Best regards, Alexey Smirnov.
There are a number of 32 to 16 bit address conversion
functions in the "devhelp" group.
You may have to download the device driver development
kit from http://service.software.ibm.com/ddk/ to get the
headers that you need.
Lorne Sunley
>There are a number of 32 to 16 bit address conversion
>functions in the "devhelp" group.
Yes. Use them! Also, if you're programming for Aurora (or Warp 4 FP13+) you
can use the KEE functions, too. See Holger Veit's web page. DON'T just
assume that you can do some "quick and dirty" translation from 16:16<->32 bit
addresses, since this fails with packed code selectors.
Watcom (with a few bugs, which version are you using?) will do the
translations automatically if you prototype the 16 bit functions
properly (ie: with _Far16).
Can you give a code example of something that is failing?