* [PATCH 1/4] ASoC: rt5645: Redefine format config for rt5650
@ 2015-03-27 12:19 Bard Liao
2015-03-27 12:19 ` [PATCH 2/4] ASoC: rt5645: Set use_single_rw flag for regmap Bard Liao
` (3 more replies)
0 siblings, 4 replies; 5+ messages in thread
From: Bard Liao @ 2015-03-27 12:19 UTC (permalink / raw)
To: broonie, lgirdwood
Cc: oder_chiou, alsa-devel, lars, john.lin, Bard Liao, jason.ym.lin, flove
rt5650 and rt5645 use different register bits for format configuration.
This patch modifies rt5645_hw_params and rt5645_set_dai_fmt to support
both codecs.
Signed-off-by: Bard Liao <bardliao@realtek.com>
---
Mark,
I am think if we should use different codec driver for rt5645 and
rt5650. In fact, they are different Die. What do you think?
---
---
sound/soc/codecs/rt5645.c | 39 +++++++++++++++++++++++++++++----------
1 file changed, 29 insertions(+), 10 deletions(-)
diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c
index 4c384a1..0133c8c 100644
--- a/sound/soc/codecs/rt5645.c
+++ b/sound/soc/codecs/rt5645.c
@@ -2049,7 +2049,7 @@ static int rt5645_hw_params(struct snd_pcm_substream *substream,
{
struct snd_soc_codec *codec = dai->codec;
struct rt5645_priv *rt5645 = snd_soc_codec_get_drvdata(codec);
- unsigned int val_len = 0, val_clk, mask_clk;
+ unsigned int val_len = 0, val_clk, mask_clk, dl_sft;
int pre_div, bclk_ms, frame_size;
rt5645->lrck[dai->id] = params_rate(params);
@@ -2063,6 +2063,16 @@ static int rt5645_hw_params(struct snd_pcm_substream *substream,
dev_err(codec->dev, "Unsupported frame size: %d\n", frame_size);
return -EINVAL;
}
+
+ switch (rt5645->codec_type) {
+ case CODEC_TYPE_RT5650:
+ dl_sft = 4;
+ break;
+ default:
+ dl_sft = 2;
+ break;
+ }
+
bclk_ms = frame_size > 32;
rt5645->bclk[dai->id] = rt5645->lrck[dai->id] * (32 << bclk_ms);
@@ -2075,13 +2085,13 @@ static int rt5645_hw_params(struct snd_pcm_substream *substream,
case 16:
break;
case 20:
- val_len |= RT5645_I2S_DL_20;
+ val_len = 0x1;
break;
case 24:
- val_len |= RT5645_I2S_DL_24;
+ val_len = 0x2;
break;
case 8:
- val_len |= RT5645_I2S_DL_8;
+ val_len = 0x3;
break;
default:
return -EINVAL;
@@ -2093,7 +2103,7 @@ static int rt5645_hw_params(struct snd_pcm_substream *substream,
val_clk = bclk_ms << RT5645_I2S_BCLK_MS1_SFT |
pre_div << RT5645_I2S_PD1_SFT;
snd_soc_update_bits(codec, RT5645_I2S1_SDP,
- RT5645_I2S_DL_MASK, val_len);
+ (0x3 << dl_sft), (val_len << dl_sft));
snd_soc_update_bits(codec, RT5645_ADDA_CLK1, mask_clk, val_clk);
break;
case RT5645_AIF2:
@@ -2101,7 +2111,7 @@ static int rt5645_hw_params(struct snd_pcm_substream *substream,
val_clk = bclk_ms << RT5645_I2S_BCLK_MS2_SFT |
pre_div << RT5645_I2S_PD2_SFT;
snd_soc_update_bits(codec, RT5645_I2S2_SDP,
- RT5645_I2S_DL_MASK, val_len);
+ (0x3 << dl_sft), (val_len << dl_sft));
snd_soc_update_bits(codec, RT5645_ADDA_CLK1, mask_clk, val_clk);
break;
default:
@@ -2116,7 +2126,16 @@ static int rt5645_set_dai_fmt(struct snd_soc_dai *dai, unsigned int fmt)
{
struct snd_soc_codec *codec = dai->codec;
struct rt5645_priv *rt5645 = snd_soc_codec_get_drvdata(codec);
- unsigned int reg_val = 0;
+ unsigned int reg_val = 0, pol_sft;
+
+ switch (rt5645->codec_type) {
+ case CODEC_TYPE_RT5650:
+ pol_sft = 8;
+ break;
+ default:
+ pol_sft = 7;
+ break;
+ }
switch (fmt & SND_SOC_DAIFMT_MASTER_MASK) {
case SND_SOC_DAIFMT_CBM_CFM:
@@ -2134,7 +2153,7 @@ static int rt5645_set_dai_fmt(struct snd_soc_dai *dai, unsigned int fmt)
case SND_SOC_DAIFMT_NB_NF:
break;
case SND_SOC_DAIFMT_IB_NF:
- reg_val |= RT5645_I2S_BP_INV;
+ reg_val |= (1 << pol_sft);
break;
default:
return -EINVAL;
@@ -2158,12 +2177,12 @@ static int rt5645_set_dai_fmt(struct snd_soc_dai *dai, unsigned int fmt)
switch (dai->id) {
case RT5645_AIF1:
snd_soc_update_bits(codec, RT5645_I2S1_SDP,
- RT5645_I2S_MS_MASK | RT5645_I2S_BP_MASK |
+ RT5645_I2S_MS_MASK | (1 << pol_sft) |
RT5645_I2S_DF_MASK, reg_val);
break;
case RT5645_AIF2:
snd_soc_update_bits(codec, RT5645_I2S2_SDP,
- RT5645_I2S_MS_MASK | RT5645_I2S_BP_MASK |
+ RT5645_I2S_MS_MASK | (1 << pol_sft) |
RT5645_I2S_DF_MASK, reg_val);
break;
default:
--
1.8.1.1.439.g50a6b54
^ permalink raw reply related [flat|nested] 5+ messages in thread* [PATCH 2/4] ASoC: rt5645: Set use_single_rw flag for regmap
2015-03-27 12:19 [PATCH 1/4] ASoC: rt5645: Redefine format config for rt5650 Bard Liao
@ 2015-03-27 12:19 ` Bard Liao
2015-03-27 12:19 ` [PATCH 3/4] ASoC: rt5645: Use update_bits for bit control Bard Liao
` (2 subsequent siblings)
3 siblings, 0 replies; 5+ messages in thread
From: Bard Liao @ 2015-03-27 12:19 UTC (permalink / raw)
To: broonie, lgirdwood
Cc: oder_chiou, alsa-devel, lars, john.lin, Bard Liao, jason.ym.lin, flove
RT5645 doesn't support auto incrementing writes so driver should set
the use_single_rw flag for regmap.
Signed-off-by: Bard Liao <bardliao@realtek.com>
---
sound/soc/codecs/rt5645.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c
index 0133c8c..f9edf09 100644
--- a/sound/soc/codecs/rt5645.c
+++ b/sound/soc/codecs/rt5645.c
@@ -2633,7 +2633,7 @@ static struct snd_soc_codec_driver soc_codec_dev_rt5645 = {
static const struct regmap_config rt5645_regmap = {
.reg_bits = 8,
.val_bits = 16,
-
+ .use_single_rw = true,
.max_register = RT5645_VENDOR_ID2 + 1 + (ARRAY_SIZE(rt5645_ranges) *
RT5645_PR_SPACING),
.volatile_reg = rt5645_volatile_register,
--
1.8.1.1.439.g50a6b54
^ permalink raw reply related [flat|nested] 5+ messages in thread* [PATCH 3/4] ASoC: rt5645: Use update_bits for bit control
2015-03-27 12:19 [PATCH 1/4] ASoC: rt5645: Redefine format config for rt5650 Bard Liao
2015-03-27 12:19 ` [PATCH 2/4] ASoC: rt5645: Set use_single_rw flag for regmap Bard Liao
@ 2015-03-27 12:19 ` Bard Liao
2015-03-27 12:19 ` [PATCH 4/4] ASoC: rt5645: Restore HP depop setting in HP off Bard Liao
2015-03-27 16:51 ` [PATCH 1/4] ASoC: rt5645: Redefine format config for rt5650 Mark Brown
3 siblings, 0 replies; 5+ messages in thread
From: Bard Liao @ 2015-03-27 12:19 UTC (permalink / raw)
To: broonie, lgirdwood
Cc: oder_chiou, alsa-devel, lars, john.lin, Bard Liao, jason.ym.lin, flove
In codec bias level off, we need to disable gate mode with MCLK
for power saving. It is set by one bit. We don't need to write
while register for that.
Signed-off-by: Bard Liao <bardliao@realtek.com>
---
sound/soc/codecs/rt5645.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c
index f9edf09..b6d5b95 100644
--- a/sound/soc/codecs/rt5645.c
+++ b/sound/soc/codecs/rt5645.c
@@ -2396,7 +2396,8 @@ static int rt5645_set_bias_level(struct snd_soc_codec *codec,
case SND_SOC_BIAS_OFF:
snd_soc_write(codec, RT5645_DEPOP_M2, 0x1100);
- snd_soc_write(codec, RT5645_GEN_CTRL1, 0x0128);
+ snd_soc_update_bits(codec, RT5645_GEN_CTRL1,
+ RT5645_DIG_GATE_CTRL, 0);
snd_soc_update_bits(codec, RT5645_PWR_ANLG1,
RT5645_PWR_VREF1 | RT5645_PWR_MB |
RT5645_PWR_BG | RT5645_PWR_VREF2 |
--
1.8.1.1.439.g50a6b54
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH 4/4] ASoC: rt5645: Restore HP depop setting in HP off
2015-03-27 12:19 [PATCH 1/4] ASoC: rt5645: Redefine format config for rt5650 Bard Liao
2015-03-27 12:19 ` [PATCH 2/4] ASoC: rt5645: Set use_single_rw flag for regmap Bard Liao
2015-03-27 12:19 ` [PATCH 3/4] ASoC: rt5645: Use update_bits for bit control Bard Liao
@ 2015-03-27 12:19 ` Bard Liao
2015-03-27 16:51 ` [PATCH 1/4] ASoC: rt5645: Redefine format config for rt5650 Mark Brown
3 siblings, 0 replies; 5+ messages in thread
From: Bard Liao @ 2015-03-27 12:19 UTC (permalink / raw)
To: broonie, lgirdwood
Cc: oder_chiou, alsa-devel, lars, john.lin, Bard Liao, jason.ym.lin, flove
This driver will set RT5645_DEPOP_MAN bit in headphone power up
depop process. We need to restore it in headphone power down
process. Otherwise, we will get headphone noise when push button
function is enabled.
Signed-off-by: Bard Liao <bardliao@realtek.com>
---
sound/soc/codecs/rt5645.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c
index b6d5b95..69528ae 100644
--- a/sound/soc/codecs/rt5645.c
+++ b/sound/soc/codecs/rt5645.c
@@ -1270,6 +1270,8 @@ static void hp_amp_power(struct snd_soc_codec *codec, int on)
snd_soc_update_bits(codec, RT5645_PWR_ANLG1,
RT5645_PWR_HP_L | RT5645_PWR_HP_R |
RT5645_PWR_HA, 0);
+ snd_soc_update_bits(codec, RT5645_DEPOP_M2,
+ RT5645_DEPOP_MASK, 0);
}
}
}
--
1.8.1.1.439.g50a6b54
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH 1/4] ASoC: rt5645: Redefine format config for rt5650
2015-03-27 12:19 [PATCH 1/4] ASoC: rt5645: Redefine format config for rt5650 Bard Liao
` (2 preceding siblings ...)
2015-03-27 12:19 ` [PATCH 4/4] ASoC: rt5645: Restore HP depop setting in HP off Bard Liao
@ 2015-03-27 16:51 ` Mark Brown
3 siblings, 0 replies; 5+ messages in thread
From: Mark Brown @ 2015-03-27 16:51 UTC (permalink / raw)
To: Bard Liao
Cc: oder_chiou, alsa-devel, lars, lgirdwood, john.lin, jason.ym.lin, flove
[-- Attachment #1.1: Type: text/plain, Size: 849 bytes --]
On Fri, Mar 27, 2015 at 08:19:06PM +0800, Bard Liao wrote:
> rt5650 and rt5645 use different register bits for format configuration.
> This patch modifies rt5645_hw_params and rt5645_set_dai_fmt to support
> both codecs.
Applied all, thanks.
> I am think if we should use different codec driver for rt5645 and
> rt5650. In fact, they are different Die. What do you think?
Well, it depends on how big the differences are for software really. If
they are mostly register compatible then it's just going to be a few
differences and the same code will work for both with small bits of per
device code. If there are too many differences then you end up with
more code handling those differences than is shared and it becomes
better to split the drivers. Right now this doesn't look too different
but I don't know what other changes you might need.
[-- Attachment #1.2: Digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]
[-- Attachment #2: Type: text/plain, Size: 0 bytes --]
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2015-03-27 16:50 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-03-27 12:19 [PATCH 1/4] ASoC: rt5645: Redefine format config for rt5650 Bard Liao
2015-03-27 12:19 ` [PATCH 2/4] ASoC: rt5645: Set use_single_rw flag for regmap Bard Liao
2015-03-27 12:19 ` [PATCH 3/4] ASoC: rt5645: Use update_bits for bit control Bard Liao
2015-03-27 12:19 ` [PATCH 4/4] ASoC: rt5645: Restore HP depop setting in HP off Bard Liao
2015-03-27 16:51 ` [PATCH 1/4] ASoC: rt5645: Redefine format config for rt5650 Mark Brown
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.