On Thursday, 96/11/14, Leslie Rhorer wrote to David Noon about "Named
Pipes" as follows:

LR>         Sure, that's the basic idea.  All the synchronization should be 
LR> controlled by the OS/2 program.  Do you think it would be 
LR> best to let the VDM do its own internal timing after 
LR> initial synchronization, or have the OS/2 program produce a 
LR> heartbeat every 100ms?

Hi Les,

I would think the simplest approach would be for the timing to be
internal to the VDM. Providing the DOS program yields its timeslices
when its polling fails, thus keeping the OS/2 program active, there
should be no contention for the CPU and so timer events should be quite
punctual. A granularity of 100 msec should be easily achieved, since
that is 3 whole timeslices and a little more too.

LR>  DN> I'll fiddle around with QBASIC, since I don't own a copy of Power
LR>  DN> BASIC, to try it out here.
LR> 
LR>         'Should work, athough any DOS compiler should be OK.

Well, I've had to resort to a little bit of assembler, since the
semaphore API's seem to need 32-bit registers and 16-bit compilers
won't always generate such code. I need to contact IBM via Compuserve
to make sure I'm using the right calling sequence for the semaphore
stuff, but we are progressing.

Regards

Dave


 * KWQ/2 1.2i * PATH=C:\DOS;C:\DOS\RUN;C:\WIN\CRASH\DOS;C:\ME\DEL\WIN;C:\ME\USE\O

--- Maximus/2 3.01