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

[Johannes Doerfert via Phabricator via Openmp-commits]

jdoerfert added a comment.

can we have a test for this?



================
Comment at: openmp/runtime/src/kmp_dispatch.cpp:96
+  return (float)(static_cast<int>(num * 100 + 0.5)) / 100;
+}
+
----------------
floats are not double precision


================
Comment at: openmp/runtime/src/kmp_dispatch.cpp:154
+  p_extra = (type == KMP_HW_CORE_TYPE_CORE) ? p_extra : 0;
+#endif
+}
----------------
Probably easier to read if the non weighted case was handled first.


================
Comment at: openmp/runtime/src/kmp_dispatch.cpp:558
+      }
+#endif
       pr->u.p.count = init;
----------------
Is it really helping that we use a macro to elide the code? It makes the impl more complex for sure.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D152955/new/

https://reviews.llvm.org/D152955