Re: [syzbot] BUG: unable to handle kernel paging request in cap_capable

[Nikolay Aleksandrov <nikolay@nvidia.com>]

On 04/08/2021 20:28, syzbot wrote:
> Hello,
> 
> syzbot found the following issue on:
> 
> HEAD commit:    b820c114eba7 net: fec: fix MAC internal delay doesn't work
> git tree:       net-next
> console output: https://syzkaller.appspot.com/x/log.txt?x=11fbdd7a300000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=5e6797705e664e3b
> dashboard link: https://syzkaller.appspot.com/bug?extid=79f4a8692e267bdb7227
> compiler:       gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.1
> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=127e4952300000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=16fef2aa300000
> 
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+79f4a8692e267bdb7227@syzkaller.appspotmail.com
> 
> netdevsim netdevsim0 netdevsim1: set [1, 0] type 2 family 0 port 6081 - 0
> netdevsim netdevsim0 netdevsim2: set [1, 0] type 2 family 0 port 6081 - 0
> netdevsim netdevsim0 netdevsim3: set [1, 0] type 2 family 0 port 6081 - 0
> BUG: unable to handle page fault for address: fffff3008f71a93b
> #PF: supervisor read access in kernel mode
> #PF: error_code(0x0000) - not-present page
> PGD 0 P4D 0 
> Oops: 0000 [#1] PREEMPT SMP KASAN
> CPU: 0 PID: 8445 Comm: syz-executor610 Not tainted 5.14.0-rc3-syzkaller #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> RIP: 0010:cap_capable+0xa0/0x280 security/commoncap.c:83
[snip]
> 
> 
> ---
> This report is generated by a bot. It may contain errors.
> See https://goo.gl/tpsmEJ for more information about syzbot.
> syzbot engineers can be reached at syzkaller@googlegroups.com.
> 
> syzbot will keep track of this issue. See:
> https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
> syzbot can test patches for this issue, for details see:
> https://goo.gl/tpsmEJ#testing-patches
> 

+CC Arnd

I think this bug is also from the recent ndo_do_ioctl conversion, it seems
nothing checks if the device is an actual bridge so one could call br_ioctl_call()
for any device. I'll add this to my list, it's a simple fix.

Sorry for the delay, I will send both bridge ioctl fixes tomorrow.

Thanks,
 Nik