>why not just make an interface to the vga bios and add some draw line
>/ draw circle / fill / and similar functions, this would be much smaller
>than a full blown vgalib
The VGA BIOS is not reentrant, so you'd have to disable interrupts while
calling it, which would play havoc with the process granularity. Also, it's
not much help on EGA or CGA systems.
Also, I think a framebuffer device is probably not much help. Remember that we
won't support mmap(), so there wouldn't be a fast way of accessing the
framebuffer (can you imaging drawing graphics with read() and write()?
Ewww...) We could have kernel support for a graphics VC, and a syscall that
returned the address of it to applications. The applications would then write
directly into the framebuffer using, I don't know, the ES segment or something.
--
+- David Given ---------------McQ-+ "`Aplysia californica' is your taxonomic
| Work: [EMAIL PROTECTED] | nomenclature.
| Play: [EMAIL PROTECTED] | A slug, by any other name, is still a slug
+- http://wired.st-and.ac.uk/~dg -+ by nature." --- drushel on a.f.c
