Index: External/CUDA/CMakeLists.txt
===================================================================
--- External/CUDA/CMakeLists.txt
+++ External/CUDA/CMakeLists.txt
@@ -46,6 +46,7 @@
 macro(create_local_cuda_tests VariantSuffix)
   create_one_local_test(axpy axpy.cu)
   create_one_local_test(empty empty.cu)
+  create_one_local_test(printf printf.cu)
   create_one_local_test(future future.cu)
 endmacro()
 
Index: External/CUDA/printf.cu
===================================================================
--- /dev/null
+++ External/CUDA/printf.cu
@@ -0,0 +1,13 @@
+#include <iostream>
+
+__global__ void kernel(char a, short b, int c, float d, double e) {
+  const char* f = "abcdefg";
+  printf("kernel_printf %c %d %d %f %f %s\n", a, b, c, d, e, f);
+}
+
+int main(int argc, char* argv[]) {
+  kernel<<<2, 8>>>('a', 2, 3, 4, 5);
+  cudaDeviceSynchronize();
+  cudaDeviceReset();
+  return 0;
+}
Index: External/CUDA/printf.reference_output
===================================================================
--- /dev/null
+++ External/CUDA/printf.reference_output
@@ -0,0 +1,17 @@
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+kernel_printf a 2 3 4.000000 5.000000 abcdefg
+exit 0