PDP-11/05 microcode dump?

Noel Chiappa jnc at mercury.lcs.mit.edu
Sat May 7 09:50:52 CDT 2016


    > From: Steven Malikoff

    > I was lucky to find an original 11/05 print set dated 1973. For what
    > it's worth, the microcode listing in my doc is Revision B 

Do note that a lot of the PROMs on those boards aren't actually 'microcode',
and aren't covered in that listing. For instance, on the M7261 (control)
board, there are 7 which aren't 'microcode' (list drawn from the 11/05
article on the Computer History wiki):

  A01A2 = E12 = Bus Request -> Grant processing
  A02A2 = E30 = Internal address decode (first stage)
  A07A1 = E68 = Internal address decode (second stage)
  A09A1 = E69 = Internal address decode (second stage)
  A09A2 = E101 = Branch utest service
  A13A1 = E90 = Internal interrupt acknowledge
  A14A1* = E100 = Console switch control

and 10 which are:

  A04A2 = E92 = Next instruction (high bits)
  A05A2 = E93 = Processor Status Word control
  A07A2* = E95 = Bus control
  A10A2 = E103 = Next instruction (low bits)
  A11A2 = E104 = ALU operation select
  A12A2 = E105 = Branch utest
  A13A2 = E106 = Multiplexor control
  A14A2 = E107 = Bus control
  A15A2 = E94 = ALU control
  A16A2 = E96 = Miscellaneous

And I haven't included the 11 PROM chips on the M7260 (Data paths) board,
none of which are 'microcode'. So those 'microcode' listings only cover about
1/3 of the PROM chips in the CPU; so one can't use just the microcode
revision level to tell you what's what.

E.g there are two chips different between the C and E etch levels of the
M7261: A07A2 and A14A1 (marked with a '*' above); one is 'microcode', one
isn't.


BTW, when you say that the microcode listings in your 1973 print set are
"Revision B", are you referring to the "Microprogram Flow", "Microprogram
Symbolic Listing", or "Microprogram Binary Listing", because they can be at
different revision levels (given in the box in the extreme lower right
corner)?

E.g. the ones in the KD11-B prints in the GT40 print set (dated February
1973) are 'B', 'C' and 'C', respectively; the hard-copy set I have (not dated
explicitly, but apparently mid-1972, given the modification date on the
'Index' sheet) has 'B', 'B' and 'B'.

    > it is *exactly* the same printout as in the the Bitsavers doc Revision
    > C

You mean the July 1976 set, the ones with microcode revision levels (as
above) of 'C', 'E', and 'E', right?

That is M7261 etch revision 'F', which uses quite a few different PROM chips
from the earlier ones, so I'd be fairly surprised if the microcode was
actually identical. I think you'd have to look at every bit to be sure; the
data on which chips changed, here:

  http://gunkies.org/wiki/PDP-11/05#Control_PROMs

would allow anyone who wanted to actually do that to focus in on specific
columns of the microprogram to look for differences.

(In fact, the 'F' etch rev has one less PROM chip than the 'E' etch rev, but
I suspect - i.e. I haven't checked the exact function of each chip on those
etch levels - that it's not a micro-program chip, though: there's one less
32x8 PROM chip, and those are generally used for control functions, the
microprogram chips are all 256x4.)

	Noel


More information about the cctalk mailing list