[PATCH] travis: add checkpatch analysis

Roman Stratiienko roman.stratiienko at globallogic.com
Fri Mar 22 20:45:44 CET 2019


Thank you for feedback. After testing of this patch I faced some challenges:

1. checkpatch.pl from kernel tree always different. Old kernel checkpatch
doesn't support git option. So we need to find a way to use ommon utility
for all kernel or just enable checkpatch only on most recent kernel
2. Even if there is no checkpatch errors, but warnings still present, exit
code will be false. And I can't find command line option to override it.
3. Getting delta commits in master..next range isn't a trivial task. Travis
clone only single branch with - - depth=50 option. Unshallowing entire
repository will take some amount of time, that will slow down build process
and it is not what me want from CI system.



вт, 19 мар. 2019 г., 20:15 Jan Kiszka <jan.kiszka at siemens.com>:

> On 19.03.19 15:59, Roman Stratiienko wrote:
> > Currently I see a lot of checkpatch warnings, that can not be fixed in
> xenomai
> > project.
> > But checkpatch errors usually could be fixed, so we should try I think.
> >
>
> I don't disagree on the general goal of clean coding style. I just don't
> want to
> open too many fronts at the same time. If the checker nags us with reports
> that
> no one reads in the end, that will not help. And we know that there
> corners like
> rtnet which needs more than some smaller tweaks (and will currently only
> generate errors unless you break the pre-existing style).
>
> If we should agree on using automated style checking, I would then
> redefine the
> range to master..next, because that is what we are currently adding.
>
> Jan
>
> >
> > On Tue, Mar 19, 2019 at 4:45 PM Jan Kiszka <jan.kiszka at siemens.com
> > <mailto:jan.kiszka at siemens.com>> wrote:
> >
> >     On 19.03.19 15:28, roman.stratiienko--- via Xenomai wrote:
> >      > From: Roman Stratiienko <roman.stratiienko at globallogic.com
> >     <mailto:roman.stratiienko at globallogic.com>>
> >      >
> >      > Check last 10 commits with checkpatch.pl <http://checkpatch.pl>
> before build
> >      >
> >      > Signed-off-by: Roman Stratiienko <
> roman.stratiienko at globallogic.com
> >     <mailto:roman.stratiienko at globallogic.com>>
> >      > ---
> >      >   .travis.yml | 1 +
> >      >   1 file changed, 1 insertion(+)
> >      >
> >      > diff --git a/.travis.yml b/.travis.yml
> >      > index 0cd9e571f..6fb604bf2 100644
> >      > --- a/.travis.yml
> >      > +++ b/.travis.yml
> >      > @@ -78,6 +78,7 @@ before_script:
> >      >     - popd
> >      >
> >      >   script:
> >      > +  - ${KDIR}/scripts/checkpatch.pl <http://checkpatch.pl> --git
> HEAD~10..HEAD
> >      >     - scripts/prepare-kernel.sh --ipipe=/tmp/ipipe.patch
> --arch=${ARCH}
> >     --linux=${KDIR}
> >      >     - pushd ${KDIR}
> >      >     - make -j $(nproc) olddefconfig
> >      >
> >
> >     Hmm, not convinced about enforced checking. That will fail the
> build, even if
> >     there might be a legitimate reason for some exception, no?
> >
> >
> >
> >
> >
> >
> >     Jan
> >
> >     --
> >     Siemens AG, Corporate Technology, CT RDA IOT SES-DE
> >     Corporate Competence Center Embedded Linux
> >
> >
> >
>
> --
> Siemens AG, Corporate Technology, CT RDA IOT SES-DE
> Corporate Competence Center Embedded Linux
>


More information about the Xenomai mailing list