[PATCH v8] ARM: EXYNOS: Use MCPM call-backs to support S2R on Exynos5420

[Abhilash Kesavan]

Hi Javier,

On Mon, Oct 6, 2014 at 5:29 PM, Javier Martinez Canillas
<javier at dowhile0.org> wrote:
> Hello Abhilash,
>
> On Mon, Oct 6, 2014 at 11:08 AM, Abhilash Kesavan
> <kesavan.abhilash at gmail.com> wrote:
>>>>> diff --git a/arch/arm/mach-exynos/platsmp.c b/arch/arm/mach-exynos/platsmp.c
>>>>> index adb36a8..222aa3c 100644
>>>>> --- a/arch/arm/mach-exynos/platsmp.c
>>>>> +++ b/arch/arm/mach-exynos/platsmp.c
>>>>> @@ -137,6 +137,18 @@ void exynos_cpu_power_down(int cpu)
>>>>>   */
>>>>>  void exynos_cpu_power_up(int cpu)
>>>>>  {
>>>>> +       if (cpu == 0 && (of_machine_is_compatible("samsung,exynos5420") ||
>>>>> +               of_machine_is_compatible("samsung,exynos5800"))) {
>>>>> +               /*
>>>>> +                * Bypass power down for CPU0 during suspend. Check for
>>>>> +                * the SYS_PWR_REG value to decide if we are suspending
>>>>> +                * the system.
>>>>> +                */
>>>>> +               int val = __raw_readl(pmu_base_addr +
>>>>> +                               EXYNOS5_ARM_CORE0_SYS_PWR_REG);
>>>>> +               if (!(val & S5P_CORE_LOCAL_PWR_EN))
>>>>> +                       return;
>>>>> +       }
>>>>>         pmu_raw_writel(S5P_CORE_LOCAL_PWR_EN,
>>>>>                         EXYNOS_ARM_CORE_CONFIGURATION(cpu));
>>>>>  }
>>>>
>>>> The above block of code should be in exynos_cpu_power_down(). Can you
>>>> re-test s2r after modifying it ? I have to convert a __raw_readl to
>>>> pmu_raw_readl in this patch as well. Once Vikas re-works the base s2r
>>>> patch I will re-post a v9.
>>>>
>>>
>>> Thanks, for the information but unfortunately even after this change I
>>> still have the issue reported in [0]. I also removed the lines from
>>> exynos5420_pm_resume() that you asked Vikash [1].
>>
>> I have just tested this on the 5420 that I have and it resumes fine.
>> The difference could be in the bootloader we are using. I'll send out
>> details of my current setup so that you can cross-check if everything
>> is the same.
>
> Thanks a lot for sending me the patches you are using to test.
>
> But after looking at the differences I noticed that your patches are
> not removing the lines that set the CPU0 low power states at the start
> of exynos5420_pm_resume() as you asked Vikas [1] but instead your
> version of this patch does not have the code that does the same
> operation after the early_wakeup label like is made in $subject.

The reason for my asking Vikas to remove the lines that restore the
core low power state register from the base s2r patch was because it
should be part of the mcpm s2r patch where the actual setting of the
register takes place.
Regarding the location of those lines we should be restoring the
register as early as possible, so my next version would have it at the
beginning of exynos5420_pm_resume.
>
> IOW, this [2] is what I had to change on top of $subject to have s2r
> working. After that change, both suspending to ram and resuming due
> the RTC IRQ alarm being triggered and the core pm_test case works well
> on my Exynos5420 Peach Pit.

That's good. Thanks for testing !

Regards,
Abhilash
>
>>
>> Regards,
>> Abhilash
>
> Best regards,
> Javier
>
> [0]: http://www.spinics.net/lists/arm-kernel/msg367615.html
> [1]: http://www.spinics.net/lists/linux-samsung-soc/msg37561.html
> [2]: http://pastebin.com/raw.php?i=XSxJH3Ys