lkml.org 
[lkml]   [2011]   [Feb]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH resend] video: omap24xxcam: Fix compilation
Hallo David,

Am 25.02.2011 00:36, schrieb David Cohen:
> On Mon, Feb 21, 2011 at 2:21 PM, Felipe Balbi <balbi@ti.com> wrote:
>> On Mon, Feb 21, 2011 at 02:09:07PM +0200, David Cohen wrote:
>>> On Mon, Feb 21, 2011 at 9:36 AM, Felipe Balbi <balbi@ti.com> wrote:
>>>> Hi,
>>>>
>>>> On Sat, Feb 19, 2011 at 06:04:58PM +0200, David Cohen wrote:
>>>>>> I have to disagree. The fundamental problem is the circular dependency
>>>>>> between those two files:
>>>>>>
>>>>>> sched.h uses wait_queue_head_t defined in wait.h
>>>>>> wait.h uses TASK_* defined in sched.h
>>>>>>
>>>>>> So, IMO the real fix would be clear out the circular dependency. Maybe
>>>>>> introducing <linux/task.h> to define those TASK_* symbols and include
>>>>>> that on sched.h and wait.h
>>>>>>
>>>>>> Just dig a quick and dirty to try it out and works like a charm
>>>>> We have 2 problems:
>>>>> - omap24xxcam compilation broken
>>>>> - circular dependency between sched.h and wait.h
>>>>>
>>>>> To fix the broken compilation we can do what the rest of the kernel is
>>>>> doing, which is to include sched.h.
>>>>> Then, the circular dependency is fixed by some different approach
>>>>> which would probably change *all* current usage of TASK_*.
>>>> considering that 1 is caused by 2 I would fix 2.
>>>>
>>>>> IMO, there's no need to create a dependency between those issues.
>>>> There's no dependency between them, it's just that the root cause for
>>>> this problem is a circular dependency between wait.h and sched.h
>>> I did a try to fix this circular dependency and the comment I got was
>>> to include sched.h in omap24xxcam.c file:
>>> http://marc.info/?l=linux-omap&m=129828637120270&w=2
>>>
>>> I'm working to remove v4l2 internal device interface from omap24xxcam
>>> and then I need this driver's compilation fixed.
>>> The whole kernel is including sched.h when wake_up*() macro is used,
>>> so this should be our first solution IMO.
>>> As I said earlier, no need to make this compilation fix be dependent
>>> of wait.h fix (if it's really going to be changed).
>>>
>>> I think we should proceed with this patch.
>> I would wait to hear from Ingo or Peter who are the maintainers for that
>> part, but fine by me.
> How about to proceed with this patch?
>
> Regards,
>
> David
>

I got a message that the patch is queued at

http://git.linuxtv.org/media_tree.git for_v2.6.39


Thanks Mauro.


Thomas






\
 
 \ /
  Last update: 2011-02-25 08:03    [from the cache]
©2003-2011 Jasper Spaans