diff options
| -rw-r--r-- | docs/index.rst | 1 | ||||
| -rw-r--r-- | docs/perfetto.rst | 110 | ||||
| -rw-r--r-- | src/tool/pps/cfg/cpu.cfg | 20 | ||||
| -rw-r--r-- | src/tool/pps/cfg/gpu.cfg | 29 |
4 files changed, 160 insertions, 0 deletions
diff --git a/docs/index.rst b/docs/index.rst index 56d47af1249..3689baa7eff 100644 --- a/docs/index.rst +++ b/docs/index.rst @@ -71,6 +71,7 @@ and other operating systems. osmesa debugging perf + perfetto extensions application-issues viewperf diff --git a/docs/perfetto.rst b/docs/perfetto.rst new file mode 100644 index 00000000000..16af6802d26 --- /dev/null +++ b/docs/perfetto.rst @@ -0,0 +1,110 @@ +Perfetto Tracing +================ + +Mesa has experimental support for `Perfetto <https://perfetto.dev>`__ for +GPU performance monitoring. Perfetto supports multiple +`producers <https://perfetto.dev/docs/concepts/service-model>`__ each with +one or more data-sources. Perfetto already provides various producers and +data-sources for things like: + +- CPU scheduling events (``linux.ftrace``) +- CPU frequency scaling (``linux.ftrace``) +- System calls (``linux.ftrace``) +- Process memory utilization (``linux.process_stats``) + +As well as various domain specific producers. + +The mesa perfetto support adds additional producers, to allow for visualizing +GPU performance (frequency, utilization, performance counters, etc) on the +same timeline, to better understand and tune/debug system level performance: + +- pps-producer: A systemwide daemon that can collect global performance + counters. +- mesa: Per-process producer within mesa to capture render-stage traces + on the GPU timeline, track events, etc. + +Run +--- + +To capture a trace with perfetto you need to take the following steps: + +1. Build perfetto from sources available at ``subprojects/perfetto`` following + `this guide <https://perfetto.dev/docs/quickstart/linux-tracing>`__. + +2. Create a `trace config <https://perfetto.dev/#/trace-config.md>`__, which is + a json formatted text file with extension ``.cfg``, or use one of the config + files under the ``src/tool/pps/cfg`` directory. More examples of config files + can be found in ``subprojects/perfetto/test/configs``. + +3. Change directory to ``subprojects/perfetto`` and run a + `convenience script <https://perfetto.dev/#/running.md>`__ to start the + tracing service: + + .. code-block:: console + + cd subprojects/perfetto + CONFIG=<path/to/gpu.cfg> OUT=out/linux_clang_release ./tools/tmux -n + +4. Start other producers you may need, e.g. ``pps-producer``. + +5. Start ``perfetto`` under the tmux session initiated in step 3. + +6. Once tracing has finished, you can detach from tmux with :kbd:`Ctrl+b`, + :kbd:`d`, and the convenience script should automatically copy the trace + files into ``$HOME/Downloads``. + +7. Go to `ui.perfetto.dev <https://ui.perfetto.dev>`__ and upload + ``$HOME/Downloads/trace.protobuf`` by clicking on **Open trace file**. + +8. Alternatively you can open the trace in `AGI <https://gpuinspector.dev/>`__ + (which despite the name can be used to view non-android traces). + +Troubleshooting +--------------- + +Tmux +~~~~ + +If the convenience script ``tools/tmux`` keeps copying artifacts to your +``SSH_TARGET`` without starting the tmux session, make sure you have ``tmux`` +installed in your system. + +.. code-block:: console + + apt install tmux + +Missing counter names +~~~~~~~~~~~~~~~~~~~~~ + +If the trace viewer shows a list of counters with a description like +``gpu_counter(#)`` instead of their proper names, maybe you had a data loss due +to the trace buffer being full and wrapped. + +In order to prevent this loss of data you can tweak the trace config file in +two different ways: + +- Increase the size of the buffer in use: + + .. code-block:: javascript + + buffers { + size_kb: 2048, + fill_policy: RING_BUFFER, + } + +- Periodically flush the trace buffer into the output file: + + .. code-block:: javascript + + write_into_file: true + file_write_period_ms: 250 + + +- Discard new traces when the buffer fills: + + .. code-block:: javascript + + buffers { + size_kb: 2048, + fill_policy: DISCARD, + } diff --git a/src/tool/pps/cfg/cpu.cfg b/src/tool/pps/cfg/cpu.cfg new file mode 100644 index 00000000000..0824b2aa6e3 --- /dev/null +++ b/src/tool/pps/cfg/cpu.cfg @@ -0,0 +1,20 @@ +buffers { + size_kb: 65536 + fill_policy: RING_BUFFER +} + +data_sources { + config { + name: "linux.sys_stats" + target_buffer: 0 + sys_stats_config { + meminfo_period_ms: 100 + meminfo_counters: MEMINFO_MEM_AVAILABLE + + stat_period_ms: 100 + stat_counters: STAT_CPU_TIMES + } + } +} + +duration_ms: 10000 diff --git a/src/tool/pps/cfg/gpu.cfg b/src/tool/pps/cfg/gpu.cfg new file mode 100644 index 00000000000..eaef4df402a --- /dev/null +++ b/src/tool/pps/cfg/gpu.cfg @@ -0,0 +1,29 @@ +buffers { + size_kb: 1024 + fill_policy: RING_BUFFER +} + +data_sources { + config { + name: "gpu.counters.i915" + gpu_counter_config { + counter_period_ns: 1000000000 + } + counter_ids: 0 + counter_ids: 1 + counter_ids: 2 + counter_ids: 3 + counter_ids: 4 + } +} + +data_sources { + config { + name: "gpu.counters.panfrost" + gpu_counter_config { + counter_period_ns: 1000000000 + } + } +} + +duration_ms: 16000 |