Messages in this thread | | | Subject | Re: [PATCH RFT v1 1/3] drm/panfrost: enable devfreq based the "operating-points-v2" property | From | Robin Murphy <> | Date | Wed, 8 Jan 2020 14:20:21 +0000 |
| |
[ +Sudeep ]
On 08/01/2020 12:38 pm, Martin Blumenstingl wrote: > Hi Robin, > > On Wed, Jan 8, 2020 at 12:18 PM Robin Murphy <robin.murphy@arm.com> wrote: >> >> On 07/01/2020 11:06 pm, Martin Blumenstingl wrote: >>> Decouple the check to see whether we want to enable devfreq for the GPU >>> from dev_pm_opp_set_regulators(). This is preparation work for adding >>> back support for regulator control (which means we need to call >>> dev_pm_opp_set_regulators() before dev_pm_opp_of_add_table(), which >>> means having a check for "is devfreq enabled" that is not tied to >>> dev_pm_opp_of_add_table() makes things easier). >> >> Hmm, what about cases like the SCMI DVFS protocol where the OPPs are >> dynamically discovered rather than statically defined in DT? > where can I find such an example (Amlogic SoCs use SCPI instead of > SCMI, so I don't think that I have any board with SCMI support) or > some documentation? > (I could only find SCPI clock and CPU DVFS implementations, but no > generic "OPPs for any device" implementation)
On closer inspection I think this applies to the SCPI DVFS protocol too[1]. AIUI the fact that neither is wired up to a devfreq driver yet is merely due to lack of demand and suitable systems to develop/test on so far - the panfrost devfreq code is only now looking like the first viable upstream use-case ;)
Robin.
[1] http://infocenter.arm.com/help/topic/com.arm.doc.dui0922g/BABFEBCD.html
| |