Bug 121857

Summary:	scsi-host with AM53C974 doesn't work
Product:	[openSUSE] SUSE LINUX 10.0	Reporter:	Peter Huyoff <peter.huyoff>
Component:	Kernel	Assignee:	POWERTeam@SuSE <power-bugs>
Status:	RESOLVED WONTFIX	QA Contact:	E-mail List <qa-bugs>
Severity:	Normal
Priority:	P5 - None
Version:	RC 1
Target Milestone:	---
Hardware:	PowerPC
OS:	Other
Whiteboard:
Found By:	Other	Services Priority:
Business Priority:		Blocker:	---
Marketing QA Status:	---	IT Deployment:	---
Attachments:	my .config

Description Peter Huyoff 2005-10-08 22:05:11 UTC

The TEKRAM DC390 Hostadapter works fine under MorphOS (PegasosII/G3). In the
past, under Debian Sarge (Kernel 2.4.7), i've no problems with the DC390. Under
SUSE10.0-ppc-RC1 something goes wrong. The following output ist found in
/var/log/boot.msg:

...
<6>scsi0 : Tekram DC390/AM53C974 V2.1d 2004-05-27
<3>DC390: DMA Blast aborted unfinished!
<5>  Vendor:           Model:                   Rev:     
<5>  Type:   Direct-Access                      ANSI SCSI revision: 00
<4>DC390: Abort command (pid 918, Device 04-00)
<4>DC390: SRB: Xferred 00000000, Remain 60000000, State 00000100, Phase 01
<4>DC390: AdpaterStatus: 00, SRB Status 00
<4>DC390: Status of last IRQ (DMA/SC/Int/IRQ): 0081c418
<4>DC390: Register dump: SCSI block:
<4>DC390: XferCnt  Cmd Stat IntS IRQS FFIS Ctl1 Ctl2 Ctl3 Ctl4
<4>DC390:  ffffa0   90   03   c4   00   82   0f   48   08   80
<4>DC390: FIFO: 30 30
<4>DC390: Register dump: DMA engine:
<4>DC390: Cmd   STrCnt    SBusA    WrkBC    WrkAC Stat SBusCtrl
<4>DC390:  83 60000000 705d911f 60000000 705d911f   00 031a4700
<4>DC390: Register dump: PCI Status: 0200
<4>DC390: In case of driver trouble read Documentation/scsi/tmscsim.txt
<6>DC390: Aborted pid 918
<4>DC390: Abort command (pid 918, Device 04-00)
<4>DC390: SRB: Xferred 00000000, Remain 00000000, State 00000000, Phase 04
<4>DC390: AdpaterStatus: 00, SRB Status 00
<4>DC390: Status of last IRQ (DMA/SC/Int/IRQ): 0081c418
<4>DC390: Register dump: SCSI block:
<4>DC390: XferCnt  Cmd Stat IntS IRQS FFIS Ctl1 Ctl2 Ctl3 Ctl4
<4>DC390:  ffffa0   90   00   c0   00   00   0f   48   08   80
<4>DC390: Register dump: DMA engine:
<4>DC390: Cmd   STrCnt    SBusA    WrkBC    WrkAC Stat SBusCtrl
<4>DC390:  00 60000000 705d911f 60000000 705d911f   00 03080000
<4>DC390: Register dump: PCI Status: 0200
<4>DC390: In case of driver trouble read Documentation/scsi/tmscsim.txt
<6>DC390: Aborted pid 918
<6>scsi: Device offlined - not ready after error recovery: host 0 channel 0 id 4
lun 0
<3>scsi0 (4:0): rejecting I/O to offline device
<3>scsi0 (4:0): rejecting I/O to offline device
<3>scsi0 (4:0): rejecting I/O to offline device
<5>sda : READ CAPACITY failed.
<4>sda : status=0, message=00, host=0, driver=06 
<4>sda : sense not available. 
<3>scsi0 (4:0): rejecting I/O to offline device
<3>sda: asking for cache data failed
<3>sda: assuming drive cache: write through
<5>Attached scsi disk sda at scsi0, channel 0, id 4, lun 0
<4>scsi_id[928]: 0:0:4:0: cannot open /tmp/scsi-maj8-min0-928: No such device or
address
<4>scsi_id[929]: 0:0:4:0: cannot open /tmp/scsi-maj8-min0-929: No such device or
address
...

Greetings, Peter Huyoff

Comment 1 Peter Huyoff 2005-10-08 22:13:50 UTC

Created attachment 52031 [details]
my .config

Comment 2 Olaf Hering 2005-10-10 14:32:09 UTC

you mean 2.four.7 works ok, but 2.six.13 does not? thats lot of changes.
Perhaps you mean 2.6.7. If so, please grab the kernel source 2.6.7 from
kernel.org and attach a diff between the plain kernel and the debian patched kernel.

Comment 3 Peter Huyoff 2005-10-11 19:05:39 UTC

(In reply to comment #2)
> you mean 2.four.7 works ok, but 2.six.13 does not? thats lot of changes.
> Perhaps you mean 2.6.7. If so, please grab the kernel source 2.6.7 from
> kernel.org and attach a diff between the plain kernel and the debian patched
kernel.

It was Kernel 2.4.27, that works without Problems. I've reviewed my old Notices
and found, that Kernel 2.6.8 causes Problems.

Here is the Link (german language) about that:

http://www.pegasosforum.de/viewtopic.php?t=1089&highlight=dc390

Yet, this Problem looks more like an Termination-Failure. At this time, i've
checked the Termination and found it ok - and with the older Kernel and under
MOS the SCSI-Host worked well.

Greetings, Peter

Comment 4 Olaf Hering 2005-11-08 08:51:32 UTC

I suggest you mail linux-scsi@vger.kernel.org 
I cant help you with that one.