[libcxx-commits] [libcxx] [libcxx] Adjust inline assembly constraints for the AMDGPU target (PR #101747)

[Joseph Huber via libcxx-commits]

================
@@ -291,17 +291,27 @@ struct is_same<T, T> { enum {value = 1}; };
 // when optimizations are enabled.
 template <class Tp>
 inline Tp const& DoNotOptimize(Tp const& value) {
-    asm volatile("" : : "r,m"(value) : "memory");
-    return value;
+  // The `m` constraint is invalid in the AMDGPU backend.
+#  if defined(__AMDGPU__) || defined(__NVPTX__)
+  asm volatile("" : : "r"(value) : "memory");
+#  else
+  asm volatile("" : : "r,m"(value) : "memory");
+#  endif
+  return value;
 }
 
 template <class Tp>
 inline Tp& DoNotOptimize(Tp& value) {
-#if defined(__clang__)
+  // The `m` and `r` output constraint is invalid in the AMDGPU backend as well
+  // as i8 / i1 arguments, so we just capture the pointer instead.
+#  if defined(__AMDGPU__)
----------------
jhuber6 wrote:

This is how `libcxx` does it apparently.

https://github.com/llvm/llvm-project/pull/101747