Subject: Re: kpsignal2 panic
To: Matt Thomas <matt@3am-software.com>
From: Jonathan Stone <jonathan@DSG.Stanford.EDU>
List: tech-kern
Date: 10/06/2003 12:40:10
>For the most part, I dislike KASSERT().  It tells you it didn't like
>something but you have no easy way to figure out what.  A printf or
>panic is very much superior in making the diagnosis.

"What Matt said".

while working on netipsec/ I noticed FreeBSD has a two-argument
KASSERT(): first is the boolean expression and second is a
paren-enclosed printf() argument list.  Much nicer.  maybe we could
copy that as KASSERT2() (to save a pass through the whole kernel tree
adding a desecriptive argument)

Or should we just do that anyway?