Don't pass the --enable-new-dtags flag to clang, just the LLVM's default
Change-Id: I2bd06181c490e1c0fec3241a404b546de3ebe970
[ROCm/hip commit: 5e7b7e11a5]
It doesn't seem to provide any functionality, probably leftover from some experimental codes.
Change-Id: I1720ccf29e057d7f2e18a7dfd1adccd3d51fa80c
[ROCm/hip commit: cb084f4072]
The tracer tool needs to remember the begin timestamps for API
callbacks, and uses a thread_local std::stack for that purpose.
The issue with thread_local objects is that they are destructed
before anything else when the main thread exits. To work around
that issue, we use a "safe" stack in the roctracer API.
Use the same "safe" stack in the tracer tool.
Change-Id: I0d69d4eb44f0205f4102d0d5ef9803a1ec1800a5
[ROCm/roctracer commit: b664937ebd]
The tracer tool needs to remember the begin timestamps for API
callbacks, and uses a thread_local std::stack for that purpose.
The issue with thread_local objects is that they are destructed
before anything else when the main thread exits. To work around
that issue, we use a "safe" stack in the roctracer API.
Use the same "safe" stack in the tracer tool.
Change-Id: I0d69d4eb44f0205f4102d0d5ef9803a1ec1800a5
rocprof errors out with the following message:
symbol lookup 'KernelNameRef' failed: libamdhip64.so.5: undefined \
symbol: KernelNameRef
The HipLoader is incorrectly looking for a KernelNameRef symbol
instead of hipKernelNameRef.
Fixed the typo: KernelNameRef -> hipKernelNameRef.
Change-Id: Ia4860e1669707b0c83d67e71b78d362b07a6aaa7
[ROCm/roctracer commit: a287f20961]
rocprof errors out with the following message:
symbol lookup 'KernelNameRef' failed: libamdhip64.so.5: undefined \
symbol: KernelNameRef
The HipLoader is incorrectly looking for a KernelNameRef symbol
instead of hipKernelNameRef.
Fixed the typo: KernelNameRef -> hipKernelNameRef.
Change-Id: Ia4860e1669707b0c83d67e71b78d362b07a6aaa7
- More to come in later commit, below is just tidying some stuff up
- clang-tidy
- mpi_gotcha quiet about not finding funcs
- update to new papi config
- sampling block_samples / unblock_samples
- disable calling component's sample functions within sampler
- release doesn't strip library
- remove HSA and ROCP env variables from modulefile / setup-env
- preliminary support for LD_PRELOAD usage
- default sampling rate is 300 interrupts / second
- fixes various deadlock issues at startup
[ROCm/rocprofiler-systems commit: 8f36620e29]
- More to come in later commit, below is just tidying some stuff up
- clang-tidy
- mpi_gotcha quiet about not finding funcs
- update to new papi config
- sampling block_samples / unblock_samples
- disable calling component's sample functions within sampler
- release doesn't strip library
- remove HSA and ROCP env variables from modulefile / setup-env
- preliminary support for LD_PRELOAD usage
- default sampling rate is 300 interrupts / second
- fixes various deadlock issues at startup
Make sure all calls calling cudaMalloc (including devCommSetup) are
called before the last bootstrapBarrier. That way, we avoid calls to
cudaMalloc be blocked by a NCCL kernel launched on another GPU by
another thread which completed init faster.
Resolve#623.
[ROCm/rccl commit: ecab28a7c9]
Make sure all calls calling cudaMalloc (including devCommSetup) are
called before the last bootstrapBarrier. That way, we avoid calls to
cudaMalloc be blocked by a NCCL kernel launched on another GPU by
another thread which completed init faster.
Resolve#623.
- Fix minor bug in helper test function in Unit_hipEvent.cc
- Move negative test from Unit_hipEventElapsedTime_DifferentDevices to Unit_hipEventRecord_Negative
- Add Unit_hipEventElapsedTime_NotReady_Negative test when function is called on event that has not been completed
[ROCm/hip-tests commit: f72a34b807]
- Fix minor bug in helper test function in Unit_hipEvent.cc
- Move negative test from Unit_hipEventElapsedTime_DifferentDevices to Unit_hipEventRecord_Negative
- Add Unit_hipEventElapsedTime_NotReady_Negative test when function is called on event that has not been completed
- Fix minor bug in helper test function in Unit_hipEvent.cc
- Move negative test from Unit_hipEventElapsedTime_DifferentDevices to Unit_hipEventRecord_Negative
- Add Unit_hipEventElapsedTime_NotReady_Negative test when function is called on event that has not been completed
[ROCm/hip commit: 5bf0dbe5ad]
- Fix minor bug in helper test function in Unit_hipEvent.cc
- Move negative test from Unit_hipEventElapsedTime_DifferentDevices to Unit_hipEventRecord_Negative
- Add Unit_hipEventElapsedTime_NotReady_Negative test when function is called on event that has not been completed
* Implement printing out of device count to standard output
* Implement printing out of device attributes retrieved by hipDeviceGetAttribute.
* Implement printing of device properties retrieved by hipGetDeviceProperties.
[ROCm/hip-tests commit: 51a22bb085]
* Implement printing out of device count to standard output
* Implement printing out of device attributes retrieved by hipDeviceGetAttribute.
* Implement printing of device properties retrieved by hipGetDeviceProperties.
* Implement printing out of device count to standard output
* Implement printing out of device attributes retrieved by hipDeviceGetAttribute.
* Implement printing of device properties retrieved by hipGetDeviceProperties.
[ROCm/hip commit: 8cb701166d]
* Implement printing out of device count to standard output
* Implement printing out of device attributes retrieved by hipDeviceGetAttribute.
* Implement printing of device properties retrieved by hipGetDeviceProperties.
- Add testing for null stream in Unit_hipStreamQuery_WithNoWork
- Add testing for null stream in Unit_hipStreamQuery_WithFinishedWork
[ROCm/hip-tests commit: 4795cc6265]
- Add testing for null stream in Unit_hipStreamQuery_WithNoWork
- Add testing for null stream in Unit_hipStreamQuery_WithFinishedWork
[ROCm/hip commit: 4afc7866b5]
- Fix bug in calculation of normal priority in Unit_hipStreamGetPriority_StreamsWithCUMask tests
- Delete duplicate test Unit_hipStreamGetPriority_InvalidPriorityPointer
- Separate Unit_hipStreamAddCallback_ParamTst into positive and negative test
- Remove Unit_hipStreamDestroy_Negative_DoubleDestroy test for Nvidia devices due to error
[ROCm/hip-tests commit: 968e21cd1f]
- Fix bug in calculation of normal priority in Unit_hipStreamGetPriority_StreamsWithCUMask tests
- Delete duplicate test Unit_hipStreamGetPriority_InvalidPriorityPointer
- Separate Unit_hipStreamAddCallback_ParamTst into positive and negative test
- Remove Unit_hipStreamDestroy_Negative_DoubleDestroy test for Nvidia devices due to error
- Fix bug in calculation of normal priority in Unit_hipStreamGetPriority_StreamsWithCUMask tests
- Delete duplicate test Unit_hipStreamGetPriority_InvalidPriorityPointer
- Separate Unit_hipStreamAddCallback_ParamTst into positive and negative test
- Remove Unit_hipStreamDestroy_Negative_DoubleDestroy test for Nvidia devices due to error
[ROCm/hip commit: d60b65b5dc]
- Fix bug in calculation of normal priority in Unit_hipStreamGetPriority_StreamsWithCUMask tests
- Delete duplicate test Unit_hipStreamGetPriority_InvalidPriorityPointer
- Separate Unit_hipStreamAddCallback_ParamTst into positive and negative test
- Remove Unit_hipStreamDestroy_Negative_DoubleDestroy test for Nvidia devices due to error
- Add hipDeviceGetStreamPriorityRange.cc file with a test for simple device stream priority range check
- Add newly created file into CMakeLists.txt
[ROCm/hip-tests commit: 26d316a250]
- Add hipDeviceGetStreamPriorityRange.cc file with a test for simple device stream priority range check
- Add newly created file into CMakeLists.txt
[ROCm/hip commit: 66471b01c9]
- Add hipEventSynchronize.cc file with positive unit tests: Simple recorded event synchronization and without event recording
- Add newly created file into CMakeLists.txt
[ROCm/hip-tests commit: bc83dae816]
- Add hipEventSynchronize.cc file with positive unit tests: Simple recorded event synchronization and without event recording
- Add newly created file into CMakeLists.txt
- Add hipEventSynchronize.cc file with positive unit tests: Simple recorded event synchronization and without event recording
- Add newly created file into CMakeLists.txt
[ROCm/hip commit: 7b8feec246]
- Add hipEventSynchronize.cc file with positive unit tests: Simple recorded event synchronization and without event recording
- Add newly created file into CMakeLists.txt
Janek van Oirschot
Siu Chi Chan
ROCm CI Service Account
Laurent Morichetti
Wen-Heng (Jack) Chung
Jonathan R. Madsen
Sylvain Jeaugey
nives-vukovic
Mirza Halilčević