On 17/06/2016 16:08, Steven Rostedt wrote: >> > +/* >> > + * Tracepoints for VM enters >> > + */ >> > +TRACE_EVENT(kvm_enter, >> > + TP_PROTO(struct kvm_vcpu *vcpu), >> > + TP_ARGS(vcpu), >> > + TP_STRUCT__entry( >> > + __field(unsigned long, pc) >> > + ), >> > + >> > + TP_fast_assign( >> > + __entry->pc = vcpu->arch.pc; >> > + ), >> > + >> > + TP_printk("PC: 0x%08lx", >> > + __entry->pc) >> > +); >> > + >> > +TRACE_EVENT(kvm_reenter, >> > + TP_PROTO(struct kvm_vcpu *vcpu), >> > + TP_ARGS(vcpu), >> > + TP_STRUCT__entry( >> > + __field(unsigned long, pc) >> > + ), >> > + >> > + TP_fast_assign( >> > + __entry->pc = vcpu->arch.pc; >> > + ), >> > + >> > + TP_printk("PC: 0x%08lx", >> > + __entry->pc) >> > +); >> > + >> > +TRACE_EVENT(kvm_out, >> > + TP_PROTO(struct kvm_vcpu *vcpu), >> > + TP_ARGS(vcpu), >> > + TP_STRUCT__entry( >> > + __field(unsigned long, pc) >> > + ), >> > + >> > + TP_fast_assign( >> > + __entry->pc = vcpu->arch.pc; >> > + ), >> > + >> > + TP_printk("PC: 0x%08lx", >> > + __entry->pc) >> > +); > > Please combine the above TRACE_EVENT()s to use a single > DECLARE_EVENT_CLASS() and three DEFINE_EVENT()s. James, I've committed the patch already, so please send a follow up. Thanks, Paolo