Welcome to the release notes for the ROCm platform.
A new functionality rocblas-gemm-tune and an environment variable ROCBLAS_TENSILE_GEMM_OVERRIDE_PATH are added to rocBLAS in the ROCm 5.7.1 release.
rocblas-gemm-tune is used to find the best-performing GEMM kernel for each GEMM problem set. It has a command line interface, which mimics the --yaml input used by rocblas-bench. To generate the expected --yaml input, profile logging can be used, by setting the environment variable ROCBLAS_LAYER4.
For more information on rocBLAS logging, see Logging in rocBLAS, in the API Reference Guide.
An example input file: Expected output (note selected GEMM idx may differ): Where the far right values (solution_index) are the indices of the best-performing kernels for those GEMMs in the rocBLAS kernel library. These indices can be directly used in future GEMM calls. See rocBLAS/samples/example_user_driven_tuning.cpp for sample code of directly using kernels via their indices.
If the output is stored in a file, the results can be used to override default kernel selection with the kernels found, by setting the environment variable ROCBLAS_TENSILE_GEMM_OVERRIDE_PATH, where points to the stored file.
For more details, refer to the rocBLAS Programmer's Guide.
ROCm 5.7.1 is a point release with several bug fixes in the HIP runtime.
The hipPointerGetAttributes API returns the correct HIP memory type as hipMemoryTypeManaged for managed memory.
| Library | Version |
|---|---|
| hipBLAS | 1.1.0 |
| hipCUB | 2.13.1 |
| hipFFT | 1.0.12 |
| hipSOLVER | 1.8.1 ⇒ 1.8.2 |
| hipSPARSE | 2.3.8 |
| MIOpen | 2.19.0 |
| rocALUTION | 2.1.11 |
| rocBLAS | 3.1.0 |
| rocFFT | 1.0.24 |
| rocm-cmake | 0.10.0 |
| rocPRIM | 2.13.1 |
| rocRAND | 2.10.17 |
| rocSOLVER | 3.23.0 |
| rocSPARSE | 2.5.4 |
| rocThrust | 2.18.0 |
| rocWMMA | 1.2.0 |
| Tensile | 4.38.0 |
hipSOLVER 1.8.2 for ROCm 5.7.1
- Fixed conflicts between the hipsolver-dev and -asan packages by excluding hipsolver_module.f90 from the latter