[clang] [CIR][CUDA] Add host-side surface registration support (PR #225146)

[via cfe-commits]

================
@@ -1,9 +1,21 @@
 // REQUIRES: x86-registered-target
 // REQUIRES: nvptx-registered-target
+
 // RUN: %clang_cc1 -fclangir -std=c++11 -fcuda-is-device -triple nvptx64-nvidia-cuda -emit-cir -o - %s | FileCheck --check-prefix=CIR-DEVICE %s
 // RUN: %clang_cc1 -fclangir -std=c++11 -fcuda-is-device -triple nvptx64-nvidia-cuda -emit-llvm -o - %s | FileCheck --check-prefix=LLVM-DEVICE %s
 // RUN: %clang_cc1 -std=c++11 -fcuda-is-device -triple nvptx64-nvidia-cuda -emit-llvm -o - %s | FileCheck --check-prefix=OGCG-DEVICE %s
 
+// RUN: echo -n "GPU binary would be here." > %t
+// RUN: %clang_cc1 -fclangir -std=c++11 -triple x86_64-unknown-linux-gnu \
+// RUN:   -target-sdk-version=12.3 -fcuda-include-gpubinary %t \
+// RUN:   -emit-cir -o - %s | FileCheck --check-prefix=CIR-HOST %s
+// RUN: %clang_cc1 -fclangir -std=c++11 -triple x86_64-unknown-linux-gnu \
+// RUN:   -target-sdk-version=12.3 -fcuda-include-gpubinary %t \
+// RUN:   -emit-llvm -o - %s | FileCheck --check-prefix=LLVM-HOST %s
+// RUN: %clang_cc1 -std=c++11 -triple x86_64-unknown-linux-gnu \
+// RUN:   -target-sdk-version=12.3 -fcuda-include-gpubinary %t \
+// RUN:   -emit-llvm -o - %s | FileCheck --check-prefix=OGCG-HOST %s
----------------
Siya-05 wrote:

Updated the host checks to reuse LLVM-HOST. 

https://github.com/llvm/llvm-project/pull/225146