lkml.org 
[lkml]   [2010]   [Mar]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: [Xen-devel] [PATCH 5/7] xen: Make event channel work with PV extension of HVM
Date
On Tuesday 02 March 2010 09:38:21 Jeremy Fitzhardinge wrote:
> On 03/01/2010 01:38 AM, Sheng Yang wrote:
> > +
> > + x86_platform.calibrate_tsc = xen_tsc_khz;
> > + x86_platform.get_wallclock = xen_get_wallclock;
> > + x86_platform.set_wallclock = xen_set_wallclock;
> > +
> > + pv_apic_ops = xen_apic_ops;
> > +#ifdef CONFIG_X86_LOCAL_APIC
> > + /*
> > + * set up the basic apic ops.
> > + */
> > + set_xen_basic_apic_ops();
> > + apic->write = xen_hvm_pv_evtchn_apic_write;
>
> I'd just change the xen_apic_write to remove the WARN_ON, since you
> don't seem to care about it either.

So which code base I should make these patches against? We expect the patchset
can be accepted in the Linux upstream soon after you pick it up.
>
> >
> > - exit_idle();
> > - irq_enter();
> > + /*
> > + * If is PV featured HVM, these have already been done
> > + */
> > + if (likely(!xen_hvm_pv_evtchn_enabled())) {
> > + exit_idle();
> > + irq_enter();
> > + }
>
> In that case, rather than putting this conditional in the hot path, make
> an inner __xen_evtchn_do_upcall which is wrapped by the PV and HVM
> variants which do the appropriate things. (And drop the pt_regs arg, I
> think.)

Seems we still need pt_regs to for handle_irq()?

--
regards
Yang, Sheng


\
 
 \ /
  Last update: 2010-03-04 06:39    [W:0.175 / U:0.140 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site