Bug#141769: prep and boot-floppies (was bugid 141769)
ronald.iemenschot@bull.nl writes:
> I hope that you will take the time to read the rest.
> Hereby I will sent you again my test results sofar for the powerpc.
> I hope someone helps me out here.
I'm going to skip your potato reports since I'm only concerned with
woody. I know you're probbaly just trying to get *anything* installed
but we can only fix things in the software itself using the woody
versions.
Also, please retain the CC to the bug itself -- this way we can track
the report.
> Machine 1
> Bull Estrella 200 (Motorala Powerstack with utah board)
> PREP method PPCBUG firmware.
> PowerPC with 601 ? a 100 MHZ processor
Is this the machine you were using for the woody attempts?
> New attempt
> Woody
> Software used : Debian3.0.19PPCBeta
What is that? Is it possible to use boot-floppies 3.0.22 (the latest) ?
> Booting of the net with tftp, using root from floppy.
I don't quite understand this. If you boot on the net, you should
be able to use TFTP to get the root floppy, no?
> Saying that there are problems with the root floppy
> Booting from cdrom
> Saying that there are problems with the root floppy
What is the exact message?
> Did not test a compleet boot from floppy, this has to be done.
> If the results are known I will let you know if you are interested
Very interested.
> And finally my own machine
>
> Machine 2
> Bull Estrella 300 (Motorala Powerstack with utah board)
> PREP method openfirmware.
> PowerPC with 604e 200 MHZ processor
> Difference with the real motorola is the on-board vga which a Bull does not
> have.
[..]
> Second try
>
> Software used : Debian potato 2.2.3
>
> During the boot phase (flop,net and cdrom) I get the following messages
> concerning the scsi controller :
>
> ncr53c8xx : At PCI bus 0, device 12, function 0
> ncr53c8xx : setting PCI_COMMAND_IO PCI_COMMAND_MEMORY...
> ncr53c8xx : setting PCI_COMMAND_MASTER...(fix up)
> ncr53c8xx : changing PCI_LATENCY_TIMER from 0 to 80
> ncr53c8xx : 53c825a detected
> ncr53c825a-0 : rev 0x13 on pci bus 0 device 12 function 0 irq 15
> ncr53c825a-0 : ID 7, fast-io, parity checking
> scsi0 : ncr53c8xx-3.4.1-200000726
> scsi1: 1 host
> scsi aborting commmand due to timeout : pid 0, scsi0, channel 0, id 0, lun
> 0, test unit ready 00 00 00 00
> ncr53c8xx_abort : pid 0 serial_number=1 serial_number_timeout=1
>
> Than the machine hangs.
>
> The same applies to the woody release 3.0.19 beta when booting from the net
> or cdrom.
> Same message.
Ok, there might be a kernel boot-argument that you can use to work
around this. Again, I'm only interested really in results using the
latest boot-floppies.
This is a kernel issue and should be filed against
'kernel-image-2.2.20-prep'. Can you do this?
> The difference between DebianPrep6 and potato or woody is that after the
> scsi1 message the DebianPrep6 correctly detects the scsi devices on the
> bus.
>
> Last, but not least : You should know that it's had successfully running
> AIX on it, before starting to put linux on it.
> So there are no harware errors.
>
> Question 1) What has changed for the scsi controller between the releases ?
I can't say -- you'd have to look at the differences between the
kernel-image-* package we use in potato and kernel-image-2.2.20-prep
in woody.
> Question 2) Is there a way to disable the probing of the serial_number ?
/me doesn't know
> Question 3) Can I run the install in debugging mode to provide some more
> info ?
Possibly.
> Question 4) What can I do more to help you to help me ?
> Question 5) Is it nessesary to update the bug report and how ?
Yes, I would like a new bug specifically representing the boot error
using woody versions against kernel-image-2.2.20-prep. Include as
much hardware info as you can and also the mention that it runs fine
under AIX is very good information.
We will leave this bug here representing the *other* problems you are
having (which there seems to be a different problem) with loading the
root disk.
--
...Adam Di Carlo..<adam@onshore-devel.com>...<URL:http://www.onshored.com/>
--
To UNSUBSCRIBE, email to debian-boot-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Reply to: