MI350 Fix L2 cache to HBM read counters/metrics (#2501)

* Fix rocprofiler-sdk metrics definition

* Use TCC_EA0_RDREQ_128B instead of TCC_BUBBLE counter for L2 cache to
  HBM counters and metrics

* Update MI350 counter definitions
    * FETCH_SIZE
    * BANDWIDTH_EA

* Update MI350 metrics definitions
    * System Speed of Light, L2-Fabric Read BW
    * Roofline Plot Points, AI (Arithmetic Intensity) HBM
    * Roofline Performance Rates, HBM Bandwidth

* Remove redundant definition for gfx950 and fix BANDWIDTH_EA definition

Test HBM bandwidth metric for memcopy workload

* Add memcopy.cpp workload

* Add metric validation test suite to validate HBM Bandwidth metric for
  memcopy workload

* Move gpu_soc() to test_utils.py for better re-usability

* Update TUI analysis config

* Fix hbm bandwidth formula for mi350 in calc_ai_profile

Co-authored-by: Alysa Liu <Alysa.Liu@amd.com>
Этот коммит содержится в:
vedithal-amd
2026-01-23 15:56:24 -05:00
коммит произвёл GitHub
родитель 809eca7616
Коммит aa5dfb98f9
24 изменённых файлов: 528 добавлений и 223 удалений
+2 -2
Просмотреть файл
@@ -38,8 +38,8 @@ python3 -m pip install -r requirements.txt
## Testing
Populate the <username> variable in `docker/docker-compose.customrocmtest.yml`.
Populate the <tarball_name> variable in `docker/Dockerfile.customrocmtest` based on latest TheRock nightly build information.
Populate the <usename> variable in `docker/docker-compose.customrocmtest.yml`.
Populate the <rocm_build_image> variable in `docker/Dockerfile.customrocmtest` based on latest ROCm CI build information.
To quickly get the environment (bash shell) for building and testing, run the following commands:
* `cd docker`