Fix rocprofiler usage in ROCm >= 5.5.x
- starting with ROCm 5.5.0, rocprofiler throws exception if OnLoad + dlopen librocprofiler
- CI skipped for this PR since CI does not support GPU usage (tested locally)
[ROCm/rocprofiler-systems commit: 223536896b]
Fix rocprofiler usage in ROCm >= 5.5.x
- starting with ROCm 5.5.0, rocprofiler throws exception if OnLoad + dlopen librocprofiler
- CI skipped for this PR since CI does not support GPU usage (tested locally)
kokkosp update
- disable tracing deep copies by default because inconsistent begin/end corrupts call-stack hierarchy
- tracing deep copy operations can be enabled via `OMNITRACE_KOKKOSP_DEEP_COPY`
[ROCm/rocprofiler-systems commit: 262f1e9299]
kokkosp update
- disable tracing deep copies by default because inconsistent begin/end corrupts call-stack hierarchy
- tracing deep copy operations can be enabled via `OMNITRACE_KOKKOSP_DEEP_COPY`
- disable deprecated function use warning
- disalbe size_t to .type' warning
- disable conversion from 'type1' to 'type2' warning
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I64161fd37cf56de3d132102103267ae8da40193a
- Replace dynamic allocation with embedded struct in hip graph
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: Idfe3f7393fa50d6e510d4c88f779408f96d3accb
Initial implementation for hipMemPoolExportToShareableHandle,
hipMemPoolImportFromShareableHandle,
hipMemPoolExportPointer and hipMemPoolImportPointer
Change-Id: I0ebdc48e9163b394ded560adca6c38bbc5aee7d1
This has slowly become less and less reliable on more and more ASICs,
so just blacklist it altogether. Using wall clock for performance
is not a reliable method for testing performance, so skip it to avoid
more failure reports on various systems.
Change-Id: I1a5744604e4620bc7675a629d146ba4ffba669d2
This has slowly become less and less reliable on more and more ASICs,
so just blacklist it altogether. Using wall clock for performance
is not a reliable method for testing performance, so skip it to avoid
more failure reports on various systems.
Change-Id: I1a5744604e4620bc7675a629d146ba4ffba669d2
[ROCm/ROCR-Runtime commit: 9a22bade89]
- Use HIP_USE_SDMA_QUERY env var toggle for new API use. Env var is 0 by
default
Change-Id: If725a0c41e15f78a1a6c3f47942954fe9240b4db
[ROCm/clr commit: 8d193c32bb]
This reverts commit 2e664d2492.
Reason for revert: Performance regressions and failures observed. Need to investigate those and before re-applying patch
Change-Id: I42ba0605797f9bdcfb5d5102927dd01405cf05e3
[ROCm/clr commit: 8047d8e3e8]
This reverts commit a4c5c44008.
Reason for revert: Performance regressions and failures observed. Need to investigate those and before re-applying patch
Change-Id: I42ba0605797f9bdcfb5d5102927dd01405cf05e3
Explicitly mention that IPC handles can only be created on GPU agents.
Change-Id: I19bc3578d6e5243c795bf6fbf981ea4bd3bfc2e8
[ROCm/ROCR-Runtime commit: 5e4490f180]
Remove docker hiplibsdk from amdgpu-install
- amdgpu-install use case hiplibsdk is not necessary and bloats the install
- same as above for package rocm-hip-sdk
[ROCm/rocprofiler-systems commit: 2ce0cb4a19]
Remove docker hiplibsdk from amdgpu-install
- amdgpu-install use case hiplibsdk is not necessary and bloats the install
- same as above for package rocm-hip-sdk
Will replace rocm-ocl-icd with packaging of khronos opencl-icd-loader
Change-Id: I26da58e80fd1426b9908d0b524f93b17ab27626d
[ROCm/clr commit: a4bba6d705]