Subj : COM Port in Use
To : Robert Wolfe
From : Mike Luther
Date : Tue Mar 29 2016 09:56 am
Well Maybe I can explain a bit but who knows?
RW> Hi all!
RW> Well, it looks like I got OS/2 Warp 4 installed in a VM and the BBS
RW> working with SIO 1.60d and VMODEM. Unfortunately, whenever I go to
RW> start the BBS, I am getting the popup saying that the virtual com port
RW> is in use when it really isn't. If I ignore through the error boxes,
RW> the BBS starts up just find and runs normally.
RW> Unfortunately, this popup is an annoyance when trying to run a front
RW> end mailer (ViaMail in this case) and it goes to load the BBS and the
RW> BBS exits back out to the mailer when a caller logs off.
RW> --- ViaMAIL!/WC v2.00
RW> * Origin: ViaMAIL! - Lightning Fast Mailer for Wildcat! (1:261/20)
I think few people have much knowledge of how OS/2 handles actual IRQ numbers
and ring level work that are really needed for carrying OS/2 back to way before
even DOS operations for some things. For example, 'normal' IRQ top level stuff
only applies to IRQ 'numbers' up to what I think I recall are number 8. That's
not enough to work with some things that we need for IRQ level numbers which go
up even way beyond number 8 especially when OS/2 is 'later' dealing with
multiple CPU operations that, as I think I recall this, are up to as many as
sixteen separate CPU's which then must be able to be coordinated with each
other as to what each CPU is doing that must be absolutely understood by any
action even down to COMM ports so that things don't violate the order of
operation for multiple CPU stuff that grew up long after the original ring
level and machine language code which I was originally working with LONG before
there was OS/2 even there.
OK, If, for example, as I do even today in many operations, use Warp 4 and MP4
later in the same computer for multiple COMM port stuff that does include BBS
operations, as well as my custom ham and commercial radio total operating stuff
that controls even the antenna direction and so on as well as integrated BBS
operations, I still use SIO very well. However,when I also totally log even
telephone line use in REAL TIME to all of this with special total digital phone
line trace work even to CNID track operations in sync with radio station
control, total real time double entry bookkeeping and even GPS location for
everywhere your fingertips are touching a body in medical work, I notice this,
By the way, all this is in my over 1,000,000 lines of source code work that
still goes down to ring level and machine language stuff even today!
When I boot the system, I obviously have to use SIO to set up VMWARE for both
BBS and even total telephone line real-time logging for all of this. When I
FIRST TURN ON the special phone line monitoring for every thing a phone company
does on the line for digital use to trace where it went, I notice that the
actual FIRST IRQ port which is tracked is port 184 if my memory is correct.
That's when SIO is first set up going to the special digital tracking device
that I'm sure none of you folks ever use. Well then on any NEXT access to the
'normal' COMM port, say number 3 or 4 that the special little unit is listening
to every phone line in the whole place for every touch to any number or
computer generated snoop on any line in real-time, the actual port that is used
is NOT 184, is say, port number 176. And that can VERY confuse ANYTHING down
to even pure CPU or multiple CPU computer unit operations. Which, to my
knowledge, ONLY OS/2 or down to even memory block operations can ever do and
Windows has NEVER been able to do after Billy Boy stole HeathDOS from them and
took Gordon Letwin away to bridge the lost of what is now Windows from IBM.
I have NO WAY of tracing this to what you folks are complaining about here but
suspect this sort of thing is exactly what may be causing this. Worse,unless
you can get the actual people that are now using whatever to run everyting in
what they all think is their neat hidden snoop for everything that is touched
down to evey CPM in NSA or whatever country, you will, I suspect never be able
to find it unless you can totally snoop them as well in a never to be able to
decrypt trick.
