On Sunday, 96/11/24, Carsten Wimmer wrote to David Noon about "Magic
code for DosSleep" as follows:

CW> int 2Fh func 1680h worked great under Warp 2.x and 3..  But
CW> under Warp 4 I get 100% cpu-load again, even with int 2Fh..
CW> Any ideas?

Hi Carsten,

I'm still sorting it out under 3.0's VMB support. It'll be another
week or so before I get 4.0 installed. [Have to move NT out of the
way.]

At present, my main concern is with VMB sessions. These hammer the CPU
even when I have an ISR installed to issue the HLT.

Since I received no reply from this echo, I ended up wading through
object code using DEBUG in a VDM. The magic code hasn't changed, but
the register convention has. The AX register _must_ contain zero under
Warp. It is no longer the number of milliseconds for DosSleep() to
suspend the VDM.

However, it still doesn't work in a VMB session. [Mutter, mutter]
[Note that that is "mutter" in English, not a reference to my mother
in German. ... :-) ]

Regards

Dave


 * KWQ/2 1.2i * It works fine except when I am in Windows.
--- Maximus/2 3.01