20 Oct
2011
20 Oct
'11
1:12 a.m.
I've been assembling from source code most of the HP diagnostics for the HP2100 and
HP1000 systems. To do so, I used Eric Smith's original "asm21" - and found
that a number of diagnostics didn't assemble correctly. I updated it - and
subsequently Terry Newton joined the project and made more updates. Between us, over the
course of several months, we've managed to debug asm21 to the place where we have not
found any more errors in several additional months of use.
asm21 runs under both *NIX and Windows environments as it is written entirely in Perl.
Te get a copy, do an anonymous FTP to bickleywest.com and cd to the asm21 directory.
Download (binary) asm21_107.zip
If you're using Firefox, Chrome or Opera, simply use:
ftp://bickleywest.com/asm21
to get to the directory, then right click on asm21_107.zip to download it.
BTW: Terry created a great testing environment - scripts and software that allowed us to
assemble source code and subsequently check asm21's binary output against HP's
binaries which had been created under one of their OSs.
Cheers,
Lyle
--
Lyle Bickley, AF6WS
Bickley Consulting West Inc.
http://bickleywest.com
"Black holes are where God is dividing by zero"
20 Oct
20 Oct
2:04 a.m.
Hi
When I got my 21MX box, I quickly wrote a disassembler and assembler in Forth.
I used these to fiddle some but not used it for some time.
It of course has all the quarkiness of a Forth assembler.
Being Forth, it is like a macro assembler on steroids.
I'd expect a perl assembler
to have similar capabilities.
Dwight
From: lbickley at bickleywest.com
I've been assembling from source code most of the HP diagnostics for the HP2100 and
HP1000 systems. To do so, I used Eric Smith's original "asm21" - and found
that a number of diagnostics didn't assemble correctly. I updated it - and
subsequently Terry Newton joined the project and made more updates. Between us, over the
course of several months, we've managed to debug asm21 to the place where we have not
found any more errors in several additional months of use.
asm21 runs under both *NIX and Windows environments as it is written entirely in Perl.
Te get a copy, do an anonymous FTP to bickleywest.com and cd to the asm21 directory.
Download (binary) asm21_107.zip
If you're using Firefox, Chrome or Opera, simply use:
ftp://bickleywest.com/asm21
to get to the directory, then right click on asm21_107.zip to download it.
BTW: Terry created a great testing environment - scripts and software that allowed us to
assemble source code and subsequently check asm21's binary output against HP's
binaries which had been created under one of their OSs.
Cheers,
Lyle
--
Lyle Bickley, AF6WS
Bickley Consulting West Inc.
http://bickleywest.com
"Black holes are where God is dividing by zero"
9:48 a.m.
That's really awesome Lyle. I had started the same project a few years ago,
got half way through it and never finished it. My aim was to produce a
single CD that contained source, object, and manual for each diagnostic in
the library. Glad to see someone working on this!
10:20 a.m.
FYI - I think where I ran in to headscratching on the project was... .it was
important for the diagnostics to show .lst files in the assembled version,
as it was incredibly handy when running the diags to be able to pick up the
object listing and see exactly what instructions were at what address and
what should be going on at that point in time. And I made great headway till
I got to the diagnostics which were relocatable, and this rendered the
object listing "less useful". Hitting that mild bump-in-the-road caused me
to move to other projects ;)
So many bits, so little media :>
J
11:17 a.m.
On Thursday 20 October 2011, Jay West wrote:
FYI - I think where I ran in to headscratching on the
project was... .it was
important for the diagnostics to show .lst files in the assembled version,
as it was incredibly handy when running the diags to be able to pick up the
object listing and see exactly what instructions were at what address and
what should be going on at that point in time. And I made great headway till
I got to the diagnostics which were relocatable, and this rendered the
object listing "less useful". Hitting that mild bump-in-the-road caused me
to move to other projects ;)
Unfortunately, at this juncture, "asm21" only handles absolute. The good news is
that with rare exception almost all of the standalone diagnostics are absolute.
The most complex tasks we had to deal with were 1) HP software that sometimes uses invalid
structure - most specifically with regard to the "OCT" mnemonic, 2)
adding/fixing multiword instructions such as those in FP, EIG, DMS, etc. and 3) adding
pseudo ops used by HP in their diagnostics such as RAM, byte pseudo ops, etc.
I have found that having the listings of the diagnostics is a must. In addition, most of
the diagnostics have a good summary on how to run the diagnostic at the beginning of the
listing. The summary is "cryptic". But once you understand them, they are
generally all you need.
BTW: "asm21" also creates a nice cross reference listing by default.
Cheers,
Lyle
So many bits, so little media :>
J
--
Lyle Bickley, AF6WS
Bickley Consulting West Inc.
http://bickleywest.com
"Black holes are where God is dividing by zero"
11:43 a.m.
On Wed, Oct 19, 2011 at 10:12 PM, Lyle Bickley <lbickley at bickleywest.com> wrote:
I've been assembling from source code most of the
HP diagnostics for the HP2100 and HP1000 systems. ?To do so, I used Eric Smith's
original "asm21" - and found that a number of diagnostics didn't assemble
correctly. I updated it - and subsequently Terry Newton joined the project and made more
updates. Between us, over the course of several months, we've managed to debug asm21
to the place where we have not found any more errors in several additional months of use.
Do you have copies of the machine readable source for the diagnostics?
I have seen the scanned hardcopy listings of the source but don't
think I've noticed copies of the original machine readable source on
the net. Did this come from the collection of software Al and the
museum acquired from HP, or from some other source?
One of these days I'll finally get around to running the diagnostics
on my 2113E and 2117F.
-Glen
8:17 p.m.
On Thursday 20 October 2011, Glen Slick wrote:
On Wed, Oct 19, 2011 at 10:12 PM, Lyle Bickley
<lbickley at bickleywest.com> wrote:
bitsavers.org has a treasure trove of HP source code available at:
http://bitsavers.org/bits/HP/tapes/
There's a certain amount of work one has to do in order to create a clean source file
and binary:
1) Find the software/diagnostics you want in the text catalogs
2) Typically, a single piece of software will be spread over a number of source segments
3) Strip the source segments of high order bits and zero characters (Might want to use
cvhp.c code I wrote below)
4) Concatenate the "striped" segments into a single text file (easy under
Linux)
5) Assemble same using "asm21" or SIMH running BCS or RTS (or a "real"
HP 21xx).
cvhp.c source
-------------
#include <stdio.h>
main()
{
int c;
while ((c = getchar()) != EOF) {
if (c) putchar(c & 0x7f);
}
}
I've been assembling from source code most of
the HP diagnostics for the HP2100 and HP1000 systems. To do so, I used Eric Smith's
original "asm21" - and found that a number of diagnostics didn't assemble
correctly. I updated it - and subsequently Terry Newton joined the project and made more
updates. Between us, over the course of several months, we've managed to debug asm21
to the place where we have not found any more errors in several additional months of use.
Do you have copies of the machine readable source for the diagnostics?
I have seen the scanned hardcopy listings of the source but don't
think I've noticed copies of the original machine readable source on
the net. Did this come from the collection of software Al and the
museum acquired from HP, or from some other source?
One of these days I'll finally get around to running the diagnostics
on my 2113E and 2117F.
Go for it ;-)
Cheers,
Lyle
--
Lyle Bickley, AF6WS
Bickley Consulting West Inc.
http://bickleywest.com
"Black holes are where God is dividing by zero"
8:37 p.m.
On Thu, Oct 20, 2011 at 5:17 PM, Lyle Bickley <lbickley at bickleywest.com> wrote:
One limitation is that I have an 800BPI 7970B with the HP specific
non-Pertec inteface which I can't connect to any other systems, and
the other Pertec interface tape drives which I can connect to other
systems to write tapes are 1600/6250 BPI. So I can't write an 800BPI
diagnostic tape myself, unless I hack up some way to do that directly
One of these
days I'll finally get around to running the diagnostics
on my 2113E and 2117F.
Go for it ;-)
from the HP system, which I suppose I could do if I
actually spent
some time working on it...
-Glen
21 Oct
21 Oct
9:23 a.m.
New subject: DEC H960 Rack Wanted
I am looking for a DEC H960 rack, Prefer it be in the USA. Any help in
finding one would be appreciated. Thanks!
Todd Paisley
11:36 a.m.
Glen wrote....
One limitation is that I have an 800BPI 7970B with the
HP specific
non-Pertec inteface which I can't connect to any other systems, and
the other Pertec interface tape drives which I can connect to other
systems to write tapes are 1600/6250 BPI. So I can't write an 800BPI
diagnostic tape myself, unless I hack up some way to do that directly
from the HP system, which I suppose I could do if I actually spent
some time working on it...
Don't you have disc on the system? You can make a bootable diagnostics disc
cartridge, that's what I use typically.
J
20 Oct
20 Oct
2:07 p.m.
Hi,
On Thu, Oct 20, 2011 at 7:12 AM, Lyle Bickley <lbickley at bickleywest.com> wrote:
I've been assembling from source code most of the
HP diagnostics for the HP2100 and HP1000 systems. ?To do so, I used Eric Smith's
original "asm21" - and found that a number of diagnostics didn't assemble
correctly. I updated it - and subsequently Terry Newton joined the project and made more
updates. Between us, over the course of several months, we've managed to debug asm21
to the place where we have not found any more errors in several additional months of use.
asm21 runs under both *NIX and Windows environments as it is written entirely in Perl.
Does it require a specific version of perl? Or some perl libraries / extensions?
tingo at kg-v2$ uname -a
FreeBSD kg-v2.kg4.no 8.2-STABLE FreeBSD 8.2-STABLE #4: Sat Jul 16
02:07:14 CEST 2011 root at kg-v2.kg4.no:/usr/obj/usr/src/sys/GENERIC
amd64
tingo at kg-v2$ perl -v
This is perl, v5.10.1 (*) built for amd64-freebsd-thread-multi
Copyright 1987-2009, Larry Wall
Perl may be copied only under the terms of either the Artistic License or the
GNU General Public License, which may be found in the Perl 5 source kit.
Complete documentation for Perl, including FAQ lists, should be found on
this system using "man perl" or "perldoc perl". If you have access to
the
Internet, point your browser at http://www.perl.org/, the Perl Home Page.
tingo at kg-v2$ ./asm21_107
Use of uninitialized value $listing in substitution (s///) at
./asm21_107 line 336.
Use of uninitialized value $listing in string eq at ./asm21_107 line 338.
Use of uninitialized value in string eq at ./asm21_107 line 338.
^C
tingo at kg-v2$ ./asm21_107 -?
Can't open -?: No such file or directory at ./asm21_107 line 367.
Can't open -?: No such file or directory at ./asm21_107 line 367.
tingo at kg-v2$ ./asm21_107 -h
Can't open -h: No such file or directory at ./asm21_107 line 367.
Can't open -h: No such file or directory at ./asm21_107 line 367.
tingo at kg-v2$ ./asm21_107 --help
Can't open --help: No such file or directory at ./asm21_107 line 367.
Can't open --help: No such file or directory at ./asm21_107 line 367.
There was little documentation included...
--
Regards,
Torfinn Ingolfsen
4659
days inactive
4660
days old
10 comments
6 participants
participants (6)
-
dkelvey＠hotmail.com -
glen.slick＠gmail.com -
jwest＠classiccmp.org -
lbickley＠bickleywest.com -
paisley＠erols.com -
tingox＠gmail.com