On Thu, 3 Jan 2013 10:13:04 +0100, you wrote:
Well, tried it:
[...]
?LNKUGS 7 undefined global symbols
NTFONC 456353
GIVSTC 456516
NTDPRV 460114
GETSTC 456543
NTONPJ 456373
NTFSEC 456354
PCBPCV 461503
Perhaps I might have found a solution. I seem to remember that once I read
something somewhere (maybe on alt.sys.pdp10) about certain cases in which
ANF-10 support couldn't be excluded even if unneeded. So I tried a build with
it and I obtained NO undefined symbols. I have no way to test the resulting
monitor, but maybe it could work. BTW, I noticed that even with those seven
undefined symbols the executable file got created anyway, so maybe they are
not really needed for the monitor to work... I Don't know.
Here is the full transcription of my attempt (some values are obviously just
placeholders for better chosen ones):
| .r mongen
|
| Write MONGEN.MIC (YES,NO) : no
| MONGEN for 704 monitors
|
| MONGEN will generate any known TOPS-10 configuration. For information
| about which configurations are supported, please refer to the SPD.
|
| /HELP (PROMPT,SHORT,LONG) ]:
| Which GEN (SYS,F) :
| Output(DSK:SYSCNF.MAC):
| DECsystem10 (KL,KS) : ks
| System name : KMC test
| CPU0 serial number (1-10000) : 999
| Exclude monitor overhead from user run time (YES,NO) :
| Account verification (YES,NO) :
|
| Autoconfigured device options:
| Include Ethernet service (NO,YES,PROMPT):
| Prompt for RP0x/RM0x Massbus disks (NO,YES):
| Include TM02/TM03 magtapes (NO,YES,PROMPT): yes
| Include UNIBUS/LP20 printers (NO,YES,PROMPT):
| Include UNIBUS/CD20 card readers (NO,YES,PROMPT):
|
| Local terminal configuration:
| DZ-11 TTY lines (0-32) : 8
|
| OPR octal line number (CTY,0-7) :
| Data set lines
|
|
| Number of Jobs (1-511) : 120
| Maximum pages of core for each job (0,0-16384) :
| Total amount of system core in K (512-4096) : 4096
| Clock ticks per second (60,50) :
| Allow jobs to be locked in core (YES,NO) :
| Guaranteed number of pages for unlocked jobs (0,0-16384) : 10
| Number of pages reserved for locking jobs in Exec Virtual Memory (10,0-?):50
| Number of high priority interrupt queues (0,0-15): 15
| Meter (NO,YES) :
| KASER (NO,YES) :
| MPXSER (YES,NO) :
| PSISER (YES,NO) :
| IPCF (YES,NO) :
| ENQ/DEQ (YES,NO) :
| Disk sets (ALL) :
| Configure non-autoconfigured hardware (NO,YES) :
| Number of KMC/DUP Lines (0,0-2) : 1
| Type of line for KDP0 (ANF10,DECNET,USER,IBM) : decnet
| Number of PTYs (20,1-500) :
|
| Network software (YES,NO) :
| Node name : test
| Number of remote TTYs (480,0-480) :
|
| ANF-10 software (YES,NO) :
| Node name (TEST) :
| Node number of central site (1,1-77) :
| Remote terminals (YES,NO) : no
| Remote card readers (YES,NO) : no
| Remote line printers (YES,NO) : no
| Remote paper tape punches (NO,YES) : no
| Remote paper tape readers (NO,YES) : no
| Remote plotters (NO,YES) : no
| Remote DN8x DDCMP devices (YES,NO) : no
| Remote data entry terminals (YES,NO) : no
| Remote task-to-task (YES,NO) : no
| Number of connects (256,1-512) :
|
| DECnet software (YES,NO) :
| Node name (TEST) :
| Area number of central site (1,1-63) :
| Node number of central site (1,1-1023) :
| Router type (ROUTING,NONROUTING) :
| Transmit password (DECNET20) :
| Remote terminals (YES,NO) :
|
| Decimal "symbol,value"
|
|
| Octal "symbol,value"
|
|
| SIXBIT "symbol,value"
|
|
| Type "device-mnemonic,PI-channel" for special devices
|
|
| Type "device-mnemonic,PI-channel,no.-of-devices"
|
|
| Type "device-mnemonic,PI-channel,highest-ac-to-save"
|
|
| Type "ersatz-device,P,Pn,search-list-type"
|
|
| Type "command-name,dispatch,flags,unique-bits,PRVTAB-name"
|
|
| Terminal class :
|
| Additional terminals for classes:
|
| Class to extend :
|
| System default terminal type (TTY) :
| File DSK:SYSCNF.MAC Closed
|
| Which GEN (SYS,F) : ^Z
|
| EXIT
|
| .COMPILE/COMPILE SYSCNF+<COMMON,COMDEV,COMMOD>
| MACRO: COMMON
| MACRO: COMDEV
| MACRO: COMMOD
|
| EXIT
|
| .r link
|
| *TESTM/SAVE/NOINITIAL/HASH:13K = /LOCALS -
| #COMMON,COMDEV,COMMOD,TOPS10[10,7,MON,KS]/SEARCH-
| #/PATCH:200/COUNTERS/GO
| [LNKRLC Reloc. ctr. initial value current value limit value
| .LOW. 0 72204 340000
| .HIGH. 340000 673131 720000
| .INIT. 740000 765112 771000
| .TWIC. 771000 772401 1000000
| Absolute code loaded]
|
| EXIT
|
| .
My gess was the distributed TOPS10.REL does not contain all the required
modules, so it should be rebuilt somehow. The "A" appendix in the software
installation manual is supposed to explain that, but it has not been of much
help...
No, I'm quite sure about this: as long as you build a monitor with a standard
set of features, you do not need to rebuild TOPS10.REL from scratch.
The Software Installation Guide is quite clear about that (page 9-16):
| If you are using a standard combination of feature-test switch settings and
| DIGITAL-supplied sources, you do not need to assemble the remainder of the
| bundled monitor modules. These modules have already been assembled and
| combined in a monitor library file named TOPS10.REL.
| [...]
| If you are not using a standard combination of feature-test switch settings
| or if you have made changes to the monitor sources, you must assemble all
| the monitor sources, creating a new monitor library file.
http://pdp-10.trailing-edge.com/tops10v704_docc/01/10,7/docupd/sig.mem.html
Let's hope that the ANF-10 trick is enough! :)
G.
El 03/01/2013, a les 11:40, Johnny Billquist <bqt at softjar.se> va escriure:
On 2013-01-03 10:56, Jordi Guillaumes i Pons wrote:
El 03/01/2013, a les 10:42, Johnny Billquist <bqt at softjar.se> va escriure:
The date/time prompt is just in the script. It's totally optional as such.
That said, simh do emulate an 11/93 or 11/94 just fine, which do have a toy.
Thanks! I assume I will need to SYSGEN for the /93...
Not really. RSX detects if it is an 11/93, and enables the additional switches to the TIM command based on that.
And STARTUP.CMD also notices, and sets the clock based on the TOY. However, until you edit STARTUP.CMD, it will still ask about the time.
If you've generated a system with all the bells and whistles, the doing a new SYSGEN won't really change anything.
It works. Thanks a lot!
Jordi Guillaumes i Pons
jg at jordi.guillaumes.name
HECnet: BITXOV::JGUILLAUMES
On 2013-01-03 10:56, Jordi Guillaumes i Pons wrote:
El 03/01/2013, a les 10:42, Johnny Billquist <bqt at softjar.se> va escriure:
The date/time prompt is just in the script. It's totally optional as such.
That said, simh do emulate an 11/93 or 11/94 just fine, which do have a toy.
Thanks! I assume I will need to SYSGEN for the /93...
Not really. RSX detects if it is an 11/93, and enables the additional switches to the TIM command based on that.
And STARTUP.CMD also notices, and sets the clock based on the TOY. However, until you edit STARTUP.CMD, it will still ask about the time.
If you've generated a system with all the bells and whistles, the doing a new SYSGEN won't really change anything.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
El 03/01/2013, a les 8:26, Jordi Guillaumes i Pons <jg at jordi.guillaumes.name> va escriure:
Jordi Guillaumes i Pons
Barcelona - Catalunya - Europa
El 03/01/2013, a les 0:10, "Rob Jarratt" <robert.jarratt at ntlworld.com> va escriure:
I have been
struggling particularly with PDP10, trying to get TOPS-10 to use it, but I
really don't know how and my attempt to build a new monitor with DMR11
emulation (almost identical to DMC11) failed at the link stage, but I don't
know why.
That is what happens here too... My guess is it is necessary to build a modified TOPS10.REL. But without the docs it can be quite difficult.
I have just found there is a bug in the current HEAD version that prevents to use the DZ multiplexer in the pdp10 simulator (it seems to work in the pdp11 and VAXen though). Work-in-progress software is wonderful :)
Jordi Guillaumes i Pons
jg at jordi.guillaumes.name
HECnet: BITXOV::JGUILLAUMES
El 03/01/2013, a les 10:42, Johnny Billquist <bqt at softjar.se> va escriure:
The date/time prompt is just in the script. It's totally optional as such.
That said, simh do emulate an 11/93 or 11/94 just fine, which do have a toy.
Thanks! I assume I will need to SYSGEN for the /93...
Jordi Guillaumes i Pons
jg at jordi.guillaumes.name
HECnet: BITXOV::JGUILLAUMES
On 2013-01-03 08:29, Jordi Guillaumes i Pons wrote:
Jordi Guillaumes i Pons
Barcelona - Catalunya - Europa
El 03/01/2013, a les 3:17, Johnny Billquist <bqt at softjar.se> va escriure:
Or else don't worry about an incorrect date at startup. Unattended boot of RSX is trivial.
Ouch... I had the impression that SIMH lacked the TOY needed to boot without that date/time prompt.
The date/time prompt is just in the script. It's totally optional as such.
That said, simh do emulate an 11/93 or 11/94 just fine, which do have a toy.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
El 03/01/2013, a les 9:33, G. <gerry77 at mail.com> va escriure:
The Software Installation Guide lists a bunch of MONGEN questions related to
DECnet on KS10s, asking for KMC/DUP or DMR hardware characteristics. A quick
Google search seems to confirm that both KMC and DMR are very similar to DMC
and software compatible with it. DMR should be a DMC enhancement/replacement.
Well, tried it:
(...)
Number of KMC/DUP Lines (0,0-2) : 1
Type of line for KDP0 (ANF10,DECNET,USER,IBM) : decnet
Number of PTYs (20,1-476) :
Network software (YES,NO) : yes
Node name : bitxr4
Number of remote TTYs (456,0-456) :
ANF-10 software (YES,NO) : no
DECnet software (YES,NO) : yes
Node name (BITXR4) :
Area number of central site (1,1-63) : 7
Node number of central site (1,1-1023) : 904
Router type (ROUTING,NONROUTING) : routing
Transmit password (DECNET20) :
Remote terminals (YES,NO) :
(...)
At link time I get some undefined symbols:
. r link
*newsys/save/noinitial/hash:13k = /locals -
#common,comdev,commod,tops10[10,7,mon,ks]/search-
#/patch:200/counters/go
[LNKRLC Reloc. ctr. initial value current value limit value
.LOW. 0 71001 344000
.HIGH. 344000 664031 720000
.INIT. 740000 764725 771000
.TWIC. 771000 772401 1000000
Absolute code loaded]
?LNKUGS 7 undefined global symbols
NTFONC 456353
GIVSTC 456516
NTDPRV 460114
GETSTC 456543
NTONPJ 456373
NTFSEC 456354
PCBPCV 461503
My gess was the distributed TOPS10.REL does not contain all the required modules, so it should be rebuilt somehow. The "A" appendix in the software installation manual is supposed to explain that, but it has not been of much help...
Jordi Guillaumes i Pons
jg at jordi.guillaumes.name
HECnet: BITXOV::JGUILLAUMES
On Thu, 03 Jan 2013 09:33:55 +0100, you wrote:
There is a DECnet-10 tape, but it mostly contains software for front-end based
DECnet solutions (KL10 only), e.g. PDP-11s in disguise called DN20s and the
like. The only useful file from the DECnet tape is NML.EXE which can be found
elsewhere too. Anyway, the DECnet tape is BB-X116C-BB and is available on
Trailing Edge; there is also a newer BB-X116D-BB, which is somewhat hidden,
but can be located using the search by filename function... :P
I stand corrected: the "hidden" tape is another one (Cobol for TOPS-20, if I'm
not wrong). DECnet tapes are simply there in the Trailing Edge repository, and
even more than two if we count older ones not for 7.04... :)
Sorry,
G.
-----Original Message-----
From: owner-hecnet at Update.UU.SE [mailto:owner-hecnet at Update.UU.SE]
On Behalf Of Jordi Guillaumes i Pons
Sent: 03 January 2013 07:27
To: hecnet at Update.UU.SE
Subject: Re: [HECnet] DMC11 in next simh version... looks nice, and
question
about TOPS-10
Jordi Guillaumes i Pons
Barcelona - Catalunya - Europa
El 03/01/2013, a les 0:10, "Rob Jarratt" <robert.jarratt at ntlworld.com> va
escriure:
I have been
struggling particularly with PDP10, trying to get TOPS-10 to use it,
but I really don't know how and my attempt to build a new monitor with
DMR11 emulation (almost identical to DMC11) failed at the link stage,
but I don't know why.
That is what happens here too... My guess is it is necessary to build a
modified TOPS10.REL. But without the docs it can be quite difficult.=
I have had communication from a PDP10 expert who makes me think it is *much*
more complex than just rebuilding the monitor. The DMR11 may well have been
in the PDP11 front end and I am not sure how that is done in SIMH at the
moment.
Regards
Rob