Movatterモバイル変換


[0]ホーム

URL:


Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
Appearance settings

[DEPRECATED] Moved to ROCm/rocm-libraries repo

License

NotificationsYou must be signed in to change notification settings

ROCm/hipBLASLt

Repository files navigation

Caution

The hipBLASLt repository is retired, please use theROCm/rocm-libraries repository

hipBLASLt is a library that provides general matrix-matrix operations. It has a flexible API that extendsfunctionalities beyond a traditional BLAS library, such as adding flexibility to matrix data layouts, inputtypes, compute types, and algorithmic implementations and heuristics.

Note

The published hipBLASLt documentation is available athipBLASLt in an organized, easy-to-read format, with search and a table of contents. The documentation source files reside in the hipBLASLt/docs folder of this repository. As with all ROCm projects, the documentation is open source. For more information, seeContribute to ROCm documentation.

hipBLASLt uses the HIP programming language with an underlying optimized generator as its backendkernel provider.

After you specify a set of options for a matrix-matrix operation, you can reuse these for differentinputs. The general matrix-multiply (GEMM) operation is performed by thehipblasLtMatmul API.

The equation is:

$$D = Activation(alpha \cdot op(A) \cdot op(B) + beta \cdot op(C) + bias)$$

Whereop( ) refers to in-place operations, such as transpose and non-transpose, andalpha andbeta are scalars.

The activation function supports GELU, ReLU, and Swish (SiLU). the bias vector matches matrix D rows andbroadcasts to all D columns.

For the supported data types, seeSupported data types.

Documentation

Full documentation for hipBLASLt is available atrocm.docs.amd.com/projects/hipBLASLt.

Run the steps below to build documentation locally.

cd docspip3 install -r sphinx/requirements.txtpython3 -m sphinx -T -E -b html -d _build/doctrees -D language=en. _build/html

Alternatively, build with CMake:

cmake -DBUILD_DOCS=ON ...

Requirements

To install hipBLASLt, you must meet the following requirements:

Required hardware:

  • gfx90a card
  • gfx94x card
  • gfx110x card

Required software:

Build and install

You can build hipBLASLt using theinstall.sh script:

# Clone hipBLASLt using gitgit clone https://github.com/ROCmSoftwarePlatform/hipBLASLt# Go to hipBLASLt directorycd hipBLASLt# Run requirements.txt in folder tensilelitepython3 -m pip install -r tensilelite/requirements.txt# 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

NOTE: To build hipBLASLt for ROCm <= 6.2, pass the--legacy_hipblas_direct flag toinstall.sh

Unit tests

All unit tests are located inbuild/release/clients/staging/. To build these tests, you must buildhipBLASLt with--clients.

You can find more information at the following links:

TensileLite Host Library Tests

To build and run TensileLite Host Library Tests, use the following commands:

 cd tensilelite && mkdir build && cd build cmake -DTENSILE_DISABLE_CTEST=OFF -DCMAKE_BUILD_TYPE=RelWithDebInfo  -DCMAKE_CXX_COMPILER=/opt/rocm/bin/amdclang++ -DTensile_ROOT=$(pwd)/../Tensile ../HostLibraryTests make -j ./TensileTests

Contribute

If you want to submit an issue, you can do so onGitHub.

To contribute to our repository, you can create a GitHub pull request.


[8]ページ先頭

©2009-2025 Movatter.jp