MLOAD
Description:
MLOAD is a transient command used to convert a "HEX" file (the
output of an assembler) into an executable "COM" file.
In its simplest form, MLOAD takes a program that is in valid Intel
hexadecimal format machine code and converts it into a command file that
can be executed. The file becomes filename.COM (the hexadecimal file is
filename.HEX). This form may be used as a replacement to the Digital
Research "LOAD" command.
The program may also be used to overlay portions of a "COM" file
with patches assembled to a new "HEX" file.
Syntax:
mload [<outnam=>][<file1.type>,]<file2>[,<file3...>] [bias]
<outnam> is the optional output file name.
<file1.type> is an optional non-HEX file to be patched by
subsequently named HEX files. Specifying the filetype enables this
function.
<file2> specifies the file to be loaded, unless a file1.type has
been specified. In that case, it specifies a HEX file to load.
<file3> specifies HEX file(s) to load.
<bias> is the offset within the saved image to apply when loading
the file.
MLOAD with no arguments prints a small help message. This
message is also printed whenever a command line syntax error
occurs.
ZCPR-style du specifications are fully supported, for both
input and output files. The following command lines are
permissible:
b3>mload a4:myfile.com=0:bigfil,b6:patch1,c9:patch2
a6>mload b5:=c3:mdm717.com,mdmpatch
Only the optional drive and user spec and the primary filename
may be specified. The output filetype is derived exclusively from
the 3-byte string at 103h within MLOAD.
If no output filename is specified, it will be derived from
the first input filename, with filetype of "COM", if not otherwise
specified (this default filetype may be patched directly into MLOAD
via DDT or with MLOAD itself, using a patch file - its location is
at 103H in MLOAD.COM). Note that a command line of the form
"C:=<FILENAME>" will place the output file on the "C" drive with
the same primary filename as the input file.
This feature obviating any need to use DDT. The first (and
only the first) filespec (after the "=", if used) may be non-HEX
and the filetype must be specified. Examples:
1. "mload MYPROG" loads MYPROG.HEX and writes the output to
MYPROG.COM.
2. "mload ws.com,wspatch" loads WS.COM, overlays it with
WSPATCH.HEX, and writes the output to WS.COM.
3. "mload MEXTEST=MEX114.COM,MXO-US13" loads MEX114.COM,
overlays it with MXO-US13.HEX, and writes the output file to
MEXTEST.COM. Note that this is the recommended technique because
it preserves the original file!
4. "mload ws.ovr,ovrpatch" loads WS.OVR and patches it with
"OVRPATCH.HEX".
Filenames may contain drive/user specs, and must not contain
wildcards. Input filenames must be separated by comas, and a
space is required between the last filename and the optional bias.
A load information summary is printed at the successful conclusion
of the load. Any errors in loading will generally include the name of
the file in question.
This program is a replacement for the CP/M "LOAD" program. Why
replace "LOAD"? well... LOAD.COM has a few deficiencies. For example,
if the hex file's origin is above 100h, LOAD.COM prepends blank space
to the output file to insure it will work as a CP/M transient. It
doesn't care if the file is not intended as a CP/M transient and it also
doesn't like hex records with mixed load addresses. For example, one
that loads below a previous record; which is a perfectly legitimate
happenstance. Also, LOAD.COM can load only one program at a time, and
has no provision for a load bias in the command specification. Finally,
there is no provision for user specification of the output file name.
In its simplest form, MLOAD's syntax is identical to LOAD.COM.
There should be no problem in learning to use the new program. The only
significant difference here is that, under LOAD.COM, all files are
output starting at 100h, even if they originate elsewhere. MLOAD
outputs starting at the hex file origin (actually, the first hex record
specifies the output load address). The bias option may be used to
override this.
An example should clarify this. Suppose you have a file that loads
at 1000h. LOAD.COM would save an output file that begins at 100h and
loads past 1000h (to wherever the program ends). MLOAD will save an
output file starting from 1000h only. If, for some reason, you need the
file to start at 100h in spite of its 1000h origin (I can think of
several circumstances where this would be necessary), you'd have to
specify a bias to MLOAD. Using this example, "MLOAD MYFILE 0F00"
would do.
Note that this program re-initializes itself each time it is run.
If your system supports a direct branch to the TPA (via a zero-length
.COM file, or the ZCPR "GO" command), you may safely re-execute MLOAD.
Error Messages:
Command line syntax error
Error in command line, re-enter and try again.
Ambiguous file name: <filename.ext> not allowed.
Ambiguous files are not allowed.
File <filename.ext> not found.
Input file could not be found, check spelling and try again.
Disk full.
No room on disk for output file.
Directory full.
No room in directory for output file.
Premature end-of-file in <filename.ext>
Error in input file, correct error, and try again.
Checksum error in <filename.ext>
Error in input file, correct error, and try again.
Can't close <filename.ext>
Problem closing file, disk may be in Read/Only status.
Memory full while loading <filename.ext>
Error in input file, correct error, and try again.
Format error in file <filename.ext>
Error in input file, correct error, and try again.
Writing <filename.ext>, nothing loaded