[clang] [llvm] [AMDGPU][GFX12.5] Reimplement monitor load as an atomic operation (PR #177343)

[Pierre van Houtryve via cfe-commits]

================
@@ -0,0 +1,51 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 5
+
+// REQUIRES: amdgpu-registered-target
+// RUN: %clang_cc1 -O3 -triple amdgcn-unknown-unknown -target-cpu gfx1250 -fcuda-is-device -emit-llvm -o - %s | FileCheck %s
----------------
Pierre-vh wrote:

The templated function is inlined, without opts this test has allocas + a function call to the template instance, but the template instance isn't visible in the test, so I can't see the arguments were correctly passed to the intrinsics

https://github.com/llvm/llvm-project/pull/177343