linux-kernel - Re: [PATCH 06/14] kbuild: define PYTHON2 and PYTHON3 variables instead of PYTHON

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20180206131017.GF31804@kroah.com>
Date:   Tue, 6 Feb 2018 05:10:17 -0800
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     Masahiro Yamada <yamada.masahiro@...ionext.com>
Cc:     Linux Kbuild mailing list <linux-kbuild@...r.kernel.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Nicolas Pitre <nicolas.pitre@...aro.org>,
        "Luis R . Rodriguez" <mcgrof@...e.com>,
        Randy Dunlap <rdunlap@...radead.org>,
        Ulf Magnusson <ulfalizer@...il.com>,
        Sam Ravnborg <sam@...nborg.org>,
        Michal Marek <michal.lkml@...kovi.net>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Tony Luck <tony.luck@...el.com>, linux-ia64@...r.kernel.org,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Fenghua Yu <fenghua.yu@...el.com>
Subject: Re: [PATCH 06/14] kbuild: define PYTHON2 and PYTHON3 variables
 instead of PYTHON

On Tue, Feb 06, 2018 at 07:44:22PM +0900, Masahiro Yamada wrote:
> 2018-02-06 18:34 GMT+09:00 Greg Kroah-Hartman <gregkh@...uxfoundation.org>:
> > On Tue, Feb 06, 2018 at 09:34:46AM +0900, Masahiro Yamada wrote:
> >> The variable 'PYTHON' allows users to specify a proper executable
> >> name in case the default 'python' does not work.  However, this does
> >> not address the case where both Python 2 and Python 3 scripts are
> >> used in one system.
> >>
> >> PEP 394 (https://www.python.org/dev/peps/pep-0394/) provides a
> >> convention for Python scripts portability.  Here is a quotation:
> >>
> >>   In order to tolerate differences across platforms, all new code
> >>   that needs to invoke the Python interpreter should not specify
> >>   'python', but rather should specify either 'python2' or 'python3'.
> >>   This distinction should be made in shebangs, when invoking from a
> >>   shell script, when invoking via the system() call, or when invoking
> >>   in any other context.
> >>
> >> arch/ia64/scripts/unwcheck.py is apparently written in Python 2, so
> >> it should be invoked by 'python2'.
> >>
> >> It is legitimate to use 'python' for scripts compatible with both
> >> Python 2 and Python 3, but this is rare (at least I do not see the
> >> case in kernel tree).  You do not need to make efforts to write your
> >> scripts in that way.  Anyway, Python 2 will retire in 2020.
> >>
> >> This commit is needed for my new scripts written in Python 3.
> >>
> >> Signed-off-by: Masahiro Yamada <yamada.masahiro@...ionext.com>
> >> ---
> >>
> >>  Makefile           | 5 +++--
> >>  arch/ia64/Makefile | 2 +-
> >>  2 files changed, 4 insertions(+), 3 deletions(-)
> >>
> >> diff --git a/Makefile b/Makefile
> >> index 11aff0f..c4e935c 100644
> >> --- a/Makefile
> >> +++ b/Makefile
> >> @@ -384,7 +384,8 @@ GENKSYMS  = scripts/genksyms/genksyms
> >>  INSTALLKERNEL  := installkernel
> >>  DEPMOD               = /sbin/depmod
> >>  PERL         = perl
> >> -PYTHON               = python
> >> +PYTHON2              = python2
> >> +PYTHON3              = python3
> >
> > Is this going to break any systems that were previous setting PYTHON?
> >
> > I like this change, and feel it is the correct thing to do, but having a
> > "fallback" might be needed here.
> >
> > Could you do what the perf makefile does and do something like:
> >         override PYTHON := $(call get-executable-or-default,PYTHON,$(PYTHON2))
> > or is it really not an issue as only ia64 seems to care about this?
> >
> 
> As far as I see, ia64 is the only instance that has used this ever.
> 
> (the perf Makefile defines PYTHON by itself, so should not be a problem.)
> 
> 
> If people expect the backward-compatibility for this, I can do like follows:
> 
> # backward compatibility for 'PYTHON'
> PYTHON2  := $(if $(PYTHON), $(PYTHON), python2)

That's true, and who knows if python3 is running on ia64 :)

> Another (unlikely) possible breakage is
> 'python2' may not be installed on users' system.
> 
> I believe this is rare, but if needed, I could do like follows
> at the cost of ugliness.
> 
> 
> PYTHON2 := $(if $(PYTHON), $(PYTHON), \
>         $(shell python2 --version 2>/dev/null && echo python2 || echo python))

Ick, I guess that might work.  Try getting your patch set into the 0-day
system to see how it handles things.  I know it builds for ia64.

But this is a real tiny question overall, and should not derail this
patchset at all :)

thanks,

greg k-h