lkml.org 
[lkml]   [2022]   [Mar]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRE: [PATCH] net: ipv6: fix invalid alloclen in __ip6_append_data
Date
From: Tadeusz Struk
> Sent: 08 March 2022 00:02
>
> Syzbot found a kernel bug in the ipv6 stack:
> LINK: https://syzkaller.appspot.com/bug?id=205d6f11d72329ab8d62a610c44c5e7e25415580
>
> The reproducer triggers it by sending an invalid message via sendmmsg() call,
> which triggers skb_over_panic, and crashes the kernel:
>
> skbuff: skb_over_panic: text:ffffffff84647fb4 len:65575 put:65575
> head:ffff888109ff0000 data:ffff888109ff0088 tail:0x100af end:0xfec0
> dev:<NULL>
>
> ------------[ cut here ]------------
> kernel BUG at net/core/skbuff.c:113!
> PREEMPT SMP KASAN
> CPU: 1 PID: 1818 Comm: repro Not tainted 5.17.0-rc7-dirty #9
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1.fc35
> RIP: 0010:skb_panic+0x173/0x175
> RSP: 0018:ffffc900015bf3b8 EFLAGS: 00010282
> RAX: 0000000000000090 RBX: ffff88810e848c80 RCX: 0000000000000000
> RDX: ffff88810fd84300 RSI: ffffffff814fa5ef RDI: fffff520002b7e69
> RBP: ffffc900015bf420 R08: 0000000000000090 R09: 0000000000000000
> R10: ffffffff814f55f4 R11: 203a666675626b73 R12: ffffffff855bff80
> R13: ffffffff84647fb4 R14: 0000000000010027 R15: ffffffff855bf420
> FS: 0000000000c8b3c0(0000) GS:ffff88811b100000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 0000000020000040 CR3: 0000000106b68000 CR4: 0000000000150ea0
> Call Trace:
> <TASK>
> skb_put.cold+0x23/0x23
> __ip6_append_data.isra.0.cold+0x396/0xe3a
> ip6_append_data+0x1e5/0x320
> rawv6_sendmsg.cold+0x1618/0x2ba9
> inet_sendmsg+0x9e/0xe0
> sock_sendmsg+0xd7/0x130
> ____sys_sendmsg+0x381/0x8a0
> ___sys_sendmsg+0x100/0x170
> __sys_sendmmsg+0x26c/0x3b7
> __x64_sys_sendmmsg+0xb2/0xbd
> do_syscall_64+0x35/0xb0
> entry_SYSCALL_64_after_hwframe+0x44/0xae
>
> The reproducer can be found here:
> LINK: https://syzkaller.appspot.com/text?tag=ReproC&x=1648c83fb00000
> This can be fixed by increasing the alloclen in case it is smaller than
> fraglen in __ip6_append_data().
>
>
> Reported-by: syzbot+e223cf47ec8ae183f2a0@syzkaller.appspotmail.com
> Signed-off-by: Tadeusz Struk <tadeusz.struk@linaro.org>
> ---
> net/ipv6/ip6_output.c | 7 +++++++
> 1 file changed, 7 insertions(+)
>
> diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
> index 4788f6b37053..622345af323e 100644
> --- a/net/ipv6/ip6_output.c
> +++ b/net/ipv6/ip6_output.c
> @@ -1629,6 +1629,13 @@ static int __ip6_append_data(struct sock *sk,
> err = -EINVAL;
> goto error;
> }
> + if (unlikely(alloclen < fraglen)) {
> + if (printk_ratelimit())
> + pr_warn("%s: wrong alloclen: %d, fraglen: %d",
> + __func__, alloclen, fraglen);
> + alloclen = fraglen;
> + }
> +

Except that is a valid case, see a few lines higher:

alloclen = min_t(int, fraglen, MAX_HEADER);
pagedlen = fraglen - alloclen;

You need to report the input values that cause the problem later on.

David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)

\
 
 \ /
  Last update: 2022-03-08 03:58    [W:0.065 / U:2.028 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site