Re: KASAN: use-after-free Read in rtnetlink_put_metrics

From: Sabrina Dubroca <sd@queasysnail.net>
To: Cong Wang <xiyou.wangcong@gmail.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>,
	syzbot+41f9c04b50ef70c66947@syzkaller.appspotmail.com,
	christian.brauner@ubuntu.com, David Miller <davem@davemloft.net>,
	David Ahern <dsahern@gmail.com>, Florian Westphal <fw@strlen.de>,
	Jiri Benc <jbenc@redhat.com>, Kirill Tkhai <ktkhai@virtuozzo.com>,
	LKML <linux-kernel@vger.kernel.org>,
	lucien xin <lucien.xin@gmail.com>,
	Linux Kernel Network Developers <netdev@vger.kernel.org>,
	syzkaller-bugs@googlegroups.com
Subject: Re: KASAN: use-after-free Read in rtnetlink_put_metrics
Date: Wed, 1 Aug 2018 10:15:37 +0200	[thread overview]
Message-ID: <20180801081537.GA31982@bistromath.localdomain> (raw)
In-Reply-To: <CAM_iQpURKOMAaPfs9v7FzhqqQPzj71sQD2LS0eU8cuqTML-YoA@mail.gmail.com>

2018-07-31, 16:03:13 -0700, Cong Wang wrote:
> On Tue, Jul 31, 2018 at 6:41 AM Sabrina Dubroca <sd@queasysnail.net> wrote:
> >
> > 2018-07-31, 05:41:56 -0700, Eric Dumazet wrote:
> > >
> > >
> > > On 07/31/2018 05:31 AM, syzbot wrote:
> > > > Hello,
> > > >
> > > > syzbot found the following crash on:
> > > >
> > > > HEAD commit:    61f4b23769f0 netlink: Don't shift with UB on nlk->ngroups
> > > > git tree:       net
> > > > console output: https://syzkaller.appspot.com/x/log.txt?x=14a9de58400000
> > > > kernel config:  https://syzkaller.appspot.com/x/.config?x=ffb4428fdc82f93b
> > > > dashboard link: https://syzkaller.appspot.com/bug?extid=41f9c04b50ef70c66947
> > > > compiler:       gcc (GCC) 8.0.1 20180413 (experimental)
> > > >
> > > > Unfortunately, I don't have any reproducer for this crash yet.
> > [...]
> >
> > > Probably also caused by :
> > >
> > >
> > > commit df18b50448fab1dff093731dfd0e25e77e1afcd1
> > > Author: Sabrina Dubroca <sd@queasysnail.net>
> > > Date:   Mon Jul 30 16:23:10 2018 +0200
> > >
> > >     net/ipv6: fix metrics leak
> >
> > Yeah, I'm looking into both those reports :/
> 
> Looks like this commit is completely unnecessary,
> fib6_drop_pcpu_from() calls fib6_info_release()
> which calls fib6_info_destroy_rcu(), so this metrics
> will be released twice...

kmemleak disagrees:

unreferenced object 0xffff88006b605080 (size 96):
  comm "ip", pid 433, jiffies 4294889793 (age 74.844s)
  hex dump (first 32 bytes):
    00 00 00 00 f4 01 00 00 00 00 00 00 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<000000002650e4e2>] ip6_route_info_create+0x770/0x4050
    [<000000000a8d4c52>] ip6_route_add+0x18/0x90
    [<00000000474d669c>] inet6_rtm_newroute+0xeb/0x100
    [<0000000019fb732d>] rtnetlink_rcv_msg+0x3b5/0xb40
    [<000000006f891e19>] netlink_rcv_skb+0x137/0x380
    [<0000000070451985>] netlink_unicast+0x47f/0x6e0
    [<000000004487d656>] netlink_sendmsg+0x7a7/0x10c0
    [<0000000089fdf5ae>] sock_sendmsg+0xac/0x160
    [<00000000aae19c54>] ___sys_sendmsg+0x6e0/0xbb0
    [<00000000a3906352>] __sys_sendmsg+0xdc/0x230
    [<00000000c7c8548a>] do_syscall_64+0x15d/0x740
    [<000000007dfdad73>] entry_SYSCALL_64_after_hwframe+0x49/0xbe
    [<000000003adb705a>] 0xffffffffffffffff


ip netns add peerA
ip link add eth0 netns peerA type veth peer name ethA
ip -net peerA link set eth0 up
ip -net peerA link set lo up
ip -net peerA a a fec0:A::1/64 dev eth0
ip -net peerA route add fec0:B::/64 via fec0:A:: mtu 1300

ip netns exec peerA nc fec0:B::1 1234

ip -net peerA route del fec0:B::/64
ip netns del peerA

-- 
Sabrina