From owner-freebsd-stable  Sun Oct 26 05:09:08 1997
Return-Path: <owner-freebsd-stable>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id FAA03910
          for stable-outgoing; Sun, 26 Oct 1997 05:09:08 -0800 (PST)
          (envelope-from owner-freebsd-stable)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id FAA03905
          for <stable@FreeBSD.ORG>; Sun, 26 Oct 1997 05:09:05 -0800 (PST)
          (envelope-from bde@zeta.org.au)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.7/8.6.9) id AAA21503; Mon, 27 Oct 1997 00:03:13 +1100
Date: Mon, 27 Oct 1997 00:03:13 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199710261303.AAA21503@godzilla.zeta.org.au>
To: anton@urc.ac.ru, stable@FreeBSD.ORG
Subject: Re: Kernel's nfs dramatically slows down a diskless workstation
Sender: owner-freebsd-stable@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>When I compile and install a new kernel for one, it becomes very-very slow.
>Systat -vmstat shows that amount of interrupts take up to 78% of time, and up
>to 700 interrupts per second are produced by ed0 installed on IRQ10.
>...
>pretty enough. Because of nfs.h and nfs_vnopts.h differ in only function
>prototypes and calls, possibly the critical changes are located in nfs_bio.c.

This seems to be fixed in rev.1.42 of nfs_bio.c in -current (ld does a
lot of small writes), and perhaps the i/o size of MAXBSIZE in 2.2.5 is
too big.

Bruce