On 26.03.2018 11:20, David Hildenbrand wrote: > Manually having to use cpu_synchronize_state() is error prone. I don't > think that the performance impact is that huge if we simply synchronize > the state on every kvm_arch_handle_exit() call. This makes the code > easier to maintain. > > We now also call it (although not neded) for > - KVM_EXIT_S390_RESET -> s390_reipl_request() > - KVM_EXIT_DEBUG -> kvm_arch_handle_debug_exit() > - unmanagable/unimplemented intercepts > - ICPT_WAITPSW -> s390_handle_wait() -> cpu gets halted
Just noticed that this one is actually also already called :) > - ICPT_CPU_STOP -> do_stop_interrupt() -> cpu gets halted > - Scenarios where we inject an operation exception > - handle_sigp() on the source CPU And this one, too. > - handle_stsi() -- Thanks, David / dhildenb
