Re: [syzbot] WARNING in internal_create_group

[Greg Kroah-Hartman <gregkh@linuxfoundation.org>]

On Thu, Sep 16, 2021 at 06:55:13AM +0000, Joel Stanley wrote:
> On Wed, 11 Aug 2021 at 21:39, Pavel Skripkin <paskripkin@gmail.com> wrote:
> >
> > On 7/21/21 7:30 AM, Christoph Hellwig wrote:
> > > On Wed, Jul 21, 2021 at 11:37:03AM +0800, Hillf Danton wrote:
> > >> On Tue, 20 Jul 2021 11:53:27 -0700
> > >> >syzbot has found a reproducer for the following issue on:
> > >> >
> > >> >HEAD commit:    8cae8cd89f05 seq_file: disallow extremely large seq buffer..
> > >> >git tree:       upstream
> > >> >console output: https://syzkaller.appspot.com/x/log.txt?x=116f92ec300000
> > >> >kernel config:  https://syzkaller.appspot.com/x/.config?x=7273c75708b55890
> > >> >dashboard link: https://syzkaller.appspot.com/bug?extid=9937dc42271cd87d4b98
> > >> >syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=15fc287c300000
> > >> >C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=178cbf6a300000
> > >
> > > <snip>
> > >
> > >> >WARNING: CPU: 0 PID: 8435 at fs/sysfs/group.c:116 internal_create_group+0x911/0xb20 fs/sysfs/group.c:116
> > >
> > > <snip>
> > >
> > >> The device_add(ddev) in register_disk() may fail but it proceeds to register
> > >> block queue even at the failure ... this falls in the class of known issue
> > >> given the comment line.
> > >>
> > >>  * FIXME: error handling
> > >>  */
> > >> static void __device_add_disk(struct device *parent, struct gendisk *disk,
> > >
> > > Yes, Luis is working on actually fixing this - but it requires changes
> > > to every single block driver.  How does a cap on the seq_buf size
> > > propagate here, though?
> > >
> >
> > Hi!
> >
> > I've looked into this, and, I think, we can add sanity check for
> > first_minor. If user will pass too big index (syzbot's repro passes
> > 1048576) this value will be shifted to part_shift and then truncated to
> > byte in __device_add_disk() and assigned to dev->devt. User may be
> > confused about why he passed 1048576, but sysfs warns about duplicate
> > creation of /dev/block/43:0
> >
> > So, these type of errors can be handled before passing wrong values to
> > sysfs API like this:
> >
> > diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
> > index c38317979f74..600e9bab5d43 100644
> > --- a/drivers/block/nbd.c
> > +++ b/drivers/block/nbd.c
> > @@ -1725,7 +1725,17 @@ static int nbd_dev_add(int index)
> >         refcount_set(&nbd->refs, 1);
> >         INIT_LIST_HEAD(&nbd->list);
> >         disk->major = NBD_MAJOR;
> > +
> > +       /* Too big first_minor can cause duplicate creation of
> > +        * sysfs files/links, since first_minor will be truncated to
> > +        * byte in __device_add_disk().
> > +        */
> >         disk->first_minor = index << part_shift;
> > +       if (disk->first_minor > 0xff) {
> > +               err = -EINVAL;
> > +               goto out_free_idr;
> > +       }
> > +
> >         disk->minors = 1 << part_shift;
> >         disk->fops = &nbd_fops;
> >         disk->private_data = nbd;
> >
> 
> This one got backported to v5.10.65, and causes a warning on boot:
> 
> [    7.114976] ------------[ cut here ]------------
> [    7.116811] WARNING: CPU: 0 PID: 1 at block/blk-mq.c:3045
> blk_mq_release+0x84/0x114
> [    7.117510] Modules linked in:
> [    7.118593] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.10.65 #196
> [    7.118691] Hardware name: Generic DT based system
> [    7.119088] Backtrace:
> [    7.119675] [<8093aff0>] (dump_backtrace) from [<8093b294>]
> (show_stack+0x20/0x24)
> [    7.120052]  r7:00000009 r6:60000153 r5:00000000 r4:80e75938
> [    7.120223] [<8093b274>] (show_stack) from [<80940938>]
> (dump_stack+0xc8/0xe4)
> [    7.120298] [<80940870>] (dump_stack) from [<80123174>] (__warn+0xe8/0x154)
> [    7.120355]  r7:00000009 r6:00000be5 r5:804bc594 r4:80b53c80
> [    7.120415] [<8012308c>] (__warn) from [<8093b9b4>]
> (warn_slowpath_fmt+0x6c/0xe4)
> [    7.120465]  r7:00000009 r6:00000be5 r5:80b53c80 r4:00000000
> [    7.120537] [<8093b94c>] (warn_slowpath_fmt) from [<804bc594>]
> (blk_mq_release+0x84/0x114)
> [    7.120594]  r8:80b53c80 r7:857b7390 r6:00000001 r5:80ea6efd r4:00000000
> [    7.120656] [<804bc510>] (blk_mq_release) from [<804ad65c>]
> (blk_release_queue+0xb8/0x128)
> [    7.120756]  r9:00000001 r8:80eee400 r7:00000000 r6:857b7390
> r5:00000001 r4:857b73d8
> [    7.120836] [<804ad5a4>] (blk_release_queue) from [<8052f87c>]
> (kobject_put+0xc8/0x210)
> [    7.120891]  r7:00000000 r6:00000000 r5:80e751bc r4:857b73d8
> [    7.120948] [<8052f7b4>] (kobject_put) from [<804a9b00>]
> (blk_put_queue+0x1c/0x20)
> [    7.120998]  r7:00000000 r6:857b3800 r5:00000000 r4:857b3860
> [    7.121055] [<804a9ae4>] (blk_put_queue) from [<804c2784>]
> (disk_release+0xb0/0x118)
> [    7.121118] [<804c26d4>] (disk_release) from [<805f717c>]
> (device_release+0x40/0xb4)
> [    7.121168]  r7:00000000 r6:00000000 r5:00000000 r4:857b3860
> [    7.121224] [<805f713c>] (device_release) from [<8052f87c>]
> (kobject_put+0xc8/0x210)
> [    7.121265]  r5:80e81154 r4:857b3860
> [    7.121318] [<8052f7b4>] (kobject_put) from [<804c1c74>] (put_disk+0x24/0x28)
> [    7.121368]  r7:ffffffea r6:00000008 r5:857b3800 r4:857bca00
> [    7.121440] [<804c1c50>] (put_disk) from [<806286f4>]
> (nbd_dev_add+0x214/0x27c)
> [    7.121670] [<806284e0>] (nbd_dev_add) from [<80d22f80>]
> (nbd_init+0xec/0x120)
> [    7.121740]  r10:80ec7000 r9:80c06b34 r8:80d39834 r7:00000000
> r6:80e82aa8 r5:00000009
> [    7.121777]  r4:00000000
> [    7.121842] [<80d22e94>] (nbd_init) from [<80102364>]
> (do_one_initcall+0x50/0x274)
> [    7.121893]  r7:00000000 r6:00000007 r5:8116d180 r4:80d22e94
> [    7.121956] [<80102314>] (do_one_initcall) from [<80d012e8>]
> (kernel_init_freeable+0x1b8/0x240)
> [    7.122005]  r7:80d39854 r6:00000007 r5:8116d180 r4:80d5e788
> [    7.122067] [<80d01130>] (kernel_init_freeable) from [<80947bac>]
> (kernel_init+0x18/0x130)
> [    7.122126]  r10:00000000 r9:00000000 r8:00000000 r7:00000000
> r6:00000000 r5:80947b94
> [    7.122161]  r4:00000000
> [    7.122218] [<80947b94>] (kernel_init) from [<80100168>]
> (ret_from_fork+0x14/0x2c)
> [    7.122413] Exception stack(0x810b7fb0 to 0x810b7ff8)
> [    7.122936] 7fa0:                                     00000000
> 00000000 00000000 00000000
> [    7.123287] 7fc0: 00000000 00000000 00000000 00000000 00000000
> 00000000 00000000 00000000
> [    7.123545] 7fe0: 00000000 00000000 00000000 00000000 00000013 00000000
> [    7.123728]  r5:80947b94 r4:00000000
> [    7.124011] ---[ end trace d69e5842dc8c9352 ]---
> 
> There's been a bit going on in this driver since v5.10, so I assume
> it's missing some dependent changes.

Ugh, ok, I'll just go revert this for now in here, and in 5.4.y to be
safe.

thanks,

greg k-h