================
@@ -824,9 +831,15 @@ SIAtomicAddrSpace
SIMemOpAccess::toSIAtomicAddrSpace(unsigned AS) const {
return SIAtomicAddrSpace::OTHER;
}
+// TODO: Consider moving single-wave workgroup->wavefront scope relaxation to
an
----------------
krzysz00 wrote:
Yeah, there's a world where, instead of doing this IR rewrite, we instead just
don't emit the waits when we know we're in a single-wave no-DMA context. I
figure it's equivalent to
> If [we have the prerequisites], wavefront scope is a valid refinement of
> workgroup scope
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
