Re: nForce 4, SATA Drive only runs at UDMA33?

Here's a recap of all the things I've tried and
discovered in a bunch of testing today.

Tried mkIII "m" patches and that doesn't show atapici1
or atapici2 - they just show as GENERIC with drives as
UDMA33

Tried mkIII "n" patches and then atapici1 shows as
nForce4 with SATA drives but has further problems
detailed below.  atapici2 always shows up in dmesg as
GENERIC no matter what.

Tried custom kernel, disabling all other parts of ATA
with no difference.
# ATA and ATAPI devices
device		ata
device		atadisk		# ATA disk drives
#device		ataraid		# ATA RAID drives
#device		atapicd		# ATAPI CDROM drives
#device		atapifd		# ATAPI floppy drives
#device		atapist		# ATAPI tape drives
options 	ATA_STATIC_ID	# Static device numbering

Tried disabling the standard IDE ports in bios,
turning off DMA on SATA, and other bios tweaks with no
changes.

With the "n" patches I get randomly alternating:
lockup, Fatal trap 12 panic, or full boot but then it
can't find the root filesystem as the disk is showing
as detached.

If I get lucky and it goes most of the way I get
results like the following:
ata2: CONNECT REQUESTED
ata2: DISCONNECT REQUESTED
...  (lots of those!)
ad6: 70911MB <WDC WD740GD-00FLA2 31.08F31> at
ata3-master SATA 150
ad6: detached
ata3: DISCONNECTED
ata2: CONNECTED
ata2: SATA connect status=00000000
ata2: DISCONNECTED
ata3: CONNECTED
ata3: SATA connect ready time=0ms
...
Mounting root from ufs:/dev/ad6s1a
setrootbyname failed
ffs_mountroot: can't find rootvp
root mount failed:6

Is there something else I should try to help in
debugging this further?

Is there anything in -CURRENT that would help this to
work better than 5-STABLE plus the ATA mkIII "n"
patches?

Thanks for all the help!
--Alan Bryan






--- Doug White <dwhite_at_gumbysoft.com> wrote:
> On Fri, 20 May 2005, alan bryan wrote:
> 
> >
> > --- Doug White <dwhite_at_gumbysoft.com> wrote:
> > > Can you post the output of "pciconf -lv"? The
> nForce
> > > IDE controller is
> > > properly detected, but it looks like there's
> another
> > > one in the system.
> > > Looking at the spec for the system it may be the
> > > proprietary nVidia RAID
> > > controller.  The pciconf output should help us
> > > identify if thats the
> > > issue.
> >
> > FYI: RAID features are disabled in the BIOS, I'm
> just
> > trying to get a single SATA drive to work at full
> > speed.  The other drive in this system is IDE and
> that
> > seems to be working at proper speed.  Thanks for
> the
> > help!
> 
> I guess turning off the RAID converts the chips into
> standard SATA
> controllers. I'll have to look into that. An nForce
> 4 machine recently
> appeared at work, so I'll see what I can get it to
> do.
> 
> > atapci0_at_pci0:6:0:       class=0x01018a
> card=0x50361297
> > chip=0x005310de rev=0xa2 hdr=0x00
> >     vendor   = 'NVIDIA Corporation'
> >     class    = mass storage
> >     subclass = ATA
> > atapci1_at_pci0:7:0:       class=0x010185
> card=0x50361297
> > chip=0x005410de rev=0xa3 hdr=0x00
> >     vendor   = 'NVIDIA Corporation'
> >     class    = mass storage
> >     subclass = ATA
> > atapci2_at_pci0:8:0:       class=0x010185
> card=0xcb8410de
> > chip=0x005510de rev=0xa3 hdr=0x00
> >     vendor   = 'NVIDIA Corporation'
> >     class    = mass storage
> >     subclass = ATA
> 
> It looks like sos added support for atapci1 and 2 in
> this listing in the
> ATAmkIII patchset.  While that patchset is in
> -CURRENT you'll have to
> apply the -stable patches yourself. Search the list
> archives for the
> location, soren posts it now and again.
> 
> -- 
> Doug White                    |  FreeBSD: The Power
> to Serve
> dwhite_at_gumbysoft.com          |  www.FreeBSD.org
> 


		
__________________________________ 
Yahoo! Mail Mobile 
Take Yahoo! Mail with you! Check email on your mobile phone. 
http://mobile.yahoo.com/learn/mail