Re: [PATCH v4 07/13] x86/hyperlaunch: obtain cmdline from device tree

[dmkhn@proton.me]

On Wed, Apr 23, 2025 at 03:08:18PM +0200, Jan Beulich wrote:
> On 23.04.2025 15:01, Alejandro Vallejo wrote:
> > On Fri Apr 18, 2025 at 11:53 PM BST, dmkhn wrote:
> >> On Thu, Apr 17, 2025 at 01:48:29PM +0100, Alejandro Vallejo wrote:
> >>> --- a/xen/common/domain-builder/fdt.h
> >>> +++ b/xen/common/domain-builder/fdt.h
> >>> @@ -9,6 +9,30 @@ struct boot_info;
> >>>  /* hyperlaunch fdt is required to be module 0 */
> >>>  #define HYPERLAUNCH_MODULE_IDX 0
> >>>
> >>> +static inline int __init fdt_cmdline_prop_size(const void *fdt, int offset)
> >>> +{
> >>> +    int ret;
> >>> +
> >>> +    fdt_get_property_by_offset(fdt, offset, &ret);
> >>
> >> Perhaps something like
> >>
> >>        (void)fdt_get_property_by_offset...
> >>
> >> since there's no need to check the return value?
> >
> > I vaguely seem to remember doing something like that a few years ago
> > (because it does show intent and many linters require it) and being told
> > not to. But maybe I misremember. I'm definitely happy to use that
> > convention here and later unless someone pushes back for some reason.
> 
> Unless we settle on the need for such for Misra's sake, I'd like to ask
> to avoid them. We generally try to avoid casts as much as possible. We
> then also shouldn't add ones like suggested here.

Ack.

> 
> Jan