[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CANpmjNMmObBWzHqcQ2VHru2Tdu8k0KOkVUE2upPitsOh3OoF3Q@mail.gmail.com>
Date: Thu, 14 Jan 2021 20:54:42 +0100
From: Marco Elver <elver@...gle.com>
To: Nick Desaulniers <ndesaulniers@...gle.com>
Cc: Dmitry Vyukov <dvyukov@...gle.com>,
Nathan Chancellor <natechancellor@...il.com>,
syzbot <syzbot+76880518931d755473cf@...kaller.appspotmail.com>,
LKML <linux-kernel@...r.kernel.org>,
syzkaller-bugs <syzkaller-bugs@...glegroups.com>,
clang-built-linux <clang-built-linux@...glegroups.com>
Subject: Re: upstream build error (12)
On Thu, 14 Jan 2021 at 19:07, Nick Desaulniers <ndesaulniers@...gle.com> wrote:
> On Thu, Jan 14, 2021 at 4:38 AM 'Dmitry Vyukov' via Clang Built Linux
> <clang-built-linux@...glegroups.com> wrote:
> >
> > On Thu, Jan 14, 2021 at 12:32 PM Dmitry Vyukov <dvyukov@...gle.com> wrote:
> > >
> > > On Tue, Jan 12, 2021 at 8:41 PM Nathan Chancellor
> > > <natechancellor@...il.com> wrote:
> > > >
> > > > Would it be possible for clang-built-linux@...glegroups.com to be CC'd
> > > > when there is a build error and the compiler is clang? Especially if
> > > > clang is hitting an assertion.
> > >
> > > Hi Nathan,
> > >
> > > I am adding functionality to CC specific emails on build errors on
> > > specific instances:
> > > https://github.com/google/syzkaller/pull/2388
> >
> > I've updated configs to CC clang-built-linux@...glegroups.com. This
> > should take effect on the next build failure (hopefully).
>
> Thanks Dmitry, (and thanks Marco for updating syzbot's clang).
>
> Marco, can you update:
> https://github.com/google/syzkaller/blob/master/docs/syzbot.md#crash-does-not-reproduce
We have:
> compiler: clang version 11.0.0 (https://github.com/llvm/llvm-project.git ca2dcbd030eadbf0aa9b660efe864ff08af6e18b)
Which is in the list of compilers:
https://storage.googleapis.com/syzkaller/clang-11-prerelease-ca2dcbd030e.tar.xz
> I wasn't able to reproduce the reported compiler crash with ToT llvm
> and the reported config, so I assume it was a mid-release of llvm-11
> bug that has since been fixed. Do we need to report to syzbot to
> close the issue? I did a quick skim of
> https://github.com/google/syzkaller/blob/master/docs/syzbot.md, but it
> looks like communication with syzbot is based on commit messages?
Given we this is likely due to a prerelease compiler, we can just use:
#syz invalid
^^ it should pick this up.
Thanks,
-- Marco
Powered by blists - more mailing lists