Re: [Outreachy kernel] Trying to use Coccinelle

[Julia Lawall <julia.lawall@inria.fr>]

On Mon, 16 Mar 2020, Stefano Brivio wrote:

> On Mon, 16 Mar 2020 15:27:38 +0100 (CET)
> Julia Lawall <julia.lawall@inria.fr> wrote:
>
> > On Mon, 16 Mar 2020, Stefano Brivio wrote:
> >
> > > On Fri, 13 Mar 2020 10:40:46 +0100
> > > Lourdes Pedrajas <lu@pplo.net> wrote:
> > >
> > > > [...]
> > > >
> > > > Indeed, this is all the output: (this command is run in the sources root
> > > > directory)
> > > >
> > > > make coccicheck MODE=patch J=8 M=drivers/staging/android/
> > > >
> > > > Please check for false positives in the output before submitting a patch.
> > > > When using "patch" mode, carefully review the patch before submitting it.
> > > >
> > > > coccicheck failed
> > > > Makefile:1740: recipe for target 'coccicheck' failed
> > > > make: *** [coccicheck] Error 255
> > >
> > > Hah, I think I got the same issue on Debian. I installed coccinelle via
> > > opam, then:
> > >
> > > ---
> > > $ make coccicheck MODE=report M=drivers/staging/android/
> > >
> > > Please check for false positives in the output before submitting a patch.
> > > When using "patch" mode, carefully review the patch before submitting it.
> > >
> > > coccicheck failed
> > > make: *** [Makefile:1740: coccicheck] Error 255
> > > ---
> > >
> > > Debugging: looking at makefiles, I thought about using V=1:
> > >
> > > ---
> > > $ make V=1 coccicheck MODE=report M=drivers/staging/android/
> > > bash ./scripts/coccicheck
> > >
> > > Please check for false positives in the output before submitting a patch.
> > > When using "patch" mode, carefully review the patch before submitting it.
> > >
> > > Processing alloc_cast.cocci
> > > with option(s) " --no-includes --include-headers"
> > >
> > > Message example to submit a patch:
> > >  Remove casting the values returned by memory allocation functions
> > >  like kmalloc, kzalloc, kmem_cache_alloc, kmem_cache_zalloc etc.
> > >
> > >  The semantic patch that makes this report is available
> > >  in scripts/coccinelle/api/alloc/alloc_cast.cocci.
> > >
> > >  More information about semantic patching is available at
> > >  http://coccinelle.lip6.fr/
> > >
> > > Semantic patch information:
> > >  This makes an effort to find cases of casting of values returned by
> > >  kmalloc, kzalloc, kcalloc, kmem_cache_alloc, kmem_cache_zalloc,
> > >  kmem_cache_alloc_node, kmalloc_node and kzalloc_node and removes
> > >  the casting as it is not required. The result in the patch case may
> > >  need some reformatting.
> > >
> > > Running (4 in parallel): /home/sbrivio/.opam/default/bin/spatch -D report --no-show-diff --very-quiet --cocci-file ./scripts/coccinelle/api/alloc/alloc_cast.cocci --no-includes --include-headers --patch . --dir drivers/staging/android/ -I ./arch/x86/include -I ./arch/x86/include/generated -I ./include -I ./arch/x86/include/uapi -I ./arch/x86/include/generated/uapi -I ./include/uapi -I ./include/generated/uapi --include ./include/linux/kconfig.h --jobs 4 --chunksize 1
> > > coccicheck failed
> > > make: *** [Makefile:1740: coccicheck] Error 255
> > > ---
> > >
> > > so, yeah, spatch fails, but no error (see below) isn't propagated (even
> > > with V=1)
> > >
> > > Julia, do you think it's worth trying to fix this? Any quick idea?
> >
> > I will try to find the problem.  I think there is a way to get debug
> > information.
>
> One idea might be, in run_cmd_parmap(), to avoid redirecting stderr
> here:
> 	        $@ 2>>$DEBUG_FILE
>
> if DEBUG_FILE is not set (that is, it's /dev/null) and we detect V=1 --
> otherwise, spatch output is thrown away.
>
> That works for me, but I'm not sure how elegant you would consider that.

Actually, I was just going to try to fix the semantic patch that caused
the problem.  I don't really maintain make coccicheck, but I will suggest
it to the person who does.  Thanks.

julia