home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.os2.programmer
- Path: sparky!uunet!gatech!darwin.sura.net!zaphod.mps.ohio-state.edu!caen!batcomputer!msiadmin.cit.cornell.edu!bai
- From: bai@msiadmin.cit.cornell.edu (Dov Bai-MSI Visitor)
- Subject: Re: Division by 0 crashes OS2/2.0
- Message-ID: <1992Sep4.154434.10678@tc.cornell.edu>
- Sender: news@tc.cornell.edu
- Nntp-Posting-Host: msiadmin.cit.cornell.edu
- Organization: /usr/local/lib/news/organization
- References: <1992Sep3.131306.13751@tc.cornell.edu> <92Sep04.024410.14613@acs.ucalgary.ca> <92248.101602HXC6@psuvm.psu.edu>
- Date: Fri, 4 Sep 1992 15:44:34 GMT
- Lines: 30
-
- In article <92248.101602HXC6@psuvm.psu.edu> HXC6@psuvm.psu.edu (H. Choe) writes:
-
-
- >... Oh, if one suspects a hardware-related problem, I would suggest
- >that he may use f2c to check whether he can duplicate the problem.
-
- Another good check to do without resorting to C is to try the
- PARANOIA ( this evil spirit in action again -:) program from NELIB. It
- "specializes" in cases like that.
-
- >BTW, f2c with GCC/2 2.2 is very nice and fast in floating point calc.
- >Although it is a little slower than WATCOM, it is not slower than
- >f2c with GCC/EMX in my experience (GCC/2 2.1 was a bit slower).
-
- The main problem I had with f2c is poor debugging. You have to debug
- a barely readable C-code. It is all right if you just want to port
- a mainframe/workstation code with little or no changes, but I think
- that for serious development GDB (the GNU debugger) is not efficient.
-
- Also, I think ( not sure )that floating point exceptions are disabled
- in the EMX port. This does not assist in debugging numerical codes
- either.
-
- By the way, how are the debugging facilities of the WATCOM compiler ?
- How are the compared to MS CodeView ? I would appreciate any information
- on it.
-
- Thanks,
- Dov
-
-