111@xry111.site>

[-- Attachment #2: log.gz --]
[-- Type: application/gzip, Size: 59246 bytes --]