ollie lho <[EMAIL PROTECTED]> writes: > On Thu, 2002-06-27 at 22:18, Ronald G Minnich wrote: > > On 27 Jun 2002, ollie lho wrote: > > > I am also wondering why we don't have something in a more OO fashion. > > > For example, we have different "classes" of objects like CPU, > > > northbridge, southbridge, mainboard and superio, but we did not > > > encapsulate them in anyway (except for superio as we do have struct > > > nsuperio). And we are solving the portability problem with an > > > "intendend namespace clash". Would it be better to to put them in > > > the same way at least as nsuperio, would that be more generic and > > > reusable ?? > > > > exactly. The nsuperio thing is the way we should be going, because: > > 1) boards with multiple northbridges will happen (micron 21PAD supports > > this) > > 2) With the P4 cache-as-ram trick, we could actually support multiple > > motherboards with DIFFERENT northbridge parts, if we go to the > > nsuperio-like architecture.
Assuming you can magically detect which motherboard you are on, and the code bloat is not a problem. Despite proving the cache-as-ram trick works it looks like it isn't one of those things you can be certain will work between cpu revisions. What I would rather see long term is a C or C-like language with a compiler that just uses registers. And I also want to look into what it would take to compress the rest of LinuxBIOS besides the memory setup code. Though maybe I should just become a terror to all of the strings and printfs present, and see if there is any code I can remove. > > Ollie, if you want to start to set something like this up for northbridge > > I'm interested. > > > > Actually, this was part of my design (now dream ??) of the general > purpose bootloader before I crashed my HD. Now I have to rethink > it allover again. Good Luck. For me I've go some ideas, but unless we fork off a development and stable branch, I'm going to be go there as gradually as I can. Probably something like the nsuperio architecture, but looked into the a device tree, like we have for pci. Anyway, ollie what were the issues with grub. Just skimming the code it doesn't look terribly hideous. Maybe the devil is in the details. It looks like a port of grub could be done, without a hideous amount of effort. Grub certainly isn't my favorite bootloader, but having a cheap menu to select between different kernels on the disk would be nice. Eric
