[Freedreno] [PATCH v4 4/4] arm64/dts/qcom/sm8250: remove assigned-clock-rate property for mdp clk

[Dmitry Baryshkov]

On Mon, 7 Mar 2022 at 19:05, Vinod Polimera <vpolimer at qti.qualcomm.com> wrote:
>
> > WARNING: This email originated from outside of Qualcomm. Please be wary
> > of any links or attachments, and do not enable macros.
> >
> > On Sat, 5 Mar 2022 at 00:49, Doug Anderson <dianders at chromium.org>
> > wrote:
> > > On Thu, Mar 3, 2022 at 4:16 PM Dmitry Baryshkov
> > > <dmitry.baryshkov at linaro.org> wrote:
> > > >
> > > > On Fri, 4 Mar 2022 at 02:56, Stephen Boyd <swboyd at chromium.org>
> > wrote:
> > > > >
> > > > > Quoting Dmitry Baryshkov (2022-03-03 15:50:50)
> > > > > > On Thu, 3 Mar 2022 at 12:40, Vinod Polimera
> > <quic_vpolimer at quicinc.com> wrote:
> > > > > > >
> > > > > > > Kernel clock driver assumes that initial rate is the
> > > > > > > max rate for that clock and was not allowing it to scale
> > > > > > > beyond the assigned clock value.
> > > > > > >
> > > > > > > Drop the assigned clock rate property and vote on the mdp clock as
> > per
> > > > > > > calculated value during the usecase.
> > > > > > >
> > > > > > > Fixes: 7c1dffd471("arm64: dts: qcom: sm8250.dtsi: add display
> > system nodes")
> > > > > >
> > > > > > Please remove the Fixes tags from all commits. Otherwise the
> > patches
> > > > > > might be picked up into earlier kernels, which do not have a patch
> > > > > > adding a vote on the MDP clock.
> > > > >
> > > > > What patch is that? The Fixes tag could point to that commit.
> > > >
> > > > Please correct me if I'm wrong.
> > > > Currently the dtsi enforces bumping the MDP clock when the mdss
> > device
> > > > is being probed and when the dpu device is being probed.
> > > > Later during the DPU lifetime the core_perf would change the clock's
> > > > rate as it sees fit according to the CRTC requirements.
> > >
> > > "Currently" means _before_ ${SUBJECT} patch lands, right? Since
> > > ${SUBJECT} patch is removing the bump to max.
> >
> > Yes. 'Before this patch'.
> >
> > >
> > >
> > > > However it would happen only when the during the
> > > > dpu_crtc_atomic_flush(), before we call this function, the MDP clock
> > > > is left in the undetermined state. The power rails controlled by the
> > > > opp table are left in the undetermined state.
> > > >
> > > > I suppose that during the dpu_bind we should bump the clock to the max
> > > > possible freq and let dpu_core_perf handle it afterwards.
> > >
> > > Definitely feels like seeing the clock to something predictable during
> > > the initial probe makes sense. If it's just for the initial probe then
> > > setting it to max (based on the opp table) seems fine.
> >
> > Vinod, could you please implement it?
> >
> > > I think an
> > > earlier version of this series set it to max every time we did runtime
> > > resume. We'd have to have a good reason to do that.
> >
> > Yes, this is correct. Based on the comments I had the impression that
> > there was a suggestion that the place for the calls was wrong. Most
> > probably I was instead projecting my own thoughts.
> >
> I had discussed internally with the team. Traditionally, mdp clk vote during
> probe/bind is required when display is turned on in bootloader and persists
> till first update in kernel.

Not each and every board has a display setup in the bootloader. For
example the RB5 I have here doesn't support setting up the display.
Not to mention that we should tell Linux, which vote is cast,
otherwise the .sync_state can turn respective votes off.

> As in chromebook, timing engine will be turned
> off during depthcharge exit and as there is no display transition from
> bootloader to kernel, mdp clk can be voted based on the calculated value
> during framework update and does not required vote during probe/bind.

Generally Linux should not depend on the bootloader setup. You can not
be sure. What if we kexec next kernel?

-- 
With best wishes
Dmitry