mcdonald@aries.scs.uiuc.edu (Doug McDonald) (02/24/90)
I am trying to use Windows programs to do data collection on an experiment in my lab, so I can do data analysis at the same time. One of the programs needs to be interrupt driven. I wrote a program the included the necessary routine in the Windows program itself, and it ran under Windows 2.03 (using fixed segments) but failed under Windows 386. Putting the interrupt code in a TSR to run before Windows caused it to work in Windows 386. I communicate with it through a fixed location in low memory (presently 0x228) in which a pointer to the data structure in the TSR is stored. It seems to work fine. At this point please note --- ***** !!!!!! This program does not have to be portable, it does not have to run on future versions of Windows and nobody but me is ever going to try to read it!!! !!!!! ***** --- It just has to reliably take data, which it seems to do. Does anybody have any suggestions for testing things like this? How have other people done such things? Are things like this documented anywhere in the vast and obscure realms of the SDK books? Doug McDonald (mcdonald@aries.scs.uiuc.edu