Added _starpu_mpi_redux_wrapup_data detection so that automatic MPI reduction can be used for partial pivoting (see also RUNTIME_ipiv_reducek).

PRUVOST Florent authored 1 year ago and Mathieu Faverge committed 1 year ago

Change name of internal libraries cudablas and hipblas to gpucublas and gpuhipblas to avoid confusion with external libraries that can have the same name

COJEAN Terry authored 8 years ago and Mathieu Faverge committed 2 years ago

Integrate the use of parallel tasks for chameleon within POTRF algorithm. Note that this requires StarPU
New cluster interface based on the hwloc vocabulary : CHAMELEON_CLUSTER_LEVEL=hardware-level:number-of-clusters and CHAMELEON_CLUSTER_SHOW to print the clusters

CMake modifications to avoid cmake projects dependents to rediscover all our dependencies at configure time

Matthieu KUHN authored 2 years ago and Mathieu Faverge committed 2 years ago

Limit automatically the number of submitted tasks in LU no pivoting as a function of the lookahead number and of the number of blocks.

Upgrade the version number to make the distinction between the release files, and the files from the master

THIBAULT Samuel authored 4 years ago and Mathieu Faverge committed 4 years ago

This will give more flexibility to StarPU.

THIBAULT Samuel authored 4 years ago and Mathieu Faverge committed 4 years ago

starpu_mpi_interface_datatype_node_register will allow StarPU-MPI to use
NUMA buffers and GPUDirect.

Add CHAM_tile_t interface

PRUVOST Florent authored 6 years ago and Mathieu Faverge committed 6 years ago

    improve user's doc style
    add some explanation about the code structure in the main doxygen
    fix coverage in analysis sonarqube
    add a release job to automatize release publication
    change the cdash server to my.cdash.org which is more recent and reactive than cdash.inria.fr
    fix the version number, dates and authors in headers to prepare the release 0.9.2