From mboxrd@z Thu Jan  1 00:00:00 1970
To: 9fans@cse.psu.edu
Subject: Re: [9fans] Plan9 in VMware?
From: rog@vitanuova.com
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Message-Id: <20010214114414.E9120199D5@mail.cse.psu.edu>
Date: Wed, 14 Feb 2001 12:49:03 +0000
Topicbox-Message-UUID: 63fd2a84-eac9-11e9-9e20-41e7f4b1d025

forsyth wrote:
> >>As far as your other reservations, VESA defines a protected mode interface
> >>which requires only (from memory):
>
> true, but i believe roger found that many cards don't implement that.

to be more accurate, i didn't manage to find a single card that
implemented it, despite trying a fair old few.

the main portion of the VESA stuff i wrote was a load of 8086 code
(using those delicious cpp macros found in boot/pc/x16.h) to set up the
graphics mode, show a test card and query the user that it was ok.

it also filched the default font from the BIOS (hence no need for extra
font code in b.com) and passed info on the card (and framebuffer address,
etc) through to the main kernel.

for cards that didn't implement a linear frame buffer (a dying breed
now, i hope) i managed to find buffer switching code out there on the
net which did the job for many graphics cards.

i did rely on inferno's software mouse cursor though - how difficult
would it be to put one back into plan 9?

i do like the idea of simulating the BIOS code...  that approach could
get rid of a lot of the above dross.  it's kind of baroque though, an
entire x86 emulator just to get around a crappy software interface!

  cheers,
    rog.