Re: hung DS10 - WAS: Please test new kernel-image-2.6.10-alpha packages
On Fri, Feb 04, 2005 at 09:23:54AM -0500, Toni L. Harbaugh-Blackford [Contr] wrote:
>
> qla1280: QLA1040 found on PCI bus 0, dev 15
> scsi(1:0): Resetting SCSI BUS
> scsi1 : QLogic QLA1040 PCI to SCSI Host Adapter
> Firmware version: 7.65.00, Driver version 3.25
> Vendor: COMPAQ Model: BD036659CC Rev: 3B00
> Type: Direct-Access ANSI SCSI revision: 03
> scsi(1:0:0:0): Sync: period 10, offset 12, Wide, Tagged queuing: depth 255
The above are indication that the qla1280 driver has found the controller
and recognized the disk on the bus. It sees the chipset as 1040, which is
the most common found installed on factory DS10s.
> ERROR: SCSI host `isp1020' has no error handling
> ERROR: This is not a safe way to run your SCSI host
> ERROR: The error handling must be added to this driver
> fffffc00015d7df8 0000000000000000 fffffffc0038c500 fffffffc0035e920
> fffffffc0035e920 fffffc003fe0c000 fffffffc0035e920 fffffc003fe0c000
> fffffc0000000000 fffffc00015d7e40 fffffc00015d7e40 fffffc003fe0c000
> fffffc00005a4500 fffffc003fe0c000 fffffffc0035208c fffffffc0035e920
> fffffc00005a4500 fffffffc0035e920 fffffffc002ec04c fffffffc0035e920
> fffffc00005a4500 fffffffc0035e9f0 fffffc00005a4500 fffffc00005a48e8
> Trace:
> [<fffffc000034e0c4>] sys_init_module+0x284/0x3e0
> [<fffffc0000310b34>] entSys+0xa4/0xc0
These messages are triggered by invocation of the old qlogicisp driver;
they are meant to "nag" the author into providing a driver with the
requested level of error handling, which will most likely never happen
since there is no longer an author, or one who wishes to do so... :-\
> qlogicisp : new isp1020 revision ID (5)
> scsi2 : QLogic ISP1020 SCSI on PCI bus 00 device 78 irq 39 I/O base 0x8400
Obviously the qlogicisp driver is attempting to probe/scan for controllers
it handles, and it has found a 1040 (aka a 1020 rev 5 chip).
It identifies the controller it has found as "device 78"; the "78" is
really a (hex) device/function pair, constructed via:
#define PCI_DEVFN(slot,func) ((((slot) & 0x1f) << 3) | ((func) & 0x07))
and thus we can see, the "slot" it is in is 0x0f or decimal 15, so it is
the SAME device that the qla1280 driver has already claimed!
This, I believe, it the cause of the hang, as you've now got 2 drivers
laying claim to the same device...
Now, I believe both the qla1280 and qlogicisp driver must be coming
from the INITRD, as already suggested; do you have a log of what
drivers were placed into the INITRD when you rebuilt it?
Regardless, you've got to find a way to prevent both drivers from
being present.
Good luck.
--Jay++
---------------------------------------------------------------
Jay A Estabrook HPTC - LINUX support
Hewlett-Packard Company - ZKO2-3/N30 (603) 884-0301
110 Spit Brook Road, Nashua NH 03062 Jay.Estabrook@hp.com
---------------------------------------------------------------
Reply to: