NetNews Usenet Archive 1992 #18/NN_1992_18.iso/spool/comp/unix/ultrix/6378

Text File | 1992-08-20 | 1.6 KB | 35 lines

Newsgroups: comp.unix.ultrix
Path: sparky!uunet!haven.umd.edu!darwin.sura.net!mips!mips!decwrl!pa.dec.com!decuac!hussar.dco.dec.com!mjr
From: [e-mail] (Marcus J. Ranum)
Subject: Re: malloc
Message-ID: <[e-mail]>
Sender: [e-mail] (USENET News System)
Nntp-Posting-Host: hussar.dco.dec.com
Organization: Digital Equipment Corporation, Washington ULTRIX Resource Center
References: <[e-mail]>
Date: Fri, 21 Aug 1992 02:16:10 GMT
Lines: 22

[e-mail] writes:

>I am having trouble finding a problem with dynamicly allocated memory.
>Specificlly I will get unexplicable bus errors inside malloc after LOTS of
>dynamic memory manipulation. I put a layer of tracking software between my
>program and malloc, which should have cought such things as freeing memory that
>wasn't malloced, writing beyond the block of allocated memory and some other
>stuff. I just ended up crashing inside molloc in a differn't place.
>Does anyone have source for malloc that I could use to build a debug version
>and look at the internal structures and try to figure out what is happening??

    You could try using the mprof code from (as usual) decuac.dec.com,
in /pub/sources/mprof-3.0.tar.Z

    Mprof's pretty cool. It has a version of malloc that keeps memory
allocation related statistics, including some Very Cool stack unrolling
code that is wizard for helping find memory leaks. There is some rudimentary
checking in there for dup free()s and whatnot, and I'd dearly love for
some hero of the revolution to add tombstones and all that, but it isn't
going to be me. :)

mjr.