Re: [PATCH net 1/2] net: pse-pd: disable IRQ before freeing PI data in unregister

[Carlo Szelinsky <github@szelinsky.de>]

Hi Simon,

Thanks, this all makes sense.

> > Did I miss a path where the worker reaches pcdev->pi?
>
> I am concerned that this can occur if pse_send_ntf_worker()
> calls pse_control_put(). In which case __pse_control_release()
> may run, which accesses psec->pcdev->pi[psec->id].admin_state_enabled.

You're right, I was wrong. The worker does reach pcdev->pi through
pse_control_put() -> __pse_control_release(). For v2 I'll move
cancel_work_sync(&pcdev->ntf_work) before pse_release_pis() and fix
the patch 1 commit message, which wrongly claims the worker never
touches pcdev->pi.

> I wonder if you may have missed the AI-generated review of it that I
> forwarded.

I did, thanks for the hint. I went over it again more carefully
now. :-) Both [High] points seem to be valid, I'll reply to them inline 
on the 2/2 review thread so the discussion stays where it was raised.

> For these last three pre-existing items I think it's best to handle
> them separately.

Agreed, I'll send those on their own later.

Thanks,
Carlo