diff mbox series

dt: rockchip: rk3399: Add dynamic power coefficient for GPU

Message ID 20210319110511.24787-1-daniel.lezcano@linaro.org (mailing list archive)
State New
Headers show
Series dt: rockchip: rk3399: Add dynamic power coefficient for GPU | expand

Commit Message

Daniel Lezcano March 19, 2021, 11:05 a.m. UTC
The DTPM framework is looking for upstream SoC candidates to share the
power numbers.

We can see around different numbers but the one which seems to be
consistent with the initial post for the values on the CPUs can be
found in the patch https://lore.kernel.org/patchwork/patch/810159/

I don't know the precision of this value but it is better than
nothing.

Hopefully, one day SoC vendors will be more generous with the power
numbers at least for the SoC which are from the previous generation
and give the community the opportunity to develop power based
frameworks.
---
 arch/arm64/boot/dts/rockchip/rk3399.dtsi | 1 +
 1 file changed, 1 insertion(+)

Comments

Robin Murphy March 19, 2021, 12:17 p.m. UTC | #1
On 2021-03-19 11:05, Daniel Lezcano wrote:
> The DTPM framework is looking for upstream SoC candidates to share the
> power numbers.
> 
> We can see around different numbers but the one which seems to be
> consistent with the initial post for the values on the CPUs can be
> found in the patch https://lore.kernel.org/patchwork/patch/810159/

The kernel hacker in me would be more inclined to trust the BSP that the 
vendor actively supports than a 5-year-old patch that was never pursued 
upstream. Apparently that was last updated more recently:

https://github.com/rockchip-linux/kernel/commit/98d4505e1bd62ff028bd79fbd8284d64b6f468f8

The ex-mathematician in me can't even comment either way without 
evidence that whatever model expects to consume this value is even 
comparable to whatever "arm,mali-simple-power-model" is. The way the 
latter apparently needs an explicit "static" coefficient as well as a 
"dynamic" one, and the value here being nearly 3 times that of a 
similarly-named one in active use downstream (ChromeOS appears to still 
be using the values from before the above commit), certainly incline me 
to think they may not be...

> I don't know the precision of this value but it is better than
> nothing.

But is it? If it leads to some throttling mechanism kicking in and 
crippling GPU performance because it's massively overestimating power 
consumption, that would be objectively worse for most users, no?

Robin.

> Hopefully, one day SoC vendors will be more generous with the power
> numbers at least for the SoC which are from the previous generation
> and give the community the opportunity to develop power based
> frameworks.
> ---
>   arch/arm64/boot/dts/rockchip/rk3399.dtsi | 1 +
>   1 file changed, 1 insertion(+)
> 
> diff --git a/arch/arm64/boot/dts/rockchip/rk3399.dtsi b/arch/arm64/boot/dts/rockchip/rk3399.dtsi
> index edbbf35fe19e..1ab1d293d2e9 100644
> --- a/arch/arm64/boot/dts/rockchip/rk3399.dtsi
> +++ b/arch/arm64/boot/dts/rockchip/rk3399.dtsi
> @@ -1933,6 +1933,7 @@
>   		interrupt-names = "job", "mmu", "gpu";
>   		clocks = <&cru ACLK_GPU>;
>   		#cooling-cells = <2>;
> +		dynamic-power-coefficient = <977>;
>   		power-domains = <&power RK3399_PD_GPU>;
>   		status = "disabled";
>   	};
>
Daniel Lezcano March 19, 2021, 2:35 p.m. UTC | #2
Hi Robin,

On 19/03/2021 13:17, Robin Murphy wrote:
> On 2021-03-19 11:05, Daniel Lezcano wrote:
>> The DTPM framework is looking for upstream SoC candidates to share the
>> power numbers.
>>
>> We can see around different numbers but the one which seems to be
>> consistent with the initial post for the values on the CPUs can be
>> found in the patch https://lore.kernel.org/patchwork/patch/810159/
> 
> The kernel hacker in me would be more inclined to trust the BSP that the
> vendor actively supports than a 5-year-old patch that was never pursued
> upstream. Apparently that was last updated more recently:
> 
> https://github.com/rockchip-linux/kernel/commit/98d4505e1bd62ff028bd79fbd8284d64b6f468f8

Yes, I've seen this value also.

> The ex-mathematician in me can't even comment either way without
> evidence that whatever model expects to consume this value is even
> comparable to whatever "arm,mali-simple-power-model" is. >
> The way the
> latter apparently needs an explicit "static" coefficient as well as a
> "dynamic" one, and the value here being nearly 3 times that of a
> similarly-named one in active use downstream (ChromeOS appears to still
> be using the values from before the above commit), certainly incline me
> to think they may not be...

Sorry, I'm missing the point :/

We dropped in the kernel any static power computation because as there
was no value, the resulting code was considered dead. So we rely on the
dynamic power only.

>> I don't know the precision of this value but it is better than
>> nothing.
> 
> But is it? If it leads to some throttling mechanism kicking in and
> crippling GPU performance because it's massively overestimating power
> consumption, that would be objectively worse for most users, no?

No because there is no sustainable power specified for the thermal zones
related to the GPU.
Robin Murphy March 19, 2021, 6:05 p.m. UTC | #3
On 2021-03-19 14:35, Daniel Lezcano wrote:
> 
> Hi Robin,
> 
> On 19/03/2021 13:17, Robin Murphy wrote:
>> On 2021-03-19 11:05, Daniel Lezcano wrote:
>>> The DTPM framework is looking for upstream SoC candidates to share the
>>> power numbers.
>>>
>>> We can see around different numbers but the one which seems to be
>>> consistent with the initial post for the values on the CPUs can be
>>> found in the patch https://lore.kernel.org/patchwork/patch/810159/
>>
>> The kernel hacker in me would be more inclined to trust the BSP that the
>> vendor actively supports than a 5-year-old patch that was never pursued
>> upstream. Apparently that was last updated more recently:
>>
>> https://github.com/rockchip-linux/kernel/commit/98d4505e1bd62ff028bd79fbd8284d64b6f468f8
> 
> Yes, I've seen this value also.
> 
>> The ex-mathematician in me can't even comment either way without
>> evidence that whatever model expects to consume this value is even
>> comparable to whatever "arm,mali-simple-power-model" is. >
>> The way the
>> latter apparently needs an explicit "static" coefficient as well as a
>> "dynamic" one, and the value here being nearly 3 times that of a
>> similarly-named one in active use downstream (ChromeOS appears to still
>> be using the values from before the above commit), certainly incline me
>> to think they may not be...
> 
> Sorry, I'm missing the point :/
> 
> We dropped in the kernel any static power computation because as there
> was no value, the resulting code was considered dead. So we rely on the
> dynamic power only.

Right, so a 2-factor model is clearly not identical to a 1-factor model, 
so how do we know that a value for one is valid for the other, even if 
it happens to have a similar name? I'm not saying that it is or isn't; I 
don't know. If someone can point to the downstream coefficient 
definition being identical to the upstream one then great, let's use 
that as justification. If not, then the justification of one arbitrary 
meaningless number over any other is a bit misleading.

>>> I don't know the precision of this value but it is better than
>>> nothing.
>>
>> But is it? If it leads to some throttling mechanism kicking in and
>> crippling GPU performance because it's massively overestimating power
>> consumption, that would be objectively worse for most users, no?
> 
> No because there is no sustainable power specified for the thermal zones
> related to the GPU.
OK, that's some reassurance at least. Does the exact value have any 
material effect? If not, what's to stop us from using an obviously 
made-up value like 1, and saying so?

Robin.
Daniel Lezcano March 19, 2021, 6:38 p.m. UTC | #4
On 19/03/2021 19:05, Robin Murphy wrote:
> On 2021-03-19 14:35, Daniel Lezcano wrote:
>>
>> Hi Robin,
>>
>> On 19/03/2021 13:17, Robin Murphy wrote:
>>> On 2021-03-19 11:05, Daniel Lezcano wrote:
>>>> The DTPM framework is looking for upstream SoC candidates to share the
>>>> power numbers.
>>>>
>>>> We can see around different numbers but the one which seems to be
>>>> consistent with the initial post for the values on the CPUs can be
>>>> found in the patch https://lore.kernel.org/patchwork/patch/810159/
>>>
>>> The kernel hacker in me would be more inclined to trust the BSP that the
>>> vendor actively supports than a 5-year-old patch that was never pursued
>>> upstream. Apparently that was last updated more recently:
>>>
>>> https://github.com/rockchip-linux/kernel/commit/98d4505e1bd62ff028bd79fbd8284d64b6f468f8
>>>
>>
>> Yes, I've seen this value also.
>>
>>> The ex-mathematician in me can't even comment either way without
>>> evidence that whatever model expects to consume this value is even
>>> comparable to whatever "arm,mali-simple-power-model" is. >
>>> The way the
>>> latter apparently needs an explicit "static" coefficient as well as a
>>> "dynamic" one, and the value here being nearly 3 times that of a
>>> similarly-named one in active use downstream (ChromeOS appears to still
>>> be using the values from before the above commit), certainly incline me
>>> to think they may not be...
>>
>> Sorry, I'm missing the point :/
>>
>> We dropped in the kernel any static power computation because as there
>> was no value, the resulting code was considered dead. So we rely on the
>> dynamic power only.
> 
> Right, so a 2-factor model is clearly not identical to a 1-factor model,
> so how do we know that a value for one is valid for the other, even if
> it happens to have a similar name? I'm not saying that it is or isn't; I
> don't know. If someone can point to the downstream coefficient
> definition being identical to the upstream one then great, let's use
> that as justification. If not, then the justification of one arbitrary
> meaningless number over any other is a bit misleading.

That's a call :)

>>>> I don't know the precision of this value but it is better than
>>>> nothing.
>>>
>>> But is it? If it leads to some throttling mechanism kicking in and
>>> crippling GPU performance because it's massively overestimating power
>>> consumption, that would be objectively worse for most users, no?
>>
>> No because there is no sustainable power specified for the thermal zones
>> related to the GPU.
> OK, that's some reassurance at least. Does the exact value have any
> material effect? 

Yes, it has when it is combined with other devices having also power
values, like the CPUs and hopefully the DMC soon.

If we can have more or less consistent power numbers for the DMC, CPU
and GPU on the rock960, with the thermal zone having these three heating
sources, we can use the DTPM framework to act on the power of the whole.

I don't know the best coefficient, 733, 977 or 1780 [1]

The value of 977 sound to me as a starting point.



[1]
https://chromium.googlesource.com/chromiumos/third_party/kernel/+/factory-gru-8652.B-chromeos-4.4/arch/arm64/boot/dts/rockchip/rk3399-gru-kevin-r3.dts

> If not, what's to stop us from using an obviously
> made-up value like 1, and saying so?
diff mbox series

Patch

diff --git a/arch/arm64/boot/dts/rockchip/rk3399.dtsi b/arch/arm64/boot/dts/rockchip/rk3399.dtsi
index edbbf35fe19e..1ab1d293d2e9 100644
--- a/arch/arm64/boot/dts/rockchip/rk3399.dtsi
+++ b/arch/arm64/boot/dts/rockchip/rk3399.dtsi
@@ -1933,6 +1933,7 @@ 
 		interrupt-names = "job", "mmu", "gpu";
 		clocks = <&cru ACLK_GPU>;
 		#cooling-cells = <2>;
+		dynamic-power-coefficient = <977>;
 		power-domains = <&power RK3399_PD_GPU>;
 		status = "disabled";
 	};