High-performance and widely portable implementation of MPI

Edit Package mpich

MPICH is a high performance and widely portable implementation of the Message
Passing Interface (MPI) standard.

The goals of MPICH are:

* to provide an MPI implementation that efficiently supports different
computation and communication platforms including commodity clusters
(desktop systems, shared-memory systems, multicore architectures),
high-speed networks and proprietary high-end computing systems
(Blue Gene, Cray)
* to enable cutting-edge research in MPI through an easy-to-extend modular
framework for other derived implementations

Refresh
Refresh
Source Files
Filename Size Changed
_constraints 0000000109 109 Bytes
_multibuild 0000000292 292 Bytes
autoconf-pull-dynamic-and-not-static-libs-from-pkg-config.patch 0000001056 1.03 KB
autogen-only-deal-with-json-yaksa-if-enabled.patch 0000001917 1.87 KB
macros.hpc-mpich 0000000612 612 Bytes
mpich-4.0.2.tar.gz 0038137945 36.4 MB
mpich.changes 0000022260 21.7 KB
mpich.spec 0000016037 15.7 KB
mpivars.csh 0000000318 318 Bytes
mpivars.sh 0000000526 526 Bytes
Latest Revision
Daniel Mach's avatar Daniel Mach (dmach) committed (revision 2)
- Revert use of internal PMIx and hwloc library (jsc#PED-2448)
  - Using external libraries disables the build of all process managers

- Disable valgrind on ppc64 to fix build dependency issues.
- Add _constraints to fix compilation running out of disk space.

- Update to 4.0.2
  - Fix support for dynamic process functionality with PMI2 clients
  - Fix non-zero appnum bug in PMI2 server in Hydra
  - Fix MPI_Op support for types created with MPI_Type_create_f90_xxx
  - Fix Level Zero properties initialization in MPL.
  - Use standard names for CPU affinity functions with POSIX
    threads.
  - Add elemental to eq/neq operators in Fortran 2008 binding

- Switch to external hwloc, and PMIx
- On 32b systems, use ofi instead of verbs flavor by default
Comments 0
openSUSE Build Service is sponsored by