Thanks for the feedback. I’ve done all the tests for proper SCSI
termination in all possible ways, even keeping a single disk, and CDROM, and
I’m still getting the error below.
Worst still. I’ve disconnected the entire SCSI chain cable. And yet,
when booting my Woody 3.0 debian from my floppy containing the boot, the error
I would be tempted to think the controller has a physical error, but it does
not explain why it was all working under AIX, and I was able to
partition/initialize all disks perfectly.
It really seems like the boot which gets something corrupt in the
drivers. Doing the diagnostics using the SMS disk reports a
perfectly functioning hardware… I’m puzzed. I’ve also swapped
CDROM drive, cables, etc.
Any hint I could try? I’d like to try an install from CDROM but I can’t
manage to find how to actually boot from a CD, only from
Sent: August-29-12 5:18
To: Bruno Gallichand; debian-ppc
Subject: Fw: PowerPC
Debian install on IBM 43P boot problems...
Sent: Wednesday, August
29, 2012 3:15 PM
Subject: Re: PowerPC Debian
install on IBM 43P boot problems...
but anyway, searching the 2.6.8 kernel code
i can find no reference the the error
SCSI host 0
abort (pid 0) timed out – resetting
SCSI bus is being reset for host 0
Ncr53c8xx_reset : pid = 0
reset_flags = 2 serial_number=(increasing numbers)
serial_number_at_timeout= (increasing numbers)
but found a reference in the ncr
driver about status returned
** Check for no
terminators or SCSI bus shorts to ground.
Read SCSI data bus, data parity bits and control
** We are expecting RESET to be TRUE
and other signals to be
so isolating the 'boot' drive and
terminating the scsi chain can help.
----- Original Message -----
Sent: Wednesday, August 29, 2012 1:32
Subject: Re: PowerPC Debian install on IBM 43P boot
> On Tue, Aug 28, 2012 at 11:29 PM, dale-bresnan <firstname.lastname@example.org> wrote:
there were two types of scsi controller on the 43P systems
onboard is/was 53c810 which has a SE/LVD drive interface.
was 53c825 which is a HVD drive interface.
> Euh, according to
my memories (and sym_dev_table in
drivers/scsi/sym53c8xx_2/sym_hipd.c seems to agree), the 53c810 is a
> Fast SCSI chip, so it is old SE, not SE/LVD. LVD was only
> Ultra2 SCSI.
> Geert Uytterhoeven -- There's lots of Linux
beyond ia32 -- email@example.com
In personal conversations with technical people, I call myself a hacker.
> when I'm talking to journalists I just say "programmer" or
-- Linus Torvalds
> To UNSUBSCRIBE, email
with a subject of "unsubscribe". Trouble? Contact firstname.lastname@example.org
Archive: [🔎] CAMuHMdVynfPmD=J2OOM3tjjB2p-VN5bC58SZSKpT+0O3iuymOA@mail.gmail.com">http://lists.debian.org/[🔎] CAMuHMdVynfPmD=J2OOM3tjjB2p-VN5bC58SZSKpT+0O3iuymOA@mail.gmail.com