Re: [PATCH] xen/arm: traps: fix MISRA C 2012 Rule 8.7 violation

[Xenia Ragiadakou <burzalodowa@xxxxxxxxx>]

On 7/4/22 11:54, Luca Fancellu wrote:
> > > > > > > As said I would vote for external linkage here but would be nice to have other
> > > > > > > developers view on this.
> > > > > > In addition to this, if we don’t want to provide a justification for those, 
> > > > > > since they seems to me
> > > > > > code related to debugging they can be removed from “production” code in some 
> > > > > > way.
> > > > > Rule 8.7 is advisory, so I think that formal justification of deviations is not 
> > > > > necessary.
> > > > Yes that is true, in that case we would only need to document it without a 
> > > > formal justification, however
> > > > if the codebase doesn’t include them (because not in production code) I guess the 
> > > > problem doesn’t exist.
> > > Having the production code using static and the non production using external 
> > > linkage would be kind of weird here.
> > > I think having them always with external linkage with a justification is the 
> > > cleanest way.
> > +1 this is what I was going to answer :).
> Yes probably I didn’t explained very well, I’m in favour for external linkage, 
> hence we are going to have an advisory to
> document.
>
> I was just thinking if we need to document that *if* the codebase doesn’t 
> include them, which is a comment not related
> to this patch so apologies for the noise on that.
>
> > Cheers,
> >
> > --
> > Julien Grall
I would like to mention that show_execution_state() is also available 
for dumping the state but probably you need them for more fine grained 
debugging.
I will wait until tomorrow in case there is further input on this and I 
will send another patch, if necessary.
--
Xenia