[PATCH] D16870: [CUDA] Tweak attribute-based overload resolution to match nvcc behavior.

[Jingyue Wu via cfe-commits]

jingyue added inline comments.

================
Comment at: test/SemaCUDA/function-overload.cu:138
@@ -149,3 +137,3 @@
   fp_t cdhp = cdh;
-  gp_t gp = g; // expected-error {{reference to __global__ function 'g' in __global__ function}}
-               // expected-note at 67 {{'g' declared here}}
+  gp_t gp = g; 
+  // expected-error at -1 {{reference to __global__ function 'g' in __global__ function}}
----------------
This line seems to have a trailing space. 


http://reviews.llvm.org/D16870