[flang-commits] [flang] [flang][cuda] Pass assumed-shape arrays of bind(c) attributes(global) kernels by base address (PR #199313)

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

================
@@ -0,0 +1,58 @@
+<!--===- docs/CUDA.md
+
+   Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+   See https://llvm.org/LICENSE.txt for license information.
+   SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+
+-->
+
+# CUDA Fortran
+
+```{contents}
+---
+local:
+---
+```
+
+Implementation notes for Flang's CUDA Fortran support.
+
+## Lowering decisions
+
+List of CUDA Fortran lowering decisions in Flang for cases where CUDA
+Fortran interoperability requires behavior that is not specified by the
+[CUDA Fortran Programming
+Guide](https://docs.nvidia.com/hpc-sdk/compilers/cuda-fortran-prog-guide/index.html)
+or by standard `BIND(C)` lowering alone.
----------------
clementval wrote:

```suggestion
```
I would remove BIND here since the list will likely have other lowering explanations that are not related. 

https://github.com/llvm/llvm-project/pull/199313