[2134] in linux-scsi channel archive
Re: NCR 53c810 and EDO RAM ??
daemon@ATHENA.MIT.EDU (Gerard Roudier)
Wed Jul 9 21:41:07 1997
Date: Wed, 9 Jul 1997 21:32:18 +0200 (MET DST)
From: Gerard Roudier <groudier@club-internet.fr>
To: Juan Cespedes <cespedes@lander.es>
cc: linux-scsi@vger.rutgers.edu, ncr53c810@colorado.edu,
Stefan Esser <se@FreeBSD.org>
In-Reply-To: <Pine.LNX.3.96.970709113441.1215A-100000@leon.lander.es>
You ask if the messages you got are harmless. Indeed they are'nt.
In my opinion the new EDO SIMMs make problems with PCI to mem and/or
mem to PCI transfers.
> ncr53c810-0:0: ERROR (81:0) (8-aa-0) (e0/13) @ (100043c:00000000).
^
Bit 1 of DSTAT register means 'illegal or reserved op code detected'.
This can happen in various situations and especially if the NCR
SCRIPT processor get problems fetching instructions from the main memory.
> ncr53c810-0-<target 0, lun 0>: phase change 6-2 1@00093cb4 resid=1.
This message means that the target switched from MESSAGE_OUT PHASE (6)
to COMMAND PHASE (2) after having accepted all messages bytes but 1 byte,
so without reporting errors with appropriate messages like MSG REJECT.
That seems very strange to me and could, in my opinion, be related to
some PCI<->mem transfer problem.
Stefan has probably better explanations than mine about the possible
causes of the problem you report.
Regards, Gerard.
PS: If you are using linux-2.0.27-29, you can update your kernel tree to
a recent driver version. That will not fix your problem but would
avoid some other potential ones.
URL : ftp://linux.wauug.org/pub/roudier/README
On Wed, 9 Jul 1997, Juan Cespedes wrote:
>
> Hi. I own a NCR53c810 SCSI card, and here are the messages
> that appeared on the screen till today: (they were all OK)
>
> ---------------------------------------------------------------------
> ncr53c8xx : at PCI bus 0, device 9, function 0
> ncr_attach: unit=0 chip=810 base=e0800000, io_port=6100, irq=10
> ncr53c810-0: using memory mapped IO at virtual address 0x2805000
> ncr53c810-0: assuming 40MHz clock
> ncr53c810-0: initial value of SCNTL3 = 03, final = 13
> ncr53c810-0: requesting shared irq 10 (dev_id=0xc068)
> ncr53c810-0: restart (scsi reset).
> scsi0 : ncr53c8xx (rel 1.12d)
> scsi : 1 host.
> Vendor: QUANTUM Model: FIREBALL1080S Rev: 1Q08
> Type: Direct-Access ANSI SCSI revision: 02
> Detected scsi disk sda at scsi0, channel 0, id 0, lun 0
> scsi : detected 1 SCSI disk total.
> ncr53c810-0-<target 0, lun 0>: FAST SCSI-2 100ns (10 Mb/sec) offset 8.
> SCSI device sda: hdwr sector= 512 bytes. Sectors= 2134305 [1042 MB] [1.0 GB]
> ---------------------------------------------------------------------
>
> I had 2 FPG SIMMs of 16Mb each, but they didn't work well, so
> today I changed them with 2 EDO SIMMs, 16Mb each. I haven't made any
> other change. Now the messages from the SCSI are:
>
> ---------------------------------------------------------------------
> ncr53c8xx : at PCI bus 0, device 9, function 0
> ncr_attach: unit=0 chip=810 base=e0800000, io_port=6100, irq=10
> ncr53c810-0: using memory mapped IO at virtual address 0x2805000
> ncr53c810-0: assuming 40MHz clock
> ncr53c810-0: initial value of SCNTL3 = 03, final = 13
> ncr53c810-0: requesting shared irq 10 (dev_id=0xc068)
> ncr53c810-0: restart (scsi reset).
> scsi0 : ncr53c8xx (rel 1.12d)
> scsi : 1 host.
> ncr53c810-0: SCSI phase error fixup: CCB address mismatch (0x0000caf8 != 0x00000000)
> Vendor: QUANTUM Model: FIREBALL1080S Rev: 1Q08
> Type: Direct-Access ANSI SCSI revision: 02
> Detected scsi disk sda at scsi0, channel 0, id 0, lun 0
> scsi : detected 1 SCSI disk total.
> ncr53c810-0: SCSI phase error fixup: CCB address mismatch (0x00093810 != 0x0000caf8)
> ncr53c810-0-<target 0, lun 0>: asynchronous.
> ncr53c810-0:0: ERROR (81:0) (8-aa-0) (e0/13) @ (100043c:00000000).
> reg: da 10 c0 13 47 e0 00 1f 01 08 00 aa 80 00 0a 00.
> ncr53c810-0: restart (fatal error).
> ncr53c810-0-<target 0, lun 0>: COMMAND FAILED (9 ff) @00093810.
> ncr53c810-0: restart (ncr dead ?).
> ncr53c810-0-<target 0, lun 0>: asynchronous.
> SCSI device sda: hdwr sector= 512 bytes. Sectors= 2134305 [1042 MB] [1.0 GB]
> ---------------------------------------------------------------------
>
> followed by *a lot* (thousands) of lines like this one:
>
> ---------------------------------------------------------------------
> ncr53c810-0-<target 0, lun 0>: phase change 6-2 1@00093cb4 resid=1.
> ---------------------------------------------------------------------
>
> Appart from this, the system appears to work ok.
>
> Does anybody know the meaning of this? Are those messages
> harmless?
>
> Thank you for you attention :)
>
> Juan Cespedes
>
>
>
