{"id":124,"date":"2015-02-06T22:38:54","date_gmt":"2015-02-06T20:38:54","guid":{"rendered":"http:\/\/paragroup.cs.uoi.gr\/wptest\/wordpress\/?page_id=124"},"modified":"2025-12-15T16:26:49","modified_gmt":"2025-12-15T14:26:49","slug":"download","status":"publish","type":"page","link":"https:\/\/paragroup.cse.uoi.gr\/wpsite\/software\/ompi\/download\/","title":{"rendered":"OMPi @ ppg downloads"},"content":{"rendered":"

\"\" The source code of OMPi is available here; the compiler has been tested on a variety of Linux, Solaris, Irix, MacOS, and Windows (WSL2\/Cygwin) machines. Actually the only system requirements are compatibility with POSIX and a native C compiler (e.g. gcc<\/em>).<\/p>\n

The most recent stable version is 3.5.2<\/strong>.<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Version<\/td>\nLink<\/td>\nNotes<\/td>\n<\/tr>\n
3.5.2<\/td>\n\u00a0Download<\/a><\/td>\n\u00bb\u00a0new numbering for OpenMP devices, compatible with >= v5.2
\n \u00bb\u00a0redesigned kernel bundling\u00a0(given as option to ompicc; enabled by default)
\n \u00bb many improvements and fixes
\n \u00a0
\n <\/em>\u00bb implementation-defined OpenMP behaviors for OMPi (HTML<\/a>,\u00a0PDF<\/a>).<\/td>\n<\/tr>\n
3.5.0<\/td>\n\u00a0Download<\/a><\/td>\n\u00bb\u00a0offloading to OpenCL GPUs (see how to use<\/a>)
\n \u00bb\u00a0offloading to Vulkan GPUs (see how to use<\/a>)
\n \u00bb\u00a0full OpenMP support (CPU+GPU) for Raspberry Pi 4\/5
\n \u00bb restructured compiler transformations and code generation
\n \u00bb switch to meson build system
\n \u00bb many improvements and fixes
\n \u00a0
\n <\/em>\u00bb implementation-defined OpenMP behaviors for OMPi (HTML<\/a>,\u00a0PDF<\/a>).<\/td>\n<\/tr>\n
3.0.0<\/td>\nDownload<\/a><\/td>\n\u00bb\u00a0remote offloading to cluster CPUs and devices
\n \u00bb task reductions
\n \u00bb many improvements and fixes<\/em>
\n \u00a0
\n \u00bb implementation-defined OpenMP behaviors for OMPi (HTML<\/a>, PDF<\/a>).<\/td>\n<\/tr>\n
2.7.0<\/td>\nDownload<\/a><\/td>\n\u00bb\u00a0 new module for\u00a0 offloading to CUDA GPUs with compute capability \u2265 3.5
\n \u00bb\u00a0full OpenMP support (CPU+GPU) for Jetson Nano 2\/4GB
\n \u00bb\u00a0support for extra worksharing loop schedules
\n \u00bb restructured compiler transformations and code generation
\n \u00bb many improvements and fixes<\/em>
\n \u00a0
\n \u00bb implementation-defined OpenMP behaviors for OMPi (HTML<\/a>, PDF<\/a>).<\/td>\n<\/tr>\n
2.5.0<\/td>\nDownload<\/a><\/td>\n\u00bb\u00a0 task dependencies
\n \u00bb\u00a0doacross loops
\n \u00bb\u00a0support OpenMP 4.5 target-related device directives and runtime
\n \u00a0\u00a0 functions; partial support for OpenMP 5.0
\n \u00bb new `mpinode’ module treats cluster nodes as OpenMP devices
\n \u00bb adaptive, compiler-assisted runtime flavors for devices
\n \u00bb affinity control and places conforming to OpenMP 5.1
\n \u00bb improvements and fixes
\n \u00a0
\n \u00bb implementation-defined OpenMP behaviors for OMPi (HTML<\/a>, PDF<\/a>).<\/td>\n<\/tr>\n
2.0.0<\/td>\nDownload<\/a><\/td>\n\u00bb initial support for OpenMP 4.0 (device, cancellation, taskgroup constructs)
\n \u00bb includes seamless support for the Parallella-16 board
\n \u00bb major reorganization of the compiler and runtime trees
\n \u00bb improvements everywhere
\n \u00bb bug fixes
\n \u00a0
\n \u00bb implementation-defined OpenMP behaviors for OMPi (HTML<\/a>, PDF<\/a>).<\/td>\n<\/tr>\n
1.2.3<\/td>\nDownload<\/a><\/td>\n\u00bb OpenMP 3.1
\n \u00bb addition of a process library (with Sys-V IPC)
\n \u00bb tasking improvements
\n \u00bb bug fixes<\/td>\n<\/tr>\n
1.2.2<\/td>\nDownload<\/a><\/td>\n\u00bb support for most parts of OpenMP 3.1
\n \u00bb tasking improvements
\n \u00bb experimental tasking support in psthr (psthreads V1.0.4 required)
\n \u00bb bug fixes<\/td>\n<\/tr>\n
1.2.0<\/td>\nDownload<\/a><\/td>\n\u00bb full OpenMP 3.0 support
\n \u00bb runtime system enhancements
\n \u00bb bug fixes
\n \u00bb dropped a few thread libraries (availabe on request)<\/td>\n<\/tr>\n
1.1.0<\/td>\nDownload<\/a><\/td>\n\u00bb OpenMP 3.0 support (most parts)
\n \u00bb restructured runtime system<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

Older, special version of the OMPi compiler for the Parallella board<\/a> have been superseded by versions \u2265 2.0.0 of OMPi. You may contact us<\/a> for further inquiries.<\/p>\n","protected":false},"excerpt":{"rendered":"

The source code of OMPi is available here; the compiler has been tested on a variety of Linux, Solaris, Irix, MacOS, and Windows (WSL2\/Cygwin) machines. Actually the only system requirements are compatibility with POSIX and a native C compiler (e.g. gcc). The most recent stable version is 3.5.2. Version Link Notes 3.5.2 \u00a0 \u00bb\u00a0new numbering […]<\/p>\n","protected":false},"author":3,"featured_media":0,"parent":118,"menu_order":0,"comment_status":"closed","ping_status":"open","template":"","meta":{"footnotes":""},"class_list":["post-124","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/paragroup.cse.uoi.gr\/wpsite\/wp-json\/wp\/v2\/pages\/124","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/paragroup.cse.uoi.gr\/wpsite\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/paragroup.cse.uoi.gr\/wpsite\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/paragroup.cse.uoi.gr\/wpsite\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/paragroup.cse.uoi.gr\/wpsite\/wp-json\/wp\/v2\/comments?post=124"}],"version-history":[{"count":57,"href":"https:\/\/paragroup.cse.uoi.gr\/wpsite\/wp-json\/wp\/v2\/pages\/124\/revisions"}],"predecessor-version":[{"id":1437,"href":"https:\/\/paragroup.cse.uoi.gr\/wpsite\/wp-json\/wp\/v2\/pages\/124\/revisions\/1437"}],"up":[{"embeddable":true,"href":"https:\/\/paragroup.cse.uoi.gr\/wpsite\/wp-json\/wp\/v2\/pages\/118"}],"wp:attachment":[{"href":"https:\/\/paragroup.cse.uoi.gr\/wpsite\/wp-json\/wp\/v2\/media?parent=124"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}