Image Map Image Map
Page 2 of 5 FirstFirst 12345 LastLast
Results 11 to 20 of 42

Thread: PS/2 Model 30 8086 BIOS dump?

  1. #11

    Default

    It seems likely the read of 61X8938 was bad. Check whatever pin goes with data line 0.

    This is of interest since it means 61X8938 probably was not some kind of European release. I am analysing the differences between the files; here is a preview of my work so far.

    Code:
               61X8938 (bit 0 missing)      61X8940 (bit 0 missing)      61X8940 (correct)
    f000:0000: 3631 5838 3833 3820 61X8838. 3631 5838 3834 3020 61X8840. 3631 5838 3934 3020 61X8940.
    f000:0048: 5858 3838 3839 3233 XX888923 5858 3838 3839 3433 XX888943 5858 3838 3939 3433 XX889943
    f000:0050: 3837 2020 2828 4243 87..((BC 3039 2020 2828 4243 09..((BC 3039 2020 2828 4343 09..((CC
    f000:e000: 3631 5838 3833 3820 61X8838. 3631 5838 3834 3020 61X8840. 3631 5838 3934 3020 61X8940.
    f000:e6f8: 0001 b400 0000 0017 ........ 0002 b400 0000 0017 ........ 0002 b400 0000 0017 ........
    f000:f1a8: 40ca 0200 fae8 ac05 @....... 40ca 0200 fa0a e475 @......u 40ca 0200 fb0a e475 @......u
    f000:f1b0: 0ae4 740a cc40 f6c2 ..t..@.. 0acd 40f6 c280 74f1 ..@...t. 0acd 40f6 c280 74f1 ..@...t.
    f000:f1b8: 8074 ee32 e4f8 80fc .t.2.... 32e4 f880 fc08 7405 2.....t. 32e4 f880 fc08 7505 2.....u.
    f000:f1c0: 0875 04e8 7a01 eae1 .u..z... e87e 00eb e480 fc15 .~...... e87e 01eb e480 fc15 .~......
    f000:f1c8: 80fc 1475 04e8 da01 ...u.... 7405 e8dd 00eb dae8 t....... 7505 e8dd 01eb dae8 u.......
    f000:f1d0: ead7 5251 521e 0656 ..RQR..V 8a05 5251 521e 0656 ..RQR..V 8a05 5351 521e 0656 ..SQR..V
    ffff:0000: ea5b e000 f031 322f .[...12/ ea5b e000 f030 322f .[...02/ ea5b e000 f030 322f .[...02/
    ffff:0008: 3032 2e38 3600 fab7 02.86... 3035 2e38 3600 fada 05.86... 3035 2f38 3700 fada 05/87...

  2. #12

    Default

    Here is good dumped of 61X8938 from my BIOS chip.

    https://mega.nz/file/3g90lICK#_oIYrV...cgJlj3LC2DDkM4

    http://www.mediafire.com/file/9mcr5viwqpefdl4/file

    61x8937 is good dumped.

    I know the following rivisions of IBM PS/2 8530, 8525 exists.


    1) 8530 (CPU : 8086)

    09/02/86 <-- First initial MCGA text font
    12/12/86 <-- Minir updated MCGA text font
    02/05/87
    01/31/89


    2) 8530 (CPU : 80286)

    08/25/88
    06/28/89


    3) 8525 (CPU : 8086)


    06/26/87 --> with P/N 00F2092 and 00F2093
    06/26/87 --> with P/N 00F2122 and 00F2123


    4) 8525 (CPU : 80286)

    ??/??/88


    I'm looking for the BIOS of IBM PS/2 8525 286.
    Last edited by fs5500; September 26th, 2020 at 02:40 AM.

  3. #13
    Join Date
    Apr 2014
    Location
    Dordrecht , Netherlands
    Posts
    235

    Default

    Did a new read of the 61X8938_A58470
    A filthy pin it was.

    Seems to be ok now.
    Curious If they are the same as the other upload 61X8938

    The new file
    61X8938_A58470_IBM87nwe26sept2020.zip
    61X8938_NweListing_Goed_web.jpg

  4. #14

    Default

    Quote Originally Posted by MauriceH View Post
    Did a new read of the 61X8938_A58470
    A filthy pin it was.

    Seems to be ok now.
    Curious If they are the same as the other upload 61X8938

    The new file
    61X8938_A58470_IBM87nwe26sept2020.zip
    61X8938_NweListing_Goed_web.jpg
    Has a valid checksum now:

    8530 ROM from 68X1687 and 68X1627: Size: 64kB Start: F000:E05B Date: 09/02/86 8525/8530
    8530 ROM from 61X8938 and 61X8937: Size: 64kB Start: F000:E05B Date: 12/12/86 8525/8530 <- this is the one supplied by MauriceH
    8530 ROM from 61X8940 and 61X8939: Size: 64kB Start: F000:E05B Date: 02/05/87 8525/8530
    8525 ROM from 00F2122 and 00F2123: Size: 64kB Start: F000:E05B Date: 06/26/87 8525 Warning: Invalid checksum C3h (-40h)
    8530 ROM from 33F4498 and 33F4499: Size: 64kB Start: F000:E05B Date: 01/31/89 8525/8530

    I am curious why the checksum for 00F2122+00F2123 is off by 40h. That would seem to indicate a single bit error of bit 6 somewhere.

  5. #15

    Default

    Here is good dumped ROM BIOS of IBM PS/2 8525 with P/N 00F2122 (8086)

    https://mega.nz/file/GotETI4Z#TkaUav...AXGBEJT9E6VMvE

    http://www.mediafire.com/file/j0bh15f4gnbky4a/file

    00F2122.BIN / SHA1:2750755f1566dc2de965c808ac8d6321d026fcbc

    *00F2123 is good dumped.

    *I still wanted ROM BIOS of IBM PS/2 8525 286 (80286)

  6. #16

    Default

    Here is the comparison of 61X8938 and 61X8940:

    Code:
    61X8938+61X8937........................|61X8940+61X8939........................
    00000000: 3631 5838 3933 3820- 61X8938 |00000000: 3631 5838 3934 3020- 61X8940 
    00000048: 5858 3838 3939 3333- XX889933|00000048: 5858 3838 3939 3433- XX889943
    00000050: 3837 2020 2828 4343- 87- ((CC|00000050: 3039 2020 2828 4343- 09- ((CC
    0000e000: 3631 5838 3933 3820- 61X8938 |0000e000: 3631 5838 3934 3020- 61X8940 
    0000e6f8: 0001 b400 0000 0017- ........|0000e6f8: 0002 b400 0000 0017- ........
    0000f1a8: 40ca 0200 fbe8 ac05- @.......|0000f1a8: 40ca 0200 fb0a e475- @......u
    0000f1b0: 0ae4 750a cd40 f6c2- ..u..@..|0000f1b0: 0acd 40f6 c280 74f1- ..@...t.
    0000f1b8: 8074 ee32 e4f8 80fc- .t.2....|0000f1b8: 32e4 f880 fc08 7505- 2.....u.
    0000f1c0: 0875 05e8 7b01 ebe1- .u..{...|0000f1c0: e87e 01eb e480 fc15- .~......
    0000f1c8: 80fc 1575 05e8 da01- ...u....|0000f1c8: 7505 e8dd 01eb dae8- u.......
    0000f1d0: ebd7 5351 521e 0656- ..SQR..V|0000f1d0: 8a05 5351 521e 0656- ..SQR..V
    0000fff0: ea5b e000 f031 322f- .[...12/|0000fff0: ea5b e000 f030 322f- .[...02/
    0000fff8: 3132 2f38 3600 fab7- 12/86...|0000fff8: 3035 2f38 3700 fada- 05/87...
    Everthing looks good.

  7. #17

    Default

    Anyone know if the Model 30 BIOS ever had a listing (or a disassembly) published anywhere?

    I suppose that's a big fat "no" at least from IBM's side, since that was right when they furiously back-tracked on "open architecture". I could have a go myself, but maybe some commented disassembly is already lying around somewhere?
    int10h.org :: :: :: blog

  8. #18

    Default

    Quote Originally Posted by VileR View Post
    Anyone know if the Model 30 BIOS ever had a listing (or a disassembly) published anywhere?

    I suppose that's a big fat "no" at least from IBM's side, since that was right when they furiously back-tracked on "open architecture". I could have a go myself, but maybe some commented disassembly is already lying around somewhere?
    I'm starting work on this at https://github.com/JoshRodd/bios8530

    The PS/2 era did not generally have ROM BIOS listing, but the PC Convertible did. The PC Convertible's BIOS would be the closest recently published thing to the Model 25/30 BIOS.

    My personal agenda is to build a PS/2 Model 25/30 emulator, specifically, an MCGA emulator, probably inside the framework of pcjs.org. My goal of disassembly of the ROM BIOS is to get a better understanding of how the MCGA works, at least in text mode.
    Last edited by MennoJoshua; September 28th, 2020 at 11:10 AM.

  9. #19

    Default

    Here's some more information about the ROM BIOSes that have been published so far:

    Code:
    rombios % ./build_roms.sh
    
       Filenames      ROM    Model                Font vectors           ROM BASIC   Copy-    Part numbers 
      Even    Odd     Date               8x16    8x8 (VGA/MCGA)   8x8      vector   rights    Even     Odd 
    68X1687 68X1627: 09/02/86 8530 8x16: 3960h 8x8V: 4960h 8x8: FA6Eh BASIC: 6000h 1981 1987 68X1645 68X1693  Part number mismatch
    61X8938 61X8937: 12/12/86 8530 8x16: 3A30h 8x8V: 4A30h 8x8: FA6Eh BASIC: 6000h 1981 1987 61X8938 61X8937
    61X8940 61X8939: 02/05/87 8530 8x16: 3A30h 8x8V: 4A30h 8x8: FA6Eh BASIC: 6000h 1981 1987 61X8940 61X8939
    00F2122 00F2123: 06/26/87 8525 8x16: 2F48h 8x8V: 3F48h 8x8: FA6Eh BASIC: 6000h 1981 1987 00F2122 00F2123
    33F4498 33F4499: 01/31/89 8530 8x16: 3A70h 8x8V: 4A70h 8x8: FA6Eh BASIC: 6000h 1981 1989 33F4498 33F4499
    Of note:
    1. 68X1687 and 68X1627 don't match the part numbers actually inside the ROMs.
    2. There is a corrupt version of 00F2122 where the even ROM has bit 2 always set from offset 0 - 2180h (0 - 10C0h in the even ROM itself).
    3. There is a corrupt version of 61X8938 where the even ROM has bit 0 always set in the entire ROM. A date of 13/12/86 means you have this corrupt copy.

    My "splice" tool checks for these specific errors so you can know if you have a bad copy of a ROM. I have more notes about these different ROM versions at https://github.com/JoshRodd/bios8530...c/manifest.txt.
    Last edited by MennoJoshua; September 28th, 2020 at 11:10 AM.

  10. #20

    Default

    Quote Originally Posted by fs5500 View Post
    3) 8525 (CPU : 8086)


    06/26/87 --> with P/N 00F2092 and 00F2093
    06/26/87 --> with P/N 00F2122 and 00F2123
    Does anyone know the differences between these two, or else have a copy of 00F2092/00F2093?

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •