[clang] [CIR][CUDA][HIP] Skip global init for device variables (PR #231971)

[Arseniy Obolenskiy via cfe-commits]

================
@@ -0,0 +1,33 @@
+// REQUIRES: amdgpu-registered-target
+// RUN: %clang_cc1 -triple spirv64-amd-amdhsa -x hip -fclangir \
+// RUN:            -fcuda-is-device -emit-llvm -I%S/../CodeGenCUDA/Inputs %s -o %t-cir.ll
+// RUN: FileCheck --check-prefixes=LLVM,SPIRV %s --input-file=%t-cir.ll
+
+// RUN: %clang_cc1 -triple spirv64-amd-amdhsa -x hip \
+// RUN:            -fcuda-is-device -emit-llvm -I%S/../CodeGenCUDA/Inputs %s -o %t.ll
+// RUN: FileCheck --check-prefixes=LLVM,SPIRV %s --input-file=%t.ll
+
+// RUN: %clang_cc1 -triple amdgcn-amd-amdhsa -x hip -fclangir \
----------------
aobolensk wrote:

Done

https://github.com/llvm/llvm-project/pull/231971