* Create VERSION.sha file while creating standalone binary
* Do not fail on VERSION.sha not found
* Add git to docker container for testing
[ROCm/rocprofiler-compute commit: 19a9365bb9]
- For H2D transfer cases, we cannot enfore ErrorNotReady behavior if the
source is unpinned. In that case the driver can sync depending on if
it pins or stages the source buffer.
Change-Id: I436a4b667bff172cfbadc58631a46a215c40fcb1
[ROCm/hip-tests commit: d8f5c2560f]
- For H2D transfer cases, we cannot enfore ErrorNotReady behavior if the
source is unpinned. In that case the driver can sync depending on if
it pins or stages the source buffer.
Change-Id: I436a4b667bff172cfbadc58631a46a215c40fcb1
1) Create getBuffer method to return pinned host memory or staging buffer
2) for D2H path use managed buffer instead of static buffer
3) use staging buffer copy for 16KB < size < 1MB
4) use pinned memory copy for size > 1MB
Change-Id: I13d4d6ab60691bc6c7724239db1e11e23f0f3dc2
A HSA_IMAGE_ENABLE_3D_SWIZZLE_DEBUG environment flag exists already to
enable/disable this. Default value is false (view3dAs2dArray = 1)
Enabling this flag will enable support for swizzles that do 3D
interleaving on GFX9, GF10 and GFX11. By default support for swizzles that
do 3D interleaving is disabled.
A HSA_IMAGE_ENABLE_3D_SWIZZLE_DEBUG environment flag exists already to
enable/disable this. Default value is false (view3dAs2dArray = 1)
Enabling this flag will enable support for swizzles that do 3D
interleaving on GFX9, GF10 and GFX11. By default support for swizzles that
do 3D interleaving is disabled.
[ROCm/ROCR-Runtime commit: 0984a1f0fd]
Changed amdsmi_get_rocm_version() to be an API in the python library only.
Updated usage and version detection
Updated path detection of librocm-core.so
Updated docs to reflect both amdsmi_get_rocm_version and amdsmi_get_lib_version() do not require initialization.
Signed-off-by: Arif, Maisam <Maisam.Arif@amd.com>
Changed amdsmi_get_rocm_version() to be an API in the python library only.
Updated usage and version detection
Updated path detection of librocm-core.so
Updated docs to reflect both amdsmi_get_rocm_version and amdsmi_get_lib_version() do not require initialization.
Signed-off-by: Arif, Maisam <Maisam.Arif@amd.com>
[ROCm/amdsmi commit: 52b3ee2dc6]
- Add rocDecode API Tracing support using domain `rocjpeg_api` in ROCPROFSYS_ROCM_DOMAINS.
- Modify existing `videodecode` and `jpegdecode` ctests to verify API tracing
- Print Perfetto values for easy debugging in verbose mode
- Convert CMake error to a warning and skip building the "decode" examples if requirements are not found
[ROCm/rocprofiler-systems commit: 3bea1d8eac]
- Add rocDecode API Tracing support using domain `rocjpeg_api` in ROCPROFSYS_ROCM_DOMAINS.
- Modify existing `videodecode` and `jpegdecode` ctests to verify API tracing
- Print Perfetto values for easy debugging in verbose mode
- Convert CMake error to a warning and skip building the "decode" examples if requirements are not found
Disable Performance_hipExtLaunchKernelGGL_QueryGPUFrequency
as it should be a local perf test.
Change-Id: I7e94743d6958ffaadae9cff297e3f2d93f9ff817
[ROCm/hip-tests commit: ff516573f2]
Resource allocated in SetUp/HsaNodeInfo::Init,
needs be delete in TearDown/HsaNodeInfo::Delete.
Signed-off-by: James Zhu <James.Zhu@amd.com>
[ROCm/ROCR-Runtime commit: f8d8b8011f]
dependabot[bot]
Arif, Maisam
Williams, Justin
vedithal-amd
Saleel Kudchadker
Rahul Manocha
Jacob Lambert
Pengda Xie
Julia Jiang
Khatri, Shweta
Mewar, Deepak
Pham, Gabriel
Galantsev, Dmitrii
Ioannis Assiouras
Tony Gutierrez
Sajina PK
Kuricheti, Mythreya
taosang2
Branislav Brzak
James Zhu
Narlo, Joseph