From owner-freebsd-hackers  Wed Aug 14 17:37:42 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id RAA05465
          for hackers-outgoing; Wed, 14 Aug 1996 17:37:42 -0700 (PDT)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id RAA05460;
          Wed, 14 Aug 1996 17:37:37 -0700 (PDT)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.12/8.6.9) id KAA11973; Thu, 15 Aug 1996 10:28:58 +1000
Date: Thu, 15 Aug 1996 10:28:58 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199608150028.KAA11973@godzilla.zeta.org.au>
To: bde@zeta.org.au, nate@mt.sri.com
Subject: Re: sio issues (silo overflows on a pentium, locked in ttywait, etc...)
Cc: freebsd-hackers@freefall.freebsd.org, msmith@atrad.adelaide.edu.au,
        ponds!rivers@dg-rtp.dg.com, ponds!rivers@freefall.freebsd.org,
        sag.space.lockheed.com!handy@dg-rtp.dg.com
Sender: owner-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>> Here is my throwaway lkm and auxiliarly program for testing bus hogging.
>..

>> Compile and run the hogtime utility.  Start processes to exercise the
>> bus hog(s), e.g., `dd bs=1024k </dev/rsd0 >/dev/null'.  Watch the
>> output from hogtime.

>For my main disk, here are typical #'s:
>min = 23, av = 38, max = 58

>And for the second disk, which is an old 40M SCSI disk which is only
>used for swap.

>min = 23, av = 25, max = 59

>Are these #'s in ms?

us.  1 ms is a long time.

They are fairly large (I guess they are typical for ISA), but not a problem.

Only the DMA burst timing should matter.  It accumulates for multiple DMA 
devices.  On my ASUS-P133 NCR'810 system:

idle:                                13, 13, 16
reading rsd0 (bs=64k)                13, 14, 19-21
reading rsd0 && active de21040       13, 14, 20-22
reading rsd0 && rfd1                 13, 15, 22-24

Bruce