FWD: qla1280 SCSI driver crash on visws

From: Andrey Panin,,,
Date: Fri Sep 12 2003 - 07:43:04 EST

Hello Jes,

qla1280 SCSI driver crashes in inetrrupt handler on visws

Please take a look.

Best regards.

----- Forwarded message from Paul Kaletta <paul@xxxxxxxxx> -----

From: Paul Kaletta <paul@xxxxxxxxx>
To: linux-visws-devel@xxxxxxxxxxxxxxxxxxxxx
X-Mailer: Ximian Evolution 1.0.8
Subject: [Linux-visws-devel] qla1280

(I'm sorry if you get this twice, but I send it out two days ago,
for some strange reason this message is not in the archives, which seem
to work now, and I didn't recieve any reply).

Hi,

I managed to get the 2.6.0-test5 kernel to work (Andrey: the part with
the VGA-Console still isn't right in test5! I had to apply your patch
for test3 manualy to make a console appear).

The qla1280 driver compiles now and recognizes my harddisk! Hurray *g*
Still something is wrong - during start-up the driver displays the
following stuff:

########
qla1280: QLA1080 found on PCI bus 1, dev 0
scsi(0): Reading NVRAM
scsi(0:0): Resetting SCSI BUS
bad: scheduling while atomic!
Call Trace:
[<c0118255>] schedule+0x3e5/0x3f0
[<c0123c63>] schedule_timeout+0x63/0xc0
[<c0123bf0>] process_timeout+0x0/0x10
[<c0214fd0>] qla1280_bus_reset+0xa0/0x110
[<c0214367>] qla1280_init_rings+0xf7/0x100
[<c0213c67>] qla1280_initialize_adapter+0x147/0x190
[<c021288c>] qla1280_do_device_init+0x22c/0x260
[<c0212a38>] qla1280_detect+0x178/0x200
[<c0352e7b>] init_this_scsi_driver+0x5b/0x110
[<c03407bb>] do_initcalls+0x2b/0xa0
[<c012b3ef>] init_workqueues+0xf/0x30
[<c01050a3>] init+0x33/0x190
[<c0105070>] init+0x0/0x190
[<c0107215>] kernel_thread_helper+0x5/0x10

scsi0 : QLogic QLA1080 PCI to SCSI Host Adapter: bus 1 device 0 irq 16
Firmware version: 8.15.00, Driver version 3.23.35
Vendor: SEAGATE Model: ST39103LW Rev: 0002
Type: Direct-Access ANSI SCSI revision: 02
scsi(0:0:0:0): Sync: period 10, offset 8, Wide, Tagged queuing: depth
255
SCSI device sda: 17783240 512-byte hdwr sectors (9105 MB)
SCSI device sda: drive cache: write back
/dev/scsi/host0/bus0/target0/lun0: p1 p2 p3
Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
#########

That looks pretty wrong to me. However I ignored it, created partitions
on the my SCSI drive, and tried to migrate my stuff from my ide drive to
the SCSI drive. The copying went pretty slow and halted frequently for
some time. After maybe 130 mb it haltet completly. I rebooted and tried
again. This time maybe after 400 megs I got this wonderful stack trace
and kernel panic:

#########
EIP is at 0x0
eax: df659074 ebx: 00000000 ecx: 00000004 edx: 00000003
esi: df634000 edi: 00000001 ebp: c033fee8 esp: c033fecc
ds: 007b es: 007b ss: 0068
Process swapper (pid: 0, threadinfo=c033e000 task=c02ecbe0)
Stack: c011831a df659074 00000003 00000000 c033e000 00000286 00000000
c033ff08
c011848e df5e1f3c 00000003 00000001 00000000 daee4c20 daee4d24
df62fa2c
c02137e5 daee4c20 00008020 c0120000 000a0000 00000001 c033e000
df62f1ac
Call Trace:
[<c011831a>] __wake_up_common+0x3a/0x60
[<c011848e>] complete+0x3e/0x70
[<c02137e5>] qla1280_done+0xa5/0x130
[<c0120000>] __check_region+0x40/0x50
[<c02132dc>] qla1280_intr_handler+0x8c/0xc0
[<c010b749>] handle_IRQ_event+0x49/0x80
[<c010bac7>] do_IRQ+0x97/0x140
[<c0106ff0>] default_idle+0x0/0x40
[<c0105000>] _stext+0x0/0x60
[<c0109d68>] common_interrupt+0x18/0x20
[<c0106ff0>] default_idle+0x0/0x40
[<c0105000>] _stext+0x0/0x60
[<c0107014>] default_idle+0x24/0x40
[<c01070a0>] cpu_idle+0x30/0x40
[<c034076c>] start_kernel+0x14c/0x160
[<c03404b0>] unknown_bootoption+0x0/0x120

Code: Bad EIP value.
<0>Kernel panic: Fatal exception in interrupt
In interrupt handler - not syncing
##########
btw. this is typed in manually. *cough*. It might contain typos.

I this a visws-issue (messy interrupt handling stuff?) or is something
in the qla1280 driver defunct?

Ciao
Paul Kaletta

----- End forwarded message -----

--
Andrey Panin | Linux and UNIX system administrator
pazke@xxxxxxxxx | PGP key: wwwkeys.pgp.net

Attachment: pgp00001.pgp
Description: PGP signature