From owner-freebsd-stable  Wed Jun 26 00:23:46 1996
Return-Path: owner-stable
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id AAA14405
          for stable-outgoing; Wed, 26 Jun 1996 00:23:46 -0700 (PDT)
Received: from zibbi.mikom.csir.co.za (zibbi.mikom.csir.co.za [146.64.24.58])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id AAA14396;
          Wed, 26 Jun 1996 00:23:39 -0700 (PDT)
Received: (from jhay@localhost) by zibbi.mikom.csir.co.za (8.7.5/8.7.3) id JAA03525; Wed, 26 Jun 1996 09:22:02 +0200 (SAT)
From: John Hay <jhay@mikom.csir.co.za>
Message-Id: <199606260722.JAA03525@zibbi.mikom.csir.co.za>
Subject: Re: Another try at the vm_pageout.c -stable diff
To: davidg@Root.COM
Date: Wed, 26 Jun 1996 09:22:01 +0200 (SAT)
Cc: dyson@freebsd.org, stable@freebsd.org
In-Reply-To: <199606260337.UAA00265@root.com> from David Greenman at "Jun 25, 96 08:37:52 pm"
X-Mailer: ELM [version 2.4ME+ PL16 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-stable@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

>    Here is yet another diff. This is the product of a discussion I just had
> with John about the nature of some of the performance problems people have
> complained about.
>    This makes a slight change to the policy for bringing pages off of the
> active queue and should make the statistics gather *much* more effective
> in some cases. Please try it out and get back to us. Thanks!
> 
> -DG
> 
> David Greenman
> Core-team/Principal Architect, The FreeBSD Project
> 
> (This includes John's other changes, so apply this to a fresh -stable copy of
> vm_pageout.c)
> 

I'm not sure how relevant this still is. I see that there was a few commits
during the night (well night for me here in SA).

David I have tried your patch, but the kernel dies with "panic: unwire:
page not in pmap" as soon as the machine starts to swap. It did not say
anything more.

I also tried the patch that John Dyson sent, but it also died. I have
written down the essentials, but I did not get a coredump.

| Fatal trap 12 page fault while in kernel mode
| fva = 0x7be05bb0
| Supervisor write
| ip = 0x8:0xf0186fa7
| The process running was top

| nm kernel | sort produces:
| f0186bd4 T _vm_map_lookup_done
| f0186c00 T _vm_map_simplify
| f0186d00 F vm_meter.o
| f0186d00 T _vmmeter
| f0186d60 T _loadav
| f0186dd0 T _vm_sysctl
| f0186eb0 T _vmtotal
| f0187060 F vm_mmap.o
| f0187068 T _sbrk
| f0187074 T _sstk

I am now running with just the "#if 0" changed to "#if 1" in vm_pageout.c
and will see how it go. We are already swapping 40MB and it still looks
good.

John
-- 
John Hay -- John.Hay@mikom.csir.co.za