Re: /usr/src/etc/rc.d/moused patch

[YONETANI Tomokazu <qhwt+dragonfly-submit@xxxxxxxxxx>]

On Wed, Apr 28, 2004 at 06:14:47PM +0200, Sascha Wildner wrote:
> I think having syscons return an error (current behavior) is correct. It 
> is /usr/src/etc/rc.d/moused which is buggy. It calls vidcontrol twelve 
> times where once would be enough. It also doesn't stop moused when it is 
> given a "stop" argument. This causes "/usr/etc/rc.d/moused restart" to 
> fail because restart just does a stop (moused is not stopped) and a 
> start (mouse already on -> error message).
> 
> Both changes are enough to stop all of vidcontrol's error messages on my 
> system. They are in my patch which is at the beginning of this thread 
> and which walt is for some reason unable to apply. So he misses the 
> moused_stop routine and still gets an error with "moused restart". I've 
> attached the patch again.

Ahh *blush*, you're right, now I understand the problem...

> Could someone else please try to apply my patch? :) It really contains 
> the correct solution. Why _not_ have syscons report an error if the 
> mouse is already on and the user tries to turn it on?

I'm not someone else:), but your latest patch(posted later in this thread)
seems to work on my system.

Sorry for interrupting the discussion by posting a pointless explanation.
Although I don't like the current behavior of vidcontrol -m, it's nothing
to do with this thread and nothing more than a bikeshed type of argument.