[linux-dvb] Patches to fix support for VisionPlus DTV (VP1020) / tuning problem

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Johannes Stezenbach wrote:
> Steffen Motzer wrote:
> 
>>These patches fix the following bugs:
>>
>>bttv-i2c.c:
>>After upgrading from 2.6.9 to 2.6.11.7, bttv failed to write to the ASIC
>>via the i2c bus. After a couple of times bttv_i2c_wait_done was called
>>the RACK flag was indicating a failed i2c operation.
> 
> 
> A number of bttv patches whent into 2.6.12-rc2. Does you stuff still
> apply there?
> 
> 
> 
>>*** bttv-i2c.c.orig	2005-04-23 00:42:40.000000000 +0200
>>--- bttv-i2c.c		2005-04-23 02:09:43.000000000 +0200
>>*************** bttv_i2c_wait_done(struct bttv *btv)
>>*** 135,143 ****
> 
> 
> Please send unified diffs only!
> 
> 
>>  	DECLARE_WAITQUEUE(wait, current);
>>  	int rc = 0;
>>  
>>  	add_wait_queue(&btv->i2c_queue, &wait);
>>! 	if (0 == btv->i2c_done)
>>! 		msleep_interruptible(20);
>>  	remove_wait_queue(&btv->i2c_queue, &wait);
>>  
>>  	if (0 == btv->i2c_done)
>>--- 135,148 ----
>>  	DECLARE_WAITQUEUE(wait, current);
>>  	int rc = 0;
>>  
>>+ 	unsigned long timeout = msecs_to_jiffies(20);
>>+ 
>>  	add_wait_queue(&btv->i2c_queue, &wait);
>>! 
>>! 	while (0 == btv->i2c_done && timeout) {
>>! 		set_current_state(TASK_INTERRUPTIBLE);
>>! 		timeout = schedule_timeout(timeout);
>>! 	}
>>  	remove_wait_queue(&btv->i2c_queue, &wait);
>>  
>>  	if (0 == btv->i2c_done)
> 
> 
> This code looks strange (before and after your change).
> Why isn't this a simple
> 
>   wait_event_interruptible_timeout(&btv->i2c_queue, btv->i2c_done,
>   				   msecs_to_jiffies(20));
> 
> ?

I tried it out, but when you suggested that, i found another problem 
that i had been going after ..
Attached is my patch.. to what you suggested..

With this the tuning speed seems to be better, and i am kind of seeing 
lesser cases of the driver getting into error recovery ..

Maybe Gerd can comment on this, as i am not too comforatble in this area 
.. The reason why i have 6, is that it is higher than the worst case 
timeout.. With larger values at least for the dst/dst_ca drivers, It 
feels kind of "sluggish" to move around .. I don't know how exactly to 
explain that, but higher values causes more chances of i2c errors and 
slower tuning ..


Manu






-------------- next part --------------
A non-text attachment was scrubbed...
Name: waitqueue-bttv-i2c.diff
Type: text/x-patch
Size: 541 bytes
Desc: not available
Url : http://www.linuxtv.org/pipermail/linux-dvb/attachments/20050423/0be4800f/waitqueue-bttv-i2c.bin

[Index of Archives]     [Linux Media]     [Video 4 Linux]     [Asterisk]     [Samba]     [Xorg]     [Xfree86]     [Linux USB]

  Powered by Linux