tra added inline comments.
================
Comment at: lib/CodeGen/CGCUDACall.cpp:1
@@ +1,2 @@
+//===----- CGCUDACall.cpp - Codegen for special CUDA calls
----------------===//
+//
----------------
We'll need to handle other builtins soon. Perhaps we can keep them in this file
and rename it CGCUDABuiltins.cpp
================
Comment at: test/CodeGenCUDA/Inputs/cuda.h:11-12
@@ -10,2 +10,4 @@
#define __launch_bounds__(...) __attribute__((launch_bounds(__VA_ARGS__)))
+#define __device_builtin__ __attribute__((device_builtin))
+#define __cudart_builtin__ __attribute__((cudart_builtin))
----------------
These are not needed to test anything relevant to this patch, IMO.
================
Comment at: test/CodeGenCUDA/printf.cu:12
@@ +11,3 @@
+// Check a simple call to printf end-to-end.
+__device__ int CheckSimple() {
+ // CHECK: [[FMT:%[0-9]+]] = load{{.*}}%fmt
----------------
You may want to add a test case for printf("constant string w/o format args").
It would check no-arguments case and would also verify that we don't get
optimized into puts().
http://reviews.llvm.org/D16372
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
