commits
also consistently use INB over inb
Add missing 0, fix hyperlink to [loader.txt](docs/loader.txt)
This can be executed by loader.asm, and then anything sent to it
will be written to dataflash in page 0.
lpt routines come from msTERM which are needed instead of the
"brecvbyte" routine in the stock firmware.
I'm using a PCI parallel port now which isn't at 0x378, so now I can
override the port at runtime.
I couldn't find any source code to this program, so this is a new
version that does the same thing based on recvdump
I'm not sure why this was causing corruption on Windows
These can easily be viewed with 'hexdump -C' and typed into the
Mailstation's hex editor.
If /code, expect to read 1Mb of data and save it to codeflash.bin.
If /data, read 512kb and save it to dataflash.bin.
For some reason the inline adding was producing invalid bytes (0xfb
instead of 0xff) but doing separate steps works.
Also only print progress every 1000 bytes, since the win32 console
is very slow and most of the time spent dumping is waiting for the
printf to run.
spew dumps the codeflash, so add another version that dumps the
dataflash
These were originally written to be compiled with the Kingswood AS80
assembler, which is binary-only and win32-only. asz80 from sdcc is
open source and works on OpenBSD.
There is no license on this code, but it was all posted to the
public mailstation Yahoo group, so I am naively assuming it is in
the public domain.
https://groups.yahoo.com/neo/groups/mailstation
Except for win32/inpout32.dll, obtained from
http://www.highrez.co.uk/downloads/inpout32/ which states:
"This product is released as open source (Freeware)."
There is no license on this code, but it was all posted to the
public mailstation Yahoo group, so I am naively assuming it is in
the public domain.
https://groups.yahoo.com/neo/groups/mailstation
Except for win32/inpout32.dll, obtained from
http://www.highrez.co.uk/downloads/inpout32/ which states:
"This product is released as open source (Freeware)."