home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.ms-windows.programmer.misc
- Path: sparky!uunet!magna!doc
- From: doc@magna.com (Matthew J. D'Errico)
- Subject: GetProcAddress/DLL -- Wierd results - Help !
- Message-ID: <Bzon8v.7Cy@magna.com>
- Organization: Magna Software Corporation
- Date: Tue, 22 Dec 1992 22:46:04 GMT
- Lines: 30
-
- This is a follow-up to a problem I've been trying to resolve...
- I STILL NEED HELP, though !
-
- I have a routine, InitLib() which is bound into several DLL's,
- and exported in the .DEF for each. What I use this for is to
- pass some data into the DLL... The InitLib() routine takes this
- data and depending upon the context, massages it, and loads it into
- Global variables which are created using GlobalAlloc w/ GMEM_SHARE...
-
- When I call LoadLibrary() in my .EXE, I follow it with a GetProcAddress()
- for "InitLib", and if it passes, I call the routine as a FARPROC
- function pointer with the appropriate data (all far pointers as well)...
-
- Here's the wierd bit:
-
- This routine, or one quite like it, is bound into several DLL's/
- In Windows 3.0, when I run the program, the data's not there, even
- though a couple of MessageBox displays in the InitLib() says it is...
-
- So I recompiled for CodeView, and when I set the breakpoint at or in
- InitLib for the DLL, I never hit it !! Even if I trace into InitLib()
- from the calling .EXE !! I *see* the breakpoint, but CV doesn't trip...
-
- Even wierder (to *me*, anyway), is that under Windows 3.1, this all
- works perfectly !! Arrrrgghh ! The Global memory's there, the
- CodeView breakpoints hit, everything's *perfect* !
-
- I need help... Desparately ! (before I really need MENTAL help!)
-
- -- Doc
-