AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update HSA async copy active signals handling (#732)

Browse Source 
* Enable INFO logging on retried CI jobs

* Update lib/rocprofiler-sdk/async_copy.cpp

- rework active_signals
  - make hsa_signal_t member variable
  - remove sync from destructor
  - replace _is_set with atomic counter
  - timeout of 30 seconds hsa_signal_wait
  - switch from relaxed to scacquire/screlease memory ordering
- improve logging and error handling
- destroy hsa signal in active_signals in async_fini

* Update lib/rocprofiler-sdk/async_copy.cpp

- active_signals::create
- change initial value of signal to 1 instead of value of completion signal
- change condition trigger of signal callback

* Update tests/counter-collection/validate.py

* Update lib/rocprofiler-sdk/async_copy.cpp

- improved logging
- fix hsa_signal_wait_scacquire_fn check

* Cleanup tests/lib/transpose/transpose.cpp

- remove huge comment block

* Appears to be working on MI200

Dependency Versions:

clr: f7b1398361  - compile mode: release

hsa-runtime: 4cd6c62f25dbbdbaa8580dd4ad8f388c98c508da - compile mode: RelWithDebug

* Update source/lib/rocprofiler-sdk/hsa/async_copy.cpp

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* Format fix

---------

Co-authored-by: Benjamin Welton <bewelton@amd.com>
Co-authored-by: Ammar ELWazir <ammar.elwazir@amd.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Ammar ELWazir <aelwazir@hpe6u-21.amd.com>
This commit is contained in:
Jonathan R. Madsen
2024-04-09 08:31:08 -05:00
committed by GitHub
parent e2d8ccad4b
commit 8c5399a68a
5 changed files with 167 additions and 91 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.github/workflows/continuous_integration.yml
+6 -1
View File
@@ -72,6 +72,7 @@ jobs:
--build-jobs 16
--site $(echo $RUNNER_HOSTNAME)-$(/opt/rocm/bin/rocm_agent_enumerator | sed -n '2 p')
--gpu-targets ${{ env.GPU_LIST }}
--run-attempt ${{ github.run_attempt }}
${{ matrix.ci-flags }}
--
-DROCPROFILER_DEP_ROCMCORE=ON
@@ -211,6 +212,7 @@ jobs:
--site $(echo $RUNNER_HOSTNAME)-$(/opt/rocm/bin/rocm_agent_enumerator | sed -n '2 p')
--gpu-targets ${{ env.GPU_LIST }}
--coverage all
--run-attempt ${{ github.run_attempt }}
--
-DCMAKE_BUILD_TYPE=${{ matrix.build-type }}
-DPython3_EXECUTABLE=$(which python3)
@@ -228,6 +230,7 @@ jobs:
--site $(echo $RUNNER_HOSTNAME)-$(/opt/rocm/bin/rocm_agent_enumerator | sed -n '2 p')
--gpu-targets ${{ env.GPU_LIST }}
--coverage tests
--run-attempt ${{ github.run_attempt }}
--
-DCMAKE_BUILD_TYPE=${{ matrix.build-type }}
-DPython3_EXECUTABLE=$(which python3)
@@ -245,6 +248,7 @@ jobs:
--site $(echo $RUNNER_HOSTNAME)-$(/opt/rocm/bin/rocm_agent_enumerator | sed -n '2 p')
--gpu-targets ${{ env.GPU_LIST }}
--coverage samples
--run-attempt ${{ github.run_attempt }}
--
-DCMAKE_BUILD_TYPE=${{ matrix.build-type }}
-DPython3_EXECUTABLE=$(which python3)
@@ -400,7 +404,8 @@ jobs:
--build-jobs 16
--site $(echo $RUNNER_HOSTNAME)-$(/opt/rocm/bin/rocm_agent_enumerator | sed -n '2 p')
--gpu-targets ${{ env.GPU_LIST }}
--memcheck=${{ matrix.sanitizer }}
--memcheck ${{ matrix.sanitizer }}
--run-attempt ${{ github.run_attempt }}
${{ matrix.ci-flags }}
--
-DCMAKE_BUILD_TYPE=${{ matrix.build-type }}