Oh, we managed to set the priviliged bit with an ordinary algol compiler and use that for access to dcalgol to call operator functions. That stopeed working when the privilege bit was remioved from the codefiles.
------Origineel bericht------
Van: Paul_Koning at Dell.com
Afzender: owner-hecnet at Update.UU.SE
Aan: hecnet at Update.UU.SE
Beantwoorden: hecnet at Update.UU.SE
Onderwerp: Re: [HECnet] PDP-11 Algol
Verzonden: 15 januari 2013 18:45
On Jan 15, 2013, at 12:15 PM, <hvlems at zonnet.nl>
wrote:
There was a port of the Burroughs Large Systems algol compiler for RT11.
I have one on a DECtape I.
It compiled test programs that I had written for, and copied from the B7700. Burroughs Extended Algol beats C Bliss hands down as tool to write compilers and os's in.
The same one, give or take some edits, I believe. Originally by Barry Folsom, modified by Greg Hosler. RSTS port by Terry Grieb and myself. And yes, originally for RT11 sounds familiar, that would certainly explain why the RSTS port was straightforward.
Burroughs Algol is nice for lots of work. For operating systems, there was Espol, which was kept more or less secret because getting access to the compiler meant you could break the OS security... (There is an Espol manual on Bitsavers.)
paul
On Tue, Jan 15, 2013 at 12:17 PM, Tim Sneddon <tim at sneddon.id.au> wrote:
GEM never generated code for the PDP-11. It did do IA-32, IA-64, Alpha,
MIPS. I supports a range of operating systems including Windows,
OpenVMS, Tru64, Linux and NonStop (although that never made it to
production).
That was my memory, also, although I thought it could also generated code for Ultrix, but I do not remember.
Clem
On 1/15/2013 12:52 PM, G. wrote:
On Tue, 15 Jan 2013 12:40:37 -0500, you wrote:
$ SHO LOG VIM
%SHOW-S-NOTRAN, no translation for logical name VIM
$ SHO LOG TMP
%SHOW-S-NOTRAN, no translation for logical name TMP
$ SHO SYM VIM
%DCL-W-UNDSYM, undefined symbol - check validity and spelling
$ sho sym/glo vim
%DCL-W-UNDSYM, undefined symbol - check validity and spelling
Obviously things aren't getting setup correctly. I wonder if I put it
all in the correct spots.
Please, don't take this as an offense (I'm not English native, so I do not
know how this would "sound" in English), but... After editing SYLOGICALS and
SYLOGIN have you rebooted the system or have you executed those definition
commands by typing them at a command prompt?
You can't offend me, I'm an idiot when it comes to VMS. :)
If you just put the definitions in the files, but haven't done anything more,
they were never considered: SYLOGICALS is read during startup and SYLOGIN
during process login.
You can define manually both logicals and symbols at the command line prompt
instead of rebooting the whole thing. But they have to be in the SY* files
too, or after the next reboot you'll have to do the manual typing again.
Yes, I did reboot. I like to always reboot after making system wide changes just to make sure I did it correctly. Nothing is worse than having to reboot and then nothing works. :)
-brian
On Tue, 15 Jan 2013 12:40:37 -0500, you wrote:
$ SHO LOG VIM
%SHOW-S-NOTRAN, no translation for logical name VIM
$ SHO LOG TMP
%SHOW-S-NOTRAN, no translation for logical name TMP
$ SHO SYM VIM
%DCL-W-UNDSYM, undefined symbol - check validity and spelling
$ sho sym/glo vim
%DCL-W-UNDSYM, undefined symbol - check validity and spelling
Obviously things aren't getting setup correctly. I wonder if I put it
all in the correct spots.
Please, don't take this as an offense (I'm not English native, so I do not
know how this would "sound" in English), but... After editing SYLOGICALS and
SYLOGIN have you rebooted the system or have you executed those definition
commands by typing them at a command prompt?
If you just put the definitions in the files, but haven't done anything more,
they were never considered: SYLOGICALS is read during startup and SYLOGIN
during process login.
You can define manually both logicals and symbols at the command line prompt
instead of rebooting the whole thing. But they have to be in the SY* files
too, or after the next reboot you'll have to do the manual typing again.
G.
On Tue, Jan 15, 2013 at 12:13 PM, <Paul_Koning at dell.com> wrote:
PL/C I remember as a hopelessly unreliable piece of junk I was forced to use at the University of Illinois, under OS/370. I don't remember it on other platforms.
I can not comment one way or the other, as I never either saw it's guts or used it for anything other than as a toy compiler on TSS. We had the real IBM compiler which is what I used there.
PL/360 was Wirth's 1967 subset language used to write AlgolW
On 1/15/2013 12:45 PM, Cory Smelosky wrote:
All of that's already up. I just need to get python working. I even installed GNV as I had the disk space...
Get python working and install the pexpect module and you'll be my hero.
Then fix my blasted system. :)
-brian
On 15 Jan 2013, at 12:42, Brian Hechinger <wonko at 4amlunch.net> wrote:
On 1/15/2013 12:39 PM, Cory Smelosky wrote:
On 15 Jan 2013, at 12:37, Brian Hechinger <wonko at 4amlunch.net> wrote:
On 1/15/2013 12:15 PM, Cory Smelosky wrote:
Sure you don't want to play with an account on my Alpha?;)
Yeah, I might just take you up on that at some point, but for now I'd really like to get something that I can haul around with me on my laptop.
Let me know. The box is going to stay on for awhile. ;)
If you set me something up I wouldn't complain. I'll need ssh/scp, vim (although that's optional) and python. :)
All of that's already up. I just need to get python working. I even installed GNV as I had the disk space...
What's the host NIC?
I'm noe of using VMware's "host only" network adapters which I believe is just a loopback adapter.
I'm tempted to spin up a loopback adapter for this, however, just to be sure.
Probably best. I didn't notice any reliability issues attaching it to a physical interface on my system.
You know what? I take that all back. I'm already using a loopback for IP connectivity. Forgot about that.
Bah.
-brian
On Jan 15, 2013, at 12:15 PM, <hvlems at zonnet.nl>
wrote:
There was a port of the Burroughs Large Systems algol compiler for RT11.
I have one on a DECtape I.
It compiled test programs that I had written for, and copied from the B7700. Burroughs Extended Algol beats C Bliss hands down as tool to write compilers and os's in.
The same one, give or take some edits, I believe. Originally by Barry Folsom, modified by Greg Hosler. RSTS port by Terry Grieb and myself. And yes, originally for RT11 sounds familiar, that would certainly explain why the RSTS port was straightforward.
Burroughs Algol is nice for lots of work. For operating systems, there was Espol, which was kept more or less secret because getting access to the compiler meant you could break the OS security... (There is an Espol manual on Bitsavers.)
paul
On 1/15/2013 12:39 PM, Cory Smelosky wrote:
On 15 Jan 2013, at 12:37, Brian Hechinger <wonko at 4amlunch.net> wrote:
On 1/15/2013 12:15 PM, Cory Smelosky wrote:
Sure you don't want to play with an account on my Alpha?;)
Yeah, I might just take you up on that at some point, but for now I'd really like to get something that I can haul around with me on my laptop.
Let me know. The box is going to stay on for awhile. ;)
If you set me something up I wouldn't complain. I'll need ssh/scp, vim (although that's optional) and python. :)
What's the host NIC?
I'm noe of using VMware's "host only" network adapters which I believe is just a loopback adapter.
I'm tempted to spin up a loopback adapter for this, however, just to be sure.
Probably best. I didn't notice any reliability issues attaching it to a physical interface on my system.
You know what? I take that all back. I'm already using a loopback for IP connectivity. Forgot about that.
Bah.
-brian
On Tue, Jan 15, 2013 at 12:13 PM, <Paul_Koning at dell.com> wrote:
I don't remember PL/360, but PL/M has nothing to do with PL/I.
PL/M was attempt at Intel to create a language for the 8080 and absolutely was a sub-set compiler. The language was designed to be a simple as possible, but to try to bring much of the PL/1 ideas into the microprocessor, without bringing in the whole language.
I was developed at Intel on their IBM mainframe originally as a cross compiler. I believe that it would eventually be able to self host on some of their development systems.
