-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Today I experienced this problem with a Cisco Aironet 350.
I just want to point out it's the first time it happens.
In fact, I still used this NIC on this kernel (2.6.1) without
any kind of problem. Attached is an extract from my log.
airo: Probing for PCI adapters
Unable to handle kernel paging request at virtual address e0b397b8
printing eip:
c01b524f
*pde = 1ed80067
*pte = 00000000
Oops: 0002 [#1]
CPU: 0
EIP: 0060:[<c01b524f>] Not tainted
EFLAGS: 00010296
EIP is at kobject_add+0x6f/0x120
eax: c0350460 ebx: e0bd379c ecx: e0b397b8 edx: e0bd37b8
esi: c0350468 edi: 00000000 ebp: e0bd3784 esp: d90b5f0c
ds: 007b es: 007b ss: 0068
Process modprobe (pid: 3780, threadinfo=d90b4000 task=ca604080)
Stack: c0350468 e0bd37a0 e0bd379c e0bd379c e0bd379c 00000000 c01b5323 e0bd379c
e0bd379c c0350400 e0bd379c c0350400 c0217a5a e0bd379c e0bd01e0 e0bd3760
00000000 e0bd37f8 d90b4000 c0217f1f e0bd3784 e0bd10a3 d90b5f8c c017d388
Call Trace:
[<c01b5323>] kobject_register+0x23/0x60
[<c0217a5a>] bus_add_driver+0x4a/0xa0
[<c0217f1f>] driver_register+0x2f/0x40
[<c017d388>] create_proc_entry+0x88/0xd0
[<c01bcecc>] pci_register_driver+0x5c/0x90
[<e0b2d0d1>] airo_init_module+0xd1/0xf8 [airo]
[<c01319bc>] sys_init_module+0x12c/0x250
[<c01091a7>] syscall_call+0x7/0xb
Code: 89 11 89 4a 04 8b 43 28 8b 38 8d 4f 48 89 c8 ba ff ff 00 00
<4>airo_cs: Unknown symbol init_airo_card
airo_cs: Unknown symbol stop_airo_card
airo_cs: Unknown symbol reset_airo_card
airo_cs: Unknown symbol init_airo_card
airo_cs: Unknown symbol stop_airo_card
airo_cs: Unknown symbol reset_airo_card
Regards.
- --
Angelo Dell'Aera 'buffer'
Antifork Research, Inc. http://buffer.antifork.org
PGP information in e-mail header
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)
iD8DBQFAFlRtpONIzxnBXKIRAuhkAJ4xsLc4IBE65CPW+2tMG7g0XKbpGwCgixVY
nES3gWlgWnWlD3KGqM4wv8Q=
=PN9/
-----END PGP SIGNATURE-----
On Tue, Jan 27, 2004 at 12:45:38PM +0000, Russell King wrote:
> On Tue, Jan 27, 2004 at 01:07:09PM +0100, Angelo Dell'Aera wrote:
> > Today I experienced this problem with a Cisco Aironet 350.
> > I just want to point out it's the first time it happens.
> > In fact, I still used this NIC on this kernel (2.6.1) without
> > any kind of problem. Attached is an extract from my log.
>
Ok, actually the bug has been fixed in 2.6.2-rc, so this problem can
finally be laid to rest. Here's the as-merged fix:
===== drivers/net/wireless/airo.c 1.80 vs 1.81 =====
--- 1.80/drivers/net/wireless/airo.c Sun Dec 7 13:22:52 2003
+++ 1.81/drivers/net/wireless/airo.c Sat Jan 10 16:04:13 2004
@@ -1027,7 +1027,6 @@
#define FLAG_802_11 7
#define FLAG_PENDING_XMIT 9
#define FLAG_PENDING_XMIT11 10
-#define FLAG_PCI 11
#define JOB_MASK 0x1ff0000
#define JOB_DIE 16
#define JOB_XMIT 17
@@ -4623,7 +4622,6 @@
return -ENODEV;
pci_set_drvdata(pdev, dev);
- set_bit (FLAG_PCI, &((struct airo_info *)dev->priv)->flags);
return 0;
}
@@ -4653,7 +4651,7 @@
#ifdef CONFIG_PCI
printk( KERN_INFO "airo: Probing for PCI adapters\n" );
- pci_module_init(&airo_driver);
+ pci_register_driver(&airo_driver);
printk( KERN_INFO "airo: Finished probing for PCI adapters\n" );
#endif
@@ -4665,22 +4663,15 @@
static void __exit airo_cleanup_module( void )
{
- int is_pci = 0;
while( airo_devices ) {
printk( KERN_INFO "airo: Unregistering %s\n", airo_devices->dev->name );
-#ifdef CONFIG_PCI
- if (test_bit(FLAG_PCI, &((struct airo_info *)airo_devices->dev->priv)->flags))
- is_pci = 1;
-#endif
stop_airo_card( airo_devices->dev, 1 );
}
remove_proc_entry("aironet", proc_root_driver);
- if (is_pci) {
#ifdef CONFIG_PCI
- pci_unregister_driver(&airo_driver);
+ pci_unregister_driver(&airo_driver);
#endif
- }
}
#ifdef WIRELESS_EXT
--
Russell King
Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/
maintainer of: 2.6 PCMCIA - http://pcmcia.arm.linux.org.uk/
2.6 Serial core
On Tue, Jan 27, 2004 at 01:07:09PM +0100, Angelo Dell'Aera wrote:
> Today I experienced this problem with a Cisco Aironet 350.
> I just want to point out it's the first time it happens.
> In fact, I still used this NIC on this kernel (2.6.1) without
> any kind of problem. Attached is an extract from my log.
AAARRRGGGHHH.
Whatever, it's a known problem with a history of at about three months.
The last message seems to be the maintainer saying that he has a patch
ready and will send to Jeff. I've no idea if that actually happened or
not.
I'm going to suggest that it gets marked as a broken and unmaintained
driver, or I'm going to bypass the maintainers and submit an _untested_
fix directly to Andrew/Linus. I don't want to do either, but...
--
Russell King
Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/
maintainer of: 2.6 PCMCIA - http://pcmcia.arm.linux.org.uk/
2.6 Serial core