From owner-freebsd-hackers  Sun Nov 23 22:51:00 1997
Return-Path: <owner-freebsd-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id WAA09279
          for hackers-outgoing; Sun, 23 Nov 1997 22:51:00 -0800 (PST)
          (envelope-from owner-freebsd-hackers)
Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id WAA09274
          for <hackers@freebsd.org>; Sun, 23 Nov 1997 22:50:56 -0800 (PST)
          (envelope-from marcs@znep.com)
Received: from znep.com (uucp@localhost)
	by scanner.worldgate.com (8.8.7/8.8.7) with UUCP id XAA23080;
	Sun, 23 Nov 1997 23:50:46 -0700 (MST)
Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with SMTP id XAA29471; Sun, 23 Nov 1997 23:49:00 -0700 (MST)
Date: Sun, 23 Nov 1997 23:49:00 -0700 (MST)
From: Marc Slemko <marcs@znep.com>
To: Dmitry Khrustalev <dima@bog.msu.su>
cc: hackers@freebsd.org
Subject: Re: Serious performance issue with 2.2.5-RELEASE  (fwd)
In-Reply-To: <Pine.BSF.3.95.971123114156.22406G-100000@alive.znep.com>
Message-ID: <Pine.BSF.3.95.971123234226.22406W-100000@alive.znep.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

While the below change will result a better fit between Apache's 
memory pools and the system malloc(), I am somewhat skeptical
that it will have any noticable impact on 99% of the systems
and even more skeptical that it could have any real impact on 
the problems of this thread.

For those that aren't familiar with Apache's memory pools, what the below
change does is make Apache use 8k as the smallest size requested from
malloc() instead of a silly 8k + ~12 bytes. 

> ---------- Forwarded message ----------
> Date: Sun, 23 Nov 1997 11:19:31 +0300 (????)
> From: Dmitry Khrustalev <dima@bog.msu.su>
> To: Jaye Mathisen <mrcpu@cdsnet.net>
> Cc: David Greenman <dg@root.com>, hackers@freebsd.org
> Subject: Re: Serious performance issue with 2.2.5-RELEASE 
> 
> 
> 
> On Sat, 22 Nov 1997, Jaye Mathisen wrote:
> 
> > 
> > I do not believe so.  What ever it is definitely appears related to
> > swapping/paging somehow.
> 
> I beleive the following patch to apache should improve situation:
> 
> --- alloc.c.old	Sun Nov 23 11:15:31 1997
> +++ alloc.c	Sun Nov 23 11:16:34 1997
> @@ -199,7 +199,8 @@
>    /* Nope. */
>  
>    min_size += BLOCK_MINFREE;
> -  return malloc_block((min_size > BLOCK_MINALLOC) ? min_size : BLOCK_MINALLOC);
> +  return malloc_block((min_size > BLOCK_MINALLOC - sizeof(union block_hdr)) ?
> +     min_size : BLOCK_MINALLOC - sizeof(union block_hdr));
>  }