Subject: Re: CVS commit: src/sys/netinet
To: <>
From: David Laight <>
List: source-changes
Date: 11/19/2003 20:25:33
> Note that the implementation of randomized IP ids is still defective,
> and should not be enabled at all (even if configured) without
> very careful deliberation. Caveat emptor.

How many copies of the 'id' routine are lurking in the kernel tree now?
More than one (even if compiled multiple times) is a mistake.

It really ought to be the same code as libc variant as well - no
point having to fix teh same buggy code several times.

Since there is no longer a syssrc, is it valid to assume that
anyone building the kernel will have a full source tree?
(or at least will grab enough bits to make it work?)


David Laight: