[BusyBox] udhcp patches/ endianess

Rainer Weikusat rainer.weikusat at sncag.com
Wed Feb 2 19:56:15 UTC 2005


Eric Lammerts <busybox at lists.lammerts.org> writes:
> On Tue, 1 Feb 2005, Rainer Weikusat wrote:
>> Eh ... first, this is the original algorithm, not the version that has
>> been modified by me. Due to the fact that the sum is
>> endian-independant (cf the relevant RFCs), it's pointless to swap the
>> bytes here.
>

[...]

> But the added config option is still totally unnecessary. The
> necessary define already exists (__BYTE_ORDER).

It exists in some non-standard glibc-header. Admittedly, the
udhcp-code is not portable to non-Linux systems, anyway (OTOH, the
timeout stuff is broken either way, so it probably doesn't make much
of a difference) but this is (for me) reason enough to avoid it. I've
already had gcc-extensions vanishing underneath my feat, so there is
either a document outside the control of the people who happen to
contribute to a certain piece of software at a given time that
requires such-and-such-a-thing, or its presence should be considered a
temporary fluctuation of the time-space-continuum :-). YMMV.

> And why is pad_octet_left() there?? And why does pad_octet_right()
> take a 'pad' argument when it's always zero anyway?

Why not?





More information about the busybox mailing list