Re: [PATCH v4 1/1] runchecks: Generalize make C={1,2} to support multiple checkers

[Knut Omang]

On Mon, 2018-02-05 at 16:03 +0900, Masahiro Yamada wrote:
> 2018-02-05 15:41 GMT+09:00 Knut Omang <knut.omang@xxxxxxxxxx>:
> > On Fri, 2018-01-19 at 11:14 +0100, Knut Omang wrote:
> >> Add scripts/runchecks which has generic support for running
> >> checker tools in a convenient and user friendly way that
> >> the author hopes can contribute to rein in issues detected
> >> by these tools in a manageable and convenient way.
> >>
> >> scripts/runchecks provides the following basic functionality:
> >>
> >> * Makes it possible to selectively suppress output from individual
> >>   checks on a per file or per subsystem basis.
> >> * Unifies output and suppression input from different tools
> >>   by providing a single unified syntax and presentation for the
> >>   underlying tools in the style of "scripts/checkpatch.pl --show-types".
> >> * Allows selective run of one, or more (or all) configured tools
> >>   for each file.
> >>
> >> In the Makefile system, the sparse specific setup has been replaced
> >> by setup for runchecks.
> >
> > Hi all,
> >
> > - Anything more I can/need to do to bring this forward?
> > - Any quiet concerns?
> >
> > I realize it is a subsystem crossing change,
> 
> Is it?  Only Kbuild this is related to.

Ok, I see!

> > and a lot going on elsewhere,
> > nevertheless I believe this is a time saver in the slightly longer run,
> > as it allows automation of checking, even without a "perfect"
> > code base to begin with.
> >
> 
> Sorry for the delay.

I understand, no problem - just was afraid it was about to get lost 
in between subsystems,

> I have not been able to find time to dive into the detail yet.
> (Actually, I tried to do that for v2 or v3, where Python code was so dirty,
> then consumed my time to figure out what the code was trying to do)

Hopefully v4 is cleaner from a Python code style point of view at least,
but let me know if you have any particular part of the code in mind wrt 
readability. Also hopefully the docs should be of help.

> I find my concern here:
> https://lkml.org/lkml/2018/1/5/497

I believe I have addressed the issues there in v4.

> Anyway, I will take a look again when I find some time.
> You do not need to take care of the detail until I request to do so.

Ok, thanks a lot for your time and the quick response now!

Best regards,
Knut