Re: [PATCH v3 2/2] usb: typec: tcpci: set ALERT_MASK register after devm_request_threaded_irq()

[Xu Yang <xu.yang_2@nxp.com>]

On Wed, Dec 18, 2024 at 07:41:49AM +0100, Emanuele Ghidoli wrote:
> 
> 
> On 18/12/2024 06:31, Xu Yang wrote:
> > On Tue, Dec 17, 2024 at 09:29:36PM +0100, Emanuele Ghidoli wrote:
> >>
> >>
> >> On 17/12/2024 17:35, Francesco Dolcini wrote:
> >>> On Tue, Dec 17, 2024 at 05:12:08PM +0800, Xu Yang wrote:
> >>>> With edge irq support, the ALERT event may be missed currently. The reason
> >>>> is that ALERT_MASK register is written before devm_request_threaded_irq().
> >>>> If ALERT event happens in this time gap, it will be missed and ALERT line
> >>>> will not recover to high level. However, we don't meet this issue with
> >>>> level irq. To avoid the issue, this will set ALERT_MASK register after
> >>>> devm_request_threaded_irq() return.
> >>>>
> >>>> Fixes: 77e85107a771 ("usb: typec: tcpci: support edge irq")
> >>>> Cc: stable@vger.kernel.org
> >>>> Signed-off-by: Xu Yang <xu.yang_2@nxp.com>
> >>>
> >>> I had an offline chat with a Emanuele (in Cc:) that worked on this a few
> >>> weeks ago and he remember that he already tried a similar approach, but
> >>> for some reason he did not work.
> >>>
> >>> He should be able to try this patch in a few days, but with the upcoming
> >>> winter holidays he might not be super responsive.
> >>>
> >>> I wonder if we could wait a little before merging this to allow this
> >>> testing to happen. Or maybe you can just test if this is working on your
> >>> setup using edge interrupts (you would need to use only one TCPCI, for
> >>> the test).
> >>>
> >>> Francesco
> >>>
> >>
> >> Hi all,
> >>
> >> I was curious, so I tested the two patches. I can confirm that if both are applied, 
> >> edge interrupts still work correctly.
> >> However, with only the first patch applied, it does not work.
> > 
> > Yes. This is an expected results. So could this be regarded as a tested-by?
> > I have tested edge irq before sending out the patches too.
> > 
> > Thanks,
> > Xu Yang
> 
> Hello Xu,
> I confirmed that the first patch introduces a regression, 
> so I agree with Francesco and Dan about merging the two patches.

Okay.

> 
> Anyway, I tested it.
> 
> Tested-by: Emanuele Ghidoli <emanuele.ghidoli@toradex.com>

Good to know.

Thanks,
Xu Yang