[1001] in linux-scsi channel archive
Re: ncr53c8xx panic
daemon@ATHENA.MIT.EDU (Gerard Roudier)
Sat Nov 23 05:59:50 1996
Date: Sat, 23 Nov 1996 12:55:41 +0000 (GMT)
From: Gerard Roudier <groudier@club-internet.fr>
To: Dan Hollis <goemon@sasami.anime.net>
cc: linux-scsi@vger.rutgers.edu, linux-raid@vger.rutgers.edu,
Stefan Esser <se@zpr.uni-koeln.de>
In-Reply-To: <Pine.LNX.3.95L01at.961123012727.19360A-100000@sasami.anime.net>
On Sat, 23 Nov 1996, Dan Hollis wrote:
> Linux 2.0.25
>
> From dmesg:
> -----------
> ncr53c8xx : at PCI bus 0, device 14, function 0
> ncr_attach: unit=0 chip=810 base=f0000000, io_port=6000, irq=12
> ncr53c810-0: using memory mapped IO at virtual address 0x488b000
> ncr53c810-0: assuming 40MHz clock
> ncr53c810-0: initial value of SCNTL3 = 03, final = 13
> ncr53c810-0: requesting shared irq 12 (dev_id=0xb10068)
> ncr53c810-0: restart (scsi reset).
> scsi0 : ncr53c8xx (rel 1.12c)
> scsi : 1 host.
> Vendor: SEAGATE Model: ST43400N Rev: 1027
> Type: Direct-Access ANSI SCSI revision: 02
> Vendor: SEAGATE Model: ST43400N Rev: 1022
> Type: Direct-Access ANSI SCSI revision: 02
> Vendor: SEAGATE Model: ST43400N Rev: 1027
> Type: Direct-Access ANSI SCSI revision: 02
> Detected scsi disk sda at scsi0, channel 0, id 1, lun 0
> Detected scsi disk sdb at scsi0, channel 0, id 2, lun 0
> Detected scsi disk sdc at scsi0, channel 0, id 3, lun 0
> ncr53c810-0-<target 1, lun 0>: FAST SCSI-2 100ns (10 Mb/sec) offset 8.
> SCSI device sda: hdwr sector= 512 bytes. Sectors= 5688447 [2777 MB] [2.8 GB]
> sda: sda1
> ncr53c810-0-<target 2, lun 0>: FAST SCSI-2 100ns (10 Mb/sec) offset 8.
> SCSI device sdb: hdwr sector= 512 bytes. Sectors= 5688447 [2777 MB] [2.8 GB]
> sdb: sdb1
> ncr53c810-0-<target 3, lun 0>: FAST SCSI-2 100ns (10 Mb/sec) offset 8.
> SCSI device sdc: hdwr sector= 512 bytes. Sectors= 5688447 [2777 MB] [2.8 GB]
> sdc: sdc1
> raid0 personality registered
> REGISTER_DEV sda1 to md0 done
> REGISTER_DEV sdb1 to md0 done
> REGISTER_DEV sdc1 to md0 done
> START_DEV md0 raid0
>
> When it crashed (I had to write this down by hand)
> ----------------
> ncr53c810-0:3: ERROR (80:48) (8-28-0) (8/13) @ (14c0:88030000).
sist = 48 bit 40 means "Arbitration complete"
bit 8 means "Gross error (overflow/undeflow)"
scsi-lines:
-----------
nc_socl = 8 SCSI-ATN
nc_sbcl = 28 SCSI-BSY + ATN
nc_sbdl = 0
14c0 is the script offset. For the moment, I have been too lazy to
disassemble script offsets, I will do it this week-end.
The source of the problem is very probably a hardware flaw.
(Cables, terminations, .... and tutti quanti)
I have experienced such "gross error" with my scsi system but rarely and
only when I use my CD/rom.
For now, I have 0.2 idea about the possible cause but I prefer first to ask
Stefan.
Help, Stefan! :)
The next messages show that error recovery is not satisfying and that I
should try to improve it.
Is it possible for you to upgrade the driver to latest revision:
ftp://tsx-11.mit.edu/pub/linux/ALPHA/scsi/ncrBsd2Linux-1.14c-src.tar.gz
ftp://linux.wauug.org/pub/roudier/ncrBsd2Linux-1.14c-src.tar.gz
That probably will not help for the problem, but it will be easier for me.
Can you try to use only 5MHz data transfer (1.14c allows to choose such
value in linux config, or use "setsync" for each device) and let me know
if it makes difference.
> script cmd =
18000000 > reg: da 10 80 13 47 08 03 1f 01 08 83 28 80 00 08 00.
> ncr53c810-0: restart (fatal error).
> ncr53c810-0-0-<target 3, lun 0>: COMMAND FAILED (9 ff) @00fed018.
> ncr53c810-0-0-<target 2, lun 0>: COMMAND FAILED (9 ff) @00fee810.
> ncr53c810-0-0-<target 1, lun 0>: COMMAND FAILED (9 ff) @00fee018.
> scsi : aborting command due to timeout : pid 6487419, scsi0, channel 0, id 3,
> lun 0 Write (6) 09 95 b0 02 00
> ncr53c8xx_abort : abort call
>
> Then the machine was locked hard. No ping, no console switching.
Gerard.
