From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dmitry Vyukov Subject: Re: WARNING in untrack_pfn Date: Wed, 8 May 2019 14:37:44 +0200 Message-ID: References: <000000000000f70a0e0571ad8ffb@google.com> <0000000000007b03eb058653caea@google.com> <20190412130112.GA8384@amd> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Return-path: In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Julia Lawall Cc: Pavel Machek , syzbot , Borislav Petkov , Dan Williams , Dave Hansen , Dave Jiang , Haozhong Zhang , "H. Peter Anvin" , jacek.anaszewski@gmail.com, LKML , linux-leds@vger.kernel.org, Andy Lutomirski , Ingo Molnar , Paolo Bonzini , Peter Zijlstra , =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= , rpurdie@rpsys.net, syzkaller-bugs , Thomas Gleixner , the arch/x86 maintainers List-Id: linux-leds@vger.kernel.org From: Julia Lawall Date: Fri, Apr 12, 2019 at 3:21 PM > On Fri, 12 Apr 2019, Pavel Machek wrote: > > > On Fri 2019-04-12 04:42:01, syzbot wrote: > > > syzbot has bisected this bug to: > > > > > > commit c68729119f4d2993bec3c9cb999ad76de5aeddba > > > Author: Julia Lawall > > > Date: Sat Jul 15 09:58:19 2017 +0000 > > > > > > leds: tlc591xx: add missing of_node_put > > > > > > bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=12cda0f3200000 > > > start commit: 23203e3f Merge branch 'akpm' (patches from Andrew) > > > git tree: upstream > > > final crash: https://syzkaller.appspot.com/x/report.txt?x=11cda0f3200000 > > > console output: https://syzkaller.appspot.com/x/log.txt?x=16cda0f3200000 > > > kernel config: https://syzkaller.appspot.com/x/.config?x=861a3573f4e78ba1 > > > dashboard link: https://syzkaller.appspot.com/bug?extid=e1a4f80c370d2381e49f > > > syz repro: https://syzkaller.appspot.com/x/repro.syz?x=17836d15400000 > > > > > > Reported-by: syzbot+e1a4f80c370d2381e49f@syzkaller.appspotmail.com > > > Fixes: c68729119f4d ("leds: tlc591xx: add missing of_node_put") > > > > > > For information about bisection process see: https://goo.gl/tpsmEJ#bisection > > > > Can you revert c68729119f4d2993bec3c9cb999ad76de5aeddba on master and > > see if it fixes the problem? Because this is hard to believe... > > Likewise, it seems hard to believe that the change is a problem. But > could the problem be elsewhere and triggered by this code? Looking at the bisection log, this looks like a hard to reproduce, flaky crash. Most likely the crash did not fire on one of the commits and that diverted bisection in the wrong direction.