BIOS Modification - DFI 586IPVG, Cyrix MII 433, K6-2/+ 450

Only for programmers and BIOS gurus with technical questions.
KachiWachi
The New Guy
Posts: 1450
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Post by KachiWachi » Tue Jun 11, 2002 9:45 pm

Got my answer to how the multiplier works...

On another note, Jan Steunebrink has added my board to the list of ones he will add K6-2 support for...any idea how this guy should perform with a K6-2+ mobile CPU @2.0V and either 66x6 for 400MHz or 75x6 for 450MHz as compared to where I am at now???

NickS
BIOS Bodhisattva
Posts: 3145
Joined: Fri May 03, 2002 10:34 am
Location: Thames Valley, UK

Post by NickS » Wed Jun 12, 2002 11:45 am

Don't keep us in suspense - how do you do the 6x ?
Tested patched BIOSes. Untested patched BIOSes.
Emails *will* be ignored unless the subject line starts "Wim's BIOS forum"

KachiWachi
The New Guy
Posts: 1450
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Post by KachiWachi » Wed Jun 12, 2002 4:52 pm

6x done internally in the AMD, remember? (2x=6x)

As for the rest, I assumed each jumper was interpreted individually when in fact they are interpreted as a whole. So the most I could ever get was 4x in my particular CPU, anything more just doesn't exist in its Multiplier "truth table".

Excited about this K6-II/III+ patch...hope it pans out. :)

NickS
BIOS Bodhisattva
Posts: 3145
Joined: Fri May 03, 2002 10:34 am
Location: Thames Valley, UK

Post by NickS » Thu Jun 13, 2002 2:32 pm

KachiWachi wrote:6x done internally in the AMD, remember? (2x=6x)
Oh yes, I got confused between that and the Cyrix. Must pay more attention in class.
Last edited by NickS on Wed Jul 24, 2002 5:27 pm, edited 1 time in total.
Tested patched BIOSes. Untested patched BIOSes.
Emails *will* be ignored unless the subject line starts "Wim's BIOS forum"

KachiWachi
The New Guy
Posts: 1450
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Post by KachiWachi » Tue Jul 23, 2002 7:25 pm

Just an update for everyone -

I just recieved patch #3 from Jan Steunebrink. My BIOS can now support the K6-II/III series processors. I am in process of looking for a K6-III+ 450, but I'm getting the hint I should settle for the K6-II+ 450 instead, mostly due to the cost factor. As soon as I get one, I will post whether the patching was successful or not. I will be running things at 66MHzx6 for 400MHz. That should be a safe mode to run in, being that I'm still using older HD's and a SCSI card which might not like the slight overclock to 37.5MHz PCI. When (if) possible, 75MHzx6 would then put me right at 450MHz mark.

Right now Jan thinks my BIOS has the 32 GB bug in it, but isn't sure. I'm going to try to borrow a 7200RPM 40GB Seagate here from work to test that...so further patching may be required... In his words, "The J.3 BIOS still limits you to 32GB IDE drives. There is a chance I can fix it but I haven't found out how yet. ;-) " Rainbow...care to give it a look-see??

Wish me luck!!

Rainbow
The UniFlasher
Posts: 3122
Joined: Wed Mar 20, 2002 4:16 pm
Location: Slovakia
Contact:

Post by Rainbow » Tue Jul 23, 2002 9:04 pm

Just send me the BIOS...
Patched and tested BIOSes are at http://wims.rainbow-software.org
UniFlash - Flash anything anywhere

KachiWachi
The New Guy
Posts: 1450
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Post by KachiWachi » Wed Jul 31, 2002 6:21 pm

Over the weekend I got version J3.2 with a fix for >32GB HD. The patched BIOS file loaded fine...all I have to do is borrow that 40GB from work to check it out. Problem is we just moved to a new facility 2 weeks ago, and they haven't unpacked it yet!!! :cry:

Jan says if the fix works, he can also fix the 64GB bug, so I will be able to put up to a 128GB in the system.

Again, wish me luck!!!

BiosMan
The BIOS Patcher
Posts: 79
Joined: Mon May 20, 2002 9:20 pm
Location: The Netherlands
Contact:

Post by BiosMan » Thu Aug 01, 2002 7:46 pm

Rainbow,

Just a technote: This BIOS has the "old style" code with insufficient space to do the 15 byte patch from Petr. Instead of providing the solution available via AWD64GB -32, I developed a new patch that leaves the original code intact and put the fix in a separate subroutine.
Here are the details:

Original:

Code: Select all

26 66 83 7D 78 00        cmp dword ptr es:[DI+78h], 0
74 1A                    je short loc_1
26 66 83 7D 78 FF        cmp dword ptr es:[DI+78h], 0FFFFFFFFh
74 12                    je short loc_1
New:

Code: Select all

26 66 83 7D 78 00        cmp dword ptr es:[DI+78h], 0
74 1A                    je short loc_1
E8 XXYY                  call Patch_32GB
90 90 90                 nop (3x)
74 12                    je short loc_1

At some free space:
           Patch_32GB    proc near
26 81 7D 7A F0 03        cmp word ptr es:[DI+7Ah], 3F0h
72 02                    jb short loc_2
B2 FF                    mov DL, 0FFh
   loc_2:
26 66 83 7D 78 FF        cmp dword ptr es:[DI+78h], 0FFFFFFFFh
C3                       retn
           Patch_32GB    endp
It's all upto KachiWachi now. I hope it works! :wink:
But I'm affraid I will miss the fun, because I'm going on vacation now! :D See you all in 4 weeks!

BiosMan.
The Unofficial AMD K6-2+ / K6-III+ page
http://web.inter.nl.net/hcc/J.Steunebrink/k6plus.htm

KachiWachi
The New Guy
Posts: 1450
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Post by KachiWachi » Wed Sep 11, 2002 3:19 am

It's been a while, but...

SUCCESS!!

Jan's patch for the 32GB bug worked. I was able to borrow a Seagate Barracuda ATA IV ST340016A, 40GB, 7200RPM Drive (finally!!) from work. Did both an Auto-Detect and an IDE HDD Auto Detection. Drive was detected as...Seagate ST340016A, LBA, Mode-4 (all I'm capable of), 40,022MB. Booted, FDISK'ed, and Formatted to 38,137.16MB (38,147MB), 32,768 bytes/allocation unit.

So I guess the next step is modification to fix the 64GB bug and try again.

Way to go JAN!! :D

On a side note...still working on the K6/+ part of this deal...trying to get what I want for the best price I can. Might have to settle for a K6-2+ instead of the III+... :cry:

KachiWachi
The New Guy
Posts: 1450
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Post by KachiWachi » Wed Oct 02, 2002 2:13 pm

OK All...

I broke down and setteled for the K6-2+...should be here by the weekend.

Full report will be posted ASAP.

KachiWachi
The New Guy
Posts: 1450
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Post by KachiWachi » Fri Oct 04, 2002 4:44 pm

OK...have good news and bad news...

I installed the K6 this morning. Set it for 2V and x2(x6) Multiplier. Booted... The CPU was recognized as a K6-2+ @400MHz...so far so good.

Updates ESCD...OK.

As soon as it reaches Verifying DMA Pool... there it hangs.

Just for fun I stepped it back down to x4 (266MHz) to see it it was a timing problem, since the Cyrix has been running there fine. No dice...

So it's back to the drawing board to check the patch.

Jan...???

NickS
BIOS Bodhisattva
Posts: 3145
Joined: Fri May 03, 2002 10:34 am
Location: Thames Valley, UK

Post by NickS » Fri Oct 04, 2002 5:33 pm

KachiWachi wrote:As soon as it reaches Verifying DMA Pool... there it hangs.
1. I think you've got DMA on the brain. 8O
2. Maybe the DMI pool is too deep.Image
OK, let's be serious. Last time this happened to me I had a conflict between PCI and ISA video cards.
Soyo wrote:This could be due to the following;
1. memory may not stable. - try use one stick of ram at a time.
2. Hardware conflicts - try remove all the unrelated add-on card.
3. Cmos data get corrupted. - Try clear Cmos.
4. Floppy drive is bad or not connected right. - make sure the cable is not loose and the red-stripe of the cable is line up with pin 1 of the Fdd drive.
Does it get past that stage if the HDD is disconnected ?
Tested patched BIOSes. Untested patched BIOSes.
Emails *will* be ignored unless the subject line starts "Wim's BIOS forum"

KachiWachi
The New Guy
Posts: 1450
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Post by KachiWachi » Fri Oct 04, 2002 8:49 pm

Re: DMA...seems so!! :roll:

Re: HDD...Dunno...I will try that...what will that prove? I assume remove the CDROM too...

BiosMan
The BIOS Patcher
Posts: 79
Joined: Mon May 20, 2002 9:20 pm
Location: The Netherlands
Contact:

Post by BiosMan » Mon Oct 07, 2002 8:05 pm

Nick,

Thanks for the tips, but this could very well be a BIOS issue.
FYI KachiWachi is using my patched BIOS for Cyrix M2 and AMD K6+ support. Everything worked oke with the Cyrix, but it hangs with a K6-2+/450. POST appears to be succesful (CPU type and speed are indicated correctly) but hangs at Verifying DMI Pool...
I will report back here if I find the bug in the BIOS. :wink:

Jan.
The Unofficial AMD K6-2+ / K6-III+ page
http://web.inter.nl.net/hcc/J.Steunebrink/k6plus.htm

janpopan
Chip off the ol' block
Posts: 82
Joined: Tue Apr 02, 2002 10:08 pm
Location: Berlin
Contact:

Post by janpopan » Tue Oct 08, 2002 10:51 pm

What does the Bios, the DMI Part, if the CPU is unknown there?
nothing or still hangs?

Sometimes it is also a good choid to disable "SystemBios cachable" ... so for the GA586HX with a fast (>233MHz) processor installed.

Maybe helpfull


Jan :)
for 32 GB+ Bios look here

http://www.jan-hill.com/bios.html

Post Reply