From owner-freebsd-hackers  Sun Apr 23 17:43:16 1995
Return-Path: hackers-owner
Received: (from majordom@localhost)
          by freefall.cdrom.com (8.6.10/8.6.6) id RAA14988
          for hackers-outgoing; Sun, 23 Apr 1995 17:43:16 -0700
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34])
          by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id RAA14974
          for <hackers@freebsd.org>; Sun, 23 Apr 1995 17:42:40 -0700
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id KAA12354; Mon, 24 Apr 1995 10:39:20 +1000
Date: Mon, 24 Apr 1995 10:39:20 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199504240039.KAA12354@godzilla.zeta.org.au>
To: FreeBSD.org!hackers@implode.root.com, toor@jsdinc.root.com
Subject: Re: benchmark hell..
Cc: hackers@FreeBSD.org
Sender: hackers-owner@FreeBSD.org
Precedence: bulk

>> > Part of the execl problem is the Sun-OS style shared libs that we
>> > use.

I think this is being fixed by slowing down the Linux libraries to
support ELF :-).

>> To the contrary, *make*ing the Linux shlib's is a pain in the butt,
>> while it's a very easy process to create a SunOS-like shlib.

Apparently the benchmark doesn't test making the libraries :-).

>efficient.  We have done lots in the mmap code to help mitigate the speed
>issues, and it would be nice if someone could take on the task of improving
>the efficiency from the ld.so viewpoint someday.

I think making libgcc static already made them 10% faster for small
programs.

Bruce