Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756789AbcJ2CgR (ORCPT ); Fri, 28 Oct 2016 22:36:17 -0400 Received: from mail-pf0-f195.google.com ([209.85.192.195]:36715 "EHLO mail-pf0-f195.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754955AbcJ2CgO (ORCPT ); Fri, 28 Oct 2016 22:36:14 -0400 Date: Fri, 28 Oct 2016 19:36:09 -0700 From: Brian Norris To: Andy Whitcroft , Joe Perches Cc: linux-kernel@vger.kernel.org, Linus Torvalds , Andrew Morton , SF Markus Elfring Subject: Re: [PATCH] checkpatch: fix uninitialized var when run with --no-tree Message-ID: <20161029023609.GA48401@google.com> References: <1477707991-140274-1-git-send-email-briannorris@chromium.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1477707991-140274-1-git-send-email-briannorris@chromium.org> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1523 Lines: 47 On Fri, Oct 28, 2016 at 07:26:31PM -0700, Brian Norris wrote: > From: Brian Norris > > If checkpatch.pl gets copied out of the tree, --no-tree shouldn't start > complaining: > > Use of uninitialized value $root in concatenation (.) or string at > /path/to/checkpatch.pl line 764. > > Let's just give the safe answer instead -- don't warn about "obsolete" > files. > > Fixes: 85b0ee18bbf8 ("checkpatch: see if modified files are marked obsolete in MAINTAINERS") > Signed-off-by: Brian Norris > --- > This is a 4.9-rc1 regression > > scripts/checkpatch.pl | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl > index a8368d1c4348..c8cd643dbc6f 100755 > --- a/scripts/checkpatch.pl > +++ b/scripts/checkpatch.pl > @@ -761,6 +761,8 @@ sub seed_camelcase_file { > sub is_maintained_obsolete { > my ($filename) = @_; > > + return 0 if (!$tree); Actually, I'm torn on this. It looks really odd to check for !$tree here, but it's the only supported case where $root shouldn't be defined. Maybe (!defined $root) is a better test? (Sorry, I did a double-take on this after I sent it.) Both would be equally correct, but I suppose the latter would be clearer. I'll send v2. Brian > + > return 0 if (!(-e "$root/scripts/get_maintainer.pl")); > > my $status = `perl $root/scripts/get_maintainer.pl --status --nom --nol --nogit --nogit-fallback -f $filename 2>&1`; > -- > 2.8.0.rc3.226.g39d4020 >