Possible Uniflash 1.47 Bug?

[miki02131]

Can anyone explain what's going on here. First let me explain: I have an award bios chip SST39F020A from an HP P2B-VT board I was trying to hotflash using an Abit BE6-II system.

I got the original bios file from HP and using Uniflash 1.47, I force-flash the chip with the bootblock included. After flashing, still the P2B-VT system wouldn't come alive.

I take a closer look at the original file and the backup image using CBROM and they seem to have almost nothing in common. Why should the original file and its backup image be different? Could this be a Uniflash bug? Could the failure be due to the peculiarity of the Abit BE6-II since Uniflash relies on the DMI pool to do its thing?

Here is what CBROM return for the components of the two files: BiOS.ROM and BACKUP.BIN. Note that backup.bin should be an image copy of bios.rom and therefore should have had the same structure and components. Here we go, any thought would be appreciated:


******** bios.rom BIOS component ********

No. Item-Name Original-Size Compressed-Size Original-File-Name
================================================================================
0. System BIOS 20000h(128.00K) 11A25h(70.54K) stn.bin
1. CPU micro code 08846h(34.07K) 04642h(17.56K) cpucode.exe
2. Other(6000:0000) 0C500h(49.25K) 05AF9h(22.74K) AWARDEXT.ROM
3. EPA LOGO 00642h(1.56K) 002A1h(0.66K) awardepa.epa
4. ACPI table 01E11h(7.52K) 00B05h(2.75K) ACPITBL.BIN
5. VRS ROM 01F67h(7.85K) 012BCh(4.68K) cav_shdw.bin
6. Other(7000:0000) 09000h(36.00K) 0533Ch(20.81K) tnt24.bin
7. OEM1 CODE 0C000h(48.00K) 07525h(29.29K) r128011.bin
8. LOGO BitMap 2583Ch(150.06K) 00A1Bh(2.53K) hpdlogo.bmp

Total compress code space = 31A25h(198.54K)
Total compressed code size = 2AE3Eh(171.56K)
Remain compress code space = 06BE7h(26.98K)

** Micro Code Information **
Update ID CPUID | Update ID CPUID | Update ID CPUID | Update ID CPUID
------------------+--------------------+--------------------+-------------------
00000000 00000000 | 0000 0000


------------------+--------------------+--------------------+-------------------
00000000 00000000 | 0000 0000

______________________________________________________________________________________________

******** backup.bin BIOS component ********

No. Item-Name Original-Size Compressed-Size Original-File-Name
================================================================================
0. System BIOS 20000h(128.00K) 11677h(69.62K) stn.bin
1. CPU micro code 0C062h(48.10K) 060F7h(24.24K) cpucode.exe
2. Other(6000:0000) 06052h(24.08K) 03701h(13.75K) AWARDEXT.ROM
3. EPA LOGO 00642h(1.56K) 002A1h(0.66K) awardepa.epa
4. ACPI table 01D8Ah(7.38K) 00B5Fh(2.84K) ACPITBL.BIN
5. VRS ROM 01F65h(7.85K) 012BBh(4.68K) cav_shdw.bin
6. Other(0800:0000) 08000h(32.00K) 04FCAh(19.95K) pci32.rom

Total compress code space = 31677h(197.62K)
Total compressed code size = 21EF4h(135.74K)
Remain compress code space = 0F783h(61.88K)

** Micro Code Information **
Update ID CPUID | Update ID CPUID | Update ID CPUID | Update ID CPUID
------------------+--------------------+--------------------+-------------------
00000000 00000000 | 0000 0000

[cp]

I got the original bios file from HP and using Uniflash 1.47, I force-flash the chip with the bootblock included. After flashing, still the P2B-VT system wouldn't come alive.

which doesn't mean the hotflash went wrong. there are some more possible failures that would cause a mainboard not to start.

I take a closer look at the original file and the backup image using CBROM and they seem to have almost nothing in common. Why should the original file and its backup image be different?

they do have some things in common. first of all they are valid award bios files because cbrom can display their contents. second they are obviously ASUS bios files because their system rom is called stn.bin and their microcode update file is called cpucode.exe (nothing to execute in there though). the bios.rom seems to be a HP bios because it obviously contains some HP logo bmp which is displayed on startup. however the files differ in some modules but that could be due to their versions or the chip didn't contain the correct bios for the mainboard in the first place (wrong flash, exchanged chip). things like that could happen easily if mainboards are bought/obtained in a non-working condition from an unreliable source.

without the files it's impossible to say if those two files were made for the same motherboard.

check the mainboard for other defects. caps and FETs of the cpu power supply are a very likely source of failure. get hardware (cpu, ram, powersupply) that works fine with other mainboard(s). get a pci post card.

[edwin]

indeed it looks like someone has tried to flash some Asus retail bios on the board and that's why it wasn't working in the first place.

[miki02131]

indeed it looks like someone has tried to flash some Asus retail bios on the board and that's why it wasn't working in the first place.

I have to admit you've nailed it right in the head. Having said that, is there a way out? How to recover from a bad flash resulting from the use of "some ASUS retail bios"? Or is it a total loss?

Thanks,

Miki.

[edwin]

well using the HP bios I pointed you at should have fixed the problem. I have one other version of the Kestral bios here, maybe that is working better, unless the board itself is physically dead in some odd way. You have tried all sorts of slower cpus and memory, right?

[miki02131]

Evasive,

I've been using exactly the one you pointed to me. I know it's the right bios. Because when the system was functional, I've used to update from version 1.0 to version 1.16. The problem now is that it doesn't seem to work with hotflashing.

I think the best bet would be to find another HP system with the same board and try to hotflash from there. In the meantime, can you send me a copy of that other kestral file that you have. I will give it a try and see if it makes any difference. If not, then I will have to wait until I can find some old HP system to see if I can fix the problem.

Note the problem system is an old HP pavilion that I use for experimental purposes. So there is not really any emergency. But it's nice to be able to fix the problem. Also it gives the opportunity to test the capabilities of Uniflash which I've never used before that event.

Please, email a copy of the other kestral file or point me where I can download it.

Thanks,

Miki.