ExecutePM4 queue full check fix

Change-Id: Id56ece6d3f5eab1ef3a2758922022f0996c1efe4


[ROCm/ROCR-Runtime commit: 8a8d7ad814]

projects/rocr-runtime/runtime/hsa-runtime/core/runtime/amd_aql_queue.cpp

@@ -799,7 +799,7 @@ void AqlQueue::ExecutePM4(uint32_t* cmd_data, size_t cmd_size_b) {
   // Obtain a queue slot for a single AQL packet.
   uint64_t write_idx = queue->AddWriteIndexAcqRel(1);
 
-  while ((write_idx - queue->LoadReadIndexRelaxed()) > queue->amd_queue_.hsa_queue.size) {
+  while ((write_idx - queue->LoadReadIndexRelaxed()) >= queue->amd_queue_.hsa_queue.size) {
     os::YieldThread();
   }