Return-Path: Delivered-To: new-httpd-archive@hyperreal.org Received: (qmail 1200 invoked by uid 6000); 21 Dec 1997 20:29:21 -0000 Received: (qmail 1193 invoked from network); 21 Dec 1997 20:29:19 -0000 Received: from twinlark.arctic.org (204.62.130.91) by taz.hyperreal.org with SMTP; 21 Dec 1997 20:29:19 -0000 Received: (qmail 7783 invoked by uid 500); 21 Dec 1997 20:40:42 -0000 Date: Sun, 21 Dec 1997 12:40:42 -0800 (PST) From: Dean Gaudet To: new-httpd@apache.org Subject: Re: calling pstrdup() on value before table_set? In-Reply-To: <349D7890.B7ED019@Golux.Com> Message-ID: X-Comment: Visit http://www.arctic.org/~dgaudet/legal for information regarding copyright and disclaimer. Organization: Transmeta Corp. MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: new-httpd-owner@apache.org Precedence: bulk Reply-To: new-httpd@apache.org I was proposing using _end as a debugging tool only. I'm happy if debugging tools work on a majority of platforms, but not all platforms. Although I am tempted to make a hack optimization for myself that won't dup string constants, detecting them using _end... because I use unix, and I don't use dynamically loaded crap. (Anyone interested in performance doesn't use dynamically loaded crap, and does -DDYNAMIC_MODULE_LIMIT=0 :) Dean On Sun, 21 Dec 1997, Rodent of Unusual Size wrote: > Dean Gaudet wrote: > > > > table_set(t, "blahblah", "boohiss"). pstrdup()ing those is a waste of > > time, they're constants. > > So I suppose I could go along with your checking against _end.. > but how does that play with mod_so and dynamically-loaded (and > -unloaded) modules? Does _end work on Win32? If it's still safe > under all circumstances, cool - otherwise I'd prefer the certainty > of copying into the table's pool. > > #ken P-)} >