Source: magma-rocm Section: math Homepage: https://icl.cs.utk.edu/magma/ Priority: optional Standards-Version: 4.7.0 Vcs-Git: https://salsa.debian.org/science-team/magma.git Vcs-Browser: https://salsa.debian.org/science-team/magma Maintainer: Debian Science Maintainers Uploaders: Mo Zhou , Cordell Bloor Build-Depends: debhelper-compat (= 13), cmake, gfortran, hipcc, libhipblas-dev, libhipsparse-dev, libopenblas-dev, librocblas-dev, librocsparse-dev Build-Depends-Indep: doxygen Package: libmagma2-rocm Section: libs Architecture: amd64 arm64 ppc64el Multi-Arch: no Depends: ${misc:Depends}, ${shlibs:Depends} Breaks: libmagma2 Replaces: libmagma2 Description: Matrix Algebra on GPU and Multicore Architectures (solib) The MAGMA project aims to develop a dense linear algebra library similar to LAPACK but for heterogeneous/hybrid architectures, starting with current "Multicore+GPU" systems. . The MAGMA research is based on the idea that, to address the complex challenges of the emerging hybrid environments, optimal software solutions will themselves have to hybridize, combining the strengths of different algorithms within a single framework. Building on this idea, we aim to design linear algebra algorithms and frameworks for hybrid manycore and GPU systems that can enable applications to fully exploit the power that each of the hybrid components offers. . This package contains the shared object libmagma.so.2 as built for the AMD ROCm backend. Package: libmagma-sparse2-rocm Section: libs Architecture: amd64 arm64 ppc64el Multi-Arch: no Depends: ${misc:Depends}, ${shlibs:Depends} Breaks: libmagma-sparse2 Replaces: libmagma-sparse2 Description: Matrix Algebra on GPU and Multicore Architectures (solib, sparse) The MAGMA project aims to develop a dense linear algebra library similar to LAPACK but for heterogeneous/hybrid architectures, starting with current "Multicore+GPU" systems. . The MAGMA research is based on the idea that, to address the complex challenges of the emerging hybrid environments, optimal software solutions will themselves have to hybridize, combining the strengths of different algorithms within a single framework. Building on this idea, we aim to design linear algebra algorithms and frameworks for hybrid manycore and GPU systems that can enable applications to fully exploit the power that each of the hybrid components offers. . This package contains the shared object libmagma_sparse.so.2 as built for the AMD ROCm backend. Package: libmagma-rocm-dev Section: libdevel Architecture: amd64 arm64 ppc64el Multi-Arch: no Depends: ${misc:Depends}, ${shlibs:Depends}, libmagma2-rocm (= ${binary:Version}), libmagma-sparse2-rocm (= ${binary:Version}), Breaks: libmagma-dev Replaces: libmagma-dev Description: Matrix Algebra on GPU and Multicore Architectures (dev) The MAGMA project aims to develop a dense linear algebra library similar to LAPACK but for heterogeneous/hybrid architectures, starting with current "Multicore+GPU" systems. . The MAGMA research is based on the idea that, to address the complex challenges of the emerging hybrid environments, optimal software solutions will themselves have to hybridize, combining the strengths of different algorithms within a single framework. Building on this idea, we aim to design linear algebra algorithms and frameworks for hybrid manycore and GPU systems that can enable applications to fully exploit the power that each of the hybrid components offers. . This package contains the headers and the static libraries. Package: libmagma-rocm-test Architecture: amd64 arm64 ppc64el Multi-Arch: no Depends: ${misc:Depends}, ${shlibs:Depends}, python3 Description: Matrix Algebra on GPU and Multicore Architectures (tests) The MAGMA project aims to develop a dense linear algebra library similar to LAPACK but for heterogeneous/hybrid architectures, starting with current "Multicore+GPU" systems. . The MAGMA research is based on the idea that, to address the complex challenges of the emerging hybrid environments, optimal software solutions will themselves have to hybridize, combining the strengths of different algorithms within a single framework. Building on this idea, we aim to design linear algebra algorithms and frameworks for hybrid manycore and GPU systems that can enable applications to fully exploit the power that each of the hybrid components offers. . This package contains the test executables. Package: libmagma-rocm-doc Section: doc Architecture: all Depends: ${misc:Depends} Description: Matrix Algebra on GPU and Multicore Architectures (doc) The MAGMA project aims to develop a dense linear algebra library similar to LAPACK but for heterogeneous/hybrid architectures, starting with current "Multicore+GPU" systems. . The MAGMA research is based on the idea that, to address the complex challenges of the emerging hybrid environments, optimal software solutions will themselves have to hybridize, combining the strengths of different algorithms within a single framework. Building on this idea, we aim to design linear algebra algorithms and frameworks for hybrid manycore and GPU systems that can enable applications to fully exploit the power that each of the hybrid components offers. . This package contains the Doxygen HTML documentation.