Re: [PATCH 1/2] hung_task: Show the blocker task if the task is hung on mutex

[Lance Yang]

On Thu, Feb 20, 2025 at 7:55 AM Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:
>
> On Thu, 20 Feb 2025 07:56:39 +0900
> Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx> wrote:
>
> > > >> this field, we don't need to take lock, though taking the wait_lock may
> > > >> still be needed to examine other information inside the mutex.
> >
> > Do we need to take it just for accessing owner, which is in an atomic?
>
> Updating the task_struct would be in the same location as the blocked_on is
> anyway. I would make it into a wrapper function that is a nop when disabled.
>
> >
> > > > But perhaps if we add a new config option for this feature, we could just
> > > > add the lock that a task is blocked on before it goes to sleep and
> > > > reference that instead. That would be easier than trying to play games
> > > > getting the lock owner from the blocked_on field.
> > >
> > > Yes, it could be a new config option. This will be a useful feature that
> > > I believe most distros will turn it on. Or we may just include that in
> > > the core code without any option.
> >
> > Do we need another option? or just extend DETECT_HUNG_TASK?
>
> DETECT_HUNG_TASK is just that, for detecting hung tasks. This adds more
> information to that, which increases the size of the task_struct not to
> mention adds code in the mutex/rwsem handlers.
>
> I would definitely make it a separate config that may depend on
> DETECT_HUNG_TASK.

Agreed. Making it a separate config option that depends on DETECT_HUNG_TASK
sounds like a reasonable approach. It could help us to identify the
root cause, but it
also adds a bit of extra overhead.

Thanks,
Lance

>
> -- Steve