Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: MSVC code gen bug?

Author: Dann Corbit

Date: 11:39:50 01/01/02

Go up one level in this thread


On January 01, 2002 at 10:42:35, Jon Dart wrote:

>It wouldn't surprise me at all if it was a bug. One way to test this is to
>compile with optimization off and see if that code runs ok.

I have also seen this simply mask bugs that were real problems.  Runs fine in
debug and fails in release.  But despite that, the actual problem is in the user
code.

>But, I'd agree with other posters that memory overruns can also cause program
>crashes and other bad behavior. For that, I'd strongly recommend you invest in a
>copy of Bounds Checker. Your eyes (and ours on this board) are not as good at
>finding these problems as Bounds Checker, which is an automated tool.

You can also find a surprising amount of stuff with static checkers like PC-Lint
by Gimpel or the free LCLint.  In particular, PC-Lint does flow analysis and
will spot a large class of memory over-writes (and also gives a few false
alarms).



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.