[Question] get_maintainer:

[Question] get_maintainer:

[Manuel Ebner]

I'm preparing a patch for get_maintainer. While testing I found out that
the order of the output can change for the same input - see below. That's
without any change to get_maintainer.pl. (state of 09d8b39563e84fc245 ).
It seems that the alternating execution of get_maintainer and
get_maintainer_patched triggers the glitch. Even if both files are the same.
I can't figure out what's the root cause. I think I change the state of the
script or it's surroundings somehow, but I can't figure out where.

Any Ideas?

Thanks
 Manuel


$ ./scripts/get_maintainer.pl scripts/get_maintainer/patch-w-signed-off.patch
Andrew Morton <akpm@linux-foundation.org> (maintainer:LIBRARY CODE)
Kees Cook <kees@kernel.org> (maintainer:GENERIC STRING LIBRARY)
Andy Shevchenko <andy@kernel.org> (reviewer:GENERIC STRING LIBRARY)
Willy Tarreau <w@1wt.eu> (maintainer:NOLIBC HEADER FILE)
"Thomas Weißschuh" <linux@weissschuh.net> (maintainer:NOLIBC HEADER FILE)
linux-kernel@vger.kernel.org (open list:LIBRARY CODE)
linux-hardening@vger.kernel.org (open list:GENERIC STRING LIBRARY)
LIBRARY CODE status: Supported
GENERIC STRING LIBRARY status: Supported

$ ./scripts/get_maintainer_patched.pl scripts/get_maintainer/patch-w-signed-off.patch
[...]

$ ./scripts/get_maintainer.pl scripts/get_maintainer/patch-w-signed-off.patch
Kees Cook <kees@kernel.org> (maintainer:GENERIC STRING LIBRARY)
Andy Shevchenko <andy@kernel.org> (reviewer:GENERIC STRING LIBRARY)
Andrew Morton <akpm@linux-foundation.org> (maintainer:LIBRARY CODE)
Willy Tarreau <w@1wt.eu> (maintainer:NOLIBC HEADER FILE)
"Thomas Weißschuh" <linux@weissschuh.net> (maintainer:NOLIBC HEADER FILE)
linux-hardening@vger.kernel.org (open list:GENERIC STRING LIBRARY)
linux-kernel@vger.kernel.org (open list:LIBRARY CODE)
GENERIC STRING LIBRARY status: Supported
LIBRARY CODE status: Supported

Re: [Question] get_maintainer:

[Manuel Ebner]

On Mon, 2026-06-08 at 10:15 +0200, Manuel Ebner wrote:
> I'm preparing a patch for get_maintainer. While testing I found out that
> the order of the output can change for the same input - see below. That's
> without any change to get_maintainer.pl. (state of 09d8b39563e84fc245 ).
> It seems that the alternating execution of get_maintainer and
> get_maintainer_patched triggers the glitch. Even if both files are the same.
> I can't figure out what's the root cause. I think I change the state of the
> script or it's surroundings somehow, but I can't figure out where.
> 
> Any Ideas?

I figured it out myself. It's explained here:
 https://lkml.org/lkml/2017/7/13/789

Sorry for the noise.
 Manuel