GDB flag (-pg) changes behaviour of code
Original Reporter info from Mantis: FPCCore
-
Reporter name: FPC core team
Original Reporter info from Mantis: FPCCore
- Reporter name: FPC core team
Description:
Programs that error out with floating point errors (e.g. Divide by zero) do not error out or error out much later when compiled with gdb support (-pg).
The attached program:
* Always errors out on the assignment of Z when compiled without -pg
* Never errors out when compiled as-is with -pg
* Errors out on the StrToFloat if sufficent other code is inserted in place of the "writeln('test')"
--Markus
Additional information:
Reporter: Markus Roberts
EMail: Markus at reality dot com
Mantis conversion info:
- Mantis ID: 5931
- OS: Linux
- Fixed in version: 2.0.2