================
@@ -334,6 +334,9 @@ class TargetCodeGenInfo {
/// Get LLVM calling convention for device kernels.
virtual unsigned getDeviceKernelCallingConv() const;
+ /// \return true if the target's launch model delivers complete work-groups.
----------------
arsenm wrote:
It's only a target property in the sense that we would want to hack this in for
non-GPU languages on the GPU targets, which is still just a language check with
a triple predicate
https://github.com/llvm/llvm-project/pull/221279
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
