[Openmp-commits] [PATCH] D152955: [OpenMP] Add skewed iteration distribution in static-steal schedule for hybrid systems

[Jonathan Peyton via Phabricator via Openmp-commits]

jlpeyton created this revision.
jlpeyton added reviewers: tlwilmar, hbae, Nawrin.
Herald added subscribers: sunshaoce, guansong, yaxunl.
Herald added a project: All.
jlpeyton requested review of this revision.
Herald added a reviewer: jdoerfert.
Herald added subscribers: openmp-commits, jplehr, sstefan1.
Herald added a project: OpenMP.

This commit adds skewed distribution of iterations in dynamic
schedule (static steal) for hybrid systems when thread affinity is assigned.
Currently, it distributes the iterations at 60:40 ratio. Consider this loop with
dynamic schedule type, `for (int i = 0; i < 100; ++i)`

In a hybrid system with 20 processes (16 CORE and 4 ATOM core), 88 iterations
will be assigned to performance cores and 12 iterations will be assigned to
efficient cores. Each thread with CORE core will process 5
iterations + extras and with ATOM core will process 3 iterations.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D152955

Files:
  openmp/runtime/src/kmp.h
  openmp/runtime/src/kmp_affinity.cpp
  openmp/runtime/src/kmp_dispatch.cpp
  openmp/runtime/src/kmp_dispatch.h
  openmp/runtime/src/kmp_global.cpp

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D152955.531488.patch
Type: text/x-patch
Size: 22077 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/openmp-commits/attachments/20230614/1d7f8265/attachment-0001.bin>