> As you realized you need to restore a checkpoint to the CPU that > created it and if a different CPU model is desired then switch CPUs > after this restoration. Perhaps all the CPUs should have the same > checkpoint state, however this isn't the way things are implemented > today and limits the type of checkpointing that can be done. Thoughts > anyone?
I think this should be largely do-able if we were to sort of merge the takeOverFrom code with the checkpoint code. It actually shouldn't be that bad since it would probably just mean moving most of the checkpoint operations over to the ProxyThreadContext object and then using the ProxyThreadContext to serialize a CPU's architectural state. That said, switchover exists and could be used in a two step (unserialize and then switch) process instead of directly unserializing to a different object type. Nate _______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev
