Path: utzoo!utgpu!attcan!uunet!husc6!bloom-beacon!mit-eddie!uw-beaver!uw-june!pardo
From: pardo@june.cs.washington.edu (David Keppel)
Newsgroups: comp.lang.c
Subject: Re: alloca wars
Keywords: alloca memory allocation
Message-ID: <5426@june.cs.washington.edu>
Date: 3 Aug 88 19:33:08 GMT
References: <3950010@eecs.nwu.edu> <62170@sun.uucp> <62363@sun.uucp> <8293@smoke.ARPA> <19895@cornell.UUCP> <5422@june.cs.washington.edu> <8301@smoke.ARPA>
Reply-To: pardo@june.cs.washington.edu (David Keppel)
Organization: U of Washington, Computer Science, Seattle
Lines: 9
Disorganization: U of Washington, Computer Science, Seattle

I wrote:
>  foo = alloc(); ... afree(foo);

Karl Haddock pointed out that afree's will not be called for any
frames whose scope is left by longjmp() of a dynamically included
(child) function.  Oh well, it seemed like a good idea at the time.
Sorry for the bandwidth.

	;-D on  ( Memory obfuscation/deallafreeing )  Pardo