starpu-contrib 1.4.1+dfsg-4 source package in Ubuntu

Changelog

starpu-contrib (1.4.1+dfsg-4) unstable; urgency=medium

  * rules: Also hardcode the maximum number of cpus and NUMA nodes.

 -- Samuel Thibault <email address hidden>  Fri, 14 Jul 2023 16:07:57 +0200

Upload details

Uploaded by:
Samuel Thibault
Uploaded to:
Sid
Original maintainer:
Samuel Thibault
Architectures:
any
Section:
libs
Urgency:
Medium Urgency

See full publishing history Publishing

Series Pocket Published Component Section
Mantic release multiverse libs

Downloads

File Size SHA-256 Checksum
starpu-contrib_1.4.1+dfsg-4.dsc 3.3 KiB 0daa92f18c861d0c4e9d15d797d50f25afac34195cda822ab6e7c4ee36ac5488
starpu-contrib_1.4.1+dfsg.orig.tar.gz 7.5 MiB d8b532fcd782ed6a7396fa5a701c3c73fbf4130db56c172af6e7d236e0ee2bc2
starpu-contrib_1.4.1+dfsg-4.debian.tar.xz 13.7 KiB 9cd38ce6b29f76262cc1e665e04d4cf06a0c94af2a91126c0210fb2a47f40ec0

No changes file available.

Binary packages built by this source

libsocl-contrib-1.4-1: Task scheduler for heterogeneous multicore machines

 StarPU is a runtime system that offers support for heterogeneous
 multicore machines. While many efforts are devoted to design efficient
 computation kernels for those architectures (e.g. to implement BLAS
 kernels on GPUs or on Cell's SPUs), StarPU not only takes care of
 offloading such kernels (and implementing data coherency across
 the machine), but it also makes sure the kernels are executed as
 efficiently as possible.
 .
 This package contains an OpenCL-compatible library interface to StarPU.
 This "contrib" version is built against CUDA.

libsocl-contrib-1.4-1-dbgsym: debug symbols for libsocl-contrib-1.4-1
libstarpu-contrib-1.4-2: Task scheduler for heterogeneous multicore machines

 StarPU is a runtime system that offers support for heterogeneous
 multicore machines. While many efforts are devoted to design efficient
 computation kernels for those architectures (e.g. to implement BLAS
 kernels on GPUs or on Cell's SPUs), StarPU not only takes care of
 offloading such kernels (and implementing data coherency across
 the machine), but it also makes sure the kernels are executed as
 efficiently as possible.
 .
 This package contains the main StarPU library
 This "contrib" version is built against CUDA.

libstarpu-contrib-1.4-2-dbgsym: debug symbols for libstarpu-contrib-1.4-2
libstarpu-contrib-dev: Task scheduler for heterogeneous multicore machines - dev

 StarPU is a runtime system that offers support for heterogeneous
 multicore machines. While many efforts are devoted to design efficient
 computation kernels for those architectures (e.g. to implement BLAS
 kernels on GPUs or on Cell's SPUs), StarPU not only takes care of
 offloading such kernels (and implementing data coherency across
 the machine), but it also makes sure the kernels are executed as
 efficiently as possible.
 .
 This package contains development headers and libraries.
 This "contrib" version is built against CUDA.

libstarpu-contrib-openmp-llvm-1.4-1: Task scheduler for heterogeneous multicore machines

 StarPU is a runtime system that offers support for heterogeneous
 multicore machines. While many efforts are devoted to design efficient
 computation kernels for those architectures (e.g. to implement BLAS
 kernels on GPUs or on Cell's SPUs), StarPU not only takes care of
 offloading such kernels (and implementing data coherency across
 the machine), but it also makes sure the kernels are executed as
 efficiently as possible.
 .
 This package contains the runtime support for LLVM OpenMP.
 This "contrib" version is built against CUDA.

libstarpu-contrib-openmp-llvm-1.4-1-dbgsym: debug symbols for libstarpu-contrib-openmp-llvm-1.4-1
libstarpu-contribfft-1.4-1: No summary available for libstarpu-contribfft-1.4-1 in ubuntu noble.

No description available for libstarpu-contribfft-1.4-1 in ubuntu noble.

libstarpu-contribfft-1.4-1-dbgsym: debug symbols for libstarpu-contribfft-1.4-1
libstarpu-contribmpi-1.4-2: Task scheduler for heterogeneous multicore machines

 StarPU is a runtime system that offers support for heterogeneous
 multicore machines. While many efforts are devoted to design efficient
 computation kernels for those architectures (e.g. to implement BLAS
 kernels on GPUs or on Cell's SPUs), StarPU not only takes care of
 offloading such kernels (and implementing data coherency across
 the machine), but it also makes sure the kernels are executed as
 efficiently as possible.
 .
 This package contains MPI extensions for StarPU.
 This "contrib" version is built against CUDA.

libstarpu-contribmpi-1.4-2-dbgsym: debug symbols for libstarpu-contribmpi-1.4-2
libstarpu-contribrm-1.4-1: Task scheduler for heterogeneous multicore machines

 StarPU is a runtime system that offers support for heterogeneous
 multicore machines. While many efforts are devoted to design efficient
 computation kernels for those architectures (e.g. to implement BLAS
 kernels on GPUs or on Cell's SPUs), StarPU not only takes care of
 offloading such kernels (and implementing data coherency across
 the machine), but it also makes sure the kernels are executed as
 efficiently as possible.
 .
 This package contains the resource management library.
 This "contrib" version is built against CUDA.

libstarpu-contribrm-1.4-1-dbgsym: debug symbols for libstarpu-contribrm-1.4-1
starpu-contrib-examples: Task scheduler for heterogeneous multicore machines - exs

 StarPU is a runtime system that offers support for heterogeneous
 multicore machines. While many efforts are devoted to design efficient
 computation kernels for those architectures (e.g. to implement BLAS
 kernels on GPUs or on Cell's SPUs), StarPU not only takes care of
 offloading such kernels (and implementing data coherency across
 the machine), but it also makes sure the kernels are executed as
 efficiently as possible.
 .
 This package contains application examples.
 This "contrib" version is built against CUDA.

starpu-contrib-examples-dbgsym: debug symbols for starpu-contrib-examples
starpu-contrib-tools: Task scheduler for heterogeneous multicore machines - tools

 StarPU is a runtime system that offers support for heterogeneous
 multicore machines. While many efforts are devoted to design efficient
 computation kernels for those architectures (e.g. to implement BLAS
 kernels on GPUs or on Cell's SPUs), StarPU not only takes care of
 offloading such kernels (and implementing data coherency across
 the machine), but it also makes sure the kernels are executed as
 efficiently as possible.
 .
 This package contains StarPU tools.
 This "contrib" version is built against CUDA.

starpu-contrib-tools-dbgsym: debug symbols for starpu-contrib-tools