Re: [XEN PATCH] xen: Add SAF deviations for MISRA C:2012 Rule 7.1

[Stefano Stabellini <sstabellini@xxxxxxxxxx>]

On Tue, 10 Oct 2023, Julien Grall wrote:
> Hi Stefano,
> 
> On 09/10/2023 23:19, Stefano Stabellini wrote:
> > > 
> > > I am not sure this is better. This is a long line to read. But this is a
> > > personal opinion.
> > > 
> > > On the technical side, can we easily teach a tool to format this kind of
> > > exception? If not, then this should not be an exception we should
> > > implement.
> > 
> > I am not sure I understand what you mean by "can we easily teach a tool
> > to format this kind of exception". Do you mean whether we can teach a
> > tool to interpret a multiline statement as a single statement?
> 
> Sorry for the wording was not clear. I was referring to tools formatting the
> code (e.g. clang-format). Hopefully, at some point, we will finally have a way
> to automatically format the code. So we need a coding style that can be easily
> be checked.
> 
> It is not clear to me whether your proposed exception would work. We may have
> to allow longer lines (and this has drawback).

Yes, that's a good point. It will probably be an issue with clang-format.


> > /* cppcheck tag */
> > line1
> > /* cppcheck tag */
> > line2
> > /* cppcheck tag */
> > line3
> > 
> > and that would end up changing the line numbers in the source files so
> > the cppcheck report wouldn't match with the original line numbers any
> > longer
> 
> Would cppcheck accepts tag at the end of the line? If so, the following would
> not modify the line count:
> 
> /* cppcheck tag */
> line1 /* added cppcheck tag */
> line2 /* added cppcheck tag */
> line3 /* added cppcheck tag */

Luca answered to a similar, more generic, question a few days ago about
Coverity: https://marc.info/?l=xen-devel&m=169657904027210

So even if we get cppcheck to work like that, we would lose Coverity
support.

It doesn't seem there are a lot of good options here. In this case Luca
came up with a good idea on how to refactor the code so we should be
good.

But it doesn't seem we'll be able to come up with a general solution to
the multiline statement problem.