[PATCH v2] drm/omap: force runtime PM suspend on system suspend

[Tony Lindgren]

* Tomi Valkeinen <tomi.valkeinen at ti.com> [200618 14:03]:
> On 18/06/2020 16:16, Tony Lindgren wrote:
> > * Tomi Valkeinen <tomi.valkeinen at ti.com> [200618 09:53]:
> > > Use SET_LATE_SYSTEM_SLEEP_PM_OPS in DSS submodules to force runtime PM
> > > suspend and resume.
> > > 
> > > We use suspend late version so that omapdrm's system suspend callback is
> > > called first, as that will disable all the display outputs after which
> > > it's safe to force DSS into suspend.
> > 
> > This looks like a nice way of doing it, would be good to have as a
> > regression fix for the -rc cycle:
> > 
> > Acked-by: Tony Lindgren <tony at atomide.com>
> 
> Thanks. Did you get the full suspend working? I still can't resume, unless I
> use pm_test to avoid the full suspend. So I'm still not sure if things
> really work.

I just tested my fixes branch at commit 80bf72598663 ("ARM: dts: am5729:
beaglebone-ai: fix rgmii phy-mode") with your patch and HDMI cable
attached and suspend and resume works for me:

Tested-by: Tony Lindgren <tony at atomide.com>

I did not have luck the last time I tried am437x-sk-evm, but that
seems like some separate regression that's been around for awhile.
I recall testing it successfully last fall at some point.

Regards,

Tony