Re: [PATCH v4 3/4] tools: add tools support for Intel CDP

[He Chen <he.chen@linux.intel.com>]

On Thu, Sep 24, 2015 at 12:22:47PM +0100, Ian Campbell wrote:
> On Thu, 2015-09-24 at 12:07 +0100, Ian Campbell wrote:
> > @@ -8517,8 +8535,19 @@ int main_psr_cat_cbm_set(int argc, char **argv)
> > >          libxl_string_list_dispose(&socket_list);
> > >          free(value);
> > >          break;
> > > +    case 'd':
> > > +        type = LIBXL_PSR_CBM_TYPE_L3_DATA;
> > > +        opt_data = 1;
> > > +        break;
> > > +    case 'c':
> > > +        type = LIBXL_PSR_CBM_TYPE_L3_CODE;
> > > +        opt_code = 1;
> > > +        break;
> > >      }
> > >  
> > > +    if (opt_data && opt_code)
> > 
> > Do you not mean !opt_data && !opt_code?
> > 
> > But also, isn't this assignment unnecessary since type is initialised to
> > the same value when it is declared?
> > 
> > In fact, because of that initialisation, aren't opt_data and opt_code
> > unnecessary, since you set type appropriately elsewhere.
> > 
> > Are -d and -c mutually exclusive, or is it expected that both can be
> > given?
> 
> Also, is there error checking for passing -c or -d when CDP is not enabled
> somewhere else?
> 
> Ian.

Yes, there is error checking in hypervisor.
hypervisor would reture an error code if -c or -d is given when CDP is
not enable.

In fact, the reture error code would be catched by
`libxl__psr_cat_log_err_msg`, and that is exactly what you and Jan
discussed in Patch 3.