Chapel 1.12.0 released

Cray Inc. and the Chapel open-source community are pleased to announce the release of version 1.12.0 of the Chapel Compiler.

The highlights of this release include:

We significantly expanded the coverage and scope of Chapel's online documentation, available at: http://chapel.cray.com/docs/1.12/

The standard library now includes new 'Barrier', 'LAPACK', and 'Spawn' modules for performing inter-task barrier synchronization, linear algebra, and multi-process spawning, respectively.

We've added a new 'chplvis' tool that supports visualizing the inter-locale communication and intra-locale concurrency in a Chapel program.

A new 'vectorizeOnly()' iterator supports parallel loops that should be implemented by a single task yet still be considered as candidates for vectorization.

New performance optimizations have been implemented to improve locality in our generated code, both within a locale, and within a locale's NUMA domains. As a result of this optimization, the stream-ep benchmark in the examples/benchmarks/hpcc directory can now be written without a 'local' block, matching or beating performance from previous releases, and typically competitive with hand-coded MPI+OpenMP versions.

The language has been expanded to support methods and iterators on types and to support the declaration of private module-level variables, functions, and sub-modules. In addition, the specification of a domain in a formal array argument is now interpreted as a constraint on the actual argument rather than an indication that it should be reindexed.

For more information on the above, or for a more complete list of changes, please refer to $CHPL_HOME/CHANGES.

This release of Chapel contains stable support for the base language, and for task and regular data parallelism using one or more locales. Data parallel operations on irregular domains and arrays are currently supported in a single-locale implementation. While performance has received a great deal of attention in Chapel's design and is generally improving from release-to-release, this release is lacking a number of crucial performance optimizations. See $CHPL_HOME/PERFORMANCE for additional notes on Chapel performance.