Re: Top MISRA violations

[Julien Grall <julien@xxxxxxx>]

On 23/04/2020 01:52, Stefano Stabellini wrote:
> On Wed, 22 Apr 2020, Julien Grall wrote:
> > > # Rule 8.4
> > >
> > > A compatible declaration shall be visible when an object or function
> > > with external linkage is defined.
> > >
> > > Example:
> > >
> > > extern int16_t count;
> > > int16_t count = 1; /* compliant */
> > >
> > > extern int16_t count = 1; /* non-compliant */
> > >
> > >
> > > ## Analysis & Potential Solution
> > >
> > > This is not a rule I would expect Xen to be violating.
> > I am afraid Xen is violating this one in quite a few places. A good
> > example is functions called from assembly such as start_xen().
> > We also have a few places where the #include declaring the prototypes
> > are not present before the implementation.
> >
> > -Wall doesn't include -Wmissing-prototypes, I did attempt it a few
> > years ago to add it but only part of the work made upstream so the
> > flag was never added.
> That is a good pointer, thank you. I can try to have a look at what
> happens when adding -Wmissing-prototypes to the build.
>
>  From my understanding, I don't think the functions called from assembly
> would count as violations, but the others would.
I don't think the violation is on the callee side. It is on the caller 
side because we are defining an external function (or variable) without 
an external prototype.
So whether the function is called from assembly (or C) would not matter 
here.
Cheers,

--
Julien Grall