[alsa-devel] [PATCH] ASoC: rt274: Disable jack report IRQ with disabling jack

[Mark Brown]

On Wed, Nov 13, 2019 at 02:55:53PM +0100, Harlozinski, Pawel wrote:
> On 11/12/2019 6:10 PM, Mark Brown wrote:
> > On Tue, Nov 12, 2019 at 02:02:37PM +0100, Pawel Harlozinski wrote:

> > >   		/* Disable jack detection */
> > >   		regmap_update_bits(rt274->regmap, RT274_EAPD_GPIO_IRQ_CTRL,
> > >   					RT274_IRQ_EN, RT274_IRQ_DIS);
> > > -
> > > +		disable_irq(rt274->i2c->irq);
> > >   		return 0;

> > Shouldn't the register update above be suppressing interrupts?
> For rt274 disable_irq is also needed, otherwise we're getting flood of irq's
> in case of not loaded machine board.

Through what mechanism is it needed?  If your machine driver is having
an impact on this I'm rather worried.

> > disable_irq() is a bit of a hammer and interferes with things like
> > possible share use.

> This irq should be handled in codec code anyway - control of jack detect
> events from non-codec code is done with set_jack.

The issue isn't that this is in CODEC code, the issue is that it's
usually very worrying to need to explicitly disable and enable an
interrupt at the controller level when the device appears to have
controls that should stop it asserting the interrupt when it's not
wanted.

> Similar solutions for jack report irq enable/disable flow are implemented in
> rt5640 and rt5651.

This may be an indication that those drivers should be improved rather
than that they should be copied.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: not available
URL: <http://mailman.alsa-project.org/pipermail/alsa-devel/attachments/20191113/470713d1/attachment.sig>