[122] in linux-scsi channel archive
NCR53c810 SCSI problems in Slackware 2.2
daemon@ATHENA.MIT.EDU (Camm Maguire)
Tue Mar 28 18:50:18 1995
Date: Tue, 28 Mar 95 13:47:47 -0500
From: camm@enhanced.com (Camm Maguire)
To: linux-scsi@vger.rutgers.edu
Please excuse me if this message is a duplicate. I'm having a few
mailer problems as well. Thanks!
=============================================================================
Greetings! I've just purchased a P75 EPS system, 16MB RAM, SCSI
controller with the NCR53c810 chip in the first PCI slot, #9GXE64
graphics card in the next PCI slot, Mitsumi CDROM off the floppy
controller, and with a Quantum Lightning 540MB SCSI drive as the only
SCSI device and hard disk in the system. The BIOS is Phoenix, with no
interrupts allocated to PCI slots 1,2,3 and IRQ 15 to slot 4.
Signalling is set to Level on all PCI slots.
I'm trying to install Slackware 2.2. On booting from the Slackware
bootdisk, the autoprobe seems to correctly identify the SCSI
controller and disk drive. The root or ramdisk is also loaded without
problem. On most attempts, I can then login and partition the drive.
The drive has a 100MB dos partition set as active, a 408MB linux
partition, and the rest a linux swap partition. Again, usually, I can
then run mkswap -c /dev/sda3 16864 and swapon /dev/sda3 without
problem. I them enter setup, and attempt to format the primary Linux
partition. The slow, checking format gives the following error. The
quick format proceeds without error, but attempts at installing the
software produce similar kernel panics. To repeat, this error was
generated by 1) booting Slackware 2.2 (scsi kernel) off a floppy, 2)
installing root disk (color144) as ramdisk, 3) mkdir /tmp/dos 4) mount
-t /dev/fd0H1440 /tmp/dos 5) cat /proc/version >/tmp/dos/log 6) cat
/proc/kmsg >>/tmp/dos/log 7) ^C 7.1) umount /tmp/dos 8) mkswap -c
/dev/sda3 16864 9) swapon /dev/sda3 10) setup 11) TARGET 12) (choose
options to slow format, with checking for bad blocks, /dev/sda2 with
inode density at 1/4096 bytes.) The partitions were created at an
earlier attempt.
I thought this may be a termination/cabling problem, but here I must
really show my ignorance. I inspected the cabling, and found 1 cable
originating on the controller card, having an intermediate connector
left open, and terminating at the disk drive. Similarly, the
controller card has a port left open which is accessible from the back
of the machine. I don't know if this is configuration is considered
terminated or not. The two *ends* of the cable are "terminated", but
I don't know if this is sufficient. I'll check out the document on
tsx-11 when its ftp becomes available again.
Please excuse the absence of the final stages of the bug reporting
format. I can't load gcc, the kernel, or anything else in order to
pinpoint the panic more accurately.
Could someone please help here?
Thanks!
=============================================================================
Linux version 1.2.1 (root@fuzzy) (gcc version 2.5.8) #3 Sun Mar 19 11:50:48 CST 1995
<6>Console: colour EGA+ 80x25, 1 virtual console (max 63)
<6>bios32_init : BIOS32 Service Directory structure at 0x000fff70
<6>bios32_init : BIOS32 Service Directory entry at 0xfdb07
<6>pcibios_init : PCI BIOS revision 2.00 entry at 0xfdb68
<6>Probing PCI hardware.
<6>Calibrating delay loop.. ok - 30.22 BogoMips
<6>Serial driver version 4.11 with no serial options enabled
<6>tty00 at 0x03f8 (irq = 4) is a 16550A
<6>tty01 at 0x02f8 (irq = 3) is a 16550A
<6>ftape: allocated 3 buffers aligned at: 00208000
<6>Floppy drive(s): fd0 is 1.44M
<6>FDC 0 is a post-1991 82077
<6>scsi-ncr53c7,8xx : at PCI bus 0, device 19, function 0
<6>scsi-ncr53c7,8xx : warning : revision of 2 is greater than 1.
<6>scsi-ncr53c7,8xx : NCR53c810 at memory 0x81000000, io 0x4000, irq 15
<6>scsi0 : using io mapped access
<6>scsi0 : using initiator ID 7
<6>scsi0 : using level active interrupts.
<6>scsi0 ; burst length 8
<6>scsi0 : using 40MHz SCSI clock
<6>scsi0 : m_to_n = 0x90, n_to_m = 0xa0, n_to_n = 0xb0
<6>scsi0 : NCR code relocated to 0x3882dc
<6>scsi0 : testing
<6>scsi0 : test 1 started
<6>scsi0 : tests complete.
<6>EATA (Extended Attachment) driver version: 2.31a
<6>developed in co-operation with DPT
<6>(c) 1993-95 Michael Neuffer neuffer@goofy.zdv.uni-mainz.de
<6>Registered HBAs:
<6>HBA no. Boardtype: Revis: EATA: Bus: BaseIO: IRQ: DMA: Ch: ID: Pr: QS: SG: CPL:
<6>scsi0 : NCR53c{7,8}xx (rel 3)
<6>scsi : 1 host.
<6> Vendor: QUANTUM Model: LIGHTNING 540S Rev: 241E
<6> Type: Direct-Access ANSI SCSI revision: 02
<6>Detected scsi disk sda at scsi0, id 0, lun 0
<6>scsi : detected 1 SCSI disk total.
<6>SCSI Hardware sector size is 512 bytes on device sda
<6>Memory: 13376k/16384k available (864k kernel code, 384k reserved, 1760k data)
<6>This processor honours the WP bit even when in supervisor mode. Good.
<6>Swansea University Computer Society NET3.019
<6>Swansea University Computer Society TCP/IP for NET3.019
<6>IP Protocols: ICMP, UDP, TCP
<6>Checking 386/387 coupling... Ok, fpu using exception 16 error reporting.
<6>Checking 'hlt' instruction... Ok.
<6>Linux version 1.2.1 (root@fuzzy) (gcc version 2.5.8) #3 Sun Mar 19 11:50:48 CST 1995
<6>Partition check:
<6> sda: sda1 sda2 sda3
<6>RAMDISK: 1474560 bytes, starting at 0x00220000
<5>VFS: Insert ramdisk floppy and press ENTER
<6>VFS: Disk change detected on device 2/28
<6>RAMDISK: Minix filesystem found at block 0
<6>RAMDISK: Loading 1440 blocks into RAM disk..........................................................................................
<6>done
<6>VFS: Mounted root (minix filesystem).
<6>VFS: Disk change detected on device 2/28
=============================================================================
(Manually copied from screen, hope all is accurate!)
unable to handle kernel NULL pointer deference at virtual address c0000038
current -> tss.cr3 = 0072d000, %cr3 = 0072d000
*pde = 00102067
*pte = 00000027
Oops : 0002
EIP : 0010: 00128320
EFLAGS: 00010202
eax: 00000010 ebx: 00000000 ecx: 00ad0d20 edx: 00000000
esi: 01800000 edi: 00000000 ebp: 0072ccf8 esp: 0072ccb4
ds: 0018 es: 0018 fs: 0026 gs: 0026 ss: 0018
Process badblocks (pid:86,process nr. 12,stackpage: 0072c0000)
Stack: 00000020 00000400 0072cf74 00ad0d5c 0000a000 000029a8 00000000 00000000
00000000 00000000 00000000 00ad0d5c 00000000 00000000 00000000 00000000
00000000 0072cc74 0012cee2 00000400 00000000 00000010 0072cd74 00eabd80
Call Trace: 0012cee2 0012862e 00150317 00127d06 0012862e 00127d06 0012862e
00112c8e 00129f10 00129e21 0011f824 0011001f 0011f824 001260c4 00110739
Code: 89 78 28 0f b7 41 0e 33 41 08 31 d2 f7 35 c0 af 1d 00 8b 35
--
Camm Maguire camm@enhanced.com
==========================================================================
"The earth is but one country, and mankind its citizens." -- Baha'u'llah
