[igt-dev] [PATCH i-g-t 8/8] gputop: Basic vendor agnostic GPU top tool

[Tvrtko Ursulin]

On 03/02/2023 16:42, Kamil Konieczny wrote:
> Hi Tvrtko,
> 
> On 2023-01-31 at 11:32:37 +0000, Tvrtko Ursulin wrote:
>> From: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
>>
>> Rudimentary vendor agnostic example of how lib_igt_drm_clients can be used
>> to display a sorted by card and usage list of processes using GPUs.
>>
>> Borrows a bit of code from intel_gpu_top but for now omits the fancy
>> features like interactive functionality, card selection, client
>> aggregation, sort modes, JSON output  and pretty engine names. Also no
>> support for global GPU or system metrics.
>>
>> On the other hand it shows clients from all DRM cards which
>> intel_gpu_top does not do.
>>
>> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
>> Cc: Rob Clark <robdclark at chromium.org>
>> Cc: Christian König <ckoenig.leichtzumerken at gmail.com>
>> Acked-by: Christian König <christian.koenig at amd.com>
> 
> I run it with:
> # ./gputop
> 
> but it do not work on my Skylake card, I see no output,
> kernel 5.19.0-29-generic, ubuntu 22.10

Odd, 5.19 should have the support. Intel_gpu_top works - it is showing 
the individual clients?

> 
> # ./lsgpu
> card0                    Intel Skylake (Gen9)              drm:/dev/dri/card0
> └─renderD128                                               drm:/dev/dri/renderD128
> 
> Please add some options like debug, version, debug with high
> verbose level, help. It seems like q or Q do not exit.

As the cover letter hints I was only set out to demonstrate an extremely 
rudimentary vendor agnostic tool. To quote the cover letter more - "..It 
also makes no effort to provide sorting modes, well any interactivity, 
or any pretty names for GPUs or engines..". I have no scope presently to 
make it better or nicer.

The tool however can serve as a starting point and people had reported 
it working as-is with a few other drivers, AMD, msm and most recently I 
believe etnaviv. So perhaps a pool of people to further improve it will 
be found there in the future.

In summary I think it's worth reviewing so that the common code gets 
extracted from intel_gpu_top into respective libraries. After that I was 
hoping other people start contributing further improvements.

Regards,

Tvrtko