OMNITRACE_ROCM_SMI_METRICS (#331)

* OMNITRACE_ROCM_SMI_METRICS

- configuration variable OMNITRACE_ROCM_SMI_METRICS for specifying which rocm-smi metrics to collect
- auto-disable metric collection when rsmi_dev_X_get returns RSMI_STATUS_NOT_SUPPORTED

* Bump version to 1.11.1

* Python formatting

* Update python/libpyomnitrace.cpp

- fix usage of substr (ignored return value)

* Update python/gui/source/gui.py

- Fix E721
  - do not compare types, for exact checks use `is` / `is not`, for instance checks use `isinstance()`

[ROCm/rocprofiler-systems commit: 15127c0d43]
This commit is contained in:
Jonathan R. Madsen
2024-02-08 07:06:23 -06:00
committed by GitHub
parent 8c8caaa1d9
commit 25ff5e3891
7 changed files with 127 additions and 35 deletions
@@ -647,6 +647,11 @@ configure_settings(bool _init)
"is collected on every available device",
"", "rocprofiler", "rocm", "hardware_counters");
OMNITRACE_CONFIG_SETTING(std::string, "OMNITRACE_ROCM_SMI_METRICS",
"rocm-smi metrics to collect: busy, temp, power, mem_usage",
"busy,temp,power,mem_usage", "backend", "rocm_smi", "rocm",
"process_sampling", "advanced");
OMNITRACE_CONFIG_SETTING(bool, "OMNITRACE_CRITICAL_TRACE_DEBUG",
"Enable debugging for critical trace", _omnitrace_debug,
"debugging", "critical_trace", "advanced");