In message <[email protected]>, James Richard Tyrer writes: > > The VGA-to-Ethernet-X11 card could include an X11 client similar to > > xterm(1). This might end up being a lot of code to put into firmware > > but it should be possible. Anyone disagree? The card's VGA BIOS > > routines would have to translate VGA BIOS calls into X11. > > Although I did not consider this, it does look like a good idea. > However, the PC spec limits the video BIOS to 64 KiB of system RAM > address space. So, you couldn't have a lot of code. > > So, this might mean another MCU would be required to run the actual X > client code.
Ouch! 64 KiB isn't much (kids today probably can't write hello world in 64K). Any workarounds for this? I'd hate to have to add a MCU just because of a 64 KiB limitation. I guess the VGAconsole wouldn't *have* to use X11. If we can't strip a X11 client down small enough, maybe Plan-9's windowing system would be smaller? Ok, here is an idea for an ugly kludge that may or may not work. We include another function, say a SCSI controller, that isn't really there. Minimal BIOS that just says no drivers are connected. Leaving most of the address space available for our X11 overflow stuff. But... how can we know where this address space is mapped? (Hey I said it was ugly, and I didn't say I liked it...) I guess before we spend too much time thinking about ugly workarounds we should get an idea of how big our BIOS code will be. Do we have a list of calls we need to provide? _______________________________________________ Open-hardware-ethervideo mailing list [email protected] http://lists.duskglow.com/mailman/listinfo/open-hardware-ethervideo
