| PALBART(1) | General Commands Manual | PALBART(1) |
palbart - BART enhanced PDP8 crossassembler
palbart [options]inputfile
This manual page documents briefly the palbart command. It is a cross-assembler to for PDP/8 assembly language programs. It will produce an output file in bin format, rim format, and using the appropriate pseudo-ops, a combination of rim and bin formats. A listing file is always produced and with an optional symbol table and/or a symbol cross-reference (concordance). The permanent symbol table can be output in a form that may be read back in so a customized permanent symbol table can be produced. Any detected errors are output to a separate file giving the filename in which they were detected along with the line number, column number and error message as well as marking the error in the listing file.
The following file name extensions are used:
.pal source code (input)
.lst assembly listing (output)
.bin assembly output in DEC's bin format (output)
.rim assembly output in DEC's rim format (output)
.err assembly errors detected (if any) (output)
.prm permanent symbol table in form suitable for reading after the EXPUNGE
pseudo-op.
A summary of options is included below.
The additional features flag enables functions that are not compatible with the PAL8 assemble. Currently only the TITLE and BANK pseudo operations are enabled with this function. There are likely other incompatibilities that are not under this flag.
Assembler error diagnostics are output to an error file and inserted in the listing file. Each line in the error file has the form
<filename>(<line>:<col>) : error: <message> at Loc = <loc>
An example error message is:
bintst.pal(17:9) : error: undefined symbol "UNDEF" at Loc =
07616
The error diagnostics put in the listing start with a two character error code (if appropriate) and a short message. A carat '^' is placed under the item in error if appropriate. An example error message is:
17 07616 3000 DCA UNDEF
UD undefined ^
18 07617 1777 TAD I DUMMY
When an indirect is generated, an at character '@' is placed after the the instruction value in the listing as an indicator as follows:
14 03716 1777@ TAD OFFPAG
Undefined symbols are marked in the symbol table listing by prepending a '?' to the symbol. Redefined symbols are marked in the symbol table listing by prepending a '#' to the symbol. Examples are:
#REDEF 04567
SWITCH 07612
?UNDEF 00000
Refer to the code for the diagnostic messages generated.
Only a minimal effort has been made to keep the listing format anything like the PAL-8 listing format.
The RIMPUNch and BINPUNch pseudo-ops do not change the binary output file type that was specified on startup. This was intentional and and allows rim formatted data to be output prior to the actual binary formatted data. On UN*X style systems, the same effect can be achieved using the "cat" command, but on DOS/Windows systems, doing this was a major chore.
The floating point input does not generate values exactly as the DEC compiler does. I worked out several examples by hand and believe that this implementation is slightly more accurate. If I am mistaken, let me know and, if possible, a better method of generating the values.
This assembler was written to support the fleet of PDP-8 systems used by the Bay Area Rapid Transit System. As of early 1997, this includes about 40 PDP-8/E systems driving the train destination signs in passenger stations.
This assembler is based on the pal assembler by:
Douglas Jones <jones@cs.uiowa.edu> and
Rich Coon <coon@convexw.convex.com>
See the symbol table for the set of pseudo-ops supported.
See the code for pseudo-ops that are not standard for PDP/8 assembly.
Refer to DEC's "Programming Languages (for the PDP/8)" for complete documentation of pseudo-ops.
Refer to DEC's "Introduction to Programming (for the PDP/8)" or a lower level introduction to the assembly language.
If you don't like it the way it works or if it doesn't work, that's tough. You're welcome to fix it yourself. That's what you get for using free software.
This is free software. There is no fee for using it. You may make any changes that you wish and also give it away. If you can make a commercial product out of it, fine, but do not put any limits on the purchaser's right to do the same. If you improve it or fix any bugs, it would be nice if you told me and offered me a copy of the new version. Gary Messenbrink <gam@rahul.net>
Version Date by Comments
v1.0 12Apr96 GAM Original
v1.1 18Nov96 GAM Permanent symbol table initialization error.
v1.2 20Nov96 GAM Added BINPUNch and RIMPUNch pseudo-operators.
v1.3 24Nov96 GAM Added DUBL pseudo-op (24 bit integer constants).
v1.4 29Nov96 GAM Fixed bug in checksum generation.
v2.1 08Dec96 GAM Added concordance processing (cross reference).
v2.2 10Dec96 GAM Added FLTG psuedo-op (floating point constants).
v2.3 2Feb97 GAM Fixed paging problem in cross reference output.
v2.4 11Apr97 GAM Fixed problem with some labels being put in cross reference
multiple times.
v2.9 23Jun13 DJG David Gesswein djg@pdp8online.com
Combined versions found online and fixed many bugs.
Used source from v2.5 03Nov07 RK and
v2.6 14Jul03 PNT.
This manual page was written by Vince Mulhollon <vlm@debian.org>, for the Debian GNU/Linux system (but may be used by others).
| June 23, 2013 |