Re: kmemcheck detected possible information leak to userspace?

["Vegard Nossum" <vegard.nossum@gmail.com>]

On Tue, Jul 1, 2008 at 12:25 PM, Evgeniy Polyakov <johnpol@2ka.mipt.ru> wrote:
>> $ addr2line -e vmlinux -i c05325c7 # packet_recvmsg
>> net/packet/af_packet.c:1093
>
> This gives:
>        copied = skb->len;
>        if (copied > len)
>        {
>                copied=len;
>                msg->msg_flags|=MSG_TRUNC;
>        }
>
>        err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
>
> So everything looks ok, but driver could setup skb->len and/or
> skb->data_len to be slightly more than it placed data. Does
> above 'uuuu' bytes are at the end of the skb->data?

In fact I didn't give you the first such message. The first one looks like this:

kmemcheck: Caught 32-bit read from uninitialized memory (c72da052)
uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu
                  ^

(with a similar stacktrace).

The u means uninitialized. So it seems not to be just the end. But I
am not sure where the end of the buffer is anyway.

But there is one thing I forgot: Is this memory initialized by DMA? If
so, the warning is bogus and I will go hide in shame.


Vegard

-- 
"The animistic metaphor of the bug that maliciously sneaked in while
the programmer was not looking is intellectually dishonest as it
disguises that the error is the programmer's own creation."
	-- E. W. Dijkstra, EWD1036