Subject: Re: uprintf() limitations?
To: None <tech-kern@netbsd.org>
From: Nick Gimbrone <Nicholas.Gimbrone@Comcast.Net>
List: tech-kern
Date: 03/13/2002 20:23:49
Thanks to all for the help on this (and the copyout()) question.
Turns out I'd shot myself in the foot...

A few months back we hired a consultant to write some code... and I was just
getting to working with that code this week... I swear, this fellow never even
compiled his code (say nothing of actually running it). He had built the ioctl()
routine to use copyout() on the kernel memory address... this being a new area
for me, I figured he must know what he was doing (nope! I should have known
better... ;-)... by the time I figured out yesterday that this code was "all
wrong" I was tired and... I ended up compounding the problem by switching from
copyout() to memcpy() (the right model), but forgetting to reverse the
source/dest arguments... which then had me questioning my sanity... sigh... it's
been a long week (already ;-).

Sorry for wasting the bandwidth...
but I must say, I am havin a wonderful time working w/ this kernel
(it is a piece of beauty ;-)

Peace -njg

p.s. A good night's sleep last night and it is all working today... ;-)