I've been attempting to interface with SimH in the least invasive way
possible, by connecting two programs together via a pseudo-TTY created by
socat connected to one of the telnet ports of a TTIX device. Something like
this: TTIX -> telnet port -> socat -> /dev/ttyPDP (pty)
This works fine for the console port. I personally like having minicom
pulled up for the serial terminal but keep the other window cluttered with
SimH backend commands. I have been able to manipulate EDIT.SV with a simple
C program to transfer text files, character by character, which is neat.
I wanted to extend this to a VC8-E emulator, again in the simplest way
possible. I am sending out the X- and Y-coordinates for the point over
TTOX0, but seem to be encountering some problems. I noticed that it likes
to get hung up on a TSF (6411) instruction, causing it to keep looping even
in single-step mode! Assuming everything is buffered, I would be under the
assumption that TSF would almost always skip with a successful device send.
[...]
Step expired, PC: 00313 (IOT 411) \ a lot more of these (~50)
Step expired, PC: 00314 (JMP 313) /
Step expired, PC: 00313 (IOT 411)
Step expired, PC: 00314 (JMP 313)
Step expired, PC: 00313 (IOT 411)
Step expired, PC: 00314 (JMP 313)
Step expired, PC: 00313 (IOT 411)
Step expired, PC: 00315 (IOT 416) <- finally skips!
By the way, I am initializing the device with a CLA; TLS (6416) at the
beginning of the program.
I noticed David Gesswein opened up a somewhat related issue, now marked as
closed, in October 2013.
https://github.com/simh/simh/issues/85
Once I'm back to the real hardware, I intend to try it there to verify. I
am running V3.9-0, by the way.
Any insight into what might be holding TSF up would be much appreciated.
Thanks,
Kyle