* Ronald G Minnich <[EMAIL PROTECTED]> [020306 15:52]: > > This is something that could be done by the hardware vendours themselfes > > if they can be made to see the importance of an instruction set > > independant initialization standard. > > forget it. We can't even get these vendors to tell us what we need to > know. You will never get them to write drivers for your bios.
We're not talking about something new, but a standard that has proved that it suits firmware needs within the last 8 years. And there are already companies that support this standard, a list of supported PCI devices is at http://msproul.rutgers.edu/macintosh/PCIcards.html Of course a free alternative to commercial Open Firmware won't immediately boost companies towards writing compliant drivers, but other free projects showed that it _is_ possible to get hardware vendors attention. PLUS i consider writing drivers for Open Firmware somewhat easier than for x86 lecacy bios, as OF provides a clean and appropriate interface without having to cope with specifics of other hardware involved. X86 lecacy bios code is by no means harder to reengineer than FCode, it's in the end just a case of doing it. To get Open Firmware to become the industry standard it should be, hardware industry has to recognize it's point, which is a hard bunch of technical and political work. Unfortunately the Open Firmware Working Group seems to have ceased existance. At least there were no announcements since '99. Probably it's possible to revive this. > BTW, we are at 15 months and counting for AMD to release materials they > have already told me should never have been NDA'd. And you want these > folks to write drivers too? That's a generic problem at the current point in time. They're all sitting on their NDA stuff until they die out. Just have a look at the current situation in the graphics adapter market. After SGI lost the battle some years ago, nowadays no company can pay the arms race anymore. They have to be presented a viable solution for that part. Whether they write crappy open firmware drivers or crappy x86 drivers, it doesn't matter in the end, just with open firmware they can drop development costs significantly, if done right. > I still think that long-term Linux makes the most sense for a bios. Not > because it is the most efficient, but because it reduces the support load > to zero. I consider this rather mid-term. But as I am doing slow development, I rather shut up until I really have something to show (Of course everyone is invited to join these efforts - http://www.freiburg.linux.de/OpenBIOS ) Ok, please forgive my visionary point of view, but what we are trying to do is not some fatuous dream, but rather a clean design which helps all of the involved parties. Why shouldnt it be possible to get at least some vendors behind that concept? Best regards, Stefan Reinauer -- Ok hex 4666 dup negate do i 4000 dup 2* negate do " *" 0 dup 2dup 1e 0 do 2swap * e >>a 2* 5 pick + -rot - j + dup dup * e >>a rot dup dup * e >>a rot swap 2dup + 10000 > if 3drop 3drop " " 0 dup 2dup leave then loop 2drop 2drop type 268 +loop cr drop 5de +loop
