From mboxrd@z Thu Jan  1 00:00:00 1970
Date: Mon,  5 Jun 2006 18:34:40 -0700
From: Roman Shaposhnick <rvs@sun.com>
To: Fans of the OS Plan 9 from Bell Labs <9fans@cse.psu.edu>
Subject: Re: [9fans] quick stupid question
Message-ID: <20060606013440.GC27456@submarine>
References: <20060606011008.GA27293@submarine>
	<018062f584bd7cfd4c738d23f969e2fe@collyer.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=koi8-r
Content-Disposition: inline
In-Reply-To: <018062f584bd7cfd4c738d23f969e2fe@collyer.net>
User-Agent: Mutt/1.4.2.1i
Topicbox-Message-UUID: 5acd7754-ead1-11e9-9d60-3106f5b1d025

On Mon, Jun 05, 2006 at 06:24:31PM -0700, geoff@collyer.net wrote:
> It looks like it's initialised in 9/pc/screen.c, for example in
> screensize(), and 9/pc/devvga.c, for example in vgaopen() and the
> routines it calls.  Look for
> 
> 	scr = &vgascreen[0];
> 
> which is a pedantic way of writing
> 
> 	scr = vgascreen;
> 
> and see how scr is used.

  Here's the thing -- it looks like everything assumes that the place
  'scr' ends up pointing to has been already initialized somehow. Every 
  sourceline that I've been able to identify using 
      "grep vgascreen  /sys/src/9/*/*" 
  does exactly what you've mentioned: 'scr = &vgascreen[0];' but the problem 
  is -- scr never gets written to, which means there's not way vgascreen could 
  be initialized. 
  
  Or am I missing something really obvious here ?

Thanks,
Roman.