feat: add test USM implicit data movement

tests/basic.cpp

@@ -146,5 +146,7 @@ int main() {
 
   sycl::free(data, q);
 
+  std::cout << "Test Passed!!!" << std::endl;
+
   return 0;
 }

tests/fig-2-2_simple-sycl-program.cpp

@@ -34,5 +34,7 @@ int main() {
     }
   }
 
+  std::cout << "Test Passed!!!" << std::endl;
+
   return 0;
 }

tests/fig-2-9_selecting-host-device.cpp

@@ -22,5 +22,8 @@ int main() {
   }
 
   std::cout << "Device is CPU: OK" << std::endl;
+
+  std::cout << "Test Passed!!!" << std::endl;
+
   return 0;
 }

tests/fig-3-4_usm-explicit-data-movement.cpp

@@ -11,23 +11,23 @@ int main() {
   sycl::queue Q;
 
   std::array<int, N> host_array;
-  int *device_array = sycl::malloc_device<int>(N, Q);
+  int* device_array = sycl::malloc_device<int>(N, Q);
 
   for (int i = 0; i < N; i++) {
     host_array[i] = N;
   }
 
   // We will learn how to simplify this example later
-  Q.submit([&](sycl::handler &h) {
+  Q.submit([&](sycl::handler& h) {
      // copy hostArray to deviceArray
      h.memcpy(device_array, &host_array[0], N * sizeof(int));
    }).wait();
 
-  Q.submit([&](sycl::handler &h) {
+  Q.submit([&](sycl::handler& h) {
      h.parallel_for(N, [=](sycl::id<1> i) { device_array[i]++; });
    }).wait();
 
-  Q.submit([&](sycl::handler &h) {
+  Q.submit([&](sycl::handler& h) {
      // copy deviceArray back to hostArray
      h.memcpy(&host_array[0], device_array, N * sizeof(int));
    }).wait();
@@ -42,5 +42,7 @@ int main() {
     }
   }
 
+  std::cout << "Test Passed!!!" << std::endl;
+
   return 0;
 }

tests/fig-3-5_usm-implicit-data-movement.cpp

@@ -0,0 +1,45 @@
+// Figure 3-5. USM implicit data movement
+// from book - Data Parallel C++
+// https://link.springer.com/book/10.1007/978-1-4842-5574-2
+
+#include <sycl/sycl.hpp>
+
+constexpr int N = 42;
+
+int main() {
+  sycl::queue Q;
+
+  int* host_array = malloc_host<int>(N, Q);
+  int* shared_array = malloc_shared<int>(N, Q);
+
+  for (int i = 0; i < N; i++) {
+    // Initialize hostArray on host
+    host_array[i] = i;
+  }
+
+  // We will learn how to simplify this example later
+  Q.submit([&](sycl::handler& h) {
+     h.parallel_for(N, [=](sycl::id<1> i) {
+       // access sharedArray and hostArray on device
+       shared_array[i] = host_array[i] + 1;
+     });
+   }).wait();
+
+  for (int i = 0; i < N; i++) {
+    // Verify that sharedArray[i] equals hostArray[i] + 1
+    if (shared_array[i] != host_array[i] + 1) {
+      std::cout << "Error at index " << i << ": expected " << (host_array[i] + 1) << ", got "
+                << shared_array[i] << std::endl;
+      free(shared_array, Q);
+      free(host_array, Q);
+      return 1;
+    }
+  }
+
+  free(shared_array, Q);
+  free(host_array, Q);
+
+  std::cout << "Test Passed!!!" << std::endl;
+
+  return 0;
+}