================
@@ -824,9 +831,15 @@ SIAtomicAddrSpace
SIMemOpAccess::toSIAtomicAddrSpace(unsigned AS) const {
return SIAtomicAddrSpace::OTHER;
}
+// TODO: Consider moving single-wave workgroup->wavefront scope relaxation to
an
----------------
Pierre-vh wrote:
> Why not do this already?
I was against it, I am really uneasy with the idea of doing it at the IR level
as it loses the initial semantics/intention.
I seem to be in the minority though, so if everyone else is confident this
should really be done at the IR level, I will not block the patch over it.
https://github.com/llvm/llvm-project/pull/224065
_______________________________________________
llvm-branch-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
