On Thu, 2010-06-24 at 20:21 +0400, Vladimir Zapolskiy wrote:
On Thu, Jun 24, 2010 at 8:05 PM, Liam Girdwood lrg@slimlogic.co.uk wrote:
On Thu, 2010-06-24 at 19:54 +0400, Vladimir Zapolskiy wrote:
Hi,
On Thu, Jun 24, 2010 at 5:49 PM, Liam Girdwood lrg@slimlogic.co.uk wrote:
On Thu, 2010-06-24 at 17:38 +0400, Vladimir Zapolskiy wrote:
@@ -559,9 +546,11 @@ static int uda134x_soc_probe(struct platform_device *pdev) codec->num_dai = 1; codec->read = uda134x_read_reg_cache; codec->write = uda134x_write; -#ifdef POWER_OFF_ON_STANDBY
codec->set_bias_level = uda134x_set_bias_level;-#endif
if (!pd->is_powered_on_standby) {codec->set_bias_level = uda134x_set_bias_level;The codec ops pointers have moved in multi-component. Can you put this check in your uda134x_set_bias_level() instead.
Not sure I understand the problem, and I do not see a simple solution for actual bugfix in patch 2/2, if this check is placed in uda134x_set_bias_level(). Please could you provide me with a hint or link to some codec source file?
I was meaning something like :-
uda134x_set_bias_level() { if (pd->is_powered_on_standby) return
/* rest of func */}
Thank you, good that I meant the same. But that makes quite impossible to set a mandatory bias level on initialization without additional code complication.
Sorry, I'm just trying to make the multi-component merge simpler here.
The multi-component branch is here :-
http://git.kernel.org/?p=linux/kernel/git/lrg/asoc-2.6.git;a=shortlog;h=refs...
I'll merge this myself later.
The multi-component branch makes codec->set_bias_level a driver level operation (that can be shared amongst other uda134x codec devices so we cant change this at runtime)
Looks like I miss something, because in codec drivers I observe codec->set_bias_level is set in probe function, like it is in uda134x.
Would it be better just to drop a condition of pd->is_powered_on_standby? What do you think?
Both looks fine.
Acked-by: Liam Girdwood lrg@slimlogic.co.uk