Re: [PATCH v8 15/17] vmx: VT-d posted-interrupt core logic handling

["Wu, Feng" <feng.wu@intel.com>]

> -----Original Message-----
> From: Dario Faggioli [mailto:dario.faggioli@citrix.com]
> Sent: Thursday, October 29, 2015 12:37 AM
> To: Jan Beulich <JBeulich@suse.com>; Wu, Feng <feng.wu@intel.com>
> Cc: Tian, Kevin <kevin.tian@intel.com>; Keir Fraser <keir@xen.org>;
> GeorgeDunlap <george.dunlap@eu.citrix.com>; Andrew Cooper
> <andrew.cooper3@citrix.com>; xen-devel@lists.xen.org
> Subject: Re: [Xen-devel] [PATCH v8 15/17] vmx: VT-d posted-interrupt core logic
> handling
> 
> On Wed, 2015-10-28 at 03:03 -0600, Jan Beulich wrote:
> > > > > On 28.10.15 at 03:58, <feng.wu@intel.com> wrote:
> 
> > > We still call arch_vcpu_block() in vcpu_block(), but we don't need
> > > to
> > > call arch_vcpu_block_cancel(v) when local_events_need_delivery()
> > > returns true. Then before VM-Entry, we can check if 'NV' field is
> > > ' pi_wakeup_vector', if it is, we change the PI status and remove
> > > it from the blocking list.
> > >
> > > Jan, if #2 is what you mean, I think it worth a try. If I don't
> > > understand
> > > your comments correctly, could you please elaborate it more? Thanks
> > > a lot!
> >
> > Ideally we'd avoid both arch_vcpu_*() calls by doing what is needed
> > in arch code (e.g. the VM entry path).
> >
> +1
> 
> > If only avoiding the cancel
> > hook is reasonably possible this way, then so be it - still better to
> > have just one hook here than having two.
> >
> +1 again
> 
> As an aside, I've spoked with some ARM people, the context being that
> they will get to implement a similar feature in the nearish future.
> 
> They said that, although they can't be sure, they don't think they'll
> be interested in arch hooks in the scheduling code and that, actually,
> having them risk being more harmful than helpful.
> 
> Of course, that does not mean that we shouldn't add them for the sake
> of this patch, if we can't avoid doing that. But if we can avoid them,
> that is perhaps one more reason for doing things that way.

Jan & Dario, thanks a lot for you guys' input. If you agree, I will remove
the arch_vcpu_block_cancel() and maybe arch_vcpu_wake_prepare()
as well, and implement the logic before VM-entry in V9, the patch will
be coming soon. Thanks again!

Thanks,
Feng 

> 
> Regards,
> Dario
> --
> <<This happens because I choose it to happen!>> (Raistlin Majere)
> -----------------------------------------------------------------
> Dario Faggioli, Ph.D, http://about.me/dario.faggioli
> Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)