[PATCH] D82324: [OPENMP]Dynamic globalization for parallel target regions.

Alexey Bataev via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Wed Jun 24 10:15:35 PDT 2020


ABataev added a comment.

In D82324#2111953 <https://reviews.llvm.org/D82324#2111953>, @jdoerfert wrote:

> I do not understand why we need the flag. As far as I can tell, it has to be on to support the standard described behavior, right? Why should we (allow to) turn it off?


It might use "slow" allocation functions, in general, since it may use malloc on the device side. Being disabled, it uses statically preallocated memory, which might be faster, if parallel target regions are not required.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D82324/new/

https://reviews.llvm.org/D82324





More information about the cfe-commits mailing list