Asus K8N-E and slow disk transfer

[Astrodog]

On Tue, 18 Jan 2005 15:11:43 -0600, Astrodog <astrodog at gmail.com> wrote:
> On Thu, 6 Jan 2005 16:52:35 +0000, Ganael Laplanche
> <ganael.laplanche at martymac.com> wrote:
> > Hi,
> >
> > Thank you for you answer :)
> >
> > I have a new 80 lead ATA cable (a brand new provided with my mobo) and my disk
> > is a single device on the channel (no slave)...
> >
> > Any (other) idea ?
> >
> > Ganaël LAPLANCHE
> > ganael.laplanche at martymac.com
> > http://www.martymac.com
> > Tel : (+33)6.84.03.57.24.
> >
> > ---------- Original Message -----------
> > From: Lars Tunkrans <lars.tunkrans at bredband.net>
> > To: Ganael Laplanche <ganael.laplanche at martymac.com>
> > Cc: freebsd-amd64 at freebsd.org
> > Sent: Wed, 05 Jan 2005 18:59:10 +0100
> > Subject: Re: Asus K8N-E and slow disk transfer
> >
> > > Ganael Laplanche wrote:
> > > > Hi all,
> > > >
> > > > I've just bought an ASUS K8N-E mobo. I'm using FreeBSD-5.3-Stable (amd64) and
> > > > suffering from *very* slow disk transfer rates. The chipset is an nforce3 and is
> > > > correctly detected at boot :
> > > >
> > > > # dmesg
> > > > [...]
> > > > atapci0: <nVidia nForce3 Pro UDMA133 controller> port
> > > > 0xffa0-0xffaf,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 at device 8.0 on pci 0
> > > > [...]
> > > >
> > > > My disk is an UDMA100 one, everything seems to be correct :
> > > > # atacontrol mode 0
> > > > Master = UDMA100
> > > > Slave  = BIOSPIO
> > > >
> > >
> > >    If you have a New and an OLD disk on  the same ATA chanell ;
> > >   Do you have an old 40 lead or an new 80 lead ATA cable ?
> > >
> > >   80 lead cable is a requirement för UDMA100 !
> > >
> > > //Lars
> > ------- End of Original Message -------
> > _______________________________________________
> > freebsd-amd64 at freebsd.org mailing list
> > http://lists.freebsd.org/mailman/listinfo/freebsd-amd64
> > To unsubscribe, send any mail to "freebsd-amd64-unsubscribe at freebsd.org"
> >
> 
> This may be a potential NForce3 Compatibility Issue.
> 
> You may try using atacontrol to force the "slave" to UDMA100 as well.
> 
> Also, at 11MB/s, its above the rate I'd expect to see if it was using
> PIO (Though, I may easily be wrong on that). Also, if you're copying
> from the disk, to the same disk, keep in mind, 11MB/s apparent, is
> 22MB/s of disc I/O.
> 
> What kind of CPU usage do you see during the copy? (time, uptime, and
> vmstat output would be helpful)
> 
> --- Harrison Grundy
> 

Disregard second to last paragraph. Just noticed you said 1MB/s,
instead of 11MB/s.
Sorry.