• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

BIOS mod for Winbook XL to see K6-2+

RamonetB

Member
Hey All!

I'm extremely new to BIOS modding.
However, I'm up for the challenge. The reason being that as far as I can tell the Latest BIOS for the Winbook XL (dated Feb 2000) DOes not have support for the AMD K6-2+.

This is very upsetting since It's such a wonderful chip.

Currently many of us at the Winbook forums (where they won't allow us to talk about such things) are very interested in seeing this chip supported. Right now I'm interested in only very basic support. That is to say, I'd love it if we could, for the moment, get a simple mod out that would recognize the K6-2+ as a castrated K6-3. (Half the on die cache). The power saving features we can do without for now. At some later point, with more experience, we can try implementing those.

Having no idea what goes into writing such code, I'm oblivious to the difficulty of this task. Any help here would be greatly appreciated!

Thanks all!

-kirk

P.S. Site or information where I could learn what goes into making a BIOS and how it works would be appreciated as well!


P.P.S. Currently working on trying to get a full copy of the BIOS. I'm told by some that I could use the AMI flash utility and that, instead of putting in the name of the bios you want to flash to, you leave the flash field blank, type in a name for the backup file, and the utility will not flash your BIOS but will create a backup of the one you currently have loaded.

Be this true?
 
Hi! Good to see you found this place and have signed on. This is the best collection of technical forums I have found. I'm sure that you will find someone who can answer those questions. It is not something I have attempted. But, I have a hunch your hypothesis is valid, and that it will work. My only concern would be heat. I had an XL, but with the Intel mobile 233 MHz CPU, and heat was not a problem. At that time, just about all AMD laptops has heat problems because AMD chips ran hot. They had no mobile version. But, I see it as a low risk, low exposure experiment since the current value of an XL is less than $200. 🙂

Anyway, welcome to Anandtech!
 
Heat generally is an issue for the K6 line of processors.
The nice thing about the K6-2+ is that it's a .18micron process running at 2.0 volts!
A 450MHz piece, when the winbook will only run at 400MHz, can be underclocked and thus decrease the voltage to 1.9! (Absolute effective according to tech docs is 1.8)

The wonderful thing about this is that the current chip, a Pentium 233 clocked at 266, only runs at 1.8 volts! So the K6-2+ will only run .1 volt higher, or at the most .2. I know this isn't all there is to heat dissipation. But the 2.0V K6-2+ or even k6-3+ is a WONDERFUL alternative to the standard K6-2 line (especially when you need some cache to utilize 128MB of RAM since the Winbook XL only sees a chachable size of 64MB!) and K6-3 line because of the heat issue!

I"ve been doing some research and have discovered a few things.

According to BIOScentral.com, once a CPU is initialized (that is to say, it receives a powerup signal when you turn the machine on and it cleans and resets it's chaches and registers) the very first thing is does is jump to a memory area FFFF_0000h. This is only 16bytes and is too small to hold the BIOS. It's basically a pointer or a Jump (JMP) command telling the processor where to go next to execute BIOS code.

According to the K6-2+ technical documents, the K6-2+ will jump to memory address FFFF_FFF0h to start executing code.

So which one is right? Or does it matter?
My first thought was that this chip is jumping to some other memory space and there's no pointer of JMP command there to tell it were to go. And that's why the system acts as though there is no CHip installed. I figured it's jumping here to receive some special intruction on what to do with its PwerNow technology. Would have been nice if this were the case.

SO I got the Data Sheets for all the K6 line. The K6-2 and K6-3 all jump to the same memory area. So that's not the issue! The good thing about this is that it brings the K6-2+ and the K6-3 closer than ever! Even the caches operate the same. (4 way associative).

The research continues.

To find the K6-2+ and K6-3+ (or any mobile K6) data sheet, go here:

http://www.amd.com/us-en/Processors/TechnicalResources/0,,30_182_739_1102^1117,00.html

Standard K6 data sheets are here:

http://www.amd.com/us-en/Processors/TechnicalResources/0,,30_182_739_1102,00.html


Question:
How exactly does one program a BIOS, from what I've been reading it responds like a Microcontroller (similar to the HC11). It sounds like assembly code... though I'd need a compiler.

-kirk
 
Some additional things:

The K6-2 DOES work in the laptop. I have personally done this myself with a K6-2 450 and got it running at 413MHz. (Don't ask, I don't know why the odd increment).


Also, I'm trying to backup my current BIOS. The BIOS updates given out by winbook is not the entire BIOS. It's a module update that the installer patches into the current BIOS. I want the WHOLE bios in some file I can edit.

How do I do this?

BIOSmods.com tells me of using a flash utility, leave the flashing rom space empty and fill in a name for the backup file. (Add some switches in the command line as well) and run the program. It doesn't flash your BIOS (since no file was selected) but DOES backup the one currently on the machine.


Will the AMI flash utility do the same? This is an AMI BIOS! Any help would be greatly appreciated!

-kirk
 
Ah yes, that flash example was done using an AWARD flash utility on an AWARD BIOS.

THe K6-3 is also reported to work. This is from the same sources that told me the K6-2 worked. Now that I have verified the K6-2 works, I have no reason to not believe the K6-3 does as well.

-kirk😱
 
Have you checked with Twinhead for a possible BIOS? They built the XL.
 
good idea, so I checked the Twinhead site, as far as I can tell the P79TX is the XL233...if so the BIOSs aren't the same... The P79TX uses Phoenix BIOS, I don't know what the particulars are, but it's not AMI. If the P79TX isn't the XL233 then which one is? Twinhead is very stingy with specs. I was trying to figure whether any of the BIOSs at Twinhead implemented support for the AMD K6 chip, but couldn't find anything.

I dug up a Hex Editor called, appropriately enough, HexEdit 2.09. I downloaded it from Simtel.net (or is it .com?). Eitherway you can view the hex data in the BIOS ROMS and the flash utilities with it...it's pretty easy to view the data, I have no idea what to do with what you can see though. ha!
 
😕

I was over at the Twinhead site myself.
I encoutered the Slimnote XL notebook which I believe is the equivalent XL. Or at elast very similar.
Even the Bios Revisiosn are the same, going back to 1.10. THough they don't have a 1.17 (which I still insist is a patch).

Specs tend to be the same except that the Slimnote has 4MB Video RAm.
And you're right, they are stingy with the specs.

Mostly I've been trying to figure out how to back up my old BIOS to have a complete latest version.

I DID come across this though:

http://web.inter.nl.net/hcc/J.Steunebrink/k6plus.htm

The individual there seems to have been making patches for other BIOSes for suppot of the chip. I've contacted him, yet have gotten no reply yet.

The search continues. Mostly right now I need to figure out how to code and manipulate the BIOS. No one is talking. From reading data sheets and devlopers guides, it all sounds like assembly language.

?????????

-kirk
 
P.S.Where's the "tired" emotion?

Try this one . . . 😛 (pant, pant!)
 


<<
I was over at the Twinhead site myself.
I encoutered the Slimnote XL notebook which I believe is the equivalent XL. Or at elast very similar.
Even the Bios Revisiosn are the same, going back to 1.10. THough they don't have a 1.17 (which I still insist is a patch).

Specs tend to be the same except that the Slimnote has 4MB Video RAm.
And you're right, they are stingy with the specs.
>>



The Slimnote XL is the P83. Judging by the fact that it has 4mb of video ram, and when you go to the BIOS update page it uses the "phlash.exe" utility to flash the Phoenix BIOS I would say that we aren't getting anywhere by digging at the Twinhead site.

Also the Twinhead site specifies the P79 as having 200/233 Pentium Processor. Maybe the P83 has the 266, or greater?

I'm gonna check out your links and see what I can do about getting a backup of the BIOS..I'm a little hesitant to fiddle around with a working laptop, but I could probably fix it somehow if things were to go wrong.
 
😛 (pant pant)

Actually, about backing up the old BIOS file, the AMI flash utility (Version 6.31 which I have if you need it, but it also comes with the Winbook BIOS download as... pf85.com I think.. edit the F.bat file. You'll see the command string) has a saving option.

Pf85 /? will list the commands. It's the "F" switch. I've tried it a few times without success. ALso, I don't know if it'll do anything if you don't put in a BIOS file to flash. So, considering you have an early BIOS, you may want to avoid this path.

The latest AMI flash utility, 8.whatever, is a very nice utility but doesn't have support for the laptop. It's too old and asks you to manually load the modules with the information. From version 6.31 I have determined some of this semi-obvious material:

Brand: AMD (amd flash technology!)
Flash Type: 29F002T

Bed time. I just "rebirthed" 2 systems by isntalling XP. No more comps!!! AHHHHHHHHHH!!!

-kirk😕
 
I got your message! (thevillageinn)
Fabulous!
I will do likewise shortly.

Sorry I didn't reply via message, but IE 6 is being a pain again and I'm rapidly considering going over to Netscape 6.2.

ANYWAY,

Some interesting things I've encountered.
After reading the bios mod page by martin, an opcode upgrade I feel is something to attack. It won't implement the powernow features but it *should* let the chip boot and run. THe trick is, I don't know the cpuidcode.

BUT, I did find this:

http://www.sandpile.org/impl/k6.htm

and more generally,

http://www.sandpile.org

A FABULOUS place about Microprocessor design and implementation.
On the first link, the K6 page, there is a listing of cpuid codes for the current K6 line! But he doesn't have the K6-2+ listed! Half the job is done, thanks to Martin as well as Intel for posting their Cpuid codes since we know what it ISN'T. If I could find an award BIOS that supports the K6-2+ and is SOCKET 7 SPECIFIC, then I could EASILY determine which cpuid it is since the rest Are known.

BTW, I haven't been able to find anything similar on AMD's site, but then both Intel and AMD have terribily structured sites I wouldn't have found it on Intel either.

Also, at AMD's site you can get developer tech docs and BIOS design application notes. In those there's spoken of certain BIOS misrepresenting or misreporting the string of the Cpu as K6-3. It in turns gives the memory address locations. ( 3 searies in the 8XXX area). The trouble with it is that the docs speak of cpuid utilities and never makes clear weather these utilities are used at boot up to help the BIOS or whether they are used at the start of a program to identify the type of chip.

At the moment, I'm after the CPUID code.

-kirk
 
And wouldn't you know it?

There's a description of CPUID on the very same sandpile site.

The pieces come together and I'm certain I have a few pieces of the puzzle I'm not putting together right.

-kirk
 
Q. ....AMD or any non-Intel processors?
A. This driver is designed for Intel IA32 microprocessors only, it will not work with AMD or any other non-Intel processors as they don't support microcode updates or they support it in a manner different from Intel's specs.



Ah well, back to the drawing board.

 
Thanks to Dan for telling me about HACKMAN. A nice little manager. While I can't compile with it, I can decompile. ANd that's just what I've done.

I have the decompiled BIOS code!!!

I've also identified the CPU identification routines. Now I have to start figuring out how to tell it that a K6-2+ is a K6-3..... then compile it, which I understand you can do at the DOS prompt using Debug. I also need to learn x86 assembly. FOrtunate thing is, I've got some friends who tell me they'd be very interested in doing likewise.

-kirk
 
Back
Top