On Thu, Oct 17, 2019 at 11:20:32AM +0200, Richard Leitner wrote:
>
> On 17/10/2019 01:23, Greg Kroah-Hartman wrote:
> > On Wed, Oct 16, 2019 at 11:35:18PM +0100, Mark Brown wrote:
> > > On Wed, Oct 16, 2019 at 03:10:25PM -0700, Greg Kroah-Hartman wrote:
> > > > On Wed, Oct 16, 2019 at 11:00:44PM +0100, Mark Brown wrote:
> > > > > On Wed, Oct 16, 2019 at 02:51:44PM -0700, Greg Kroah-Hartman wrote:
> > > > > > From: Oleksandr Suvorov <oleksandr.suvorov@xxxxxxxxxxx>
> > >
> > > > > > commit 694b14554d75f2a1ae111202e71860d58b434a21 upstream.
> > >
> > > > > > This control mute/unmute the ADC input of SGTL5000
> > > > > > using its CHIP_ANA_CTRL register.
> > >
> > > > > This seems like a new feature and not an obvious candidate for stable?
> > >
> > > > there was a long email from Richard that said:
> > > > Upstream commit 631bc8f0134a ("ASoC: sgtl5000: Fix of unmute
> > > > outputs on probe"), which is e9f621efaebd in v5.3 replaced
> > > > snd_soc_component_write with snd_soc_component_update_bits and
> > > > therefore no longer cleared the MUTE_ADC flag. This caused the
> > > > ADC to stay muted and recording doesn't work any longer. This
> > > > patch fixes this problem by adding a Switch control for
> > > > MUTE_ADC.
> > >
> > > > That's why I took this. If this isn't true, I'll be glad to drop this.
> > >
> > > That's probably not an appropriate fix for stable - it's going to add a
> > > new control which users will need to manually set (or hope their
> > > userspace automatically figures out that it should set for them, more
> > > advanced userspaces like PulseAudio should) which isn't a drop in fix.
> > > You could either drop the backport that was done for zero cross or take
> > > a new patch that clears the MUTE_ADC flag (rather than punting to
> > > userspace to do so), or just be OK with what you've got at the minute
> > > which might be fine given the lack of user reports.
> >
> > Ok, I'll gladly go drop it, thanks!
>
> Mark, thanks for the clarification! I haven't thought of breaking anything
> with the backport as it worked fine for our application.
>
> Greg, just to be sure:
>
> Are you going to drop this patch and revert e9f621efaebd for v5.3?
I dropped this patch.
If the revert is needed, please send that and I will be glad to queue it
up for the next round of releases.
thanks,
greg k-h
