[Simh] RE: Potential design for 'real time' controls

[Brian Knittel]

Here are my two bits on the current topics...

I've just done some work on the 1130 simulator involving
an outbound TCP/IP connection so anyone wanting to work on
that issue, I'd be happy to participate in the discussion.
I've found that sim_sock.c would benefit from a few tweaks
to make the process of establishing outbound connections easier.

I agree that multithreading would be a good addition to simh,
especially for things like GUIs and socket IO. I've used Windows-
specific It should be possible to write a simh threading API so that 
the same simulator code could work under both Unix and Windows. Does 
VMS support multithreading
under C? I know it has a good set of signalling objects.
I suspect that MacOS (9 and earlier) would be left out, but 
everyone's using OS X now anyway, right?

With a multithreaded approach it should be possible to
leave the existing CPU loop model as is; other threads could simply
set flags or events to indicate that they want to interact with the
CPU loop, then the CPU loop can respond when it's convenient.
Going to a one-instruction-per-call model for the CPU
I would imagine would have a significant performance hit.

Brian

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
_| _| _|  Brian Knittel
_| _| _|  Quarterbyte Systems, Inc.
_| _| _|  Tel:   1-510-559-7930
_| _| _|  Fax: 1-510-525-6889
_| _| _|  Email: brian at quarterbyte.com
_| _| _|  http://www.quarterbyte.com