Help with RT-11/Fortran
Dimitris Karangelos
Hi all,
I'm trying to use SIMH to run a simple Fortran program
originally written for an LSI-11 running RT-11 v4, with
Fortran v2.1. I found a Fortran disk image here :
http://www.headcrashers.org/comp/rx01/index.html
which however does not include the executable for the
compiler, FORTRA.SAV. I'm guessing this is generated
by FORGEN.SAV on the disks, but I can't get it to work.
I have attached the output of the SIMH session below,
if anyone has an idea of what is wrong it would be most
helpful.
Thanks in advance,
Dimitris
-------------------------------------------------------
SIMH session follows:
>PDP-11 simulator V3.5-1
>
>sim> rest my_pdp11
>RX: buffering file in memory
>
>sim> boot rk0
>
>RT-11SJ V04.00C
>
>.D 56=5015
>
> SER.TXT
> o RT-11 Version 4. RT-11 V04 provides new hardware support
> major enhancements over Version 3B.
>
> e the HELP command; it describes the new options in many
> ilities.
>
> e using a terminal that requires fill characters,
> cation 56 with a Deposit command before proceeding with
> stallation. LA36 DECwriter II and VT52 DECscope terminals
> quire such modification.
>
>
>.D 56=0
>
>.RUN FORGEN
>Answer questions with:
>"?" or carriage return (<CR>) for information,
>YES(Y) for affirmation, anything else for NO.
>
>
>A maximum of 56 lines are allowed per listing page.
>Is this acceptable? Y
>A maximum of 136 characters are allowed in a formatted
>(ASCII) record. Is this acceptable? Y
>A maximum of 6 channels may be open at a given time.
>Is this acceptable? Y
>The FORTRAN compiler can compile system-specific OPEN and CLOSE
>statement keywords for RT-11, RSX-11, and RSTS/E systems. These
>statements are currently compiled for the following system: RT-11
>Is this acceptable? Y
>FORTRAN can produce inline code for EAE, EIS or FIS hardware,
>or it can produce threaded (THR) code which is hardware independent.
>Should the compiler produce EAE code? Y
>The inline compiler can optimize for SPEED or for SIZE. These
>optimizations are mutually exclusive. Do you want
>the optimization to be for SPEED? Y
>Do you wish an inline only compiler? Y
>
>?Err 27
>in routine ".MAIN." line ?
-----------------------------------------------------------------
Jerome H. Fine
Jerome Fine replies:
I am more familiar with how E11 runs RT-11. You
can download a hobby version of E11 from:
http://www.dbit.com/pub/e11/
If you still have the same problem, let us know.
As for the FORTRA.SAV file and the library, you
might have a corrupted version. If so, you can
legally run layered products for RT-11 which
were distributed up to V05.03 of RT-11 in 1983.
There is a file which contains all of these versions
of RT-11 at:
http://classiccmp.org/PDP-11/RT-11/dists/
The file is RT11DV10.ISO.zip
After you UnZip, the container file (RT11DV10.ISO) can
be used with both SIMH and E11. Since the container file
is meant to be burned to a CD which is then READ ONLY, it
is BEST to NEVER modify RT11DV10.ISO at any time. Simply
make the file READ ONLY.
If you still have problems, I will look for a distribution
of RT-11 that you can run with SIMH.
Incidentially, I am using FORTRAN 77 to calculate functions
associated with Prime Numbers. When I use numbers as large
as 10 ** 23, I often require 128 bit accuracy which even
REAL * 8 will not provide. I am therefore writing my own
subroutines in FORTRAN 77 and MACRO-11 which allow for
REAL * 32 = Unsigned INTEGER * 16 / FRACTION * 16
which ought to solve the problem. If not, then
REAL * 64 = Unsigned INTEGER * 32 / FRACTION * 32
will almost certainly be sufficient. If anyone has some
interest, please respond! This project is to allow me
to calculate pi(x) and li(x) up to x = 10 ** 22 with
sufficient accuracy to specify li(x) to 10 ** -6.
Sincerely yours,
Jerome Fine
--
To obtain the original e-mail address, please remove
the ten characters which immediately follow the 'at'.
If you attempted to send a reply and the original e-mail
address has been discontinued due a high volume of junk
e-mail, then the semi-permanent e-mail address can be
obtained by replacing the four characters preceding the
'at' with the four digits of the current year.