feat: add hipBlas support (#94)

This commit is contained in:
旺旺碎冰冰 authored and GitHub committed 2024-01-14 11:53:42 +08:00
1 parent 5c614e4bc2
commit c6071fa82f
6 files changed
+113 -2

No files matched your search

+11
View File
@@ -117,6 +117,17 @@ cmake .. -DSD_CUBLAS=ON
cmake --build . --config Release
```
##### Using HipBLAS
This provides BLAS acceleration using the ROCm cores of your AMD GPU. Make sure to have the ROCm toolkit installed.
Windows User Refer to [docs/hipBLAS_on_Windows.md](docs%2FhipBLAS_on_Windows.md) for a comprehensive guide.
```
cmake .. -G "Ninja" -DCMAKE_C_COMPILER=clang -DCMAKE_CXX_COMPILER=clang++ -DSD_HIPBLAS=ON -DCMAKE_BUILD_TYPE=Release -DAMDGPU_TARGETS=gfx1100
cmake --build . --config Release
```
##### Using Metal
Using Metal makes the computation run on the GPU. Currently, there are some issues with Metal when performing operations on very large matrices, making it highly inefficient at the moment. Performance improvements are expected in the near future.