Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 18 Apr 1998 16:43:31 -0400
From:      Mark Mayo <mark@vmunix.com>
To:        Mike Smith <mike@smith.net.au>
Cc:        stable@FreeBSD.ORG, hardware@FreeBSD.ORG
Subject:   Re: best wdc0 flags ?
Message-ID:  <19980418164331.08912@vmunix.com>
In-Reply-To: <199804181645.JAA04953@antipodes.cdrom.com>; from Mike Smith on Sat, Apr 18, 1998 at 09:45:53AM -0700
References:  <19980418023307.34709@vmunix.com> <199804181645.JAA04953@antipodes.cdrom.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Apr 18, 1998 at 09:45:53AM -0700, Mike Smith wrote:
> > Hi. I just put a shiney new IDE (Ultra33) drive in a machine, and
> > I'm wondering what the best flags to use for the drive are.. I realize
> > that I can't get DMA on -STABLE (cvsupping right now), but with the
> > default LINT flags of:
[SNIP]
> We've been duplicating 3GB Western Digital disks the last few days under
> 2.2.6, flags 0x80ff80ff on 166MHz P5/430TX boards, and averaging about
> 5M/sec throughput.  Try  'dd if=/dev/wd1 of=/dev/null bs=1m' to get an 
> idea of your raw disk speed as opposed to filesystem throughput.

First, relavent dmesg output:

  FreeBSD 2.2.6-STABLE #0: Sat Apr 18 15:50:05 EDT 1998
  CPU: Pentium (167.07-MHz 586-class CPU)
  real memory  = 67108864 (65536K bytes)
  chip0 <Intel 82439TX PCI cache memory controller> rev 1 on pci0:0:0
  chip2 <Intel 82371AB IDE interface> rev 1 on pci0:7:1
  wdc0 at 0x1f0-0x1f7 irq 14 flags 0x80ff80ff on isa
  wdc0: unit 0 (wd0): <QUANTUM FIREBALL SE4.3A>, 32-bit, multi-block-16
  wd0: 4110MB (8418816 sectors), 14848 cyls, 9 heads, 63 S/T, 512 B/S

Now the mega-crappy performance I'm still getting:

vnode:{77}~ % dd if=/dev/wd0 of=/dev/null bs=1m
^C76+0 records in
76+0 records out
79691776 bytes transferred in 36.231349 secs (2199525 bytes/sec)

vnode:{83}~ % iozone 165 17000
...
IOZONE performance measurements:
   2438356 bytes/second for writing the file
   3048197 bytes/second for reading the file

Hmmrf. I'm sure the 3.0 read was just due to caching.. (64MB of pretty much
unused RAM).

vnode:{86}~ % bonnie -s 200
File './Bonnie.323', size: 209715200
Writing with putc()...done
Rewriting...done
Writing intelligently...done
Reading with getc()...done
Reading intelligently...done
Seeker 1...Seeker 2...Seeker 3...start 'em...done...done...done...
              -------Sequential Output-------- ---Sequential Input-- --Random--
              -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Machine    MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU  /sec %CPU
          200  1833 36.3  2384 17.4  1168  5.3  2097 31.4  2406  6.4  73.5  4.6

Man, either this drive really sucks, or something weird is going on! :-)
The motherboard is the highly recomended Abit, DIMM RAM, etc.. STREAM
is still showing memory throughput over 100MB/s, so that's working fine,
and the old 1GB SCSI disk in the system is still pumping out about 4MB/s
with the above tests..

If anyone has any ideas why the latest greatest IDE drive from Quantum is
blowing chunks all over my I/O, give me a shout! :-)

-Mark

> 
> -- 
> \\  Sometimes you're ahead,       \\  Mike Smith
> \\  sometimes you're behind.      \\  mike@smith.net.au
> \\  The race is long, and in the  \\  msmith@freebsd.org
> \\  end it's only with yourself.  \\  msmith@cdrom.com
> 

-- 
------------------------------------------------------------------------
 Mark Mayo		  				mark@vmunix.com       
 RingZero Comp.  	  		    http://www.vmunix.com/mark 

	 finger mark@vmunix.com for my PGP key and GCS code
------------------------------------------------------------------------
 "The problem is how do you build tools that understand your programs
  at a deeper semantic level." - James Gosling

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19980418164331.08912>