ASoC: set bias_level at if

Merge series from Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>:

ASoC has 2 functions to set bias level.

	(A) snd_soc_dapm_force_bias_level()
	(B) snd_soc_dapm_set_bias_level()

(A) sets dapm->bias_level, but (B) is not.
I think we should set it on both (A) and (B).
I think it is miss or bug, but Samsung is the only vendor that feels a
problem about this.

I think this patch (= [1/5]) is correct approach, but some non-samsung
vendor might get affect from this patch-set, so I added [RFC] on this
patch-set.

Furthermore, (B) cares both Card and Component, (A) cares Component only.
I guess it is the reason why it is called as "force" bias_level function.
(A) is used from each drivers, (B) is used from soc-dapm only.
I'm not 100% sure though, except special cases, each driver should use (B),
I guess ?

sound/soc/samsung/bells.c

@@ -133,8 +133,6 @@ static int bells_set_bias_level_post(struct snd_soc_card *card,
 		break;
 	}
 
-	dapm->bias_level = level;
-
 	return 0;
 }
 

sound/soc/samsung/littlemill.c

@@ -95,8 +95,6 @@ static int littlemill_set_bias_level_post(struct snd_soc_card *card,
 		break;
 	}
 
-	dapm->bias_level = level;
-
 	return 0;
 }
 

sound/soc/samsung/speyside.c

@@ -90,8 +90,6 @@ static int speyside_set_bias_level_post(struct snd_soc_card *card,
 		break;
 	}
 
-	card->dapm.bias_level = level;
-
 	return 0;
 }
 

sound/soc/samsung/tobermory.c

@@ -91,8 +91,6 @@ static int tobermory_set_bias_level_post(struct snd_soc_card *card,
 		break;
 	}
 
-	dapm->bias_level = level;
-
 	return 0;
 }
 

sound/soc/soc-dapm.c

@@ -1030,6 +1030,10 @@ static int snd_soc_dapm_set_bias_level(struct snd_soc_dapm_context *dapm,
 out:
 	trace_snd_soc_bias_level_done(dapm, level);
 
+	/* success */
+	if (ret == 0)
+		snd_soc_dapm_init_bias_level(dapm, level);
+
 	return ret;
 }