Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 19 Apr 2016 17:35:12 +0900
From: Greg KH <greg@...ah.com>
To: Marcus Meissner <meissner@...e.de>,
	Ignat Korchagin <ignat.korchagin@...il.com>
Cc: OSS Security List <oss-security@...ts.openwall.com>,
	security@...nel.org
Subject: Re: CVE Request: Linux kernel: remote buffer overflow in usbip

On Tue, Apr 19, 2016 at 10:06:43AM +0200, Marcus Meissner wrote:
> Hi,
> 
> https://github.com/torvalds/linux/commit/b348d7dddb6c4fbfc810b7a0626e8ec9e29f7cbb
> 
> commit b348d7dddb6c4fbfc810b7a0626e8ec9e29f7cbb
> Author: Ignat Korchagin <ignat.korchagin@...il.com>
> Date:   Thu Mar 17 18:00:29 2016 +0000
> 
>     USB: usbip: fix potential out-of-bounds write
> 
>     Fix potential out-of-bounds write to urb->transfer_buffer
>     usbip handles network communication directly in the kernel. When receiving a
>     packet from its peer, usbip code parses headers according to protocol. As
>     part of this parsing urb->actual_length is filled. Since the input for
>     urb->actual_length comes from the network, it should be treated as untrusted.
>     Any entity controlling the network may put any value in the input and the
>     preallocated urb->transfer_buffer may not be large enough to hold the data.
>     Thus, the malicious entity is able to write arbitrary data to kernel memory.
> 
>     Signed-off-by: Ignat Korchagin <ignat.korchagin@...il.com>
>     Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
> 
> diff --git a/drivers/usb/usbip/usbip_common.c b/drivers/usb/usbip/usbip_common.c
> index facaaf0..e40da77 100644
> --- a/drivers/usb/usbip/usbip_common.c
> +++ b/drivers/usb/usbip/usbip_common.c
> @@ -741,6 +741,17 @@ int usbip_recv_xbuff(struct usbip_device *ud, struct urb *urb)
>         if (!(size > 0))
>                 return 0;
> 
> +       if (size > urb->transfer_buffer_length) {
> +               /* should not happen, probably malicious packet */
> +               if (ud->side == USBIP_STUB) {
> +                       usbip_event_add(ud, SDEV_EVENT_ERROR_TCP);
> +                       return 0;
> +               } else {
> +                       usbip_event_add(ud, VDEV_EVENT_ERROR_TCP);
> +                       return -EPIPE;
> +               }
> +       }
> +
>         ret = usbip_recv(ud->tcp_socket, urb->transfer_buffer, size);
>         if (ret != size) {
>                 dev_err(&urb->dev->dev, "recv xbuf, %d\n", ret);
> 
> Our USB developer confirms:
> https://bugzilla.suse.com/show_bug.cgi?id=975945
> |The vulnerability is true. If an attacker can get a malicious package
> |into the connection the kernel will accept all of the data in that
> |package whether it fits into the buffer or not.
> |You can scribble about 1k into RAM, albeit at an unpredictable location.

I think Ignat already asked for a CVE for this through some other
channel, and was going to announce it in some manner.

Ignat, did you do that?

thanks,

greg k-h

Powered by blists - more mailing lists

Your e-mail address:

Please check out the Open Source Software Security Wiki, which is counterpart to this mailing list.

Powered by Openwall GNU/*/Linux - Powered by OpenVZ