On 9/22/21 9:13 AM, Paolo Abeni wrote:

On Tue, 2021-09-21 at 17:13 -0700, syzbot wrote:

quoted

Hello,

syzbot found the following issue on:

HEAD commit:    1f77990c4b79 Add linux-next specific files for 20210920
git tree:       linux-next
console output: https://syzkaller.appspot.com/x/log.txt?x=1383891d300000
kernel config:  https://syzkaller.appspot.com/x/.config?x=ab1346371f2e6884
dashboard link: https://syzkaller.appspot.com/bug?extid=3360da629681aa0d22fe
compiler:       gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2
syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1625f1ab300000
C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17eb1b3b300000

The issue was bisected to:

commit 020250f31c4c75ac7687a673e29c00786582a5f4
Author: Dmitry Kadashev [off-list ref]
Date:   Thu Jul 8 06:34:43 2021 +0000

    namei: make do_linkat() take struct filename

bisection log:  https://syzkaller.appspot.com/x/bisect.txt?x=15f5ef77300000
final oops:     https://syzkaller.appspot.com/x/report.txt?x=17f5ef77300000
console output: https://syzkaller.appspot.com/x/log.txt?x=13f5ef77300000

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+3360da629681aa0d22fe@syzkaller.appspotmail.com
Fixes: 020250f31c4c ("namei: make do_linkat() take struct filename")

I'm unsure why the bisection points to this commit. This looks like an
MPTCP specific issue, due to bad handling of MSG_WAITALL recvmsg()
flag.

This seems to happen quite often, and I'm guessing it's mostly likely
due to the test being run not always triggering the issue. Hence you end
up with git bisection results that go off into the weeds.

-- 
Jens Axboe