Next Cube address register

Does anyone know the default values for the address register for the 040, mine is running rom monitor 2.4 v65

Thanks

I'm not sure which address register you think of - the 68040 has 8 address registers (A0-A7 with A7 usually used as stack pointer). I suspect you rather mean the initial PC value?

On 68k CPUs, the initial stack pointer value and reset PC vector are read from addresses 0x0 and 0x4, respectively. In ROM v65 (a dump can be found in the nextcomputers.org file archive), the value in ROM is 0x0100001e.

Now, the ROM is initially mapped to address 0x0 and 0x01000000, so the first instruction is actually read from offset 0x1e in the ROM. Somewhere in the early ROM startup routines, the ROM mapping at 0x0 is removed IIRC.

hth, Michael

Quote from: "cuby"I'm not sure which address register you think of - the 68040 has 8 address registers (A0-A7 with A7 usually used as stack pointer). I suspect you rather mean the initial PC value?

On 68k CPUs, the initial stack pointer value and reset PC vector are read from addresses 0x0 and 0x4, respectively. In ROM v65 (a dump can be found in the nextcomputers.org file archive), the value in ROM is 0x0100001e.

Now, the ROM is initially mapped to address 0x0 and 0x01000000, so the first instruction is actually read from offset 0x1e in the ROM. Somewhere in the early ROM startup routines, the ROM mapping at 0x0 is removed IIRC.

hth, Michael

Thanks, Yes, refering to the 8 address registers A0-A7, as I did by accident modify few in the rom monitor mode, and then my board stop working

could you please help me if you can?


Thanks

Quote from: "NeXTnewbe"

Quote from: "cuby"I'm not sure which address register you think of - the 68040 has 8 address registers (A0-A7 with A7 usually used as stack pointer). I suspect you rather mean the initial PC value?

On 68k CPUs, the initial stack pointer value and reset PC vector are read from addresses 0x0 and 0x4, respectively. In ROM v65 (a dump can be found in the nextcomputers.org file archive), the value in ROM is 0x0100001e.

Now, the ROM is initially mapped to address 0x0 and 0x01000000, so the first instruction is actually read from offset 0x1e in the ROM. Somewhere in the early ROM startup routines, the ROM mapping at 0x0 is removed IIRC.

hth, Michael

Thanks, Yes, refering to the 8 address registers A0-A7, as I did by accident modify few in the rom monitor mode, and then my board stop working

could you please help me if you can?

The effect of modifying the address registers is only temporary until (at most) the next reboot. There should be no way to cause permanent damage by changing A0-A7. What does the board do now, nothing at all? Has it stopped working immediately after changing the registers?

-- Michael

Hi

I still having the same issues, I got some pictures, and I would like to know if anyone with a Cube can just look on their configuration and just let me know whats wrong



Thanks

The settings of the address and data registers are completely irrelevant, the system sets them correctly before booting. Try to enable the power-on system test and the DRAM tests using the "p" command, perhaps your hardware is flaky?

One thing that worries me is the error message in http://www.flickr.com/photos/53827255@N05/4979917284/in/photostream/ - does this show up when the system tries to boot?

Exception #2 is an access fault, so the system seems to access an invalid address when executing the machine instruction at address 0x1006ea2. IIRC the 128 kB of address space starting at 0x01000000 are the alternative address space for the ROM. So I suspect either your boot ROM (which is actually an EPROM) is defective or - perhaps - some data structure on disk confuses the boot ROM (if this message occurs when the NeXT is trying to boot).

Which version of the boot ROM do you have? Perhaps I can find out some details about what exactly might fail... the nextcomputers.org file archive has images of many NeXT boot ROMs, so if you have access to an EPROM burner, you could try burning a new firmware ROM.

Apparently it was a problem in the initial boot settings, I did enable the test and now it seems it's fine, also other issue had was that I did not have an scsi terminator



Thanks

maybe your RAM configuration was wrong.
I suspect there is some EEPROM memory somewhere for some registers (RAM configuration, IP Address), it may reside in the device memory space at x02000000

there are some memory registers in x020c00xx that are set on boot by the rom. I don't know if this space is battery backed up or just some RAM