WARNING: THIS SITE IS A MIRROR OF GITHUB.COM / IT CANNOT LOGIN OR REGISTER ACCOUNTS / THE CONTENTS ARE PROVIDED AS-IS / THIS SITE ASSUMES NO RESPONSIBILITY FOR ANY DISPLAYED CONTENT OR LINKS / IF YOU FOUND SOMETHING MAY NOT GOOD FOR EVERYONE, CONTACT ADMIN AT ilovescratch@foxmail.com
Skip to content

Conversation

@nealcrook
Copy link

Hi,
I was in the marked for a 6809 emulator and, resisting the temptation to start one from scratch I thought that I would try to build on the work you have done on exec09. I realise that you have not worked on this project for ages, but I wanted to check and see whether you are interested in pulling any of my changes into your repository. If you are, I will work to maintain the existing code structure etc. If not, I'll be a bit more relaxed about what I change.
regards,
Neal
Berkshire, UK.

configure and build stand-alone. Add simple build instructions to README.
work on watchpoints as well as breakpoints.
…p_machine

to print machine name and device list. Change printing of map so that it only
prints on interesting boundaries (basically, when anything changes). Add info
command that calls dump_machine. Update documentation accordingly. Tweak
serial.c to stop eon2 from core-dumping.
Changed the image file reader to auto-detect the file type (currently
supports hex and s19). Reworked the image file readers to cope with both
Unix and DOS line endings and to do a bit more checking (which also reduced
compilation warnings).
…age: map

ioexpand over a ROM, use one slot for I/O and map the remaining slots to
the "underlying" ROM by supplying an offset when mapping each slot.
…fault()

to fault() and move it to machine.c. Revise README accordingly.
turn off echo on i/o available to target system; target system is now
responsible for echo.
…outines

to fix-up keyboard codes for target machines that expect a DOS-like 0x0d as
the line ending.
…t the

write-only registers do not corrupt the ROM. Tidy up cpu_write.
yet. Updates to command.c from Erik, so that byte stream is printed
for list and trace-dump in addition to disassembly.
that is invoked (if it exists) by the previously unimplemented
du command.
currently start up properly using .hex files as they load over
the I/O and corrupt it. Works OK with .bin files. Problem with
.hex files will be fixed by defining a "startup" phase when they
are loaded. Lots of nasty hacks in this code at the moment; need
to think through some proper fixes to the problems exposed here.
Neal Andrew Crook and others added 30 commits September 13, 2016 11:54
Always exit with success - for now. Need this so that I can terminate the
simulator on reaching a cycle limit. Want a cleaner way of doing this in
order to run the emulator in a scripted loop for test4gcc6809.
…ve value

on branch. Don't know why I never saw this before.. nasty.
…nto trunk"

This reverts commit 0730ad9, reversing
changes made to f39131a.
…th long

branch that made PC go to ffffc156 instead of c156. Don't know why this didn't
come to light long ago.
There’s no /lib on OS X
…iated

with the memory mapping. NitrOS9 now gets through SysGo.
in disassembly. Thanks to Ole for the idea/code.
enabled; and vice-versa. Currently do not support enable of perodic
IRQ and FIRQ interrupts at the same time.
with the read-modify-write behaviour of the CLR instruction.. but only
when used with extended addressing.
Rx interrupts when the FIFO is non-empty. Has only undergone
brief testing but seems to work well enough not to break
polled code and to allow NITROS9 L2 to boot to the shell,
and FUZIX to continue to boot.
…t interrupts

in the mc09 machine. Scarily clean now! FUZIX and NITROS9 L1 (both of which
use timer and UART interrupts) both boot and appear to work correctly.
interrupts were enabled, regardless of whether a character was available.
Now, only indicate an Rx interrupt when interrupts are enabled AND the
Rx FIFO is non-empty.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants