On 2022-01-21, Dave Froble <davef@tsoft-inc.com> wrote:
>
> I'll admit to learning much from overlays, but, it was still torture ..
>

I learnt a lot from overlays, such as never wanting to go near the
bloody things ever again in my life. :-)

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

On 1/24/22 09:00, Simon Clubley wrote:
> On 2022-01-21, Dave Froble <davef@tsoft-inc.com> wrote:
>>
>> I'll admit to learning much from overlays, but, it was still torture ..
>>
>
> I learnt a lot from overlays, such as never wanting to go near the
> bloody things ever again in my life. :-)

Funny how I always hear this. I have done lots of overlays on
Ultrix-11 (even for the kernel!) and never had any serious problems
with them. I even like them. :-)

bill

On 2022-01-24, Bill Gunshannon <bill.gunshannon@gmail.com> wrote:
> On 1/24/22 09:00, Simon Clubley wrote:
>> On 2022-01-21, Dave Froble <davef@tsoft-inc.com> wrote:
>>>
>>> I'll admit to learning much from overlays, but, it was still torture ..
>>>
>>
>> I learnt a lot from overlays, such as never wanting to go near the
>> bloody things ever again in my life. :-)
>
> Funny how I always hear this. I have done lots of overlays on
> Ultrix-11 (even for the kernel!) and never had any serious problems
> with them. I even like them. :-)
>

You poor devil. Have you considered psychiatric help ? :-)

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

On Monday, January 24, 2022 at 4:16:42 PM UTC+11, Johnny Billquist wrote:
[snip]

> Well, I should say that it might not work in RSTS/E. RSTS/E were more
> limited than RSX in many ways. This might very well be one of those cases.
> I think that in RSX, the interactive stuff makes use of an additional
> process. And that's not really possible to do with RSTS/E since each
> terminal only have one process.
>
> But I haven't tried (or can't remember). I haven't played much in RSTS/E
> in a long time.

It does work under RSTS/E too - just like RSX11M-Plus.

$ sys/a

RSTS V10.1-L PDP11/70 PiDP11 status at 25-Jan-22, 09:11 AM Up: 3 00:23:41

Job Who Where What Size State Run-Time RTS
7 1,42 KB0 DCL 4K ^C 0.3 DCL
8 11,42 KB13! SYSTAT 17K RN Lck 49.3 ...RSX
$ basic/bp2

PDP-11 BASIC-PLUS-2 V2.7-00

BASIC2

print ert$(i%) for i%=1% to 5%

??Bad directory for device
?Illegal file name
?Account or device in use
?No room for user on device
?Can't find file or account

BASIC2

old ox3d.b2s

BASIC2

compile ox3d/obj/chain/nowarn
OX3D 09:12 AM 25-Jan-22

BASIC2

build ox3d

BASIC2

$ switch dcl
$ tkb @ox3d
$ dir ox3d.*/fu

Name .Typ Size Prot Access Date Time Clu RTS Pos Op/rr
SY:[11,42]
OX3D .BAC 19C <124> 19-Feb-96 19-Feb-96 10:45 AM 64 BASIC 3062 0/0
OX3D .BAS 7 < 60> 19-Feb-96 19-Feb-96 10:45 AM 64 BASIC 3063 0/0
OX3D .B2S 7 < 60> 25-Jan-22 25-Jan-22 08:44 AM 64 ...RSX 5202 0/0
OX3D .OBJ 27 < 60> 25-Jan-22 25-Jan-22 09:12 AM 64 ...RSX 5216 0/0
RF:VAR=128 FO:SEQ USED:27:390 RECSI:126 CC:IMP
OX3D .CMD 1 < 60> 25-Jan-22 25-Jan-22 09:12 AM 64 ...RSX 5217 0/0
OX3D .ODL 1 < 60> 25-Jan-22 25-Jan-22 09:12 AM 64 ...RSX 5219 0/0
OX3D .TSK 48C <124> 25-Jan-22 25-Jan-22 09:13 AM 64 RSX 4993 0/0
RF:FIX FO:SEQ USED:49:0 RECSI:512

Total of 110 blocks in 7 files in SY:[11,42]

$ type ox3d.cmd
SY:OX3D=SY:OX3D/MP
UNITS = 13
ASG = SY:5:6:7:8:9:10:11:12
EXTTSK = 512
//
$ type ox3d.odl
.ROOT USER
USER: .FCTR SY:OX3D-LIBR
LIBR: .FCTR LB:[1,1]BP2OTS/LB
.END
$ run ox3d.tsk

MOVE 1 ?
8 DINGO::KB13 OX3D+...RSX KB(0R) 12(64)K+0K 57.1(+10.5)
1,1,1
O.K. - ( 1 , 1 , 1 )
MY MOVE 1 IS TO CELL ( 4 , 4 , 4 )

O . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . X

MOVE 2 ?

Tony

Den 2022-01-24 kl. 18:24, skrev Bill Gunshannon:
> On 1/24/22 09:00, Simon Clubley wrote:
>> On 2022-01-21, Dave Froble <davef@tsoft-inc.com> wrote:
>>>
>>> I'll admit to learning much from overlays, but, it was still torture ..
>>>
>>
>> I learnt a lot from overlays, such as never wanting to go near the
>> bloody things ever again in my life. :-)
>
> Funny how I always hear this.  I have done lots of overlays on
> Ultrix-11 (even for the kernel!) and never had any serious problems
> with them.  I even like them.  :-)
>
> bill
>
>

I never thought anything bad/good about them, the overlay feature
was just that, a feature of RSX-11M. I think the most "overlayed"
image I built was using both the RMS libraries and the DTR-11
callable library. It was mainly DTR-11 that added most overlays,
as I remember it. I think the FMS libs was used also...

On Friday, January 21, 2022 at 4:51:45 AM UTC-8, VAXman- wrote:

(snip, I wrote)

> >The one I remember is that quoting on the DCL command line is
> >different from quoting in COM file of DCL commands.
> >Seems to be that it should be consistent between them.
> Short of where DCL reads its input (terminal or file via an RMS $GET), I don't
> see why/where processing should be any different.

As I remember it, there are some places where you need triple quotes,
in place of regular quotes. I believe it is related to symbols and logical names.

On 25/01/2022 11:35, gah4 wrote:
> On Friday, January 21, 2022 at 4:51:45 AM UTC-8, VAXman- wrote:
>
> (snip, I wrote)
>
>>> The one I remember is that quoting on the DCL command line is
>>> different from quoting in COM file of DCL commands.
>
>>> Seems to be that it should be consistent between them.
>
>> Short of where DCL reads its input (terminal or file via an RMS $GET), I don't
>> see why/where processing should be any different.
>
> As I remember it, there are some places where you need triple quotes,
> in place of regular quotes. I believe it is related to symbols and logical names.
>

IIRC if a DCL command file bulds and writes a (sub-)command file that handles symbols, there is use for more than three
double-quotes. (do I remember groups of six?)

--
Mike R.
Home: http://alpha.mike-r.com/
QOTD: http://alpha.mike-r.com/qotd.php
No Micro$oft products were used in the URLs above, or in preparing this message.
Recommended reading: http://www.catb.org/~esr/faqs/smart-questions.html#before
and: http://alpha.mike-r.com/jargon/T/top-post.html
Missile address: N31.7624/E34.9691

Den 2022-01-25 kl. 11:36, skrev Henry Crun:
> On 25/01/2022 11:35, gah4 wrote:
>> On Friday, January 21, 2022 at 4:51:45 AM UTC-8, VAXman- wrote:
>>
>> (snip, I wrote)
>>
>>>> The one I remember is that quoting on the DCL command line is
>>>> different from quoting in COM file of DCL commands.
>>>> Seems to be that it should be consistent between them.
>>> Short of where DCL reads its input (terminal or file via an RMS $GET), I
>>> don't
>>> see why/where processing should be any different.
>>
>> As I remember it, there are some places where you need triple quotes,
>> in place of regular quotes.   I believe it is related to symbols and
>> logical names.
>>
>
> IIRC if a DCL command file bulds and writes a (sub-)command file that
> handles symbols, there is use for more than three double-quotes. (do I
> remember groups of six?)
>

Sure, if you need them. No problem to find examples. This is from the
Oracle Rdb "Bugcheck dump summarizer":

$ if error_area_name .nes. ""
$ then
$ call get_more_details -
"""Storage area """"''error_area_name'""""""/exact" 0 8
$ endif

In article <ssonkm$cn7$1@dont-email.me>, =?UTF-8?Q?Jan-Erik_S=c3=b6derholm?= <jan-erik.soderholm@telia.com> writes:
>Den 2022-01-25 kl. 11:36, skrev Henry Crun:
>> On 25/01/2022 11:35, gah4 wrote:
>>> On Friday, January 21, 2022 at 4:51:45 AM UTC-8, VAXman- wrote:
>>>
>>> (snip, I wrote)
>>>
>>>>> The one I remember is that quoting on the DCL command line is
>>>>> different from quoting in COM file of DCL commands.
>>>>> Seems to be that it should be consistent between them.
>>>> Short of where DCL reads its input (terminal or file via an RMS $GET), I
>>>> don't
>>>> see why/where processing should be any different.
>>>
>>> As I remember it, there are some places where you need triple quotes,
>>> in place of regular quotes.   I believe it is related to symbols and
>>> logical names.
>>>
>>
>> IIRC if a DCL command file bulds and writes a (sub-)command file that
>> handles symbols, there is use for more than three double-quotes. (do I
>> remember groups of six?)
>>
>
>Sure, if you need them. No problem to find examples. This is from the
>Oracle Rdb "Bugcheck dump summarizer":
>
>
>$ if error_area_name .nes. ""
>$ then
>$ call get_more_details -
> """Storage area """"''error_area_name'""""""/exact" 0 8
>$ endif

Passing a quoted string with a "quoted" string embedded in it. How's this any
different in a command file from the command line, other than that you can not
pass it to the subroutine?

--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.

Den 2022-01-25 kl. 14:08, skrev VAXman-@SendSpamHere.ORG:
> In article <ssonkm$cn7$1@dont-email.me>, =?UTF-8?Q?Jan-Erik_S=c3=b6derholm?= <jan-erik.soderholm@telia.com> writes:
>> Den 2022-01-25 kl. 11:36, skrev Henry Crun:
>>> On 25/01/2022 11:35, gah4 wrote:
>>>> On Friday, January 21, 2022 at 4:51:45 AM UTC-8, VAXman- wrote:
>>>>
>>>> (snip, I wrote)
>>>>
>>>>>> The one I remember is that quoting on the DCL command line is
>>>>>> different from quoting in COM file of DCL commands.
>>>>>> Seems to be that it should be consistent between them.
>>>>> Short of where DCL reads its input (terminal or file via an RMS $GET), I
>>>>> don't
>>>>> see why/where processing should be any different.
>>>>
>>>> As I remember it, there are some places where you need triple quotes,
>>>> in place of regular quotes.   I believe it is related to symbols and
>>>> logical names.
>>>>
>>>
>>> IIRC if a DCL command file bulds and writes a (sub-)command file that
>>> handles symbols, there is use for more than three double-quotes. (do I
>>> remember groups of six?)
>>>
>>
>> Sure, if you need them. No problem to find examples. This is from the
>> Oracle Rdb "Bugcheck dump summarizer":
>>
>>
>> $ if error_area_name .nes. ""
>> $ then
>> $ call get_more_details -
>> """Storage area """"''error_area_name'""""""/exact" 0 8
>> $ endif
>
> Passing a quoted string with a "quoted" string embedded in it. How's this any
> different in a command file from the command line, other than that you can not
> pass it to the subroutine?
>

Yes, right. The fact that it in this case happend to be in a CALL is not
important. My point is that there is no magic about it, it just follows the
DCL quoting rules.

On 2022-01-24 18:24, Bill Gunshannon wrote:
> On 1/24/22 09:00, Simon Clubley wrote:
>> On 2022-01-21, Dave Froble <davef@tsoft-inc.com> wrote:
>>>
>>> I'll admit to learning much from overlays, but, it was still torture ..
>>>
>>
>> I learnt a lot from overlays, such as never wanting to go near the
>> bloody things ever again in my life. :-)
>
> Funny how I always hear this.  I have done lots of overlays on
> Ultrix-11 (even for the kernel!) and never had any serious problems
> with them.  I even like them.  :-)

There is a big difference between overlays in Ultrix (or BSD) and RSX
(well, TKB).

They are indeed very simple to use in Ultrix, but they are also much
less powerful.

Johnny

On 2022-01-24 23:22, Jan-Erik Söderholm wrote:
> Den 2022-01-24 kl. 18:24, skrev Bill Gunshannon:
>> On 1/24/22 09:00, Simon Clubley wrote:
>>> On 2022-01-21, Dave Froble <davef@tsoft-inc.com> wrote:
>>>>
>>>> I'll admit to learning much from overlays, but, it was still torture ..
>>>>
>>>
>>> I learnt a lot from overlays, such as never wanting to go near the
>>> bloody things ever again in my life. :-)
>>
>> Funny how I always hear this.  I have done lots of overlays on
>> Ultrix-11 (even for the kernel!) and never had any serious problems
>> with them.  I even like them.  :-)
>>
>> bill
>>
>>
>
> I never thought anything bad/good about them, the overlay feature
> was just that, a feature of RSX-11M. I think the most "overlayed"
> image I built was using both the RMS libraries and the DTR-11
> callable library. It was mainly DTR-11 that added most overlays,
> as I remember it. I think the FMS libs was used also...

The "worst" of it is when you need to juggle things a lot in order to
fit it all in. That can be quite an exercise. Multiple co-trees,
figuring out dependencies, and writing the ODL file for it all.

When I backported DUNGEON V3.2 to RSX, I really had to work hard on
that. Probably the most complex thing I've ever done. I think I ended up
with 10 co-trees, and all kind of cruft. I would agree that they are not
the most fun, but you sortof live with them and accept them. Sure, I can
understand people who felt relief when moving to VMS, and never having
to care about that again.

Johnny

On 1/27/2022 8:52 AM, Johnny Billquist wrote:
> On 2022-01-24 23:22, Jan-Erik Söderholm wrote:
>> Den 2022-01-24 kl. 18:24, skrev Bill Gunshannon:
>>> On 1/24/22 09:00, Simon Clubley wrote:
>>>> On 2022-01-21, Dave Froble <davef@tsoft-inc.com> wrote:
>>>>>
>>>>> I'll admit to learning much from overlays, but, it was still torture ..
>>>>>
>>>>
>>>> I learnt a lot from overlays, such as never wanting to go near the
>>>> bloody things ever again in my life. :-)
>>>
>>> Funny how I always hear this. I have done lots of overlays on
>>> Ultrix-11 (even for the kernel!) and never had any serious problems
>>> with them. I even like them. :-)
>>>
>>> bill
>>>
>>>
>>
>> I never thought anything bad/good about them, the overlay feature
>> was just that, a feature of RSX-11M. I think the most "overlayed"
>> image I built was using both the RMS libraries and the DTR-11
>> callable library. It was mainly DTR-11 that added most overlays,
>> as I remember it. I think the FMS libs was used also...
>
> The "worst" of it is when you need to juggle things a lot in order to fit it all
> in. That can be quite an exercise. Multiple co-trees, figuring out dependencies,
> and writing the ODL file for it all.
>
> When I backported DUNGEON V3.2 to RSX, I really had to work hard on that.
> Probably the most complex thing I've ever done. I think I ended up with 10
> co-trees, and all kind of cruft. I would agree that they are not the most fun,
> but you sortof live with them and accept them. Sure, I can understand people who
> felt relief when moving to VMS, and never having to care about that again.
>
> Johnny

Well, it's more than just "relief". Though there is a whole big bunch of that.
It's the capability to get on with the "job" without all the jumping through
hoops. With the expanded address space, it is not required to think about
fitting the task into memory, it's simply (as if) the straight progression of
what's required.

--
David Froble Tel: 724-529-0450
Dave Froble Enterprises, Inc. E-Mail: davef@tsoft-inc.com
DFE Ultralights, Inc.
170 Grimplin Road
Vanderbilt, PA 15486

Johnny Billquist <bqt@softjar.se> wrote:
> On 2022-01-24 23:22, Jan-Erik Söderholm wrote:
>> Den 2022-01-24 kl. 18:24, skrev Bill Gunshannon:
>>> On 1/24/22 09:00, Simon Clubley wrote:
>>>> On 2022-01-21, Dave Froble <davef@tsoft-inc.com> wrote:
>>>>>
>>>>> I'll admit to learning much from overlays, but, it was still torture ..
>>>>>
>>>>
>>>> I learnt a lot from overlays, such as never wanting to go near the
>>>> bloody things ever again in my life. :-)
>>>
>>> Funny how I always hear this.  I have done lots of overlays on
>>> Ultrix-11 (even for the kernel!) and never had any serious problems
>>> with them.  I even like them.  :-)
>>>
>>> bill
>>>
>>>
>>
>> I never thought anything bad/good about them, the overlay feature
>> was just that, a feature of RSX-11M. I think the most "overlayed"
>> image I built was using both the RMS libraries and the DTR-11
>> callable library. It was mainly DTR-11 that added most overlays,
>> as I remember it. I think the FMS libs was used also...
>
> The "worst" of it is when you need to juggle things a lot in order to
> fit it all in. That can be quite an exercise. Multiple co-trees,
> figuring out dependencies, and writing the ODL file for it all.
>
> When I backported DUNGEON V3.2 to RSX, I really had to work hard on
> that. Probably the most complex thing I've ever done. I think I ended up
> with 10 co-trees, and all kind of cruft. I would agree that they are not
> the most fun, but you sortof live with them and accept them. Sure, I can
> understand people who felt relief when moving to VMS, and never having
> to care about that again.
>
> Johnny
>

For big programs with many complex overlays it was hell to debug overlay
tree calling mistakes. You should only call to branches up the current tree
or down towards the root of the current branch. If you called across to
another parallel branch it all worked until you tried to return from the
call , where the rest of the calling routine had been overlaid and
obliterated. I remember several late nights debugging such for an early
digital mapping system on RSX11M.

Moving to the flat address space of VMS was a delight!

--
Paul at the paulhardy.net domain

On 2022-01-28 13:09, Paul Hardy wrote:
> Johnny Billquist <bqt@softjar.se> wrote:
>> The "worst" of it is when you need to juggle things a lot in order to
>> fit it all in. That can be quite an exercise. Multiple co-trees,
>> figuring out dependencies, and writing the ODL file for it all.
>>
>> When I backported DUNGEON V3.2 to RSX, I really had to work hard on
>> that. Probably the most complex thing I've ever done. I think I ended up
>> with 10 co-trees, and all kind of cruft. I would agree that they are not
>> the most fun, but you sortof live with them and accept them. Sure, I can
>> understand people who felt relief when moving to VMS, and never having
>> to care about that again.
>>
>> Johnny
>>
>
> For big programs with many complex overlays it was hell to debug overlay
> tree calling mistakes. You should only call to branches up the current tree
> or down towards the root of the current branch. If you called across to
> another parallel branch it all worked until you tried to return from the
> call , where the rest of the calling routine had been overlaid and
> obliterated. I remember several late nights debugging such for an early
> digital mapping system on RSX11M.

Well, technically you can't jump between branches. Not directly. (The
namespace for a different branch isn't visible, so you cannot refer to it.)

But I know what you mean. If you call up (down?) towards the root, you
can from there go down another branch, and then when you are returning,
at some point you will return to a branch that isn't in memory, and all
hell breaks loose.
And I agree, this is not nice to try and figure out.

I do have an application that actually is doing this, deliberately. But
the routine in the root is made aware from which branch it was called,
and before returning, it makes sure that branch is back in memory again.
But it's not the most fun thing...

> Moving to the flat address space of VMS was a delight!

Yes. From the overlay landscape it was. That was, however, also the path
towards huge software and libraries where you no longer can understand
what it all does. And from that perspective I prefer overlays... :-)

(Not to mention there are a bunch of things that I just feel are better
in RSX than VMS, but of course there are things better in VMS as well...)

Johnny