hipBLASLt is a library that provides general matrix-matrix operations with a flexible API and extends funtionalities beyond traditional BLAS library. hipBLASLt is exposed APIs in HIP programming language with an underlying optimized generator as a backend kernel provider.
This library adds flexibility in matrix data layouts, input types, compute types, and also in choosing the algorithmic implementations and heuristics through parameter programmability. After a set of options for the intended GEMM operation are identified by the user, these options can be used repeatedly for different inputs. The GEMM operation of hipBLASLt is performed by hipblasLtMatmul API.
The equation is as below:
Where op(A)/op(B) refers to in-place operations such as transpose/non-transpose, and alpha, beta are scalars. Acitivation function supports Gelu, Relu. Bias vector match matrix D rows and broadcast to all D columns.
Here are data type supported list:
A | B | C | D | Compute(Scale) |
---|---|---|---|---|
fp32 | fp32 | fp32 | fp32 | fp32 |
fp16 | fp16 | fp16 | fp16 | fp32 |
bf16 | bf16 | bf16 | bf16 | fp32 |
Run the steps below to build documentation locally.
cd docs
pip3 install -r .sphinx/requirements.txt
python3 -m sphinx -T -E -b html -d _build/doctrees -D language=en . _build/html
- gfx90a card
- Git
- CMake 3.16.8 or later
- python3.7 or later
- python3.7-venv or later
- AMD ROCm 5.5 or later
- hipBLAS (for the header file)
You can build hipBLASLt using the install.sh script
# Clone hipBLASLt using git
git clone https://github.com/ROCmSoftwarePlatform/hipBLASLt
# Go to hipBLASLt directory
cd hipBLASLt
# Run install.sh script
# Command line options:
# -h|--help - prints help message
# -i|--install - install after build
# -d|--dependencies - install build dependencies
# -c|--clients - build library clients too (combines with -i & -d)
# -g|--debug - build with debug flag
./install.sh -idc
To build unit tests, hipBLASLt has to be built with --clients.
All unit tests are in path build/release/clients/staging/.
Please check these links for more information.
hipblaslt-test
hipblaslt-bench
example_XXX
Please use the issue tracker for bugs and feature requests.
The license file can be found in the main repository.