Branch: refs/heads/master
Revision: 78c9b5c
Author: ben-albrecht
Log Message:
Merge pull request #15976 from rahulghangas/LoopTiledMM
Loop tiled matrix multiplication
Parallel matrix multiplication based on loop tiling.
Adapted from Chapel PRK DGEMM
- [x] Correctness Testing
- [x] Performance graph comparing reindexing overhead
Performance graphs against Naive approach(not needed anymore as a local dot perf test already exists)
[Contributed by @rahulghangas]
[Reviewed by @ben-albrecht && @e-kayrakli]]
Modified Files:
M modules/packages/LinearAlgebra.chpl
Compare: https://github.com/chapel-lang/chapel/compare/ccd9602361f4...78c9b5c29b06