41 lines
1.3 KiB
Text
41 lines
1.3 KiB
Text
|
Be sure to read the svgalib.FAQ man 7 svgalib.faq
|
||
|
|
|
||
|
|
Have to decide what exactly is G320x200x256 (VGA Mode 13h or SVGA
|
||
|
|
packed-pixel). A separate Mode X would be nice too.
|
||
|
|
|
||
|
|
A set of functions to handle page-flipping would be nice. (e.g. a
|
||
|
|
vga_setvideopage() that returns the new video start address)
|
||
|
|
|
||
|
|
Support for more chipsets?
|
||
|
|
|
||
|
|
Clean up keyword parsing code.
|
||
|
|
|
||
|
|
Kernel support for page-flipping.
|
||
|
|
|
||
|
|
automatic console restore on device close.
|
||
|
|
|
||
|
|
really reliable vc switching.
|
||
|
|
|
||
|
|
There are races in the mouse code when the mouse device is closed in the
|
||
|
|
signal handler on a vc switch. Again, probably kernel support is due.
|
||
|
|
|
||
|
|
start marking functions that will become obsolete.
|
||
|
|
|
||
|
|
Move away from hard-coded mode numbers:
|
||
|
|
|
||
|
|
Think what would happen if a program is compiled to call
|
||
|
|
vga_setmode(G1280x1024x256) and we change the mode number to
|
||
|
|
something else... This has happened before. The solution is either
|
||
|
|
to provide a getmodenum(xres, yres, colordepth, layout) function, or
|
||
|
|
a setmode(xres, yres, colordepth, layout) function or vow not to
|
||
|
|
change the definitions of the mode numbers (i.e. new modes can only
|
||
|
|
be added to the end of the list).
|
||
|
|
|
||
|
|
More RamDAC support. The S3 code should have no problem with 868, Trio and
|
||
|
|
9xx chips except that boards that uses those chips usually uses
|
||
|
|
newer/unsupported RamDACs as well.
|
||
|
|
|
||
|
|
Multiple-adaptor support.
|
||
|
|
|
||
|
|
IBM RGB52x >8bpp support.
|