From owner-cvs-lib  Thu Sep 26 19:45:21 1996
Return-Path: owner-cvs-lib
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id TAA26936
          for cvs-lib-outgoing; Thu, 26 Sep 1996 19:45:21 -0700 (PDT)
Received: from silvia.HIP.Berkeley.EDU (ala-ca16-06.ix.netcom.com [204.32.168.134])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id TAA26023;
          Thu, 26 Sep 1996 19:43:14 -0700 (PDT)
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.7.6/8.6.9) id TAA18544; Thu, 26 Sep 1996 19:42:55 -0700 (PDT)
Date: Thu, 26 Sep 1996 19:42:55 -0700 (PDT)
Message-Id: <199609270242.TAA18544@silvia.HIP.Berkeley.EDU>
To: peter@freebsd.org
CC: CVS-committers@freebsd.org, cvs-all@freebsd.org, cvs-lib@freebsd.org
In-reply-to: <199609261422.HAA08895@freefall.freebsd.org> (peter)
Subject: Re: cvs commit:  src/lib/libfakegnumalloc Makefile
From: asami@freebsd.org (Satoshi Asami)
Sender: owner-cvs-lib@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

 *   Remove the nuke of /usr/lib/libgnumalloc*, this is closer to "How It Was

I thougt we agreed that this was the right thing to do.  If ANY
version of libgnumalloc.so.* exists in /usr/lib, it will be picked up
by the linker.

Most of the complaints came from people who didn't notice the commit
message urging them to update rc.

 *   Before", hopefully it won't cause as much hassle.  This now entirely
 *   confines it's activities to /usr/lib/compat/lib{fake,}gnumalloc.so.2.0

Besides, the new rm is redundant:

===
+         rm -f libgnumalloc.so.${SHLIB_MAJOR}.${SHLIB_MINOR} && \
          ln -fs libfakegnumalloc.so.${SHLIB_MAJOR}.${SHLIB_MINOR} \
                 libgnumalloc.so.${SHLIB_MAJOR}.${SHLIB_MINOR} )
===

(Note the "f" in ln's flags.)

Satoshi