kernel panic: page fault

[Robert Watson]

On Wed, 5 Apr 2006, Kazuaki Oda wrote:

> I've read the source code:
>
>
> 	/*
> 	 * XXXRW: Time wait state for inpcb has been recycled, but inpcb is
> 	 * still present.  This is undesirable, but temporarily necessary
> 	 * until we work out how to handle inpcb's who's timewait state has
> 	 * been removed.
> 	 */
> 	if (tw == NULL)
> 		goto drop;
>
> <snip>
>
> drop:
> 	INP_UNLOCK(tw->tw_inpcb);
> 	m_freem(m);
> 	return (0);
>
>
> Hmm, it seems to be null pointer dereference because tw is NULL...

Indeed.  I've inserted a NULL check here.  Thanks again!

Robert N M Watson