[Clang] Change 'gpuintrin.h' to use target specific address spceas

Summary:
A recent patch allowed the target specific address spcaces to be handled
correctly. The one downside here is that we no long get semantic errors
for initializers, but that will error in the backend anyway.

Author: jhuber6

clang/lib/Headers/amdgpuintrin.h

@@ -24,11 +24,11 @@ _Pragma("omp begin declare target device_type(nohost)");
 _Pragma("omp begin declare variant match(device = {arch(amdgcn)})");
 
 // Type aliases to the address spaces used by the AMDGPU backend.
-#define __gpu_private __attribute__((opencl_private))
-#define __gpu_constant __attribute__((opencl_constant))
-#define __gpu_local __attribute__((opencl_local))
-#define __gpu_global __attribute__((opencl_global))
-#define __gpu_generic __attribute__((opencl_generic))
+#define __gpu_private __attribute__((address_space(5)))
+#define __gpu_constant __attribute__((address_space(4)))
+#define __gpu_local __attribute__((address_space(3)))
+#define __gpu_global __attribute__((address_space(1)))
+#define __gpu_generic __attribute__((address_space(0)))
 
 // Attribute to declare a function as a kernel.
 #define __gpu_kernel __attribute__((amdgpu_kernel, visibility("protected")))

clang/lib/Headers/nvptxintrin.h

@@ -24,11 +24,11 @@ _Pragma("omp begin declare target device_type(nohost)");
 _Pragma("omp begin declare variant match(device = {arch(nvptx64)})");
 
 // Type aliases to the address spaces used by the NVPTX backend.
-#define __gpu_private __attribute__((opencl_private))
-#define __gpu_constant __attribute__((opencl_constant))
-#define __gpu_local __attribute__((opencl_local))
-#define __gpu_global __attribute__((opencl_global))
-#define __gpu_generic __attribute__((opencl_generic))
+#define __gpu_private __attribute__((address_space(5)))
+#define __gpu_constant __attribute__((address_space(4)))
+#define __gpu_local __attribute__((address_space(3)))
+#define __gpu_global __attribute__((address_space(1)))
+#define __gpu_generic __attribute__((address_space(0)))
 
 // Attribute to declare a function as a kernel.
 #define __gpu_kernel __attribute__((nvptx_kernel, visibility("protected")))