Re: [PATCH v1] perf drm_pmu: Add a tool like PMU to expose DRM information

[Namhyung Kim <namhyung@kernel.org>]

Hi Ian,

On Mon, Feb 10, 2025 at 11:17:27PM -0800, Ian Rogers wrote:
> DRM clients expose information through usage stats as documented in
> Documentation/gpu/drm-usage-stats.rst (available online at
> https://docs.kernel.org/gpu/drm-usage-stats.html). Add a tool like
> PMU, similar to the hwmon PMU, that exposes DRM information.

Probably better to put this link in a file header comment.


> For example on a tigerlake laptop:
> ```
> $ perf list drm
> 
> List of pre-defined events (to be used in -e or -M):
> 
> drm:
>   drm-active-stolen-system0
>        [Total memory active in one or more engines. Unit: drm_i915]
>   drm-active-system0
>        [Total memory active in one or more engines. Unit: drm_i915]
>   drm-engine-capacity-video
>        [Engine capacity. Unit: drm_i915]
>   drm-engine-copy
>        [Utilization in ns. Unit: drm_i915]
>   drm-engine-render
>        [Utilization in ns. Unit: drm_i915]
>   drm-engine-video
>        [Utilization in ns. Unit: drm_i915]
>   drm-engine-video-enhance
>        [Utilization in ns. Unit: drm_i915]
>   drm-purgeable-stolen-system0
>        [Size of resident and purgeable memory bufers. Unit: drm_i915]
>   drm-purgeable-system0
>        [Size of resident and purgeable memory bufers. Unit: drm_i915]
>   drm-resident-stolen-system0
>        [Size of resident memory bufers. Unit: drm_i915]
>   drm-resident-system0
>        [Size of resident memory bufers. Unit: drm_i915]
>   drm-shared-stolen-system0
>        [Size of shared memory bufers. Unit: drm_i915]
>   drm-shared-system0
>        [Size of shared memory bufers. Unit: drm_i915]
>   drm-total-stolen-system0
>        [Size of shared and private memory. Unit: drm_i915]
>   drm-total-system0
>        [Size of shared and private memory. Unit: drm_i915]
> ```
> 
> System wide data can be gathered:
> ```
> $ perf stat -x, -I 1000 -e drm-active-stolen-system0,drm-active-system0,drm-engine-capacity-video,drm-engine-copy,drm-engine-render,drm-engine-video,drm-engine-video-enhance,drm-purgeable-stolen-system0,drm-purgeable-system0,drm-resident-stolen-system0,drm-resident-system0,drm-shared-stolen-system0,drm-shared-system0,drm-total-stolen-system0,drm-total-system0
> 1.000904910,0,bytes,drm-active-stolen-system0,1,100.00,,
> 1.000904910,0,bytes,drm-active-system0,1,100.00,,
> 1.000904910,36,capacity,drm-engine-capacity-video,1,100.00,,
> 1.000904910,0,ns,drm-engine-copy,1,100.00,,
> 1.000904910,1472970566175,ns,drm-engine-render,1,100.00,,
> 1.000904910,0,ns,drm-engine-video,1,100.00,,
> 1.000904910,0,ns,drm-engine-video-enhance,1,100.00,,
> 1.000904910,0,bytes,drm-purgeable-stolen-system0,1,100.00,,
> 1.000904910,38199296,bytes,drm-purgeable-system0,1,100.00,,
> 1.000904910,0,bytes,drm-resident-stolen-system0,1,100.00,,
> 1.000904910,4643196928,bytes,drm-resident-system0,1,100.00,,
> 1.000904910,0,bytes,drm-shared-stolen-system0,1,100.00,,
> 1.000904910,1886871552,bytes,drm-shared-system0,1,100.00,,
> 1.000904910,0,bytes,drm-total-stolen-system0,1,100.00,,
> 1.000904910,4643196928,bytes,drm-total-system0,1,100.00,,
> 2.264426839,0,bytes,drm-active-stolen-system0,1,100.00,,
> ```
> 
> Or for a particular process:
> ```
> $ perf stat -x, -I 1000 -e drm-active-stolen-system0,drm-active-system0,drm-engine-capacity-video,drm-engine-copy,drm-engine-render,drm-engine-video,drm-engine-video-enhance,drm-purgeable-stolen-system0,drm-purgeable-system0,drm-resident-stolen-system0,drm-resident-system0,drm-shared-stolen-system0,drm-shared-system0,drm-total-stolen-system0,drm-total-system0 -p 200027
> 1.001040274,0,bytes,drm-active-stolen-system0,6,100.00,,
> 1.001040274,0,bytes,drm-active-system0,6,100.00,,
> 1.001040274,12,capacity,drm-engine-capacity-video,6,100.00,,
> 1.001040274,0,ns,drm-engine-copy,6,100.00,,
> 1.001040274,1542300,ns,drm-engine-render,6,100.00,,
> 1.001040274,0,ns,drm-engine-video,6,100.00,,
> 1.001040274,0,ns,drm-engine-video-enhance,6,100.00,,
> 1.001040274,0,bytes,drm-purgeable-stolen-system0,6,100.00,,
> 1.001040274,13516800,bytes,drm-purgeable-system0,6,100.00,,
> 1.001040274,0,bytes,drm-resident-stolen-system0,6,100.00,,
> 1.001040274,27746304,bytes,drm-resident-system0,6,100.00,,
> 1.001040274,0,bytes,drm-shared-stolen-system0,6,100.00,,
> 1.001040274,0,bytes,drm-shared-system0,6,100.00,,
> 1.001040274,0,bytes,drm-total-stolen-system0,6,100.00,,
> 1.001040274,27746304,bytes,drm-total-system0,6,100.00,,
> 2.016629075,0,bytes,drm-active-stolen-system0,6,100.00,,
> ```

I've tested it briefly.

  $ ./perf stat -e drm-engine-render,drm-total-system0 -a sleep 1
  
   Performance counter stats for 'system wide':
  
   2,869,492,628,815 ns    drm-engine-render                                                     
       2,777,497,600 bytes drm-total-system0                                                     
  
         1.004182447 seconds time elapsed

It seems the numbers are quite big.

  $ ./perf stat -e drm-engine-render,drm-total-system0 -aA sleep 1
  
   Performance counter stats for 'system wide':
  
  CPU0    2,870,871,280,238 ns    drm_i915/drm-engine-render/                                           
  CPU1        <not counted> ns    drm_i915/drm-engine-render/                                           
  CPU2        <not counted> ns    drm_i915/drm-engine-render/                                           
  CPU3        <not counted> ns    drm_i915/drm-engine-render/                                           
  CPU0        2,750,578,688 bytes drm_i915/drm-total-system0/                                           
  CPU1        <not counted> bytes drm_i915/drm-total-system0/                                           
  CPU2        <not counted> bytes drm_i915/drm-total-system0/                                           
  CPU3        <not counted> bytes drm_i915/drm-total-system0/                                           
  
         1.001678363 seconds time elapsed

Ok, it only reads from the CPU0.  But I guess there are some
duplications.  Have you checked drm-client-id?

  $ ./perf stat -e drm-engine-render -a --per-thread sleep 1
  
   Performance counter stats for 'system wide':
  
              Xorg-6900       852,545,872,646 ns    drm-engine-render                                                     
       Xorg:disk$0-6901       852,545,872,646 ns    drm-engine-render                                                     
          Xorg:sh0-6902       852,545,872,646 ns    drm-engine-render                                                     
      Xorg:traceq0-6904       852,545,872,646 ns    drm-engine-render                                                     
        Xorg:gdrv0-6906       852,545,872,646 ns    drm-engine-render                                                     
       InputThread-6946       852,545,872,646 ns    drm-engine-render                                                     
       gnome-shell-7127       808,521,145,191 ns    drm-engine-render                                                     
      pool-spawner-7146       808,521,145,191 ns    drm-engine-render                                                     
             gmain-7147       808,521,145,191 ns    drm-engine-render                                                     
             gdbus-7149       808,521,145,191 ns    drm-engine-render                                                     
      dconf worker-7150       808,521,145,191 ns    drm-engine-render                                                     
         JS Helper-7151       808,521,145,191 ns    drm-engine-render                                                     
         JS Helper-7152       808,521,145,191 ns    drm-engine-render                                                     
      ...

Trying record..

  $ ./perf record -e drm-engine-render sleep 1
  failed to mmap with 9 (Bad file descriptor)

I think you can fail evsel__open() if attr.sample_period != 0.

> 
> As with the hwmon PMU, high numbered PMU types are used to encode
> multiple possible "DRM" PMUs. The appropriate fdinfo is found by
> scanning /proc and filtering which fdinfos to read with stat. To avoid
> some unneeding scanning, events not starting with "drm-" are ignored.

It's sad that it should scan /proc whenever it reads the event but I
don't think we have other options.


> The patch builds on commit 57e13264dcea ("perf pmus:
> Restructure pmu_read_sysfs to scan fewer PMUs") so that only if full
> wild carding is being done, that the drm PMUs will be read.

Can you please add a test case?

Thanks,
Namhyung