>I'm working on a project (a packet classification framework) that uses
>the PFIL_HOOKS mechanism.  However, the placement of the PFIL_HOOKS
>filtering points is problematic.
>
>Specifically, the filtering point is located *after* the ip_len and
>ip_off fields are converted to host order in the ip_input() case, and
>*before* they are converted to network byte order in the ip_output()
>case.

	if we need to revisit endian swap in ip_input, i'd like to see
	no endian swap in kernel.  raw ip input should swap for backward
	compatibility reasons, and that's all we need to swap.

itojun