Re: [PATCH v2] arm64: dts: r8a7795: pmu: switch to Cortex specific device nodes

[Dirk Behme <dirk.behme@de.bosch.com>]

On 05.02.2016 10:54, Simon Horman wrote:
> [CC shiny new linux-renesas-soc ML]


Yes, thanks. I've to learn and remember this, too :)


> On Wed, Feb 03, 2016 at 06:20:52PM +0100, Dirk Behme wrote:
>> On 16.01.2016 15:13, Dirk Behme wrote:
>>> Instead of using the generic armv8-pmuv3 compatibility use the more
>>> specific Cortex A57 compatibility.
>>>
>>> Signed-off-by: Dirk Behme <dirk.behme@gmail.com>
>>> ---
>>> Changes in v2: Drop the not yet merged Cortex A53 part.
>>>
>>>   arch/arm64/boot/dts/renesas/r8a7795.dtsi | 4 ++--
>>>   1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/arch/arm64/boot/dts/renesas/r8a7795.dtsi b/arch/arm64/boot/dts/renesas/r8a7795.dtsi
>>> index 266c5de..a82bce8 100644
>>> --- a/arch/arm64/boot/dts/renesas/r8a7795.dtsi
>>> +++ b/arch/arm64/boot/dts/renesas/r8a7795.dtsi
>>> @@ -247,8 +247,8 @@
>>>   			power-domains = <&cpg>;
>>>   		};
>>>
>>> -		pmu {
>>> -			compatible = "arm,armv8-pmuv3";
>>> +		pmu_a57 {
>>> +			compatible = "arm,cortex-a57-pmu";
>>>   			interrupts = <GIC_SPI 72 IRQ_TYPE_LEVEL_HIGH>,
>>>   				     <GIC_SPI 73 IRQ_TYPE_LEVEL_HIGH>,
>>>   				     <GIC_SPI 74 IRQ_TYPE_LEVEL_HIGH>,
>>>
>>
>>
>> Any further comments to this? If not, could this be applied?
>
> Hi Dirk,
>
> I apologise for loosing track of this one.
>
> Does this change work? I saw some discussion about using arm,cortex-a57-pmu
> not working correctly on the r8a7795 in an earlier thread.

Yes, the A57 PMU probe is successful.

What's not working with the latest Renesas BSP is the A53, because these 
cores are disabled in the firmware coming with that BSP. But that's not 
a PMU issue, it's just because the cores are disabled. Using an older 
BSP firmware enabling all cores makes the A53 PMU probe fine, too.

Best regards

Dirk