[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [ PATCH 2/2] xen: enable Virtual-interrupt delivery


  • To: Jan Beulich <JBeulich@xxxxxxxx>
  • From: "Li, Jiongxi" <jiongxi.li@xxxxxxxxx>
  • Date: Fri, 14 Sep 2012 02:31:30 +0000
  • Accept-language: en-US
  • Cc: "xen-devel@xxxxxxxxxxxxx" <xen-devel@xxxxxxxxxxxxx>
  • Delivery-date: Fri, 14 Sep 2012 02:32:22 +0000
  • List-id: Xen developer discussion <xen-devel.lists.xen.org>
  • Thread-index: AQHNkadmVtF+66B2HEqeBgsCFNgP8peJEh5w
  • Thread-topic: [Xen-devel] [ PATCH 2/2] xen: enable Virtual-interrupt delivery


> -----Original Message-----
> From: Jan Beulich [mailto:JBeulich@xxxxxxxx]
> Sent: Thursday, September 13, 2012 8:01 PM
> To: Li, Jiongxi
> Cc: xen-devel@xxxxxxxxxxxxx
> Subject: RE: [Xen-devel] [ PATCH 2/2] xen: enable Virtual-interrupt delivery
> 
> >>> On 13.09.12 at 12:13, "Li, Jiongxi" <jiongxi.li@xxxxxxxxx> wrote:
> >> Considering that the original could already have been written with
> > if/else-if, I
> >> was suggesting to expand this to your addition:
> >>
> >> if ( cpu_has_vmx_virtual_intr_delivery ) { } else if (a)
> >>   {}
> >> else if (b)
> >>   {}
> >>
> >> which will avoid any (indentation only) changes past the first of the
> >> two
> > else-if-s.
> >> Plus it would make the logic of the code more clear, at once likely
> >> making apparent that there'll now be quite a few "goto out"-s that
> >> ought to be check for being replaceable by fewer instances of them
> >> placed slightly
> > differently.
> > It is a good suggestion. But the original code is two parallel if()
> > case, not the if/else-if case, and can't be changed to if/else-if
> > case, so I just keep the original code here. :)
> 
> That's simply not true. The code before your patch is
> 
>         if ( intblk == hvm_intblk_tpr )
>         {
>             ...
>             goto out;
>         }
> 
>         if ( (intblk != hvm_intblk_none) || ... )
>         {
>             ...
>             goto out;
>         }
> 
> which can easily be re-written into and if()/else if() (due to the goto at 
> the first
> if() body's end). All you want in your patch is then to prepend another if() 
> and
> convert the initial if() into an else if() too.
> 
I get your idea now, sorry for the misunderstanding before. A new patch for 
this will be sent out
> Jan


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


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.