NetBSD/sys/dev/rasops
ad d8a02e8017 Code cleanup. Add ri_bswap member to 'struct rasops_info' to indicate that
framebuffer endianness differs from CPU (for 15,16,32-bit displays).
1999-04-13 03:02:40 +00:00
..
files.rasops
Makefile
rasops1.c Code cleanup. Add ri_bswap member to 'struct rasops_info' to indicate that 1999-04-13 03:02:40 +00:00
rasops8.c Fixed licensing due to a foobar on my behalf. 1999-04-13 00:40:07 +00:00
rasops15.c Fixed licensing due to a foobar on my behalf. 1999-04-13 00:40:07 +00:00
rasops24.c Code cleanup. Add ri_bswap member to 'struct rasops_info' to indicate that 1999-04-13 03:02:40 +00:00
rasops32.c Fixed licensing due to a foobar on my behalf. 1999-04-13 00:40:07 +00:00
rasops.c Code cleanup. Add ri_bswap member to 'struct rasops_info' to indicate that 1999-04-13 03:02:40 +00:00
rasops.h Code cleanup. Add ri_bswap member to 'struct rasops_info' to indicate that 1999-04-13 03:02:40 +00:00
README

$NetBSD: README,v 1.1 1999/04/13 00:17:57 ad Exp $

This directory contains 'rasops', a set of raster operations intended to
replace the dev/rcons/raster stuff for both wscons and rcons. It yields
significantly improved performance, supports multiple depths and color.

The stamp_mutex used by some of the colordepths is not fully atomic. It's
designed to stop stamp corruption in the event that text is printed at
interrupt time. Even if it fails, >99% of the time text will be white on
black, so it doesn't really matter. Going to splhigh for each character is
a Bad Thing.

Status of the various depths. Feel free to complete/test:

	8	tested, complete
	15/16	untested, complete (endainness issues exist?)
	32	untested, complete (endainness issues exist?)
	1	untested, incomplete for non 8,16 pel fonts
	24	untested, incomplete, assumes RGB order

- Andy Doran <ad@NetBSD.org>, April 1999