On Wed, Apr 13, 2022, Sean Christopherson wrote: > On Wed, Apr 13, 2022, Sean Christopherson wrote: > > On Wed, Apr 13, 2022, Sean Christopherson wrote: > > > On Wed, Apr 13, 2022, Sean Christopherson wrote: > > > > On Tue, Apr 12, 2022, Varad Gautam wrote: > > > > > @@ -142,3 +143,26 @@ void smp_reset_apic(void) > > > > > > > > > > atomic_inc(&active_cpus); > > > > > } > > > > > + > > > > > +void ap_init(void) > > > > Sorry for chaining these, I keep understanding more things as I read through the > > end of the series. Hopefully this is the last one. > > > > Can this be named setup_efi_rm_trampoline()? Or whatever best matches the name > > we decide on. I keep thinking APs bounce through this to do their initialization, > > but it's the BSP doing setup to prep waking the APs. > > Well that didn't take long. Just realized this isn't unique to EFI, and it also > does the waking. Maybe wake_aps()? That makes smp_init() even more confusing > (I keep thinking that it wakes APs...), but IMO smp_init() is the one that needs > a new name. *sigh* Ignore the last complaint about smp_init(), it does do SMP initialization by sending IPIs. We should really change that, but that's a future problem.
