Re: Mac 7200 (PPC) sees Promise controller.. BUT NO DRIVES

[Benjamin Herrenschmidt <benh@kernel.crashing.org>]

On Sat, 2003-04-12 at 17:14, Jessica Blank wrote:
> Well... okay. This is interesting.
> 
> Ben, your kernel patches simply don't like my machine. Maybe I'm doing
> something wrong, but I've configured around a billion trillion bajillion
> kernels in my life (since age 15) and I see nothing missing.
> 
> Plain old non-Ben-patched 2.4.20 kernels seem to boot fine on this thing.
> However, your patched kernel seems to shunt itself immediately into
> Runlevel 0 or Runlevel 6, since the last thing I see when I boot it up
> (sorry, the messages scroll by so fast I can't catch them all, and no,
> they DO NOT end up in /var/log/messages nor /var/log/syslog) is
> "Setting the System Clock using the Hardware Clock as reference..."
> ... which is generated by hwclock.sh / hwclockfirst.sh, two startup
> scripts (in /etc/init.d) which are only loaded by runlevels 0 and 6...

Make sure you have CONFIG_PPC_RTC and _not_ CONFIG_RTC and let me
know. If the kernel tries to access the legacy RTC chip on the ISA
bus, it will hang.

> This is not making me happy. :(
> 
> Any ideas what I might be doing wrong?
> 
> Again, for reference, this machine is a Mac 7200.
> 
> Oh, and I even tried removing all the PCI cards, effectively returning the
> Mac to "stock" condition (no IDE, just SCSI; no Tulip NIC, just the
> built-in MESH)... no go. Same results!
> 
> Any help?
> 
> On Fri, 11 Apr 2003, Benjamin Herrenschmidt wrote:
> 
> > On Fri, 2003-04-11 at 16:24, Jessica Blank wrote:
> > > I am having a big problem here.
> > >
> > > My nice little "CD burning server", under my bed, is crippled because the
> > > large (40GB) IDE drive in it simply does not detect.
> > >
> > > This is a KNOWN WORKING drive. I just pulled it out of a working system.
> > >
> > > I've tried compiling IDE Disk Support and the Promise controller driver
> > > (the two necessary modules, of course) as modules, and I've tried
> > > compiling them straight into the kernel. NEITHER WORKS.
> > >
> > > When I insert ide-mod and ide-disk, I get:
> > >
> > > Apr 11 10:19:28 gandalf kernel: Uniform Multi-Platform E-IDE driver
> > > Revision: 6.31
> > > Apr 11 10:19:28 gandalf kernel: ide: Assuming 33MHz system bus speed for
> > > PIO modes; override with idebus=xx
> > > Apr 11 10:19:28 gandalf kernel: PDC20246: IDE controller on PCI bus 00 dev
> > > 70
> > > Apr 11 10:19:28 gandalf kernel: PDC20246: chipset revision 1
> > > Apr 11 10:19:28 gandalf kernel: PDC20246: not 100%% native mode: will
> > > probe irqs later
> > > Apr 11 10:19:28 gandalf kernel: PDC20246: ROM enabled at 0x80810000
> > > Apr 11 10:19:28 gandalf kernel: PDC20246: (U)DMA Burst Bit DISABLED
> > > Primary PCI Mode Secondary PCI Mode.
> > > Apr 11 10:19:28 gandalf kernel:     ide2: BM-DMA at 0x0800-0x0807, BIOS
> > > settings: hde:DMA, hdf:DMA
> > > Apr 11 10:19:28 gandalf kernel:     ide3: BM-DMA at 0x0808-0x080f, BIOS
> > > settings: hdg:DMA, hdh:DMA
> > >
> > > .......however, as you can see, it does not detect the disk itself!
> >
> > Actually, I can't see as the disk probing happens later during boot...
> >
> > > This is very frustrating.
> > >
> > > And yes, I have already checked the connections to the disk. They are
> > > fine. :)
> >
> > Can you send me the complete dmesg log along with the version of the
> > kernel you are using ?
> >
> > Ben.
> >
> 
> J e s s i c a    L e a h    B l a n k
-- 
Benjamin Herrenschmidt <benh@kernel.crashing.org>