[CUDA] Regression test to make sure C++ include path are forwarded to host and device frontends.

[sfantao]

Summary

Add test to detect the C++ include paths are passed to both CUDA host and device frontends.

[sfantao]

Friendly ping.

Thanks!

[tra]

cuda-detect.cu may be a better place for this test.

test/Driver/cuda-simple.cu

27

Linker match pattern will cause problems on windows. We normally use ld{{.*}}" .

Linker options are irrelevant here, IMO, as this line is only used as a marker for the end of the list of commands produced by compiler.

[sfantao]

• Move CUDA C++ include path tests to cuda-detect.cu. Address other Art's comments.

[tra]

test/Driver/cuda-detect.cu

75–77 ↗

(On Diff #67335)

These two lines seem to be copy-pasted from cuda-simple.cu by mistake.

[sfantao]

• Remove redundant test - copied by mistake.

test/Driver/cuda-detect.cu

75–77 ↗

(On Diff #67335)

Woops... Sorry about that. Fixed in the last diff.

[tra]

LGTM.

test/Driver/cuda-simple.cu

+//
+// REQUIRES: clang-driver
+// REQUIRES: x86-registered-target
+// REQUIRES: nvptx-registered-target
+//
 // Verify that we can parse a simple CUDA file with or without -save-temps
 // http://llvm.org/PR22936
 // RUN: %clang -nocudainc -Werror -fsyntax-only -c %s
 //
 // Verify that we pass -x cuda-cpp-output to compiler after 
 // preprocessing a CUDA file
 // RUN: %clang  -Werror -### -save-temps -c %s 2>&1 | FileCheck %s
 // CHECK: "-cc1"
 // CHECK: "-E"
 // CHECK: "-x" "cuda"
 // CHECK-NEXT: "-cc1"
 // CHECK: "-x" "cuda-cpp-output"
 //
 // Verify that compiler accepts CUDA syntax with "-x cuda-cpp-output".
 // RUN: %clang -Werror -fsyntax-only -x cuda-cpp-output -c %s
+//
+// Verify that C++ include paths are passed for both host and device frontend.
+// RUN: %clang -### -target x86_64-linux-gnu %s --sysroot=%S/Inputs/ubuntu_14.04_multiarch_tree2 2>&1 \
+// RUN: | FileCheck %s --check-prefix CHECK-CXXINCLUDE
+// CHECK-CXXINCLUDE: clang{{.*}} "-cc1" "-triple" "nvptx64-nvidia-cuda" {{.*}}"-internal-isystem" "{{.+}}/include/c++/4.8"
+// CHECK-CXXINCLUDE: clang{{.*}} "-cc1" "-triple" "x86_64--linux-gnu" {{.*}}"-internal-isystem" "{{.+}}/include/c++/4.8"
+// CHECK-CXXINCLUDE: ld" {{.*}}"-m" "elf_x86_64"
 
 int cudaConfigureCall(int, int);
 __attribute__((global)) void kernel() {}
 
 void func() {
      kernel<<<1,1>>>();
 }