From mboxrd@z Thu Jan  1 00:00:00 1970
Message-ID: <c2480cf9849883f19f1de9ea9d02d769@proxima.alt.za>
To: 9fans@cse.psu.edu
Date: Fri, 29 Feb 2008 14:53:01 +0200
From: lucio@proxima.alt.za
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Subject: [9fans] Defacing RIO
Topicbox-Message-UUID: 68ff5f0c-ead3-11e9-9d60-3106f5b1d025

I regularly use $vgasize to select different conditions in
$home/bin/rc/riostart.  It struck me that RIO might be amenable to
accept a fresh geometry when started.  Although this may be
inappropriate in a new "display", it makes sense in a "window",
specially when targetting a new CPU server:

	cpu -h calc -c rio -g 1024x768 -s -i riostart

(not quite the same, but close to

	window -dx 1024 -dy 768 cpu -h calc -c rio -s -i riostart

the only difference seems to be the number of windows when done)

so I went ahead.  True to Ron's predictions on a similar subject, it
turned out to be rather easy, although there are unanswered questions.

I have yet to prepare a patch relative to "sources", but if anyone is
interested, just let me know.

The questions I know about: what happens if we try to change the
geometry (by writing to /dev/wctl) on the "root" RIO?  I presume that
/dev/wctl does not yet exist, so it's easy enough to ignore the
command, silently or otherwise, I can fix the patch to deal with this;
the other is more subtle: I assumed that global variable "screen"
would adjust "automatically" (a "resize" is presumably triggered)
following a geometry adjustment, but despite the fact that it doesn't
(I couldn't detect it or it didn't match my expectations), everything
still seems to work just fine.

Now, I do limit the geometry specification to the form HHHxWWW where
one could be considerably more generous, but I thought I'd walk before
trying to run, that was all I really required.  And, a significant
addition, I do set $vgasize to match the geometry and that may need
improving as well.

++L