[Intel-gfx] [PATCH 4/7] drm/i915/hwmon: Show device level energy usage

[Tvrtko Ursulin]

On 21/09/2022 01:24, Dixit, Ashutosh wrote:
> On Tue, 13 Sep 2022 01:50:08 -0700, Tvrtko Ursulin wrote:
>>
> 
> Hi Tvrtko,
> 
>> On 25/08/2022 14:21, Badal Nilawar wrote:
>>> From: Dale B Stimson <dale.b.stimson at intel.com>
>>>
>>> diff --git a/Documentation/ABI/testing/sysfs-driver-intel-i915-hwmon b/Documentation/ABI/testing/sysfs-driver-intel-i915-hwmon
>>> index 9a2d10edfce8..03d71c6869d3 100644
>>> --- a/Documentation/ABI/testing/sysfs-driver-intel-i915-hwmon
>>> +++ b/Documentation/ABI/testing/sysfs-driver-intel-i915-hwmon
>>> @@ -25,3 +25,11 @@ Contact:	dri-devel at lists.freedesktop.org
>>>    Description:	RO. Card default power limit (default TDP setting).
>>> 			Only supported for particular Intel i915 graphics
>>> platforms.
>>> +
>>> +What:		/sys/devices/.../hwmon/hwmon<i>/energy1_input
>>> +Date:		June 2022
>>> +KernelVersion:	5.19
>>
>> Date and kernel version will need updating throughout I think.
>>
>> But why I am here actually is to ask if there are plans to make
>> intel_gpu_top support this? It would be nice to have since it reports power
>> for integrated.
> 
> There were no plans but now Riana has an IGT patch series which exposes a
> unified inteface for rapl/hwmon (igfx/dgfx):
> 
> https://patchwork.freedesktop.org/series/108185/
> 
> So perhaps we can either have intel_gpu_top use this IGT lib or if it
> doesn't, copy some functions to intel_gpu_top.

Looks promising and would be nice yeah.

On the practicalities, first thing which comes to mind that probably it 
would need to be a separate library like igt_perf and igt_device_scan, 
since so far we have avoided linking intel_gpu_top with the whole lib_igt.

Regards,

Tvrtko