samples: eliminate cvconfig.h usage

- don't use TBB in samples

samples/gpu/multi.cpp

@@ -7,29 +7,14 @@
 #endif
 
 #include <iostream>
-#include "opencv2/cvconfig.h"
 #include "opencv2/core.hpp"
 #include "opencv2/cudaarithm.hpp"
 
-#ifdef HAVE_TBB
-#  include "tbb/tbb.h"
-#  include "tbb/task.h"
-#  undef min
-#  undef max
-#endif
-
-#if !defined(HAVE_CUDA) || !defined(HAVE_TBB)
+#if !defined(HAVE_CUDA)
 
 int main()
 {
-#if !defined(HAVE_CUDA)
-    std::cout << "CUDA support is required (CMake key 'WITH_CUDA' must be true).\n";
-#endif
-
-#if !defined(HAVE_TBB)
-    std::cout << "TBB support is required (CMake key 'WITH_TBB' must be true).\n";
-#endif
-
+    std::cout << "CUDA support is required (OpenCV CMake parameter 'WITH_CUDA' must be true)." << std::endl;
     return 0;
 }
 
@@ -39,7 +24,14 @@ using namespace std;
 using namespace cv;
 using namespace cv::cuda;
 
-struct Worker { void operator()(int device_id) const; };
+struct Worker : public cv::ParallelLoopBody
+{
+    void operator()(const Range& r) const CV_OVERRIDE
+    {
+        for (int i = r.start; i < r.end; ++i) { this->operator()(i); }
+    }
+    void operator()(int device_id) const;
+};
 
 int main()
 {
@@ -64,8 +56,8 @@ int main()
     }
 
     // Execute calculation in two threads using two GPUs
-    int devices[] = {0, 1};
-    tbb::parallel_do(devices, devices + 2, Worker());
+    cv::Range devices(0, 2);
+    cv::parallel_for_(devices, Worker(), devices.size());
 
     return 0;
 }