home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cis.ohio-state.edu!pacific.mps.ohio-state.edu!linac!att!ucbvax!NSCVAX.PRINCETON.EDU!dragon
- From: dragon@NSCVAX.PRINCETON.EDU (Mighty Firebreather)
- Newsgroups: comp.os.vms
- Subject: RE: Desperately seeking memory corruptions
- Message-ID: <009601C2.A329B0A0.11915@nscvax.princeton.edu>
- Date: 4 Sep 92 14:09:57 GMT
- Sender: usenet@ucbvax.BERKELEY.EDU
- Organization: The Internet
- Lines: 33
-
-
- Cameron Laird <claird@Neosoft.com> writes:
- >
- >What support is there for C development (C++,
- >actually, but apparently it all uses the same
- >library) in regards to memory management? I
- >think I have a needle-in-a-haystack heap error,
- >and I'm close to the end of my rope in tracking
- >it down. If I were under UNIX, I'd use
- >debug/{malloc,mallocmap}.o, Saber [Centerline],
- >and/of Purify; if this were an MSDOS project
- >I'd have a number of IDEs from which to choose;
- >what do VMS folks do?
- >--
-
- At a DECUS Symposium I attended a few years ago, someone reported
- having written his own jacket routines for malloc and free. These jacket
- routines wrote an audit trail of memory allocated and freed. He used this
- to track down a memory leak by sorting the audit trails by memory address
- and matching entries to eliminate all of the malloc entries that had
- subsequently been freed. The remaining entries pointed to the allocator of
- all memory that had not been freed.
-
- Some similar approach might work for you.
-
- *************************************************************************
- * *
- * Here, there be dragons! *
- * dragon@nscvax.princeton.edu *
- * *
- * Richard B. Gilbert *
- *************************************************************************
-
-