Re: [Xen-devel] [PATCH v4 02/24] x86: refactor psr: remove L3 CAT/CDP codes.

[Yi Sun <yi.y.sun@xxxxxxxxxxxxxxx>]

On 16-12-22 09:03:25, Jan Beulich wrote:
> >>> On 14.12.16 at 05:07, <yi.y.sun@xxxxxxxxxxxxxxx> wrote:
> > There is no construction without destruction. So, this patch
> > removes all current L3 CAT/CDP codes in psr.c. The following
> > patches will introduce the new mechanism.
> > 
> > Signed-off-by: Yi Sun <yi.y.sun@xxxxxxxxxxxxxxx>
> 
> Acked-by: Jan Beulich <jbeulich@xxxxxxxx>

Thank you!

> with one question:
> 
> > @@ -738,14 +282,10 @@ static int __init psr_presmp_init(void)
> >      if ( (opt_psr & PSR_CMT) && opt_rmid_max )
> >          init_psr_cmt(opt_rmid_max);
> >  
> > -    if ( opt_psr & PSR_CAT )
> > -        init_psr_cat();
> > -
> > -    if ( psr_cpu_prepare(0) )
> > -        psr_cat_free();
> > +    psr_cpu_prepare(0);
> >  
> >      psr_cpu_init();
> > -    if ( psr_cmt_enabled() || cat_socket_info )
> > +    if ( psr_cmt_enabled() )
> >          register_cpu_notifier(&cpu_nfb);
> 
> You retain this notifier - do you expect to need it going forward?
> Till now it was used to allocate that strange temporary object,
> which iirc we've discussed (during the earlier versions) would go
> away.
> 
> Jan

Do you mean register_cpu_notifier() or psr_cpu_prepare()? For
register_cpu_notifier(), it is needed because we need handle CPU events
through it. For psr_cpu_prepare(), it is needed too. You can see it is
used to allocate memory for feature node in patch 04.

BRs,
Sun Yi

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel