Re: [PATCH v2 3/4] ARM: dts: exynos: Add regulator suspend configuration to Odroid XU3/XU4/HC1 family

[Anand Moon]

Hi Marek / Krzysztof,

On Mon, 24 Jun 2019 at 14:31, Marek Szyprowski <m.szyprowski@xxxxxxxxxxx> wrote:
>
> Hi Krzysztof,
>
> On 2019-06-24 09:41, Krzysztof Kozlowski wrote:
> > On Mon, 24 Jun 2019 at 09:20, Marek Szyprowski <m.szyprowski@xxxxxxxxxxx> wrote:
> >> On 2019-06-23 18:02, Anand Moon wrote:
> >>> Thanks for this patch. Please add my
> >>>
> >>> Tested-by: Anand Moon <linux.amoon@xxxxxxxxx>
> >>>
> >>> [snip]
> >>>
> >>> Could you integrate below small changes into this patch.
> >>> with these below changes suspend and resume work correctly at my end.
> >>>
> >>> [1] XU4_suspendresume.patch
> >>>
> >>> As per S2MPS11B PMIC 1.2.1 Regulator (Features)
> >>> Fix the min max value for *Buck7* and *Buck8*
> >>>
> >>> -- Buck7 (VDD_1.0V_LDO) 1.5 A (1.2 V to 1.5 V, 12.5 mV step, default on 1.35 V)
> >>> -- Buck8 (VDD_1.8V_LDO) 2.5 A (1.8 V to 2.1 V, 12.5 mV step, default on 2.0 V)
> >> Could you elaborate why such change for Buck7 and Buck8 is needed?
> > Anand has here valid point - the constraints in DTS do not match
> > hardware manual. This leads to question whether voltage table in
> > driver is proper... Another point is the voltage itself. The
> > schematics describes them as at specific voltage (1.35 V and 2.0 V)
> > but after boot they are 1.2 V and 1.85 V. Maybe this shift comes from
> > the problem above.
> >
> >>> Also add suspend-off for *Buck9*
> >>> Buck9 internally controls the power of USB hub.
> >>> Adding suspend the this node help proper reset of USB hub on Odroid
> >>> XU4 / HC1/ XU3
> >>> during suspend and resume. Below it the logs from my testing.
> >> Disabling Buck9 in suspend indeed reduces the power consumed by the
> >> board during suspend-to-ram from about 80mA to as little as 7-10mA, what
> >> matches the results of OdroidXU3. Thanks for the hint!
> > Although I did not get what is the difference in the logs (Anand
> > pasted two logs but they look the same) but the power consumption is
> > reason is good enough. I would be happy to put in the changelog entire
> > consumption  difference. I can measure it on XU3-Lite but can you give
> > me the XU4 (before and after)?\
>
>
> HC1:
>
> next-20190620: 120mA (@5V)
> this patchset: 72mA (@5V)
> this patchset + fixup from Anand: 7-10mA (@5V)
>
> XU4 (SDcard):
>
> next-20190620: 88mA (@5V)
> this patchset: 74mA (@5V), sometimes 42mA (@5V)
> this patchset + fixup from Anand: 6-9mA (@5V)
>
>
> XU4 (eMMC):
>
> next-20190620: 100mA (@5V)
> this patchset: 72mA (@5V), sometimes 41mA (@5V)
> this patchset + fixup from Anand: 6-9mA (@5V)
>
>
> Best regards
> --
> Marek Szyprowski, PhD
> Samsung R&D Institute Poland
>

Thanks for this results.

Best Regards
-Anand