Source: rocwmma Section: devel Priority: optional Maintainer: Debian ROCm Team Uploaders: Talha Can Havadar Standards-Version: 4.7.4 Homepage: https://github.com/ROCm/rocWMMA Build-Depends: cmake, debhelper-compat (= 13), hipcc (>= 7.2.4~), libamdhip64-dev (>= 7.2.0~), libgtest-dev , libomp-22-dev, lld-22, librocblas-dev (>= 7.2.0~), librocm-smi-dev (>= 7.2.0~), pkg-rocm-tools (>= 0.9.8~), rocm-cmake, rocm-device-libs-22 (>= 7.2.0~) Build-Depends-Indep: dh-sequence-sphinxdoc , libjs-jquery , libjs-mathjax , libjs-sphinxdoc , libjs-underscore , python3-doxysphinx , python3-rocm-docs (>= 1.20.0-1~) , python3-sphinx-external-toc Rules-Requires-Root: no Package: librocwmma-dev Section: libdevel Architecture: all Depends: ${misc:Depends}, ${shlibs:Depends} Description: development files for rocWMMA C++ library rocWMMA is a C++ library for accelerating mixed-precision matrix multiply-accumulate (MMA) operations leveraging AMD GPU hardware. rocWMMA makes it easier to break down MMA problems into fragments and distribute block-wise MMA operations in parallel across GPU wavefronts. The API consists of a header library, that can be used to compile MMA acceleration directly into GPU kernel device code. This can benefit from compiler optimization in the generation of kernel assembly, and doesn't incur additional overhead costs of linking to external runtime libraries or having to launch separate kernels. . Development files and headers for rocwmma. Package: librocwmma-tests Architecture: amd64 arm64 Multi-Arch: foreign XB-X-ROCm-GPU-Architecture: ${rocm:GPU-Architecture} Depends: ${misc:Depends}, ${shlibs:Depends} Description: test binaries for rocWMMA C++ library rocWMMA is a C++ library for accelerating mixed-precision matrix multiply-accumulate (MMA) operations leveraging AMD GPU hardware. rocWMMA makes it easier to break down MMA problems into fragments and distribute block-wise MMA operations in parallel across GPU wavefronts. The API consists of a header library, that can be used to compile MMA acceleration directly into GPU kernel device code. This can benefit from compiler optimization in the generation of kernel assembly, and doesn't incur additional overhead costs of linking to external runtime libraries or having to launch separate kernels. . Test files for rocwmma. Package: librocwmma-tests-validate Architecture: amd64 arm64 Multi-Arch: foreign XB-X-ROCm-GPU-Architecture: ${rocm:GPU-Architecture} Depends: ${misc:Depends}, ${shlibs:Depends} Description: test binaries (with -validate) for rocWMMA C++ library rocWMMA is a C++ library for accelerating mixed-precision matrix multiply-accumulate (MMA) operations leveraging AMD GPU hardware. rocWMMA makes it easier to break down MMA problems into fragments and distribute block-wise MMA operations in parallel across GPU wavefronts. The API consists of a header library, that can be used to compile MMA acceleration directly into GPU kernel device code. This can benefit from compiler optimization in the generation of kernel assembly, and doesn't incur additional overhead costs of linking to external runtime libraries or having to launch separate kernels. . Validate test files for rocwmma. Containing test binaries with -validate suffix. Package: librocwmma-docs Section: doc Architecture: all Depends: ${misc:Depends}, ${shlibs:Depends} Description: documentation for rocWMMA C++ library rocWMMA is a C++ library for accelerating mixed-precision matrix multiply-accumulate (MMA) operations leveraging AMD GPU hardware. rocWMMA makes it easier to break down MMA problems into fragments and distribute block-wise MMA operations in parallel across GPU wavefronts. The API consists of a header library, that can be used to compile MMA acceleration directly into GPU kernel device code. This can benefit from compiler optimization in the generation of kernel assembly, and doesn't incur additional overhead costs of linking to external runtime libraries or having to launch separate kernels. . Documentation files for rocwmma.