Add gpu().hasPendingDispatch() in the SDMA path

SWDEV-234947

Change-Id: I8aa501f8755d136708b0d12ee3c30229c238660d


[ROCm/clr commit: 2a02026696]

projects/clr/rocclr/device/rocm/rocblit.cpp

@@ -660,6 +660,8 @@ bool DmaBlitManager::hsaCopy(const Memory& srcMemory, const Memory& dstMemory,
     if (val != (kInitVal - 1)) {
       LogError("Async copy failed");
       status = HSA_STATUS_ERROR;
+    } else {
+      gpu().hasPendingDispatch();
     }
   } else {
     LogPrintfError("Hsa copy from host to device failed with code %d", status);
@@ -747,6 +749,8 @@ bool DmaBlitManager::hsaCopyStaged(const_address hostSrc, address hostDst, size_
     offset += size;
   }
 
+  gpu().hasPendingDispatch();
+
   return true;
 }