Re: [PATCH 5/5] x86/xen: Don't register PV spinlock IPI when it isn't going to be used

[David Woodhouse <dwmw2@xxxxxxxxxxxxx>]

On Mon, 2021-01-04 at 12:06 -0500, boris.ostrovsky@xxxxxxxxxx wrote:
> > @@ -115,7 +115,7 @@ PV_CALLEE_SAVE_REGS_THUNK(xen_vcpu_stolen);
> >   void __init xen_init_spinlocks(void)
> >   {
> >        /*  Don't need to use pvqspinlock code if there is only 1 vCPU. */
> > -     if (num_possible_cpus() == 1 || nopvspin)
> > +     if (num_possible_cpus() == 1 || nopvspin || !xen_have_vector_callback)
> 
> 
> xen_init_spinlock() will not be called without vector callbacks so
> this test is not really necessary.

Right, that's just paranoia to make the conditions consistent and safe.

Attachment: smime.p7s
Description: S/MIME cryptographic signature