Users/taosang/SWDEV-510994 - Refractor atomics header and tests (#902)

* SWDEV-550626 - Refactor atomics header and tests

1. Introduce __HIP_ATOMIC_BACKWARD_COMPAT.
By default we define __HIP_ATOMIC_BACKWARD_COMPAT=1 to
let hip atomic functions maintain old assumptions. if
users want to adopt the new behavior, that is , by default
assume no-fine-grained no-remote-memory, then they can
define __HIP_ATOMIC_BACKWARD_COMPAT=0 and get the new
behaviour.

2. Use  __HIP_ATOMIC_BACKWARD_COMPAT_MEMORY to replace
original __HIP_FINE_GRAINED_MEMORY  in atomic header.
And apply __HIP_FINE_GRAINED_MEMORY onto all 
atomicXXX_system() functions to prevent failure on memory
allocated by hipHostMalloc().

3. Replace HIP_TEST_FINE_GRAINED_MEMORY with
HIP_TEST_ATOMIC_BACKWARD_COMPAT_MEMORY in hip-tests.

4. Fix negative test errors.
    Fix managed memory test error on memory order.
    some other minor changes.
    As a result  all originally disabled tests are enabled.

5. Add more atomics tests in some cases.

6. Reduce test time in each case.
     Reduce iteration number to 1 for tests that cost too much time.

8. Put common codes into hip_test_common.hh

projects/hip-tests/catch/include/hip_test_common.hh

@@ -40,6 +40,16 @@ THE SOFTWARE.
 #include <sys/resource.h>
 #endif
 
+#if !defined(__HIP_ATOMIC_BACKWARD_COMPAT)
+#define __HIP_ATOMIC_BACKWARD_COMPAT 1
+#endif
+
+#if defined(__has_extension) && __has_extension(clang_atomic_attributes) && __HIP_ATOMIC_BACKWARD_COMPAT
+#define HIP_TEST_ATOMIC_BACKWARD_COMPAT_MEMORY [[clang::atomic(fine_grained_memory, remote_memory)]]
+#else
+#define HIP_TEST_ATOMIC_BACKWARD_COMPAT_MEMORY
+#endif
+
 #ifdef TEST_CLOCK_CYCLE
 #define clock_function() clock64()
 #else
@@ -388,6 +398,30 @@ inline bool isP2PSupported(int& d1, int& d2) {
   return supported;
 }
 
+inline bool checkConcurrentKernels(int num_devices) {
+  for (auto i = 0; i < num_devices; ++i) {
+    HIP_CHECK(hipSetDevice(i));
+    int concurrent_kernels = 0;
+    HIP_CHECK(hipDeviceGetAttribute(&concurrent_kernels, hipDeviceAttributeConcurrentKernels, i));
+    if (!concurrent_kernels) {
+      return false;
+    }
+  }
+  if (num_devices > 1) {
+    HIP_CHECK(hipSetDevice(0));
+  }
+  return true;
+}
+
+inline bool isXnackOn() {
+  hipDeviceProp_t prop;
+  int device = 0;
+  HIP_CHECK(hipGetDevice(&device));
+  HIP_CHECK(hipGetDeviceProperties(&prop, device));
+  std::string gfxName(prop.gcnArchName);
+  return gfxName.find("xnack+") != std::string::npos;
+}
+
 inline bool areWarpMatchFunctionsSupported() {
   int matchFunctionsSupported = 1;
 #if HT_NVIDIA