On 10/14/13 09:53, Ricardo Ribalda Delgado wrote:
> Hello Hans
>
> There is a java web server that broadcast the images. :S
>
> I can see that the server does open/read/close everytime it is asked
> for an image. The error was triggered once in a while. Now I can try
> to trigger the error faster by openning multiple client windows. I
> dont know how the broken connections are handled.
This suggests multi-threaded operation. I actually wonder if the whole
dqbuf/qbuf cycle is thread-safe. Looking at videobuf2-core.c, in
__vb2_wait_for_done_vb(), while that function is waiting for an event it
doesn't hold any locks. So another thread may close the filehandle.
I think this will all work fine provided dqbuf returns an error if streaming
stopped while waiting. I'm not sure it does, though.
It might need a check like this:
if (!q->streaming)
return -ESOMEERROR;
This really needs more testing. Anyone willing to dig into this a bit more?
I won't have time for this this month.
Regards,
Hans
>
> I have tried to crash it with this script and it did not failed.
>
> while true;
>
> do
>
> for a in $(seq 100)
> do
> dd if=/dev/video of=/dev/null count=5 &
> done
>
> sleep 2
> done
>
>
> Thanks!
>
> On Mon, Oct 14, 2013 at 9:47 AM, Hans Verkuil <hverkuil@xxxxxxxxx> wrote:
>> On 10/14/13 09:42, Ricardo Ribalda Delgado wrote:
>>> Hello Hans
>>>
>>> Thanks for your feedback. I have send a patch to the list with your suggestion.
>>>
>>> If I dont apply my patch, but your your branch I still get the error:
>>
>> Thanks for testing this. I suspected that might be the case.
>>
>> What exactly is the sequence that reproduces this error? Ctrl-C in the middle of
>> a read()? Multiple threads accessing the same file handle?
>>
>> Regards,
>>
>> Hans
>>
>>>
>>>
>>> Thanks!
>>>
>>>
>>> [ 57.567944] BUG: unable to handle kernel NULL pointer dereference
>>> at 0000000000000260
>>> [ 57.567980] IP: [<ffffffffa07ab963>]
>>> __vb2_perform_fileio+0x363/0x600 [videobuf2_core]
>>> [ 57.568013] PGD 15952a067 PUD 159519067 PMD 0
>>> [ 57.568032] Oops: 0000 [#1] SMP
>>> [ 57.568045] Modules linked in: nfc qtec_xform gpio_xilinx
>>> qt5023_video videobuf2_vmalloc videobuf2_dma_sg videobuf2_memops
>>> videobuf2_core qtec_pcie qtec_cmosis qtec_white qtec_mem fglrx(PO)
>>> spi_xilinx spi_bitbang qt5023
>>> [ 57.568109] CPU: 1 PID: 1213 Comm: java Tainted: P O
>>> 3.11.0-qtec-standard+ #112
>>> [ 57.568122] Hardware name: QTechnology QT5022/QT5022, BIOS
>>> PM_2.1.0.309 X64 05/23/2013
>>> [ 57.568134] task: ffff88014aa98000 ti: ffff88014aa94000 task.ti:
>>> ffff88014aa94000
>>> [ 57.568144] RIP: 0010:[<ffffffffa07ab963>] [<ffffffffa07ab963>]
>>> __vb2_perform_fileio+0x363/0x600 [videobuf2_core]
>>> [ 57.568168] RSP: 0018:ffff88014aa95e48 EFLAGS: 00010246
>>> [ 57.568178] RAX: 0000000000000000 RBX: ffff88015694e870 RCX: 0000000000000000
>>> [ 57.568187] RDX: 0000000000000000 RSI: ffff88015694c400 RDI: ffff88015694c400
>>> [ 57.568195] RBP: ffff88014aa95e98 R08: 0000000000000000 R09: 0000000000000001
>>> [ 57.568204] R10: ffffea00055c2880 R11: ffff88014aa95fd8 R12: ffff88015694e800
>>> [ 57.568213] R13: 000000000021de00 R14: ffff88015727f0b8 R15: 000000000126ae20
>>> [ 57.568224] FS: 00007fc23bbda700(0000) GS:ffff88015ed00000(0000)
>>> knlGS:0000000000000000
>>> [ 57.568233] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
>>> [ 57.568241] CR2: 0000000000000260 CR3: 0000000156931000 CR4: 00000000000007e0
>>> [ 57.568248] Stack:
>>> [ 57.568254] ffff880159632910 0000000000000000 ffff880100000001
>>> 0000000059bd0820
>>> [ 57.568270] ffff88014aa95f50 ffff88015727f720 ffff880159632900
>>> ffff88015727f340
>>> [ 57.568285] 000000000021de00 ffff88014aa95f50 ffff88014aa95ed8
>>> ffffffffa07abe09
>>> [ 57.568300] Call Trace:
>>> [ 57.568328] [<ffffffffa07abe09>] vb2_fop_read+0xb9/0x110 [videobuf2_core]
>>> [ 57.568351] [<ffffffff81534985>] v4l2_read+0x65/0xb0
>>> [ 57.568372] [<ffffffff81166bea>] vfs_read+0x9a/0x150
>>> [ 57.568388] [<ffffffff81167649>] SyS_read+0x49/0xa0
>>> [ 57.568407] [<ffffffff81760479>] tracesys+0xd0/0xd5
>>> [ 57.568415] Code: 89 c2 44 8b 4d c0 4c 8b 45 b8 0f 8f 2d 02 00 00
>>> 85 d2 48 63 c2 0f 85 86 fd ff ff 41 83 84 24 78 03 00 00 01 31 d2 4b
>>> 8b 4c c6 48 <8b> 81 60 02 00 00 85 c0 74 05 8b 41 58 89 c2 89 43 08 80
>>> 63 10
>>> [ 57.568566] RIP [<ffffffffa07ab963>]
>>> __vb2_perform_fileio+0x363/0x600 [videobuf2_core]
>>> [ 57.568587] RSP <ffff88014aa95e48>
>>> [ 57.568594] CR2: 0000000000000260
>>> [ 57.568606] ---[ end trace d664b8a0460e8ca5 ]---
>>>
>>>
>>>
>>> On Sun, Oct 13, 2013 at 6:42 PM, Hans Verkuil <hverkuil@xxxxxxxxx> wrote:
>>>> Hi Ricardo,
>>>>
>>>> On 10/13/2013 03:03 PM, Ricardo Ribalda Delgado wrote:
>>>>> Hello
>>>>>
>>>>> These days I have been testing an app that uses the old read/write
>>>>> API. It is interfacing a videobuf2-sg driver.
>>>>>
>>>>> Once in a while I get an oops on the vb2_perform_fileio function.
>>>>>
>>>>> After digging a while I have been able to fix the bug like this:
>>>>>
>>>>> int vb2_fop_release(struct file *file)
>>>>> {
>>>>> struct video_device *vdev = video_devdata(file);
>>>>> + struct mutex *lock = vdev->queue->lock ? vdev->queue->lock : vdev->lock;
>>>>>
>>>>> + if (lock)
>>>>> + mutex_lock(lock);
>>>>> if (file->private_data == vdev->queue->owner) {
>>>>> vb2_queue_release(vdev->queue);
>>>>> vdev->queue->owner = NULL;
>>>>> }
>>>>> + if (lock)
>>>>> + mutex_unlock(lock);
>>>>> return v4l2_fh_release(file);
>>>>> }
>>>>> EXPORT_SYMBOL_GPL(vb2_fop_release);
>>>>>
>>>>> (or at least, I havent seen the oops since then).
>>>>>
>>>>> I was wondering if this a real bug on vb2 and I shall post a proper
>>>>> patch or the device driver is doing something out of spec.
>>>>>
>>>>> vb2_perform_fileio+0x372 corresponds to
>>>>>
>>>>> memset(&fileio->b, 0, sizeof(fileio->b));
>>>>>
>>>>> My theory is that:
>>>>>
>>>>> fileio = q->fileio;
>>>>> if (!fileio){
>>>>> ...
>>>>> q->fileio = NULL;
>>>>>
>>>>> can be prone to race conditions if the lock is not held on release.
>>>>
>>>> This would not surprise me.
>>>>
>>>> I have a number of vb2 patches in this branch:
>>>>
>>>> http://git.linuxtv.org/hverkuil/media_tree.git/shortlog/refs/heads/vb2-thread
>>>>
>>>> Among others it replaces the 'q->fileio = NULL' hack by a proper solution.
>>>>
>>>> It will be interesting to see if my patch series in that branch would fix your
>>>> problem as well.
>>>>
>>>> That said, I do think vb2_fop_release should take the lock regardless, although
>>>> I would put the mutex_lock inside the 'if' that checks the owner.
>>>>
>>>> Regards,
>>>>
>>>> Hans
>>>>
>>>>>
>>>>>
>>>>> [ 308.297741] BUG: unable to handle kernel NULL pointer dereference at
>>>>> 0000000000000260
>>>>> [ 308.297759] IP: [<ffffffffa07a9fd2>] vb2_perform_fileio+0x372/0x610
>>>>> [videobuf2_core]
>>>>> [ 308.297794] PGD 159719067 PUD 158119067 PMD 0
>>>>> [ 308.297812] Oops: 0000 #1 SMP
>>>>> [ 308.297826] Modules linked in: qt5023_video videobuf2_dma_sg
>>>>> qtec_xform videobuf2_vmalloc videobuf2_memops videobuf2_core
>>>>> qtec_white qtec_mem gpio_xilinx qtec_cmosis qtec_pcie fglrx(PO)
>>>>> spi_xilinx spi_bitbang qt5023
>>>>> [ 308.297888] CPU: 1 PID: 2189 Comm: java Tainted: P O 3.11.0-qtec-standard #1
>>>>> [ 308.297919] Hardware name: QTechnology QT5022/QT5022, BIOS
>>>>> PM_2.1.0.309 X64 05/23/2013
>>>>> [ 308.297952] task: ffff8801564e1690 ti: ffff88014dc02000 task.ti:
>>>>> ffff88014dc02000
>>>>> [ 308.297962] RIP: 0010:[<ffffffffa07a9fd2>] [<ffffffffa07a9fd2>]
>>>>> vb2_perform_fileio+0x372/0x610 [videobuf2_core]
>>>>> [ 308.297985] RSP: 0018:ffff88014dc03df8 EFLAGS: 00010202
>>>>> [ 308.297995] RAX: 0000000000000000 RBX: ffff880158a23000 RCX: dead000000100100
>>>>> [ 308.298003] RDX: 0000000000000000 RSI: dead000000200200 RDI: 0000000000000000
>>>>> [ 308.298012] RBP: ffff88014dc03e58 R08: 0000000000000000 R09: 0000000000000001
>>>>> [ 308.298020] R10: ffffea00051e8380 R11: ffff88014dc03fd8 R12: ffff880158a23070
>>>>> [ 308.298029] R13: ffff8801549040b8 R14: 0000000000198000 R15: 0000000001887e60
>>>>> [ 308.298040] FS: 00007f65130d5700(0000) GS:ffff88015ed00000(0000)
>>>>> knlGS:0000000000000000
>>>>> [ 308.298049] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>>>>> [ 308.298057] CR2: 0000000000000260 CR3: 0000000159630000 CR4: 00000000000007e0
>>>>> [ 308.298064] Stack:
>>>>> [ 308.298071] ffff880156416c00 0000000000198000 0000000000000000
>>>>> ffff880100000001
>>>>> [ 308.298087] ffff88014dc03f50 00000000810a79ca 0002000000000001
>>>>> ffff880154904718
>>>>> [ 308.298101] ffff880156416c00 0000000000198000 ffff880154904338
>>>>> ffff88014dc03f50
>>>>> [ 308.298116] Call Trace:
>>>>> [ 308.298143] [<ffffffffa07aa3c4>] vb2_read+0x14/0x20 [videobuf2_core]
>>>>> [ 308.298198] [<ffffffffa07aa494>] vb2_fop_read+0xc4/0x120 [videobuf2_core]
>>>>> [ 308.298252] [<ffffffff8154ee9e>] v4l2_read+0x7e/0xc0
>>>>> [ 308.298296] [<ffffffff8116e639>] vfs_read+0xa9/0x160
>>>>> [ 308.298312] [<ffffffff8116e882>] SyS_read+0x52/0xb0
>>>>> [ 308.298328] [<ffffffff81784179>] tracesys+0xd0/0xd5
>>>>> [ 308.298335] Code: e5 d6 ff ff 83 3d be 24 00 00 04 89 c2 4c 8b 45 b0
>>>>> 44 8b 4d b8 0f 8f 20 02 00 00 85 d2 75 32 83 83 78 03 00 00 01 4b 8b
>>>>> 44 c5 48 <8b> 88 60 02 00 00 85 c9 0f 84 b0 00 00 00 8b 40 58 89 c2 41
>>>>> 89
>>>>> [ 308.298487] RIP [<ffffffffa07a9fd2>] vb2_perform_fileio+0x372/0x610
>>>>> [videobuf2_core]
>>>>> [ 308.298507] RSP <ffff88014dc03df8>
>>>>> [ 308.298514] CR2: 0000000000000260
>>>>> [ 308.298526] ---[ end trace e8f01717c96d1e41 ]---
>>>>>
>>>>
>>>
>>>
>>>
>>
>
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-media" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
