Re: [PATCH 2/4] xen/scripts: add cppcheck tool to the xen-analysis.py script

[Stefano Stabellini <sstabellini@xxxxxxxxxx>]

On Thu, 1 Dec 2022, Jan Beulich wrote:
> On 01.12.2022 12:18, Luca Fancellu wrote:
> >> On 1 Dec 2022, at 08:33, Jan Beulich <jbeulich@xxxxxxxx> wrote:
> >> On 30.11.2022 21:26, Stefano Stabellini wrote:
> >>> On Wed, 30 Nov 2022, Luca Fancellu wrote:
> >>>>> I think the revert of the cppcheck integration in xen/Makefile and
> >>>>> xen/tools/merge_cppcheck_reports.py could be a separate patch. There is
> >>>>> no need to make sure cppcheck support in the xen Makefile is
> >>>>> "bisectable". That patch could have my acked-by already.
> >>>>
> >>>> Ok I will split these changes in a following patch
> >>>>
> >>>>>
> >>>>> Also the document changes introduced in this patch have my reviewed-by:
> >>>>> - docs/misra/cppcheck.txt
> >>>>> - docs/misra/documenting-violations.rst
> >>>>> - docs/misra/false-positive-cppcheck.json
> >>>>> - docs/misra/xen-static-analysis.rst
> >>>>
> >>>> Thank you, should I put those files in a separate patch with your rev-by 
> >>>> before
> >>>> this patch or this is just a comment for you to remember which file you 
> >>>> already
> >>>> reviewed?
> >>>
> >>> If Jan and the other reviewers are OK, I think you could split them out
> >>> in a separate patch and add my reviewed-by. If Jan prefers to keep it
> >>> all together in one patch, then I wrote it down so that I remember what
> >>> I have already acked :-)
> >>
> >> Docs changes being split off and going in first is okay as long as what
> >> is being documented is present behavior. If other changes are needed to
> >> make (parts of) new documentation actually correct, then it should imo
> >> go together. If new documentation describes future behavior (e.g.
> >> design docs), then of course it's fine as well to go in early, as then
> >> there simply is no code anywhere which this would (temporarily) not
> >> describe correctly.
> > 
> > Yeah I thought so, I would prefer to keep these files here otherwise I 
> > would need to
> > change them somehow and I would lose the r-by anyway.
> > 
> > Regarding the revert of cppcheck from makefile and 
> > xen/tools/merge_cppcheck_reports.py,
> > are you ok if I send a patch with only those changes? Would it be ok for 
> > you if the new patch
> > is after this one?
> 
> I don't mind you doing so, but I guess the question is mainly to people
> actually / possibly making use of those make goals.

I think it is OK -- we are not at the stage where cppcheck is used in
production-worthy pipelines yet.