Compat-Wireless Kernel Bug Report

Compat-Wireless Kernel Bug Report

[Gmeiner Maximilian]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="us-ascii", Size: 5231 bytes --]

After compiling from the bleeding-edge without any errors this is what I get in kern.log when doing modprobe ath9k:

Hardware:
02:00.0 Network controller: Atheros Communications Inc. AR9485 Wireless Network Adapter (rev 01)
	Subsystem: AzureWave Device 2086
	Kernel driver in use: ath9k
	Kernel modules: ath9

Kernel: 
3.3.0-030300rc6-generic (the same happens on 3.2 kernel

Distribution:
Ubuntu 12.04 Testing



Mar  5 21:21:31 karl kernel: [25063.046563] ath9k: Driver unloaded
Mar  5 21:21:32 karl kernel: [25063.557834] NET: Unregistered protocol family 31
Mar  5 21:22:14 karl kernel: [25105.553769] Compat-wireless backport release: compat-wireless-2012-02-28
Mar  5 21:22:14 karl kernel: [25105.553781] Backport based on linux-next.git next-20120302
Mar  5 21:22:14 karl kernel: [25105.553793] sys_init_module: 'compat'->init suspiciously returned 68, it should follow 0/-E convention
Mar  5 21:22:14 karl kernel: [25105.553799] sys_init_module: loading module anyway...
Mar  5 21:22:14 karl kernel: [25105.553811] Pid: 6188, comm: modprobe Tainted: G        WC O 3.3.0-030300rc6-generic #201203032235
Mar  5 21:22:14 karl kernel: [25105.553817] Call Trace:
Mar  5 21:22:14 karl kernel: [25105.553833]  [<ffffffff810b557b>] sys_init_module+0x21b/0x220
Mar  5 21:22:14 karl kernel: [25105.553844]  [<ffffffff816683e9>] system_call_fastpath+0x16/0x1b
Mar  5 21:22:14 karl kernel: [25105.553862] BUG: unable to handle kernel paging request at ffff8800df439014
Mar  5 21:22:14 karl kernel: [25105.560967] IP: [<ffffffff810b3138>] module_put+0x28/0x90
Mar  5 21:22:14 karl kernel: [25105.564465] PGD 1c06063 PUD 0 
Mar  5 21:22:14 karl kernel: [25105.567077] Oops: 0002 [#1] SMP 
Mar  5 21:22:14 karl kernel: [25105.569750] CPU 1 
Mar  5 21:22:14 karl kernel: [25105.569776] Modules linked in: compat(O) msr parport_pc ppdev snd_hda_codec_hdmi snd_hda_codec_realtek joydev rts5139(C) snd_hda_intel asus_nb_wmi snd_hda_codec asus_wmi sparse_keymap snd_hwdep snd_pcm snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq snd_timer snd_seq_device arc4 psmouse serio_raw i915 snd drm_kms_helper mac_hid drm soundcore snd_page_alloc i2c_algo_bit video mei(C) wmi lp parport [last unloaded: bluetooth]
Mar  5 21:22:14 karl kernel: [25105.582988] 
Mar  5 21:22:14 karl kernel: [25105.586294] Pid: 6188, comm: modprobe Tainted: G        WC O 3.3.0-030300rc6-generic #201203032235 ASUSTeK Computer Inc. UX31E/UX31E
Mar  5 21:22:14 karl kernel: [25105.588207] RIP: 0010:[<ffffffff810b3138>]  [<ffffffff810b3138>] module_put+0x28/0x90
Mar  5 21:22:14 karl kernel: [25105.591281] RSP: 0018:ffff88000e6aff28  EFLAGS: 00010282
Mar  5 21:22:14 karl kernel: [25105.594775] RAX: ffffffffa001900c RBX: ffffffffa001b020 RCX: 0000000000000000
Mar  5 21:22:14 karl kernel: [25105.598230] RDX: 0000000000000000 RSI: 0000000000000282 RDI: ffffffffa001b020
Mar  5 21:22:14 karl kernel: [25105.601637] RBP: ffff88000e6aff48 R08: 0000000000000000 R09: 0000000000000000
Mar  5 21:22:14 karl kernel: [25105.604235] R10: 0000000000000000 R11: 0000000000000001 R12: ffffffffa001b020
Mar  5 21:22:14 karl kernel: [25105.607506] R13: 0000000000ad7530 R14: 0000000000000000 R15: 0000000000405d30
Mar  5 21:22:14 karl kernel: [25105.610716] FS:  00007f61c762a700(0000) GS:ffff88013f420000(0000) knlGS:0000000000000000
Mar  5 21:22:14 karl kernel: [25105.613961] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Mar  5 21:22:14 karl kernel: [25105.617146] CR2: ffff8800df439014 CR3: 00000001364f6000 CR4: 00000000000406e0
Mar  5 21:22:14 karl kernel: [25105.620365] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
Mar  5 21:22:14 karl kernel: [25105.623530] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Mar  5 21:22:14 karl kernel: [25105.626720] Process modprobe (pid: 6188, threadinfo ffff88000e6ae000, task ffff88000260c4d0)
Mar  5 21:22:14 karl kernel: [25105.629988] Stack:
Mar  5 21:22:14 karl kernel: [25105.629992]  ffff88000e6aff48 ffffffffa001b020 0000000000000044 0000000000ad7530
Mar  5 21:22:14 karl kernel: [25105.630000]  ffff88000e6aff78 ffffffff810b5486 0000000000405d30 0000000000ada220
Mar  5 21:22:14 karl kernel: [25105.630006]  0000000000000000 0000000000ad6080 0000000000ad7530 ffffffff816683e9
Mar  5 21:22:14 karl kernel: [25105.630013] Call Trace:
Mar  5 21:22:14 karl kernel: [25105.630025]  [<ffffffff810b5486>] sys_init_module+0x126/0x220
Mar  5 21:22:14 karl kernel: [25105.630034]  [<ffffffff816683e9>] system_call_fastpath+0x16/0x1b
Mar  5 21:22:14 karl kernel: [25105.630038] Code: 44 00 00 55 48 89 e5 48 83 ec 20 48 89 5d e8 4c 89 65 f0 4c 89 6d f8 66 66 66 66 90 48 85 ff 49 89 fc 74 21 48 8b 87 48 02 00 00 <65> 48 ff 40 08 8b 05 a5 29 c1 00 4c 8b 6d 08 85 c0 75 15 41 83 
Mar  5 21:22:14 karl kernel: [25105.630089] RIP  [<ffffffff810b3138>] module_put+0x28/0x90
Mar  5 21:22:14 karl kernel: [25105.630096]  RSP <ffff88000e6aff28>
Mar  5 21:22:14 karl kernel: [25105.630099] CR2: ffff8800df439014
Mar  5 21:22:14 karl kernel: [25105.725381] ---[ end trace 5190b7aaf2d6bf84 ]---


FH JOANNEUM Gesellschaft mbH
Rechtsform/Legal form: GmbH
Firmenbuchgericht/Court of registry: Landesgericht für ZRS Graz
Firmenbuchnummer/Company registration: FN 125888 f
DVR: 0813559
UID-Nr.: ATU 42361001

[PATCH] compat: Fix module_init return type from void to int.

[Nicolas Cavallari]

086f3a1 overrode module_init to make it depend on compat, but the
overriding module_init has return type 'void', so strange things would
happen when the module loading code see that init_module() returns
positive values.  This patch makes it return the value returned by
the overriden module_init.

Signed-off-by: Nicolas Cavallari <cavallar@lri.fr>
---
On 05/03/2012 22:50, Gmeiner Maximilian wrote:
> After compiling from the bleeding-edge without any errors this is what I get in kern.log when doing modprobe ath9k

I had the same problem with a completely different architecture and device,
and this patch fixed it.

diff --git a/include/linux/compat-2.6.h b/include/linux/compat-2.6.h
index bb00cd4..5715ce5 100644
--- a/include/linux/compat-2.6.h
+++ b/include/linux/compat-2.6.h
@@ -27,10 +27,10 @@ void compat_dependency_symbol(void);
 
 #undef module_init
 #define module_init(initfn)						\
-	static void __init __init_compat(void)				\
+	static int __init __init_compat(void)				\
 	{								\
 		compat_dependency_symbol();				\
-		initfn();						\
+		return initfn();					\
 	}								\
 	int init_module(void) __attribute__((alias("__init_compat")));
 
-- 
1.7.9

Re: [PATCH] compat: Fix module_init return type from void to int.

[Johannes Berg]

On Tue, 2012-03-06 at 10:55 +0100, Nicolas Cavallari wrote:
> 086f3a1 overrode module_init to make it depend on compat, but the
> overriding module_init has return type 'void', so strange things would
> happen when the module loading code see that init_module() returns
> positive values.  This patch makes it return the value returned by
> the overriden module_init.
> 
> Signed-off-by: Nicolas Cavallari <cavallar@lri.fr>
> ---
> On 05/03/2012 22:50, Gmeiner Maximilian wrote:
> > After compiling from the bleeding-edge without any errors this is what I get in kern.log when doing modprobe ath9k
> 
> I had the same problem with a completely different architecture and device,
> and this patch fixed it.
> 
> diff --git a/include/linux/compat-2.6.h b/include/linux/compat-2.6.h
> index bb00cd4..5715ce5 100644
> --- a/include/linux/compat-2.6.h
> +++ b/include/linux/compat-2.6.h
> @@ -27,10 +27,10 @@ void compat_dependency_symbol(void);
>  
>  #undef module_init
>  #define module_init(initfn)						\
> -	static void __init __init_compat(void)				\
> +	static int __init __init_compat(void)				\
>  	{								\
>  		compat_dependency_symbol();				\
> -		initfn();						\
> +		return initfn();					\


Yikes, you're right, thanks for the patch!

Acked-by: Johannes Berg <johannes@sipsolutions.net>

johannes

Re: [PATCH] compat: Fix module_init return type from void to int.

[Luis R. Rodriguez]

On Tue, Mar 6, 2012 at 2:09 AM, Johannes Berg <johannes@sipsolutions.net> wrote:
> On Tue, 2012-03-06 at 10:55 +0100, Nicolas Cavallari wrote:
>> Signed-off-by: Nicolas Cavallari <cavallar@lri.fr>
> Acked-by: Johannes Berg <johannes@sipsolutions.net>

Applied and pushed, thanks!

  Luis