SWDEV-453422: introduce warp sync tests

The following builtins are tested:

  - __all_sync, __any_sync, __ballot_sync and __activemask
  - __match_any_sync and __match_all_sync
  - __shfl_sync, __shfl_up_sync, __shfl_down_sync, and __shfl_xor_sync

The tests for shfl (all variants) were manually validated on a CUDA setup.

NOTE:

  - Unit_hipVoteSync_All temporarily disabled on Windows (SWDEV-452308).
  - All new tests temporarily disabled on CUDA (SWDEV-453145).

Change-Id: I84b205a88aa24219d199c760793e2f19f0cf8f13

catch/include/hip_test_common.hh

@@ -170,6 +170,21 @@ static void initHipCtx(hipCtx_t* pcontext) {
 #define HIP_ARRAY hipArray_t
 #endif
 
+static inline int getWarpSize() {
+#if HT_NVIDIA
+  return 32;
+#elif HT_AMD
+  int device = -1;
+  int warpSize = -1;
+  HIP_CHECK(hipGetDevice(&device));
+  HIP_CHECK(hipDeviceGetAttribute(&warpSize, hipDeviceAttributeWarpSize, device));
+  return warpSize;
+#else
+  std::cout<<"Have to be either Nvidia or AMD platform, asserting"<<std::endl;
+  assert(false);
+#endif
+}
+
 static inline bool IsGfx11() {
 #if HT_NVIDIA
   return false;