1. Forum
  2. Usenet
  3. COMP.SYS.M68K

  • software error with address

    From [email protected]@21:1/5 to All on Sun Mar 27 12:45:10 2016
    hi my oscilloscope running 68k cpu has 4 lower + 4 upper eproms
    right now when booting the scope halts with message ''software error 1111 address 0000ah0 , what address is this related? internal cpu address or rom address?

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Michael Schwingen@21:1/5 to [email protected] on Mon Mar 28 15:59:10 2016
    On 2016-03-27, [email protected] <[email protected]> wrote:
    hi my oscilloscope running 68k cpu has 4 lower + 4 upper eproms

    right now when booting the scope halts with message ''software error 1111 address 0000ah0 , what address is this related? internal cpu address or
    rom address?

    That does not sound like anything that is based on a 68k exception, so it is probably some firmware-internal integrity check - in which case, you will
    have to analyze that firmware to find out what the exact meaning is.

    Assuming this is some kind of EPROM integrity check, you might try to read
    all the EPROMs with reduced supply voltage (in an EPROM reader or rigged-up device, eg. using an arduino) - sometimes, if the bits start to
    deteriorate, you can still read the original values when lowering the supply voltage. You could then re-program EPROMs with the corrected original contents.

    cu
    Michael

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Tom Evans@21:1/5 to [email protected] on Mon Mar 28 23:48:57 2016
    On Monday, March 28, 2016 at 6:45:11 AM UTC+11, [email protected] wrote:
    hi my oscilloscope running 68k cpu has 4 lower + 4 upper eproms
    right now when booting the scope halts with message ''software error 1111 address 0000ah0 , what address is this related? internal cpu address or rom address?

    That isn't "0000ah0", but from the PHOTO IN THE FOLLOWING it is "0000a0h", which makes a lot more sense.

    http://www.eevblog.com/forum/repair/hp-54112d/

    "Bod...", are you "Daxxin" or has someone else had the identical fault with a HP 54110D oscilloscope, and make the identical typo?

    0xA0 is in the vector area in one of the "TRAP #0 D 15 Instruction Vectors".

    Are you sure it is a 68000? The big chip on the board doesn't have any recognisable part numbers to identify it as a 68000.

    Tom

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Michael Schwingen@21:1/5 to Tom Evans on Tue Mar 29 09:18:47 2016
    On 2016-03-29, Tom Evans <[email protected]> wrote:

    Are you sure it is a 68000? The big chip on the board doesn't have any recognisable part numbers to identify it as a 68000.

    The quoted service manual on that page seems to indicate is *is* an 68k -
    the marking on the photo is a HP part number, and there are not many chips
    in that kind of DIL package.

    http://www.cpu-world.com/info/HP-parts.html

    decodes that HP number to be a MC68000L.

    cu
    Michael

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Tom Evans@21:1/5 to Michael Schwingen on Tue Mar 29 18:12:02 2016
    On Tuesday, March 29, 2016 at 8:18:48 PM UTC+11, Michael Schwingen wrote:
    http://www.cpu-world.com/info/HP-parts.html

    decodes that HP number to be a MC68000L.

    Wow. So HP private label CPUs. Weird.

    software error 1111 address 0000a0h

    No idea, but I can make some guesses. The 68k "trap" instruction causes one of 16 "software interrupts". The vector at 0x00A0 would be executed for a "TRAP #8", which has the opcode 0x4e48. That's only one bit different to an 'UNLK D0" instruction (
    unlikely, but possible). It is also close to one of the JSR instructions. So it could be "bit rot" in one of the EPROMs as previously suggested, causing execution of a TRAP instruction. It this is unused in that device then the vector might trigger an "
    unused interrupt" which could be printing that error.

    If you can get another one of those oscilloscopes I'd try swapping the EPROMs as a set. Unless they've got identical code loaded into them you wouldn't be able to swap individual chips.

    Tom

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Michael Schwingen@21:1/5 to Tom Evans on Wed Mar 30 15:35:29 2016
    On 2016-03-30, Tom Evans <[email protected]> wrote:
    On Tuesday, March 29, 2016 at 8:18:48 PM UTC+11, Michael Schwingen wrote:
    http://www.cpu-world.com/info/HP-parts.html

    decodes that HP number to be a MC68000L.

    Wow. So HP private label CPUs. Weird.

    Not only CPUs - look at the other ICs, they had private-labelled parts for nearly everything for some time.

    No idea, but I can make some guesses. The 68k "trap" instruction causes
    one of 16 "software interrupts". The vector at 0x00A0 would be executed
    for a "TRAP #8", which has the opcode 0x4e48. That's only one bit
    different to an 'UNLK D0" instruction (unlikely, but possible). It is
    also close to one of the JSR instructions. So it could be "bit rot" in
    one of the EPROMs as previously suggested, causing execution of a TRAP instruction. It this is unused in that device then the vector might
    trigger an "unused interrupt" which could be printing that error.

    Right. Checking EPROM contents would be my first attempt, too.

    cu
    Michael

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From [email protected]@21:1/5 to All on Sat Apr 9 02:26:39 2016
    Thanks all for the help , yes its for sure an 68k cpu thats is custom label
    net yet solved , what i did is downloaded a working back up of eproms and burned
    it still boot with some steps now i get msg 02C39CH error and msg say ''to try reset the machine or contact HP'' the backup was a litle updated version than mine
    because has eprom also on video card maybe incompatible?

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)