SWDEV-232807

[ROCm][TCT][HIP] cooperative stream test case is failing.

Make sure lockXfer() in the blit manager returns a valid value.
Port the latest PAL backend logic into the ROCr backend.
This change doesn't fix the issue, reported in the ticket.

Change-Id: I54101a824f49a2dcfbbf5414cb5b3af41745306d


[ROCm/clr commit: 89133a7301]

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

@@ -2300,15 +2300,15 @@ bool VirtualGPU::submitKernelInternal(const amd::NDRangeContainer& sizes, const
  */
 void VirtualGPU::submitKernel(amd::NDRangeKernelCommand& vcmd) {
   if (vcmd.cooperativeGroups() || vcmd.cooperativeMultiDeviceGroups()) {
+    // Wait for the execution on the current queue, since the coop groups will use the device queue
+    releaseGpuMemoryFence();
+
     // Get device queue for exclusive GPU access
     VirtualGPU* queue = dev().xferQueue();
 
     // Lock the queue, using the blit manager lock
     amd::ScopedLock lock(queue->blitMgr().lockXfer());
 
-    // Wait for the execution on the current queue, since the coop groups will use the device queue
-    releaseGpuMemoryFence();
-
     queue->profilingBegin(vcmd);
 
     if (vcmd.cooperativeGroups()) {