[SPAM key] - Re: memory map for RT-11 v 5

Don North north at alum.mit.edu
Tue Jul 26 22:44:02 CDT 2016

On 7/26/2016 7:17 PM, Jerome H. Fine wrote:
>> On Sunday, July 24th, 2016 at 14::29:59 -0700, Don North wrote:
>> >On 7/24/2016 8:06 AM, william degnan wrote:
>>> >On Jul 24, 2016 8:58 AM, "Jerome H. Fine" <jhfinedp3k at compsys.to> wrote:
>>>>> >On Wednesday, July 20th, 2016 at 18:02:44 - 0400, william degnan wrote:
>>>>> Is there a minimum memory requirement for RT-11 v5?  I was discussing with
>>>>> Ray Fantini about it today, unsure...anyone know if 16K will work (from
>>>>> 000000).
>>>>> Bill
>>>> You need to be more specific!  Starting with V05.00 of RT-11 in 1983,
>>>> there were a total of 17 versions released up to V05.07 in 1998, including
>>>> sub-versions V05.01B, V05.01C, and V05.04A to V05.04G of RT-11.
>>>> Up until V05.05 of RT-11, RT11SJ.SYS required the least memory which
>>>> was replaced with RT11SB.SYS for V05.06 and V05.07 of RT-11.
>>>> Attempts to boot from RT11SJ.SYS under V05.00 of RT-11 with 24K
>>>> bytes of memory were successful.  Attempts to boot with 16K bytes of
>>>> memory were unsuccessful.  An RK05 was used as the disk drive since
>>>> it is close to the smallest device driver.  The answer to your question
>>>> about using 16K bytes of memory is NO for all versions of RT-11 starting with
>>>> V05.00 or RT-11.
>>>> Attempts to boot from RT11SJ.SYS under V04.00 of RT-11 with 24K
>>>> bytes of memory were successful.  Attempts to boot with 16K bytes of
>>>> memory were also successful.  An RK05 was used as the disk drive.  The
>>>> error message "Insufficient memory" is displayed, but some useful work
>>>> might be done with just 16K bytes of memory. However, you did not
>>>> ask if useful work being done was one of the criteria?
>>>> NOTE that I used the Ersatz-11 emulator to check the above details,
>>>> so there might be a difference with actual hardware.
>>>> If you have any more questions, please ask.
>>>> Jerome Fine
>>> Thanks for the details.  I had been trying to boot rt-11 v5.3 on a 16k core
>>> 11/40 using RL11 (rl02) and it did not work.   The system was unable to
>>> complete the initialization.  CPU diagnostics passed, I could load BASIC
>>> papertape.  RL11 working correctly.  In this context I posted my question.
>>> After I posted my message here I loaded up simh and emulated an 11 with
>>> 32k.  RT-11 v5.3 disk boots.  When I re-built the system and reduced to
>>> 16k, I could not boot, bombed.
>>> One thing to remember is 16KW in a pdp11 is not the same thing that simh
>>> refers to when one sets the CPU to 16K.  WWW do not all make this
>>> distinction clearly.  I get it, just making this comment for future readers
>>> of this thread.
>>> Bill
>> RT-11 v5.03 single job monitor boots fine and runs in just 32KB (16KW) of 
>> memory.
>> You need to be more specific about how you specify the memory configuration 
>> (words vs bytes).
>> DEC routinely specified everything in KW (words) but most users and tools use 
>> KB (bytes) nowadays.
>> Note that to force RT11SJ (vs RT11FB) to boot on the below image I booted 
>> first using FB
>> in a larger memory configuration, did a:
>> to force it to boot using the SJ monitor the next time.
> NOTE that while the above command is REQUIRED to perform a
> hardware boot (from boot ROMs on real hardware or from
> within the SimH emulator.  Starting at least with V04.00 of RT-11,
> a software boot (which uses DUP.SAV) is supported from within
> RT-11 from within any disk (with the files required to support being
> a System Disk) to boot any specific monitor file (in this example
> RT11SJ.SYS) using the RT-11 command:

Bill's original issue is he wanted to boot RT-11 v5 on a 16KW 11/40 system, so 
the only
option that would work would be to boot directly to RT11SJ. Booting into XM or FB
and then rebooting into SJ via executing the BOOT command is not possible in such
a memory configuration.

> Thus even if there has been the RT-11 command to set up a
> boot block on a specific disk for a specific RT-11 monitor
> file, it is possible to override the boot block and use any
> qualified RT-11 monitor file via an RT-11 software boot
> (which uses DUP.SAV).
> It is also possible from within RT-11 to force the use of the Primary
> Boot Code (which is placed in block zero via the COPY/BOOT
> command) via the RT-11 command:
> If the user includes a specific monitor in the command:
> then the specified monitor file is ignored and the boot code in
> block zero of DL1: (placed there via the COPY/BOOT command)
> is used instead along with the actual monitor file that had been
> specified during the COPY/BOOT command.
>> Don
>> PDP-11 simulator V4.0-0 Beta        git commit id: 4065f47f
> This portion of my reply contains many questions.  If possible,
> answers to all of them would be appreciated, not just the first
> question.
> Is this the latest version of SimH?  Also, is is possible to support
> a VT100 display.  I have heard that telnet can be used, but I have
> never been able to do so.  Since I always use the RT-11 SL:
> (Single Line Editor) device driver to support a command stack,
> it is essential to support a VT100 display, etc. in order to use
> the ARROW keys to move the cursor around.  Is it possible
> to support VT100 escape sequences when SimH is being used?
> If so, how is that done in general and what specific programs
> are required along with the specific commands to implement?
> I would need to know how to do this from a 64-bit Windows 10
> point of view since that is the only system that I will have available
> after the Windows 98SE system that I currently use to run RT-11
> is no longer available.  As a default, Ersatz-11 supports running
> via the Win32 variant under a 64-bit Windows 10 system.

This was done using a relatively recent SIMH commit (ie within the last
week or so). So it is about the latest version of SIMH available.

As to the other questions, I only use SIMH in basic terminal mode, so
I can't answer any of the emulated VT100 questions.

As to your second point, my usage of SIMH is via compiling it for the CYGWIN 32b
environment running on a Windows 7 64b system. So all my usage is thru the
CYGWIN command window, or sometimes thru a telnet (via TeraTerm).

My understanding is that sometime 'soon' Windows 10 will be getting an Ubuntu
Linux subsystem so that it will run compiled linux images directly. So you will get
a linux command line terminal in a window and can run images directly.  So
theoretically a precompiled SIMH distribution for Ubuntu linux could be downloaded
and run with no compilation (or porting to Windows) necessary.  At least this is
my understanding at this point, I haven't paid too much attention because I expect
to be on Windows 7 64b until the turn of the next century. And then I will decide if
I want to upgrade to Windows 42 256b, or not. :-)

>> CPU     11/34, FPP, idle disabled, autoconfiguration enabled
>>         32KB
>> NOTE: ctrl-E to exit to SIMH monitor
>> sim> boot rl1
>> RT-11SJ (S) V05.03
>> .sho all
> If I remember correctly, the above RT-11 command may be incorrect.
> I seem to remember that the following output is displayed as a result of
> the RT-11 command:
> unless, of course, the rest of the display was omitted and there was
> much more actually output.

The above command is correct for RT-11.  I just cut off the bottom part
of the output as it was irrelevant for this post, as the only relevant issue
was the fact that RT11SJ v5.3 will boot and run in a 16KW memory size.

>> RT-11SJ (S) V05.03
>> Booted from DL1:RT11SJ
>> USR   is set SWAP
>> EXIT  is set SWAP
>> KMON  is set NOIND
>> TT    is set NOQUIET
>> ERROR is set ERROR
>> SL    is set OFF
>> EDIT  is set KED
>> KMON nesting depth is 3
>> PDP 11/34 Processor
>> 32KB of memory
>> FP11 Hardware Floating Point Unit
>> Extended Instruction Set (EIS)
>> Memory Management Unit
>> 60 Cycle System Clock
> Jerome Fine

More information about the cctalk mailing list