LSI-11/2 microcode source

[Eric Smith]

Heinz wrote:
> Alternate anyone have info on reading out the WD PMOS roms?

Reading the MICROMs wouldn't be that difficult.  But it's not the
whole story.  The control chip has two metal-mask PLAs ("translation
arrays") that can cause PC changes even when the current microinstruction
is not a branch, based on two 8-bit translation registers and various
other inputs.

I can envision two ways to dump the translation arrays.

1:  Capture a whole lot of logic analyzer traces covering both the
    Qbus and the microinstruction bus while executing a variety
    of PDP-11 software.  Write software to parse the traces,
    identify translation branches, track the expected translation
    register contents, and build up logic equations.  This is a lot
    of work and wouldn't be guarateed to come up with identical
    PLA equations, though if you run enough traces through you'd
    probably get close.

2:  Deencapsulate a control chip with fuming nitric acid, take a
    high resolution photomicrograph, and spend a lot of time studying
    it to figure out the topology of the PLAs.

> I've got 3x working M7264 LSI-11/2(quad height) with all the MICROMs
> loaded as well
> as a M7270 LSI-11 (dual height), they all run RT fine. Anyone know
> if you can just swap the MICROMS to make a WD Pascal Microegine?

You can't, because the translation arrays are different in the LSI-11,
Pascal Microengine, and WD-16.  The data path chip of the Pascal
Microengine and WD-16 are the same, but I'm not 100% certain that it's
the same as that of the LSI-11.

Also, the system-level interface of the LSI-11 and Pascal Microengine
are different.  There are four bus control signals from the MICROMs that
get decoded for bus control, and they are not used identically in the
LSI-11, Pascal Microengine, and WD-16.  Even if you swap the control
chip along with the MICROMs, the Pascal Microengine doesn't know how to
talk to a Qbus.

Eric