Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <00000000000093079705ea9aada2@google.com> <0000000000004ee9b405f82355d8@google.com>
In-Reply-To: <0000000000004ee9b405f82355d8@google.com>
From:   Aleksandr Nogikh <nogikh@google.com>
Date:   Fri, 31 Mar 2023 10:43:50 +0200
Message-ID: <CANp29Y7wQccVBGJM_jJzrszzkE9wiHOhjdYoaPtzB=0WP0BbDA@mail.gmail.com>
Subject: Re: [syzbot] [jfs?] UBSAN: shift-out-of-bounds in dbFindCtl
To:     syzbot <syzbot+7edb85bc97be9f350d90@syzkaller.appspotmail.com>
Cc:     dave.kleikamp@oracle.com, jfs-discussion@lists.sourceforge.net,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
        liushixin2@huawei.com, shaggy@kernel.org,
        syzkaller-bugs@googlegroups.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Thu, Mar 30, 2023 at 9:45=E2=80=AFPM syzbot
<syzbot+7edb85bc97be9f350d90@syzkaller.appspotmail.com> wrote:
>
> syzbot suspects this issue was fixed by commit:
>
> commit fad376fce0af58deebc5075b8539dc05bf639af3
> Author: Liu Shixin via Jfs-discussion <jfs-discussion@lists.sourceforge.n=
et>
> Date:   Thu Nov 3 03:01:59 2022 +0000
>
>     fs/jfs: fix shift exponent db_agl2size negative
>
> bisection log:  https://syzkaller.appspot.com/x/bisect.txt?x=3D146d74a5c8=
0000
> start commit:   b229b6ca5abb Merge tag 'perf-tools-fixes-for-v6.1-2022-10=
-..
> git tree:       upstream
> kernel config:  https://syzkaller.appspot.com/x/.config?x=3Da66c6c673fb55=
5e8
> dashboard link: https://syzkaller.appspot.com/bug?extid=3D7edb85bc97be9f3=
50d90
> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=3D16d5817a880=
000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=3D15d68cde88000=
0
>
> If the result looks correct, please mark the issue as fixed by replying w=
ith:
>
> #syz fix: fs/jfs: fix shift exponent db_agl2size negative

It might well be possible. The patch improves the validation of
db_agl2size parameter, which affects the execution some frames up in
the crash stack trace:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/fs/=
jfs/jfs_dmap.c?id=3D6d36c728bc2e2d632f4b0dea00df5532e20dfdab#n729

#syz fix: fs/jfs: fix shift exponent db_agl2size negative

>
> For information about bisection process see: https://goo.gl/tpsmEJ#bisect=
ion
>