azurewave / marvell ?

[Christopher Friedt]

Hi Peter,

On Wed, Jan 13, 2010 at 1:58 PM, Peter Wippich <pewi at gw-instruments.de> wrote:
> the Azurewave module is based on the Marvel 8686. They just make modules,
> not chips.

thanks for your quick reply ! Just within a couple of hours I was able
to get the module to load on my pxa device. It's working ok, but
there's a repeating bug

BUG: scheduling while atomic: lbs_main/2052/0x00000002
Modules linked in: libertas_spi libertas
[<c002fcec>] (unwind_backtrace+0x0/0xdc) from [<c0365ad4>] (schedule+0x80/0x37c)
[<c0365ad4>] (schedule+0x80/0x37c) from [<c036639c>]
(schedule_timeout+0x18/0x1ec)
[<c036639c>] (schedule_timeout+0x18/0x1ec) from [<c0366224>]
(wait_for_common+0xf0/0x1b8)
[<c0366224>] (wait_for_common+0xf0/0x1b8) from [<c01d1dc0>] (spi_sync+0x44/0x5c)
[<c01d1dc0>] (spi_sync+0x44/0x5c) from [<bf02e14c>]
(spu_write+0xb8/0xd0 [libertas_spi])
[<bf02e14c>] (spu_write+0xb8/0xd0 [libertas_spi]) from [<bf02e288>]
(if_spi_host_to_card+0xb4/0x154 [libertas_spi])
[<bf02e288>] (if_spi_host_to_card+0xb4/0x154 [libertas_spi]) from
[<bf00fd20>] (lbs_thread+0x7b8/0x98c [libertas])
[<bf00fd20>] (lbs_thread+0x7b8/0x98c [libertas]) from [<c0059ebc>]
(kthread+0x7c/0x84)
[<c0059ebc>] (kthread+0x7c/0x84) from [<c002afa0>] (kernel_thread_exit+0x0/0x8)


Also, when I run "rmmod libertas_spi" the following segfault happens.

libertas: PREP_CMD: command 0x001c failed: 1
Unable to handle kernel NULL pointer dereference at virtual address 00000000
pgd = c7bdc000
[00000000] *pgd=a6c8b031, *pte=00000000, *ppte=00000000
Internal error: Oops: 17 [#1] PREEMPT
last sysfs file: /sys/devices/platform/pxa2xx-spi.1/spi1.0/net/eth0/carrier
Modules linked in: libertas [last unloaded: libertas_spi]
CPU: 0    Not tainted  (2.6.32-dirty #64)
PC is at strcmp+0x8/0x34
LR is at sysfs_find_dirent+0x18/0x38
pc : [<c01711f4>]    lr : [<c00ea9ac>]    psr: a0000013
sp : c6c0def8  ip : 00000000  fp : bee8a254
r10: 00000000  r9 : c6c0c000  r8 : c002a108
r7 : c6c0df44  r6 : 00000880  r5 : 00000000  r4 : c7aa0f00
r3 : 00002326  r2 : 00000000  r1 : 00000000  r0 : 00000000
Flags: NzCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment user
Control: 0000397f  Table: a7bdc000  DAC: 00000015
Process rmmod (pid: 3351, stack limit = 0xc6c0c278)
Stack: (0xc6c0def8 to 0xc6c0e000)
dee0:                                                       c7aa0ed0 00000000
df00: 00000880 c00e93dc c7aa0ed0 00000000 00000000 00000000 00000000 c7bca4c0
df20: 00000880 c006c21c bf030400 bf030400 00000880 c006c6f8 00000000 c006ca68
df40: 40020000 6562696c 73617472 6970735f c6ee8100 c00917d0 40020000 00000000
df60: c7a54e80 c7ace0b0 00013008 c0170e98 00000000 c7af6334 4001f000 00001000
df80: bf030400 00000880 c6c0df8c 00000000 00013008 00000800 bee8a218 00000880
dfa0: 00000081 c0029f60 00000800 bee8a218 bee8a218 00000880 bee89998 00000000
dfc0: 00000800 bee8a218 00000880 00000081 00000000 00000000 00000000 bee8a254
dfe0: 17dd1443 bee8a210 000095c4 4010d6ac 60000010 bee8a218 00000000 00000000
[<c01711f4>] (strcmp+0x8/0x34) from [<c00ea9ac>] (sysfs_find_dirent+0x18/0x38)
[<c00ea9ac>] (sysfs_find_dirent+0x18/0x38) from [<c00e93dc>]
(sysfs_hash_and_remove+0x28/0x60)
[<c00e93dc>] (sysfs_hash_and_remove+0x28/0x60) from [<c006c21c>]
(free_notes_attrs+0x24/0x4c)
[<c006c21c>] (free_notes_attrs+0x24/0x4c) from [<c006c6f8>]
(free_module+0x38/0xdc)
[<c006c6f8>] (free_module+0x38/0xdc) from [<c006ca68>]
(sys_delete_module+0x220/0x260)
[<c006ca68>] (sys_delete_module+0x220/0x260) from [<c0029f60>]
(ret_fast_syscall+0x0/0x28)
Code: 1afffff2 e8bd8030 e1a0c000 e3a02000 (e7dc0002)
---[ end trace 0a0d0d5d616b28ff ]---


I'm using the latest libertas from wireless-testing git. If I can get
anywhere with debugging, hopefully I'll be able to send in a patch.

Cheers,

Chris