Re: Oops in 2.6.37-rc{3,4,5}

[Mauro Carvalho Chehab <mchehab@redhat.com>]

Em 21-09-2011 11:45, Mauro Carvalho Chehab escreveu:
> Em 12-12-2010 09:15, Chris Clayton escreveu:
>> On Sunday 12 December 2010, Dave Young wrote:
>>> On Fri, Dec 10, 2010 at 10:34:06PM +0000, Chris Clayton wrote:
>>>> I'm not subscribed, so please cc me on any reply.
>>>>
>>>> With rc kernels from 2.6.37, X frequently (approx 3 boots out of every 4)
>>>> fails to start. dmesg shows the oops below. I can bisect over the weekend
>>>> - probably Sunday - if no answer comes up in the meantime. I get the same
>>>> oops with rc3, rc and rc5. rc2 doesn't get as far as trying to start X.
>>>> Happy to test patches or provide additional diagnostics, but I'll be off
>>>> line soon until 20:00 or so UK time tomorrow.
>>
>> <snip>
>>
>>>>
>>>> [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
>>>> BUG: unable to handle kernel NULL pointer dereference at   (null)
>>>> IP: [<c13229ef>] __mutex_lock_slowpath+0x9f/0x170
>>>> *pdpt = 0000000034676001 *pde = 0000000000000000
>>>> Oops: 0002 [#1] PREEMPT SMP
>>>> last sysfs file: /sys/module/drm_kms_helper/initstate
>>>> Modules linked in: i915 drm_kms_helper drm fb fbdev cfbcopyarea video
>>>> backlight output cfbimgblt cfbfillrect xt_state iptable_filter
>>>> ipt_MASQUERADE iptable_nat nf_nat nf_conntrack_ipv4 nf_conntrack
>>>> nf_defrag_ipv4 saa7134_alsa tda10048 saa7134_dvb videobuf_dvb dvb_core
>>>> mt20xx tea5767 tda9887 msp3400 gspca_zc3xx gspca_main tda18271 tda8290
>>>> ir_lirc_codec tuner lirc_dev bttv i2c_algo_bit btcx_risc snd_bt87x
>>>> ir_common uhci_hcd ir_core saa7134 v4l2_common videodev v4l1_compat
>>>> ehci_hcd videobuf_dma_sg videobuf_core tveeprom evdev [last unloaded:
>>>> floppy]
>>>>
>>>> Pid: 1725, comm: X Not tainted 2.6.37-rc5+ #476 EG41MF-US2H/EG41MF-US2H
>>>> EIP: 0060:[<c13229ef>] EFLAGS: 00013246 CPU: 3
>>>> EIP is at __mutex_lock_slowpath+0x9f/0x170
>>>> EAX: 00000000 EBX: f4403410 ECX: f4403420 EDX: f4641dd8
>>>> ESI: f4403414 EDI: 00000000 EBP: f4403418 ESP: f4641dd4
>>>>  DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
>>>> Process X (pid: 1725, ti=f4640000 task=f3c67390 task.ti=f4640000)
>>>> Stack:
>>>>  f3c67390 f4403418 00000000 f46a3800 f4403410 f4fd14f0 f425f600 f4403400
>>>>  c13228a6 f4fd1000 f4fd14f0 f8841b6c c10aabb2 ffffffff 05100004 f4c34000
>>>>  f4df72c0 c10aabf7 f4403410 00000001 f425f600 ffffffed f425f604 f46bfe40
>>>> Call Trace:
>>>>  [<c13228a6>] ? mutex_lock+0x16/0x30
>>>>  [<f8841b6c>] ? bttv_open+0xac/0x280 [bttv]
>>>>  [<c10aabb2>] ? cdev_get+0x52/0x90
>>>>  [<c10aabf7>] ? exact_lock+0x7/0x10
>>>>  [<f87095a7>] ? v4l2_open+0xb7/0xd0 [videodev]
>>>>  [<c10ab2ea>] ? chrdev_open+0xda/0x1b0
>>>>  [<c10a5f25>] ? __dentry_open+0xd5/0x280
>>>>  [<c10a7068>] ? nameidata_to_filp+0x68/0x70
>>>>  [<c10ab210>] ? chrdev_open+0x0/0x1b0
>>>>  [<c10b351f>] ? do_last.clone.32+0x34f/0x5a0
>>>>  [<c10b3af3>] ? do_filp_open+0x383/0x550
>>>>  [<c10b1e58>] ? getname+0x28/0xf0
>>>>  [<c10a70c8>] ? do_sys_open+0x58/0x110
>>>>  [<c10a5d09>] ? filp_close+0x49/0x70
>>>>  [<c10a71ac>] ? sys_open+0x2c/0x40
>>>>  [<c1002d10>] ? sysenter_do_call+0x12/0x26
>>>>  [<c1320000>] ? timer_cpu_notify+0x1b4/0x233
>>>> Code: 83 78 18 63 7f b6 8d b6 00 00 00 00 8d 73 04 8d 6b 08 89 f0 e8 f3
>>>> 10 00 00 8b 43 0c 8d 54 24 04 89 44 24 08 89 53 0c 89 6c 24 04 <89> 10 8b
>>>> 04 24 ba ff ff ff ff 89 44 24 0c 89 d0 87 03 83 f8 01
>>>> EIP: [<c13229ef>] __mutex_lock_slowpath+0x9f/0x170 SS:ESP 0068:f4641dd4
>>>> CR2: 0000000000000000
>>>> ---[ end trace 5ac4e44ad0dc7959 ]---
>>
>> <snip>
>>
>>> Could you try following patch?
>>>
>>> --- linux-2.6.orig/drivers/media/video/bt8xx/bttv-driver.c	2010-11-27
>>> 11:21:30.000000000 +0800 +++
>>> linux-2.6/drivers/media/video/bt8xx/bttv-driver.c	2010-12-12
>>> 16:31:39.633333338 +0800 @@ -3291,6 +3291,8 @@ static int bttv_open(struct
>>> file *file)
>>>  	fh = kmalloc(sizeof(*fh), GFP_KERNEL);
>>>  	if (unlikely(!fh))
>>>  		return -ENOMEM;
>>> +
>>> +	mutex_init(&fh->cap.vb_lock);
>>>  	file->private_data = fh;
>>>
>>>  	/*
>>
>> Yes Dave, that's fixed it thanks. Six successful boots, which never happened 
>> without the patch.
>>
>> Tested-by: Chris Clayton <chris2553@googlemail.com>
>>
> 
> It seems we missed this patch. It looks right on my eyes, so I'm
> applying it.

My mistake. This patch were superseded by some changes at vb_lock.

> 
> Thanks,
> Mauro
> --
> To unsubscribe from this list: send the line "unsubscribe linux-media" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html