Re: [crash, bisected] I2C, kobject (ffff81003e8c4160): tried toinit an initialized object (was: Re: [GIT PULL] i2c updates for 2.6.27, round 1)

[Jean Delvare]

Hi Ingo,

On Wed, 16 Jul 2008 10:23:52 +0200, Ingo Molnar wrote:
> 
> hi Jean,
> 
> * Jean Delvare <khali@xxxxxxxxxxxx> wrote:
> 
> >       i2c-nforce2: Add support for multiplexing on the Tyan S4985
> 
> -tip random boot testing started triggering the following upstream boot 
> crash, starting two days ago when the i2c tree was merged:
> 
> [   11.772002] kobject (ffff81003e8c4160): tried to init an initialized object, something is seriously wrong.
> [   11.784004] Pid: 1, comm: swapper Not tainted 2.6.26-00007-g279e902-dirty #19896
> [   11.793651] 
> [   11.793651] Call Trace:
> [   11.793651]  [<ffffffff8054471a>] kobject_init+0x7a/0x80
> [   11.798969]  [<ffffffff805f0f9d>] device_initialize+0x2d/0xd0
> [   11.806323]  [<ffffffff805f19a1>] device_register+0x11/0x20
> [   11.811903]  [<ffffffff8087ee3b>] i2c_register_adapter+0xab/0x1d0
> [   11.817666]  [<ffffffff8087f0e6>] i2c_add_adapter+0x96/0xb0
> [   11.823761]  [<ffffffff81196372>] nforce2_s4985_init+0x272/0x370
> [   11.829358]  [<ffffffff8027516a>] ? getnstimeofday+0x3a/0xc0
> [   11.835365]  [<ffffffff802728ba>] ? ktime_get_ts+0x4a/0x60
> [   11.840853]  [<ffffffff81196100>] ? nforce2_s4985_init+0x0/0x370
> [   11.846859]  [<ffffffff81160aa3>] kernel_init+0x123/0x300
> [   11.852261]  [<ffffffff81160140>] ? early_idt_handler+0x0/0x74
> [   11.857957]  [<ffffffff80228548>] child_rip+0xa/0x12
> [   11.863781]  [<ffffffff81160140>] ? early_idt_handler+0x0/0x74
> [   11.868757]  [<ffffffff81160980>] ? kernel_init+0x0/0x300
> [   11.874160]  [<ffffffff8022853e>] ? child_rip+0x0/0x12
> [   11.879989] 
> [   11.881141] general protection fault: 0000 [1] SMP 
> [   11.881141] CPU 1 
> 
> with this config:
> 
>    http://redhat.com/~mingo/misc/config-Wed_Jul_16_02_26_04_CEST_2008.bad
> 
> today i had time to do a bisection run, and it came up with this commit:
> 
> | 279e902445557897707d325182916a6e28ba80de is first bad commit
> | commit 279e902445557897707d325182916a6e28ba80de
> | Author: Jean Delvare <khali@xxxxxxxxxxxx>
> | Date:   Mon Jul 14 22:38:21 2008 +0200
> |
> |    i2c-nforce2: Add support for multiplexing on the Tyan S4985
> 
> the crash is reproducible and i can test any fix patch in short order. 
> As a quick temporary hack i've reverted this patch for now via the patch 
> below, to keep the tests going. (Can send more details about the 
> test-system if needed.)

Probably the same problem as reported by John Stultz:
http://lkml.org/lkml/2008/7/15/720

I'll look into it once I am done with higher priority tasks (read:
baby's bath ;)).

> btw., Greg, the kobject debugging check and output is really nice, 
> informative and useful! It likely pinpoints the real bug here.

Is this something new in 2.6.26-gitX? I want to know if the bug is
something new or if it could be already present in 2.6.26 and earlier
kernels and went unnoticed so far.

Thanks,
-- 
Jean Delvare
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/