On Wed, 2015-03-11 at 14:29 +0300, Alexey Kodanev wrote:
> sysctl has sysctl.net.core.rmem_*/wmem_* parameters which can be
> set to incorrect values. Given that 'struct sk_buff' allocates from
> rcvbuf, incorrectly set buffer length could result to memory
> allocation failures. For example, set them as follows:
>
> # sysctl net.core.rmem_default=64
> net.core.wmem_default = 64
> # sysctl net.core.wmem_default=64
> net.core.wmem_default = 64
> # ping localhost -s 1024 -i 0 > /dev/null
>
Yeah, root users seem to act like monkeys these days ;)
Note the real bug is in unix_stream_sendmsg(), as it assumes
sk_sndbuf > 130 :
/* Keep two messages in the pipe so it schedules better */
size = min_t(int, size, (sk->sk_sndbuf >> 1) - 64);
The 64 here seems to assume skb overhead of linux 0.1 on a hypothetical
16bit host ;)
I guess your fix is fine, but might break applications that were
expecting that sysctl net.core.wmem_default=512 would effectively limit
queue occupancy to one skb (without crashing unix_stream_sendmsg())
As commit eea86af6b1e18d6fa did not trigger bug reports, I think we
should get your patch, and eventually adjust SOCK_MIN_SNDBUF if required
later.
Note that we should also limit max value, because of possible arithmetic
overflows for sk->sk_wmem_alloc.
A max of (1U << 31) might be reasonable ?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
