Re: [PATCH net] ppp: limit MRU to 64K

[Eric Dumazet <edumazet@google.com>]

On Mon, Nov 13, 2023 at 4:17 AM Willem de Bruijn
<willemdebruijn.kernel@gmail.com> wrote:
>
> From: Willem de Bruijn <willemb@google.com>
>
> ppp_sync_ioctl allows setting device MRU, but does not sanity check
> this input.
>
> Limit to a sane upper bound of 64KB.
>
> No implementation I could find generates larger than 64KB frames.
> RFC 2823 mentions an upper bound of PPP over SDL of 64KB based on the
> 16-bit length field. Other protocols will be smaller, such as PPPoE
> (9KB jumbo frame) and PPPoA (18190 maximum CPCS-SDU size, RFC 2364).
> PPTP and L2TP encapsulate in IP.
>
> Syzbot managed to trigger alloc warning in __alloc_pages:
>
>         if (WARN_ON_ONCE_GFP(order > MAX_ORDER, gfp))
>
>     WARNING: CPU: 1 PID: 37 at mm/page_alloc.c:4544 __alloc_pages+0x3ab/0x4a0 mm/page_alloc.c:4544
>
>     __alloc_skb+0x12b/0x330 net/core/skbuff.c:651
>     __netdev_alloc_skb+0x72/0x3f0 net/core/skbuff.c:715
>     netdev_alloc_skb include/linux/skbuff.h:3225 [inline]
>     dev_alloc_skb include/linux/skbuff.h:3238 [inline]
>     ppp_sync_input drivers/net/ppp/ppp_synctty.c:669 [inline]
>     ppp_sync_receive+0xff/0x680 drivers/net/ppp/ppp_synctty.c:334
>     tty_ldisc_receive_buf+0x14c/0x180 drivers/tty/tty_buffer.c:390
>     tty_port_default_receive_buf+0x70/0xb0 drivers/tty/tty_port.c:37
>     receive_buf drivers/tty/tty_buffer.c:444 [inline]
>     flush_to_ldisc+0x261/0x780 drivers/tty/tty_buffer.c:494
>     process_one_work+0x884/0x15c0 kernel/workqueue.c:2630
>
> With call
>
>     ioctl$PPPIOCSMRU1(r1, 0x40047452, &(0x7f0000000100)=0x5e6417a8)
>
> Similar code exists in other drivers that implement ppp_channel_ops
> ioctl PPPIOCSMRU. Those might also be in scope. Notably excluded from
> this are pppol2tp_ioctl and pppoe_ioctl.
>
> This code goes back to the start of git history.
>
> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
> Reported-by: syzbot+6177e1f90d92583bcc58@syzkaller.appspotmail.com
> Signed-off-by: Willem de Bruijn <willemb@google.com>
> ---
>  drivers/net/ppp/ppp_synctty.c | 4 ++++
>  1 file changed, 4 insertions(+)
>
> diff --git a/drivers/net/ppp/ppp_synctty.c b/drivers/net/ppp/ppp_synctty.c
> index ea261a628786..52d05ce4a281 100644
> --- a/drivers/net/ppp/ppp_synctty.c
> +++ b/drivers/net/ppp/ppp_synctty.c
> @@ -453,6 +453,10 @@ ppp_sync_ioctl(struct ppp_channel *chan, unsigned int cmd, unsigned long arg)
>         case PPPIOCSMRU:
>                 if (get_user(val, (int __user *) argp))
>                         break;
> +               if (val > U16_MAX) {
> +                       err = -EINVAL;
> +                       break;
> +               }
>                 if (val < PPP_MRU)
>                         val = PPP_MRU;
>                 ap->mru = val;
> --
> 2.43.0.rc0.421.g78406f8d94-goog
>

Reviewed-by: Eric Dumazet <edumazet@google.com>

It is unclear why ppp_sync_input() does not allocate an skb based on
@count argument though...

diff --git a/drivers/net/ppp/ppp_synctty.c b/drivers/net/ppp/ppp_synctty.c
index ebcdffdf4f0e0193635d2b479e8a9f7a32703509..79cb7916ff03c9b311109372e3d3e434e3669fa6
100644
--- a/drivers/net/ppp/ppp_synctty.c
+++ b/drivers/net/ppp/ppp_synctty.c
@@ -659,14 +659,14 @@ ppp_sync_input(struct syncppp *ap, const u8
*buf, const u8 *flags, int count)
        struct sk_buff *skb;
        unsigned char *p;

-       if (count == 0)
+       if (count <= 0)
                return;

        if (ap->flags & SC_LOG_INPKT)
                ppp_print_buffer ("receive buffer", buf, count);

        /* stuff the chars in the skb */
-       skb = dev_alloc_skb(ap->mru + PPP_HDRLEN + 2);
+       skb = dev_alloc_skb(count + PPP_HDRLEN + 2);
        if (!skb) {
                printk(KERN_ERR "PPPsync: no memory (input pkt)\n");
                goto err;