[flang-commits] [flang] [flang][cuda] Rewrite predefined variable in acc.parallel with inlined cuf function (PR #206205)

[Valentin Clement バレンタイン クレメン via flang-commits]

clementval wrote:

> Does the same issue apply to acc.serial or acc.kernels regions? They are also OpenACC compute regions, so I wonder whether this should handle those too or whether the inlining case is only expected under acc.parallel.

Yeah I was thinking about that and didn't see any use case but to be future proof I just switched to ComputeRegionOpInterface so we cover those. 

https://github.com/llvm/llvm-project/pull/206205