Weird AV:s at app shutdown on slight misuse of madExcept
Posted: Fri Apr 03, 2020 9:11 am
Hello,
Could someone test this with madExcept 5.0 version.
Try to give best details I can.
1. Debug version of app (Debug define and so on)
2. link madExcept to project but not from .dpr, add madExcept, madLinkDisAsm files to uses in main form etc
- So the IDE-integration will not patch exe-file
3. use app for a while and close it
we get super weird AV:s with guite odd callstacks, but I always suspected this was a problem, removed the linkin to the Debug version and all errors disappear.
Can anyone verify this with their APP/environment
If Author will read this would be nice some other patching detection than those two inits in .dpr file, maybe some external file beside dpr that could also control it. We would like to use single unit where all madExcept related uses clauses and code relies. (As currently) on our release version this is OK because we patch it from command line on Jenkins server after build.
-Tee-
Could someone test this with madExcept 5.0 version.
Try to give best details I can.
1. Debug version of app (Debug define and so on)
2. link madExcept to project but not from .dpr, add madExcept, madLinkDisAsm files to uses in main form etc
- So the IDE-integration will not patch exe-file
3. use app for a while and close it
we get super weird AV:s with guite odd callstacks, but I always suspected this was a problem, removed the linkin to the Debug version and all errors disappear.
Can anyone verify this with their APP/environment
If Author will read this would be nice some other patching detection than those two inits in .dpr file, maybe some external file beside dpr that could also control it. We would like to use single unit where all madExcept related uses clauses and code relies. (As currently) on our release version this is OK because we patch it from command line on Jenkins server after build.
-Tee-