30 Apr
2016
30 Apr
'16
9:36 a.m.
On Tue, 26 Apr 2016, Robert Jarratt wrote:
It turns out that I only needed to flip the bit in the
bank 8 config
register and I can now access the flashbus. I can write to the LEDs, it
turns out that writing a 0 turns the LED on, rather than off, which is why I
thought it hadn't worked.
Oh, I thought you would try 0x55 or suchlike pattern -- given that you
did not know whether it is active-high or active-low logic being used by
the circuit. And active-low is actually what I'd expect here given that
you get all the LEDs lit at reset.
I can now do test code for the NVRAM, I will start
with repeated reads to
see if I get different values on different occasions. Then I will save it
and write patterns etc. It may be a checksum that is failing of course. Or
the DROM itself, which I can't verify until I get the PLCC adapter for my
PROM programmer and someone else to read their DROM.
I just hope it's not a heisenbug of some sort and with NVRAM supposedly
now eliminated as the cause it is DROM contents indeed. The DROM part at
least can be fixed without soldering, which is always a bit of a challenge
with finer pitch components.
Thanks for all the help I have a way forward for a
little while now.
Although anyone with one of these machines who could read the DROM would be
helpful. I will be able to supply a VMS program to read it, so need to
remove it from the machine or to have a programmer.
You are welcome! Good luck tracking down a DROM image soon -- that's
beyond me I'm afraid, I only have TURBOchannel stuff really.
Maciej