01:08 <leezu> lumag: Thank you for your suggestion on reverting e17af1c9d861dc177e5b56009bd4f71ace688d97! This indeed fixes hot plug detection. I've sent out the patch to revert the commit explaining it's impact on lazor.

02:58 <bamse> lumag: the pwm in pm8916 is just an lpg without pattern and the current sink...

03:00 <bamse> z3ntu: lumag is correct, you can pinmux the mpp to be controlled via the special functions, and for some gpios/mpps one of the special functions is the output of a particluar pwm...

03:07 <bamse> z3ntu: hmm, looks like there was some dts changes that i never got upstream...and it seems i used the internal dtest lines to route the pwm signal to the mpps...

03:08 <bamse> z3ntu: https://github.com/andersson/kernel/commit/786fbb7c93b00272e86e49af85a72d7f4e229811

03:08 <bamse> z3ntu: and i think this is thereby incomplete https://github.com/andersson/kernel/commit/9236c290043a75d16aa16b01f4e120feb2ed22b0

07:40 <z3ntu> bamse: yeah that looks closer to what I think is used on pm8226. I've also seen something regarding current sink on the two MPP pins. Driver support for these dts parts should be present I imagine?

08:07 <aka_[m]> lumag: are you working on these old boards like 8974 in your free time or there are still customers for that?

09:30 <konradybcio> we're just trying to get our hands on perserverance :D

09:36 <lumag> aka_[m], in was Sunday, no customer work allowed :-)

09:38 <aka_[m]> I really regret there is no dragonboard 650

09:39 <aka_[m]> It would fly with someone like you

09:39 <lumag> seriously speaking, I was doing the HDMI PHY rework. And before finishing that I wanted to be sure that we do not break anything. On 8974 the HDMI is in semi-supported case: kernel has everything except the PLL.

09:40 <lumag> So bringing up the old patch from Brian and an old board, sitting in my closed for some time, was quite natural.

09:41 <aka_[m]> Hdmi phy is like thing of the past right?

09:42 <aka_[m]> Now it's all dp

09:48 <lumag> aka_[m], yes.

09:48 <lumag> no more native HDMI after 8996 and 8998

09:50 <lumag> At this moment we have everything in drm/msm/hdmi. Now that the probe order is mostly fixed, we can move that to the proper place - drivers/phy/qualcomm

09:51 <lumag> Doing so would allow us to reuse QMP PHY definitions, streamline some of the PHY code, remove ad-hoc API, etc.

16:46 <bamse> z3ntu: the mpps can be operated in pairs...don't remember the details though :/