> -----Original Message-----
> From: Jan Beulich <[email protected]>
> Sent: 17 September 2019 14:03
> To: Paul Durrant <[email protected]>
> Cc: [email protected]; Suravee Suthikulpanit 
> <[email protected]>; Andrew
> Cooper <[email protected]>; George Dunlap <[email protected]>; 
> Roger Pau Monne
> <[email protected]>; Jun Nakajima <[email protected]>; Kevin Tian 
> <[email protected]>;
> BorisOstrovsky <[email protected]>; Paul Durrant <[email protected]>; Wei 
> Liu <[email protected]>
> Subject: Re: [Xen-devel] [PATCH v2 4/9] x86/HVM: move NOFLUSH handling out of 
> hvm_set_cr3()
> 
> On 17.09.2019 14:45, Paul Durrant wrote:
> >> From: Xen-devel <[email protected]> On Behalf Of Jan 
> >> Beulich
> >> Sent: 17 September 2019 07:15
> >>
> >> --- a/xen/arch/x86/hvm/emulate.c
> >> +++ b/xen/arch/x86/hvm/emulate.c
> >> @@ -2080,6 +2080,8 @@ static int hvmemul_write_cr(
> >>      HVMTRACE_LONG_2D(CR_WRITE, reg, TRC_PAR_LONG(val));
> >>      switch ( reg )
> >>      {
> >> +        bool noflush;
> >> +
> >
> > I said this before... I think the tighter case-scope is better.
> 
> I recall you saying so, but I don't recall you actually making this a
> requirement to get your ack.
> 
> > Cosmetically it may look a little odd, but surely it gives the
> > compiler a better chance to optimize?
> 
> I don't think so - they're pretty good to limit the life range of
> variables (at least in not overly hairy functions) these days. The
> more narrow scopes we often ask for are more for the reader to
> easily know what the intended usage range of a variable is.

Ok. I'm not going to insist, but I would still prefer case-scope here.

Reviewed-by: Paul Durrant <[email protected]>

> 
> Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

Reply via email to