Re: [Xen-devel] [PATCH 1/8] viridian: add init hooks

[Paul Durrant <Paul.Durrant@xxxxxxxxxx>]

> -----Original Message-----
> From: Wei Liu [mailto:wei.liu2@xxxxxxxxxx]
> Sent: 25 January 2019 14:23
> To: Paul Durrant <Paul.Durrant@xxxxxxxxxx>
> Cc: Andrew Cooper <Andrew.Cooper3@xxxxxxxxxx>; xen-
> devel@xxxxxxxxxxxxxxxxxxxx; Wei Liu <wei.liu2@xxxxxxxxxx>; Jan Beulich
> <jbeulich@xxxxxxxx>; Roger Pau Monne <roger.pau@xxxxxxxxxx>
> Subject: Re: [Xen-devel] [PATCH 1/8] viridian: add init hooks
> 
> On Thu, Jan 03, 2019 at 02:18:18PM +0000, Paul Durrant wrote:
> > > -----Original Message-----
> [...]
> > >
> > > Looking at this a little more... Viridian is an x86 specific thing, so
> an
> > > extra flag in xen_arch_domainconfig would seem most appropriate. This
> > > would then need to wired into the appropriate place(s) in libxl. I'll
> have
> > > a look at how much work this is likely to be.
> >
> > Would something along the lines of the following (as yet untested and
> incomplete) patch be acceptable?
> >
> > (I've blindly coded the Ocaml $MAGIC and just hardcoded the flag in the
> stub)
> >
> 
> Not sure if your question is directed to me or not.
> 
> I think this patch is independent of this series to hand?

Yes, it was a stab at an incremental patch to do what Andy wanted.

> 
> > ---8<---
> >
> > -int hvm_domain_initialise(struct domain *d)
> > +int hvm_domain_initialise(struct domain *d, bool enable_viridian)
> >  {
> >      unsigned int nr_gsis;
> >      int rc;
> > diff --git a/xen/include/asm-x86/hvm/hvm.h b/xen/include/asm-
> x86/hvm/hvm.h
> > index 7892f98c7b..6bde613dd8 100644
> > --- a/xen/include/asm-x86/hvm/hvm.h
> > +++ b/xen/include/asm-x86/hvm/hvm.h
> > @@ -241,7 +241,7 @@ extern s8 hvm_port80_allowed;
> >  extern const struct hvm_function_table *start_svm(void);
> >  extern const struct hvm_function_table *start_vmx(void);
> >
> > -int hvm_domain_initialise(struct domain *d);
> > +int hvm_domain_initialise(struct domain *d, bool enable_viridian);
> 
> hvm_domain_initialise (and pv_domain_initialise) used to take a
> xen_arch_domain_config. I think you can introduce it back.

I'm going to side-step the whole issue for the moment. I don't think this 
series is the right place to be making the change. Yes, it out-of-lining the 
structs and thus enables the optimization, but allocating them unconditionally 
is essentially only equivalent to leaving them inline in terms of memory cost.

  Paul

> 
> Wei.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel