[PATCH 0/7] ASoC: codecs: ES8389: Modify the ES8389 driver

[PATCH 0/7] ASoC: codecs: ES8389: Modify the ES8389 driver

[Zhang Yi]

v0:
  - Modify volatile_register
  - Add MCLK_MASK to resolve the issue related to mclk_src
  - Modify configuration about clock
  - Add private members about HPF at driver and devicetree
  - Add INPUT MUX at route

Zhang Yi (7):
  ASoC: codecs: ES8389: Modify volatile_register
  ASoC: codecs: ES8389: Fix the issue about mclk_src
  ASoC: codecs: ES8389: Modify the clock table
  ASoC: codecs: ES8389: Modify the initial configuration
  ASoC: codecs: ES8389: Add private members related to HPF
  ASoC: codecs: ES8389: Add INPUTL MUX and INPUTR MUX
  ASoC: dt-bindings: ES8389: Add members about HPF and clock

 .../bindings/sound/everest,es8389.yaml        |  23 +++
 sound/soc/codecs/es8389.c                     | 141 +++++++++++++-----
 sound/soc/codecs/es8389.h                     |   6 +-
 3 files changed, 128 insertions(+), 42 deletions(-)

-- 
2.17.1

[PATCH 1/7] ASoC: codecs: ES8389: Modify volatile_register

[Zhang Yi]

Mark some registers that are not volatile as false
And modified the logic for `cache_bypass` during `8389_resume`.

Signed-off-by: Zhang Yi <zhangyi@everest-semi.com>
---
 sound/soc/codecs/es8389.c | 27 +++++++++++++++++++++++----
 1 file changed, 23 insertions(+), 4 deletions(-)

diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
index 449d9574b..be7a36f17 100644
--- a/sound/soc/codecs/es8389.c
+++ b/sound/soc/codecs/es8389.c
@@ -50,10 +50,29 @@ static const char * const es8389_core_supplies[] = {
 static bool es8389_volatile_register(struct device *dev,
 			unsigned int reg)
 {
-	if ((reg  <= 0xff))
-		return true;
-	else
+	switch (reg) {
+	case ES8389_ADCL_VOL:
+	case ES8389_ADCR_VOL:
+	case ES8389_MIC1_GAIN:
+	case ES8389_MIC2_GAIN:
+	case ES8389_DACL_VOL:
+	case ES8389_DACR_VOL:
+	case ES8389_ALC_ON:
+	case ES8389_ALC_CTL:
+	case ES8389_ALC_TARGET:
+	case ES8389_ALC_GAIN:
+	case ES8389_ADC_MUTE:
+	case ES8389_OSR_VOL:
+	case ES8389_DAC_INV:
+	case ES8389_MIX_VOL:
+	case ES8389_DAC_MIX:
+	case ES8389_ADC_RESET:
+	case ES8389_ADC_MODE:
+	case ES8389_DMIC_EN:
 		return false;
+	default:
+		return true;
+	}
 }
 
 static const DECLARE_TLV_DB_SCALE(dac_vol_tlv, -9550, 50, 0);
@@ -861,13 +880,13 @@ static int es8389_resume(struct snd_soc_component *component)
 	regcache_cache_only(es8389->regmap, false);
 	regcache_cache_bypass(es8389->regmap, true);
 	regmap_read(es8389->regmap, ES8389_RESET, &regv);
-	regcache_cache_bypass(es8389->regmap, false);
 
 	if (regv == 0xff)
 		es8389_init(component);
 	else
 		es8389_set_bias_level(component, SND_SOC_BIAS_ON);
 
+	regcache_cache_bypass(es8389->regmap, false);
 	regcache_sync(es8389->regmap);
 
 	return 0;
-- 
2.17.1

Re: [PATCH 1/7] ASoC: codecs: ES8389: Modify volatile_register

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:

Pre-existing issues:
- [High] Critical configuration registers are incorrectly marked as volatile, causing permanent state loss on suspend/resume.
- [High] Failure to check the return value of `regmap_read` leads to branching on and evaluating uninitialized stack variables.
--

commit 963a77368701e6888ce20ffbda8712f190e260d3
Author: Zhang Yi <zhangyi@everest-semi.com>

ASoC: codecs: ES8389: Modify volatile_register

This patch updates the es8389_volatile_register function to explicitly mark
certain registers as non-volatile, falling back to treating others as
volatile. It also adjusts the timing of when regcache_cache_bypass is
disabled in the es8389_resume function.

> diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
> index 449d9574b03a8..be7a36f17083f 100644
> --- a/sound/soc/codecs/es8389.c
> +++ b/sound/soc/codecs/es8389.c
> @@ -50,10 +50,29 @@ static const char * const es8389_core_supplies[] = {
>  static bool es8389_volatile_register(struct device *dev,
>  			unsigned int reg)
>  {
> -	if ((reg  <= 0xff))
> -		return true;
> -	else
> +	switch (reg) {
> +	case ES8389_ADCL_VOL:
> +	case ES8389_ADCR_VOL:
> +	case ES8389_MIC1_GAIN:
> +	case ES8389_MIC2_GAIN:
> +	case ES8389_DACL_VOL:
> +	case ES8389_DACR_VOL:
> +	case ES8389_ALC_ON:
> +	case ES8389_ALC_CTL:
> +	case ES8389_ALC_TARGET:
> +	case ES8389_ALC_GAIN:
> +	case ES8389_ADC_MUTE:
> +	case ES8389_OSR_VOL:
> +	case ES8389_DAC_INV:
> +	case ES8389_MIX_VOL:
> +	case ES8389_DAC_MIX:
> +	case ES8389_ADC_RESET:
> +	case ES8389_ADC_MODE:
> +	case ES8389_DMIC_EN:
>  		return false;
> +	default:
> +		return true;
> +	}
>  }

[Severity: High]
This is a pre-existing issue, but does defaulting to true for all other
registers omit key configuration registers from the cache?

Because volatile registers are skipped by regcache_sync() during resume,
registers such as ES8389_MASTER_MODE, TDM settings, and clock dividers will
not be restored if they are treated as volatile. Could this break the audio
formatting and cause playback or capture to fail after a suspend and resume
cycle?

[ ... ]

> @@ -861,13 +880,13 @@ static int es8389_resume(struct snd_soc_component *component)
>  	regcache_cache_only(es8389->regmap, false);
>  	regcache_cache_bypass(es8389->regmap, true);
>  	regmap_read(es8389->regmap, ES8389_RESET, &regv);
> -	regcache_cache_bypass(es8389->regmap, false);
>  
>  	if (regv == 0xff)

[Severity: High]
This is a pre-existing issue, but does the lack of a return value check on
regmap_read() leave regv uninitialized if the I2C read fails?

If the read fails, evaluating the uninitialized stack variable regv will
unpredictably dictate whether a full hardware initialization is performed.

This same pattern appears to exist in other places as well:

sound/soc/codecs/es8389.c:es8389_init() {
    ...
    regmap_read(es8389->regmap, ES8389_MAX_REGISTER, &reg);
    es8389->version = reg;
    ...
}

and

sound/soc/codecs/es8389.c:es8389_mute() {
    ...
    regmap_read(es8389->regmap, ES8389_CSM_STATE1, &regv);
    if (regv != ES8389_STATE_ON) {
    ...
}

Could these lead to inconsistent hardware states or silent audio failures?

>  		es8389_init(component);
>  	else
>  		es8389_set_bias_level(component, SND_SOC_BIAS_ON);
>  
> +	regcache_cache_bypass(es8389->regmap, false);
>  	regcache_sync(es8389->regmap);
>  
>  	return 0;

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260609071732.20939-1-zhangyi@everest-semi.com?part=1

[PATCH 2/7] ASoC: codecs: ES8389: Fix the issue about mclk_src

[Zhang Yi]

Fix the issue with incorrect modifications to mclk_src

Signed-off-by: Zhang Yi <zhangyi@everest-semi.com>
---
 sound/soc/codecs/es8389.c | 12 ++++--------
 sound/soc/codecs/es8389.h |  5 ++---
 2 files changed, 6 insertions(+), 11 deletions(-)

diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
index be7a36f17..886069be1 100644
--- a/sound/soc/codecs/es8389.c
+++ b/sound/soc/codecs/es8389.c
@@ -36,8 +36,8 @@ struct	es8389_private {
 	unsigned int sysclk;
 	int mastermode;
 
-	u8 mclk_src;
 	u8 vddd;
+	bool mclk_src;
 	int version;
 	enum snd_soc_bias_level bias_level;
 };
@@ -607,9 +607,9 @@ static int es8389_pcm_hw_params(struct snd_pcm_substream *substream,
 	regmap_update_bits(es8389->regmap, ES8389_ADC_FORMAT_MUTE, ES8389_DATA_LEN_MASK, state);
 	regmap_update_bits(es8389->regmap, ES8389_DAC_FORMAT_MUTE, ES8389_DATA_LEN_MASK, state);
 
-	if (es8389->mclk_src == ES8389_SCLK_PIN) {
+	if (es8389->mclk_src) {
 		regmap_update_bits(es8389->regmap, ES8389_MASTER_CLK,
-					ES8389_MCLK_SOURCE, es8389->mclk_src);
+					ES8389_MCLK_MASK, ES8389_MCLK_FROM_SCLK);
 		es8389->sysclk = params_channels(params) * params_width(params) * params_rate(params);
 	}
 
@@ -897,11 +897,7 @@ static int es8389_probe(struct snd_soc_component *component)
 	int ret, i;
 	struct es8389_private *es8389 = snd_soc_component_get_drvdata(component);
 
-	ret = device_property_read_u8(component->dev, "everest,mclk-src", &es8389->mclk_src);
-	if (ret != 0) {
-		dev_dbg(component->dev, "mclk-src return %d", ret);
-		es8389->mclk_src = ES8389_MCLK_SOURCE;
-	}
+	es8389->mclk_src = device_property_read_bool(component->dev, "everest,mclk-from-sclk");
 
 	for (i = 0; i < ARRAY_SIZE(es8389_core_supplies); i++)
 		es8389->core_supply[i].supply = es8389_core_supplies[i];
diff --git a/sound/soc/codecs/es8389.h b/sound/soc/codecs/es8389.h
index d21e72f87..7f1212219 100644
--- a/sound/soc/codecs/es8389.h
+++ b/sound/soc/codecs/es8389.h
@@ -116,9 +116,8 @@
 #define ES8389_TDM_SLOT               (0x70 << 0)
 #define ES8389_TDM_SHIFT              4
 
-#define ES8389_MCLK_SOURCE            (1 << 6)
-#define ES8389_MCLK_PIN               (1 << 6)
-#define ES8389_SCLK_PIN               (0 << 6)
+#define ES8389_MCLK_MASK              (3 << 6)
+#define ES8389_MCLK_FROM_SCLK         (1 << 6)
 
 /* ES8389_FMT */
 #define ES8389_S24_LE                 (0 << 5)
-- 
2.17.1

Re: [PATCH 2/7] ASoC: codecs: ES8389: Fix the issue about mclk_src

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:

Pre-existing issues:
- [High] Incorrect BCLK calculation for MCLK derivation
- [High] Missing check on the return value of regmap_read() during resume
- [High] Resource Leak: Regulators enabled during component probe are leaked on removal
--

commit 81fe94b62642e618ad9d244bb25b737cbe5017dd
Author: Zhang Yi <zhangyi@everest-semi.com>

ASoC: codecs: ES8389: Fix the issue about mclk_src

Fix the issue with incorrect modifications to mclk_src.

> diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
> index be7a36f17083f..886069be1535f 100644
> --- a/sound/soc/codecs/es8389.c
> +++ b/sound/soc/codecs/es8389.c
[ ... ]
> @@ -607,9 +607,9 @@ static int es8389_pcm_hw_params(struct snd_pcm_substream *substream,
>  	regmap_update_bits(es8389->regmap, ES8389_ADC_FORMAT_MUTE, ES8389_DATA_LEN_MASK, state);
>  	regmap_update_bits(es8389->regmap, ES8389_DAC_FORMAT_MUTE, ES8389_DATA_LEN_MASK, state);
>  
> -	if (es8389->mclk_src == ES8389_SCLK_PIN) {
> +	if (es8389->mclk_src) {
>  		regmap_update_bits(es8389->regmap, ES8389_MASTER_CLK,
> -					ES8389_MCLK_SOURCE, es8389->mclk_src);
> +					ES8389_MCLK_MASK, ES8389_MCLK_FROM_SCLK);
>  		es8389->sysclk = params_channels(params) * params_width(params) * params_rate(params);
                                                           ^^^^^^^^^^^^^^^^^^^^

[Severity: High]
This is a pre-existing issue, but does this incorrectly calculate the
physical BCLK frequency?

Using params_width() returns the sample data width instead of the physical
slot width on the wire. When mclk_src is set, es8389->sysclk acts as the MCLK
derived from SCLK.

By under-calculating the physical SCLK, the get_coeff() function is provided
with an incorrect MCLK frequency assumption, which can lead to the selection
of wrong internal clock dividers and distorted or missing audio.

Could the physical BCLK be calculated using snd_soc_params_to_bclk(params)
instead?

>  	}

Additionally, I noticed a few other pre-existing issues in this driver:

[Severity: High]
This isn't a bug introduced by this patch, but is there a missing check on the
return value of regmap_read() during resume?

In es8389_resume():

static int es8389_resume(struct snd_soc_component *component)
{
	...
	unsigned int regv;

	regcache_cache_only(es8389->regmap, false);
	regcache_cache_bypass(es8389->regmap, true);
	regmap_read(es8389->regmap, ES8389_RESET, &regv);

	if (regv == 0xff)
	...
}

If the I2C read fails during resume, the stack variable regv is left
uninitialized and evaluated against 0xff. Could this introduce undefined
behavior and lead to incorrect skipping or triggering of the hardware
re-initialization?


[Severity: High]
This isn't a bug introduced by this patch, but are the regulators enabled
during probe leaked on removal?

es8389_probe() successfully enables core regulators:

static int es8389_probe(struct snd_soc_component *component)
{
	...
	ret = regulator_bulk_enable(ARRAY_SIZE(es8389_core_supplies),
				    es8389->core_supply);
	...
}

However, es8389_remove() does not call the corresponding
regulator_bulk_disable().

Since devm_regulator_bulk_get() only automatically frees the handles but does
not implicitly disable the regulators, this leaks the regulator enable
reference count. Could this cause the power domains to remain permanently on
even when the module is unbound or removed?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260609071732.20939-1-zhangyi@everest-semi.com?part=2

[PATCH 3/7] ASoC: codecs: ES8389: Modify the clock table

[Zhang Yi]

Updated the configuration for certain frequencies
If get_coeff does not find a value that matches the condition, return -EINVAL

Signed-off-by: Zhang Yi <zhangyi@everest-semi.com>
---
 sound/soc/codecs/es8389.c | 49 +++++++++++++++++++++------------------
 1 file changed, 26 insertions(+), 23 deletions(-)

diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
index 886069be1..647d9d628 100644
--- a/sound/soc/codecs/es8389.c
+++ b/sound/soc/codecs/es8389.c
@@ -434,52 +434,54 @@ static const struct _coeff_div  coeff_div[] = {
 	{36, 576000, 16000, 0x00, 0x55, 0x84, 0xD0, 0x01, 0xC1, 0x90, 0x00, 0x00, 0x23, 0x8F, 0xBF, 0xC0, 0x1F, 0x8F, 0x01, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
 	{48, 768000, 16000, 0x02, 0x57, 0x04, 0xC0, 0x01, 0xC1, 0x90, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
 	{50, 800000, 16000, 0x00, 0x7E, 0x01, 0xD9, 0x00, 0xC2, 0x80, 0x00, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0xC7, 0x95, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
-	{64, 1024000, 16000, 0x00, 0x45, 0x24, 0xC0, 0x01, 0xD1, 0x90, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
-	{72, 1152000, 16000, 0x00, 0x45, 0x24, 0xC0, 0x01, 0xD1, 0x90, 0x00, 0x00, 0x23, 0x8F, 0xBF, 0xC0, 0x1F, 0x8F, 0x01, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
+	{64, 1024000, 16000, 0x00, 0x45, 0x24, 0xC0, 0x01, 0xC1, 0x90, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
+	{72, 1152000, 16000, 0x00, 0x45, 0x24, 0xC0, 0x01, 0xC1, 0x90, 0x00, 0x00, 0x23, 0x8F, 0xBF, 0xC0, 0x1F, 0x8F, 0x01, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
 	{96, 1536000, 16000, 0x02, 0x55, 0x84, 0xD0, 0x01, 0xC1, 0x90, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
 	{128, 2048000, 16000, 0x00, 0x51, 0x04, 0xD0, 0x01, 0xC1, 0x90, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
 	{144, 2304000, 16000, 0x00, 0x51, 0x00, 0xC0, 0x01, 0xC1, 0x90, 0x00, 0x00, 0x23, 0x8F, 0xBF, 0xC0, 0x1F, 0x8F, 0x01, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
+	{150, 2400000, 16000, 0x02, 0x7E, 0x01, 0xC9, 0x00, 0xC2, 0x80, 0x40, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0xC7, 0x95, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
 	{192, 3072000, 16000, 0x02, 0x65, 0x25, 0xE0, 0x00, 0xE1, 0x90, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
-	{256, 4096000, 16000, 0x00, 0x41, 0x04, 0xC0, 0x01, 0xD1, 0x90, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
+	{256, 4096000, 16000, 0x00, 0x41, 0x04, 0xC0, 0x01, 0xC1, 0x90, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
 	{300, 4800000, 16000, 0x02, 0x66, 0x01, 0xD9, 0x00, 0xC2, 0x80, 0x00, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0xC7, 0x95, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
 	{384, 6144000, 16000, 0x02, 0x51, 0x04, 0xD0, 0x01, 0xC1, 0x90, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
-	{512, 8192000, 16000, 0x01, 0x41, 0x04, 0xC0, 0x01, 0xD1, 0x90, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
+	{512, 8192000, 16000, 0x01, 0x41, 0x04, 0xC0, 0x01, 0xC1, 0x90, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
 	{750, 12000000, 16000, 0x0E, 0x7E, 0x01, 0xC9, 0x00, 0xC2, 0x80, 0x40, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0xC7, 0x95, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
-	{768, 12288000, 16000, 0x02, 0x41, 0x04, 0xC0, 0x01, 0xD1, 0x90, 0x40, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
-	{1024, 16384000, 16000, 0x03, 0x41, 0x04, 0xC0, 0x01, 0xD1, 0x90, 0x40, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
+	{768, 12288000, 16000, 0x02, 0x41, 0x04, 0xC0, 0x01, 0xC1, 0x90, 0x40, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
+	{1024, 16384000, 16000, 0x03, 0x41, 0x04, 0xC0, 0x01, 0xC1, 0x90, 0x40, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
 	{1152, 18432000, 16000, 0x08, 0x51, 0x04, 0xD0, 0x01, 0xC1, 0x90, 0x40, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
 	{1200, 19200000, 16000, 0x0B, 0x66, 0x01, 0xD9, 0x00, 0xC2, 0x80, 0x40, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0xC7, 0x95, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
 	{1500, 24000000, 16000, 0x0E, 0x26, 0x01, 0xD9, 0x00, 0xC2, 0x80, 0xC0, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0xC7, 0x95, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
-	{1536, 24576000, 16000, 0x05, 0x41, 0x04, 0xC0, 0x01, 0xD1, 0x90, 0xC0, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
+	{1536, 24576000, 16000, 0x05, 0x41, 0x04, 0xC0, 0x01, 0xC1, 0x90, 0xC0, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0xFF, 0x7F, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
 	{1625, 26000000, 16000, 0x40, 0x6E, 0x05, 0xC8, 0x01, 0xC2, 0x90, 0xC0, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x12, 0x31, 0x0E, 2, 2},
 	{800, 19200000, 24000, 0x07, 0x66, 0x01, 0xD9, 0x00, 0xC2, 0x80, 0x40, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0xC7, 0x95, 0x00, 0x12, 0x00, 0x1A, 0x49, 0x14, 2, 2},
 	{375, 12000000, 32000, 0x0E, 0x2E, 0x05, 0xC8, 0x00, 0xC2, 0x80, 0x40, 0x01, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x23, 0x61, 0x1B, 2, 0},
-	{600, 19200000, 32000, 0x05, 0x46, 0x01, 0xD8, 0x10, 0xD2, 0x80, 0x40, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x23, 0x61, 0x1B, 2, 2},
-	{32, 1411200, 44100, 0x00, 0x45, 0xA4, 0xD0, 0x10, 0xD1, 0x80, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
+	{600, 19200000, 32000, 0x05, 0x46, 0x01, 0xD8, 0x10, 0xC2, 0x80, 0x40, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x23, 0x61, 0x1B, 2, 2},
+	{32, 1411200, 44100, 0x00, 0x45, 0xA4, 0xD0, 0x10, 0xC1, 0x80, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
 	{64, 2822400, 44100, 0x00, 0x51, 0x00, 0xC0, 0x10, 0xC1, 0x80, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
-	{128, 5644800, 44100, 0x00, 0x41, 0x04, 0xD0, 0x10, 0xD1, 0x80, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
-	{256, 11289600, 44100, 0x01, 0x41, 0x04, 0xD0, 0x10, 0xD1, 0x80, 0x40, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
-	{512, 22579200, 44100, 0x03, 0x41, 0x04, 0xD0, 0x10, 0xD1, 0x80, 0xC0, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
-	{32, 1536000, 48000, 0x00, 0x45, 0xA4, 0xD0, 0x10, 0xD1, 0x80, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
+	{128, 5644800, 44100, 0x00, 0x41, 0x04, 0xD0, 0x10, 0xC1, 0x80, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
+	{256, 11289600, 44100, 0x01, 0x41, 0x04, 0xD0, 0x10, 0xC1, 0x80, 0x40, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
+	{512, 22579200, 44100, 0x03, 0x41, 0x04, 0xD0, 0x10, 0xC1, 0x80, 0xC0, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
+	{32, 1536000, 48000, 0x00, 0x45, 0xA4, 0xD0, 0x10, 0xC1, 0x80, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
 	{48, 2304000, 48000, 0x02, 0x55, 0x04, 0xC0, 0x10, 0xC1, 0x80, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
 	{50, 2400000, 48000, 0x00, 0x76, 0x01, 0xC8, 0x10, 0xC2, 0x80, 0x00, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
 	{64, 3072000, 48000, 0x00, 0x51, 0x04, 0xC0, 0x10, 0xC1, 0x80, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
-	{100, 4800000, 48000, 0x00, 0x46, 0x01, 0xD8, 0x10, 0xD2, 0x80, 0x00, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
+	{100, 4800000, 48000, 0x00, 0x46, 0x01, 0xD8, 0x10, 0xC2, 0x80, 0x00, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
 	{125, 6000000, 48000, 0x04, 0x6E, 0x05, 0xC8, 0x10, 0xC2, 0x80, 0x00, 0x01, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
-	{128, 6144000, 48000, 0x00, 0x41, 0x04, 0xD0, 0x10, 0xD1, 0x80, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
-	{200, 9600000, 48000, 0x01, 0x46, 0x01, 0xD8, 0x10, 0xD2, 0x80, 0x00, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
+	{128, 6144000, 48000, 0x00, 0x41, 0x04, 0xD0, 0x10, 0xC1, 0x80, 0x00, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
+	{200, 9600000, 48000, 0x01, 0x46, 0x01, 0xD8, 0x10, 0xC2, 0x80, 0x00, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
 	{250, 12000000, 48000, 0x04, 0x76, 0x01, 0xC8, 0x10, 0xC2, 0x80, 0x40, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
-	{256, 12288000, 48000, 0x01, 0x41, 0x04, 0xD0, 0x10, 0xD1, 0x80, 0x40, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
-	{384, 18432000, 48000, 0x02, 0x41, 0x04, 0xD0, 0x10, 0xD1, 0x80, 0x40, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
-	{400, 19200000, 48000, 0x03, 0x46, 0x01, 0xD8, 0x10, 0xD2, 0x80, 0x40, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
-	{500, 24000000, 48000, 0x04, 0x46, 0x01, 0xD8, 0x10, 0xD2, 0x80, 0xC0, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
-	{512, 24576000, 48000, 0x03, 0x41, 0x04, 0xD0, 0x10, 0xD1, 0x80, 0xC0, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
+	{256, 12288000, 48000, 0x01, 0x41, 0x04, 0xD0, 0x10, 0xC1, 0x80, 0x40, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
+	{384, 18432000, 48000, 0x02, 0x41, 0x04, 0xD0, 0x10, 0xC1, 0x80, 0x40, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
+	{400, 19200000, 48000, 0x03, 0x46, 0x01, 0xD8, 0x10, 0xC2, 0x80, 0x40, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
+	{500, 24000000, 48000, 0x04, 0x46, 0x01, 0xD8, 0x10, 0xC2, 0x80, 0xC0, 0x00, 0x18, 0x95, 0xD0, 0xC0, 0x63, 0x95, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
+	{512, 24576000, 48000, 0x03, 0x41, 0x04, 0xD0, 0x10, 0xC1, 0x80, 0xC0, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
 	{800, 38400000, 48000, 0x18, 0x45, 0x04, 0xC0, 0x10, 0xC1, 0x80, 0xC0, 0x00, 0x1F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x00, 0x12, 0x00, 0x35, 0x91, 0x28, 2, 2},
 	{128, 11289600, 88200, 0x00, 0x50, 0x00, 0xC0, 0x10, 0xC1, 0x80, 0x40, 0x00, 0x9F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x80, 0x12, 0xC0, 0x32, 0x89, 0x25, 2, 2},
-	{64, 6144000, 96000, 0x00, 0x41, 0x00, 0xD0, 0x10, 0xD1, 0x80, 0x00, 0x00, 0x9F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x80, 0x12, 0xC0, 0x35, 0x91, 0x28, 2, 2},
+	{64, 6144000, 96000, 0x00, 0x41, 0x00, 0xD0, 0x10, 0xC1, 0x80, 0x00, 0x00, 0x9F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x80, 0x12, 0xC0, 0x35, 0x91, 0x28, 2, 2},
 	{96, 9216000, 96000, 0x02, 0x43, 0x00, 0xC0, 0x10, 0xC0, 0x80, 0x00, 0x00, 0x9F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x80, 0x12, 0xC0, 0x35, 0x91, 0x28, 2, 2},
 	{256, 24576000, 96000, 0x00, 0x40, 0x00, 0xC0, 0x10, 0xC1, 0x80, 0xC0, 0x00, 0x9F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x80, 0x12, 0xC0, 0x35, 0x91, 0x28, 2, 2},
 	{128, 24576000, 192000, 0x00, 0x50, 0x00, 0xC0, 0x18, 0xC1, 0x81, 0xC0, 0x00, 0x8F, 0x7F, 0xBF, 0xC0, 0x3F, 0x7F, 0x80, 0x12, 0xC0, 0x3F, 0xF9, 0x3F, 2, 2},
+	{64, 12288000, 192000, 0x00, 0x41, 0x00, 0xC0, 0x18, 0xC1, 0x80, 0x00, 0x00, 0x8F, 0x7F, 0xEF, 0xC0, 0x7F, 0x7F, 0x80, 0x12, 0xC0, 0x3F, 0xF9, 0x3F, 1, 0},
 };
 
 static inline int get_coeff(u8 vddd, u8 dmic, int mclk, int rate)
@@ -657,7 +659,8 @@ static int es8389_pcm_hw_params(struct snd_pcm_substream *substream,
 		regmap_write(es8389->regmap, ES8389_SYSTEM18, coeff_div[coeff].Reg0x18);
 		regmap_write(es8389->regmap, ES8389_SYSTEM19, coeff_div[coeff].Reg0x19);
 	} else {
-		dev_warn(component->dev, "Clock coefficients do not match");
+		dev_err(component->dev, "Clock coefficients do not match");
+		return -EINVAL;
 	}
 
 	return 0;
-- 
2.17.1

Re: [PATCH 3/7] ASoC: codecs: ES8389: Modify the clock table

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] The `get_coeff` function incorrectly treats `dvdd_vol == 0` and `dmic_sel == 0` as wildcards, causing exact matches for `0` to fail and shadowing subsequent table entries.
--

commit 4e226f94a9447f64377fe0e23eb54e3295a2c321
Author: Zhang Yi <zhangyi@everest-semi.com>
ASoC: codecs: ES8389: Modify the clock table

Updated the configuration for certain frequencies and returns -EINVAL if
get_coeff does not find a match.

> diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
> index 886069be1535f..647d9d62820e6 100644
> --- a/sound/soc/codecs/es8389.c
> +++ b/sound/soc/codecs/es8389.c
> @@ -434,52 +434,54 @@ static const struct _coeff_div  coeff_div[] = {

[ ... ]

>  	{256, 24576000, 96000, 0x00, 0x40, 0x00, 0xC0, 0x10, 0xC1, 0x80, 0xC0, 0x00, 0x9F, 0x7F, 0xBF, 0xC0, 0x7F, 0x7F, 0x80, 0x12, 0xC0, 0x35, 0x91, 0x28, 2, 2},
>  	{128, 24576000, 192000, 0x00, 0x50, 0x00, 0xC0, 0x18, 0xC1, 0x81, 0xC0, 0x00, 0x8F, 0x7F, 0xBF, 0xC0, 0x3F, 0x7F, 0x80, 0x12, 0xC0, 0x3F, 0xF9, 0x3F, 2, 2},
> +	{64, 12288000, 192000, 0x00, 0x41, 0x00, 0xC0, 0x18, 0xC1, 0x80, 0x00, 0x00, 0x8F, 0x7F, 0xEF, 0xC0, 0x7F, 0x7F, 0x80, 0x12, 0xC0, 0x3F, 0xF9, 0x3F, 1, 0},
>  };
>  
>  static inline int get_coeff(u8 vddd, u8 dmic, int mclk, int rate)

[Severity: High]
This is a pre-existing issue, but does the modulo 2 arithmetic in get_coeff()
unintentionally treat 0 as a wildcard alongside 2?

Looking at get_coeff():

    vddd_det |= ~(coeff_div[i].dvdd_vol % 2) & 0x01;
    dmic_det |= ~(coeff_div[i].dmic_sel % 2) & 0x01;

Since 0 % 2 is 0, this causes ~0 & 1, which evaluates to 1.

This means a row specifying dmic_sel = 0 (like the newly added row above)
will incorrectly match requests where dmic = 1.

Similarly, if an earlier row specifies 0, will it shadow subsequent rows by
acting as a wildcard instead of an exact match?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260609071732.20939-1-zhangyi@everest-semi.com?part=3

[PATCH 4/7] ASoC: codecs: ES8389: Modify the initial configuration

[Zhang Yi]

Modify the initial configuration

Signed-off-by: Zhang Yi <zhangyi@everest-semi.com>
---
 sound/soc/codecs/es8389.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
index 647d9d628..284eb707f 100644
--- a/sound/soc/codecs/es8389.c
+++ b/sound/soc/codecs/es8389.c
@@ -793,7 +793,7 @@ static void es8389_init(struct snd_soc_component *component)
 
 	regmap_read(es8389->regmap, ES8389_MAX_REGISTER, &reg);
 	es8389->version = reg;
-	regmap_write(es8389->regmap, ES8389_ISO_CTL, 0x00);
+	regmap_write(es8389->regmap, ES8389_ISO_CTL, 0x56);
 	regmap_write(es8389->regmap, ES8389_RESET, 0x7E);
 	regmap_write(es8389->regmap, ES8389_ISO_CTL, 0x38);
 	regmap_write(es8389->regmap, ES8389_ADC_HPF1, 0x64);
@@ -845,7 +845,7 @@ static void es8389_init(struct snd_soc_component *component)
 	regmap_write(es8389->regmap, ES8389_SCLK_DIV, 0x04);
 	regmap_write(es8389->regmap, ES8389_LRCK_DIV1, 0x01);
 	regmap_write(es8389->regmap, ES8389_LRCK_DIV2, 0x00);
-	regmap_write(es8389->regmap, ES8389_OSC_CLK, 0x00);
+	regmap_write(es8389->regmap, ES8389_OSC_CLK, 0x10);
 	regmap_write(es8389->regmap, ES8389_ADC_OSR, 0x1F);
 	regmap_write(es8389->regmap, ES8389_ADC_DSP, 0x7F);
 	regmap_write(es8389->regmap, ES8389_ADC_MUTE, 0xC0);
-- 
2.17.1

Re: [PATCH 4/7] ASoC: codecs: ES8389: Modify the initial configuration

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] Calls to `regmap_read()` do not check for error return codes, leading to the use of uninitialized stack memory if the read fails.
--

commit 73684b8bb77238aedd3f4796976da733be77e2de
Author: Zhang Yi <zhangyi@everest-semi.com>

ASoC: codecs: ES8389: Modify the initial configuration

This commit updates several initialization registers in the es8389 codec.

> diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
> index 647d9d62820e6..284eb707f9f7f 100644
> --- a/sound/soc/codecs/es8389.c
> +++ b/sound/soc/codecs/es8389.c
> @@ -793,7 +793,7 @@ static void es8389_init(struct snd_soc_component *component)
>  
>  	regmap_read(es8389->regmap, ES8389_MAX_REGISTER, &reg);
>  	es8389->version = reg;

[Severity: High]
This is a pre-existing issue, but does this code handle regmap_read()
failures correctly?

If regmap_read() fails (e.g. due to I2C/SPI communication errors), it
returns a negative error code and leaves reg uninitialized. This
uninitialized stack memory is then permanently saved to es8389->version
and could affect later logic.

There appear to be a few other places in this file with the same pattern
where uninitialized stack memory might be used if regmap_read() fails:

es8389_pcm_hw_params():
	regmap_read(es8389->regmap, ES8389_DMIC_EN, &regv);
	dmic_enable = regv >> 7 & 0x01;

es8389_mute():
	regmap_read(es8389->regmap, ES8389_CSM_STATE1, &regv);
	if (regv != ES8389_STATE_ON) {

es8389_resume():
	regmap_read(es8389->regmap, ES8389_RESET, &regv);
	if (regv == 0xff)

Could we check the return value of regmap_read() in these locations to
avoid evaluating garbage data?

> -	regmap_write(es8389->regmap, ES8389_ISO_CTL, 0x00);
> +	regmap_write(es8389->regmap, ES8389_ISO_CTL, 0x56);

[ ... ]

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260609071732.20939-1-zhangyi@everest-semi.com?part=4

[PATCH 5/7] ASoC: codecs: ES8389: Add private members related to HPF

[Zhang Yi]

Add private members related to HPF.
And defines ES8389_HPF_DEFAULT

Signed-off-by: Zhang Yi <zhangyi@everest-semi.com>
---
 sound/soc/codecs/es8389.c | 18 ++++++++++++++++--
 sound/soc/codecs/es8389.h |  1 +
 2 files changed, 17 insertions(+), 2 deletions(-)

diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
index 284eb707f..2460cf977 100644
--- a/sound/soc/codecs/es8389.c
+++ b/sound/soc/codecs/es8389.c
@@ -36,6 +36,8 @@ struct	es8389_private {
 	unsigned int sysclk;
 	int mastermode;
 
+	u8 hpfl;
+	u8 hpfr;
 	u8 vddd;
 	bool mclk_src;
 	int version;
@@ -743,8 +745,8 @@ static int es8389_mute(struct snd_soc_dai *dai, int mute, int direction)
 			regmap_update_bits(es8389->regmap, ES8389_DAC_FORMAT_MUTE,
 						0x03, 0x00);
 		} else {
-			regmap_update_bits(es8389->regmap, ES8389_ADC_HPF1, 0x0f, 0x0a);
-			regmap_update_bits(es8389->regmap, ES8389_ADC_HPF2, 0x0f, 0x0a);
+			regmap_update_bits(es8389->regmap, ES8389_ADC_HPF1, 0x0f, es8389->hpfl);
+			regmap_update_bits(es8389->regmap, ES8389_ADC_HPF2, 0x0f, es8389->hpfr);
 			regmap_update_bits(es8389->regmap, ES8389_ADC_FORMAT_MUTE,
 						0x03, 0x00);
 		}
@@ -902,6 +904,18 @@ static int es8389_probe(struct snd_soc_component *component)
 
 	es8389->mclk_src = device_property_read_bool(component->dev, "everest,mclk-from-sclk");
 
+	ret = device_property_read_u8(component->dev, "everest,hpfl", &es8389->hpfl);
+	if (ret != 0) {
+		dev_dbg(component->dev, "hpfl return %d\n", ret);
+		es8389->hpfl = ES8389_HPF_DEFAULT;
+	}
+
+	ret = device_property_read_u8(component->dev, "everest,hpfr", &es8389->hpfr);
+	if (ret != 0) {
+		dev_dbg(component->dev, "hpfr return %d\n", ret);
+		es8389->hpfr = ES8389_HPF_DEFAULT;
+	}
+
 	for (i = 0; i < ARRAY_SIZE(es8389_core_supplies); i++)
 		es8389->core_supply[i].supply = es8389_core_supplies[i];
 	ret = devm_regulator_bulk_get(component->dev, ARRAY_SIZE(es8389_core_supplies), es8389->core_supply);
diff --git a/sound/soc/codecs/es8389.h b/sound/soc/codecs/es8389.h
index 7f1212219..81353dcec 100644
--- a/sound/soc/codecs/es8389.h
+++ b/sound/soc/codecs/es8389.h
@@ -105,6 +105,7 @@
 
 #define ES8389_MIC_SEL_MASK           (7 << 4)
 #define ES8389_MIC_DEFAULT            (1 << 4)
+#define ES8389_HPF_DEFAULT            (0x0A << 0)
 
 #define ES8389_MASTER_MODE_EN         (1 << 0)
 
-- 
2.17.1

Re: [PATCH 5/7] ASoC: codecs: ES8389: Add private members related to HPF

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Medium] ADC High Pass Filter (HPF) configurations are exposed as devicetree properties rather than ALSA kcontrols, violating ASoC subsystem conventions.
- [Medium] Using `device_property_read_u8()` for a scalar devicetree property creates a silent fallback trap if standard 32-bit devicetree cells are used.
--

commit 709f7c3cb7dfda3e2bb5a1b6319ef2eb235e0bae
Author: Zhang Yi <zhangyi@everest-semi.com>

ASoC: codecs: ES8389: Add private members related to HPF

This commit adds private members to manage the left and right High Pass
Filter (HPF) values on the ES8389 codec. It reads these configurations from
devicetree properties during probe and applies them in the un-mute sequence.

> diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
> index 284eb707f9f7f..2460cf9775c75 100644
> --- a/sound/soc/codecs/es8389.c
> +++ b/sound/soc/codecs/es8389.c

[ ... ]

> @@ -902,6 +904,18 @@ static int es8389_probe(struct snd_soc_component *component)
>  
>  	es8389->mclk_src = device_property_read_bool(component->dev, "everest,mclk-from-sclk");
>  
> +	ret = device_property_read_u8(component->dev, "everest,hpfl", &es8389->hpfl);

[Severity: Medium]
Devicetree is typically reserved for describing hardware topology, while
software-configurable tuning parameters like an ADC High Pass Filter are
usually exposed as ALSA kcontrols. Should these HPF values be registered as
mixer controls instead of devicetree properties so userspace can dynamically
adjust them at runtime?

[Severity: Medium]
Could using device_property_read_u8() here lead to a silent misconfiguration
if standard 32-bit devicetree cells are used?

If a devicetree specifies the property as a standard cell, for example:
    everest,hpfl = <10>;

The read_u8 function will read the first zero-padded byte (0x00) and return
success instead of an error. This means the fallback to ES8389_HPF_DEFAULT
will be bypassed, and the hardware will be configured with 0 instead. Would
it be safer to read this as a u32 using device_property_read_u32() and then
cast or bounds-check the result?

> +	if (ret != 0) {
> +		dev_dbg(component->dev, "hpfl return %d\n", ret);
> +		es8389->hpfl = ES8389_HPF_DEFAULT;
> +	}
> +
> +	ret = device_property_read_u8(component->dev, "everest,hpfr", &es8389->hpfr);
> +	if (ret != 0) {
> +		dev_dbg(component->dev, "hpfr return %d\n", ret);
> +		es8389->hpfr = ES8389_HPF_DEFAULT;
> +	}
> +

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260609071732.20939-1-zhangyi@everest-semi.com?part=5

[PATCH 6/7] ASoC: codecs: ES8389: Add INPUTL MUX and INPUTR MUX

[Zhang Yi]

Add INPUTL MUX and INPUTR MUX in route

Signed-off-by: Zhang Yi <zhangyi@everest-semi.com>
---
 sound/soc/codecs/es8389.c | 35 +++++++++++++++++++++++++++++++++--
 1 file changed, 33 insertions(+), 2 deletions(-)

diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
index 2460cf977..23e85ddfe 100644
--- a/sound/soc/codecs/es8389.c
+++ b/sound/soc/codecs/es8389.c
@@ -164,6 +164,16 @@ static const struct soc_enum alc_ramprate =
 static const struct soc_enum alc_winsize =
 	SOC_ENUM_SINGLE(ES8389_ALC_CTL, 0, 16, winsize);
 
+static const char *const es8389_adcl_mux_txt[] = {
+	"Normal",
+	"ADC2 channel to ADC1 channel",
+};
+
+static const char *const es8389_adcr_mux_txt[] = {
+	"Normal",
+	"ADC1 channel to ADC2 channel",
+};
+
 static const char *const es8389_outl_mux_txt[] = {
 	"Normal",
 	"DAC2 channel to DAC1 channel",
@@ -191,6 +201,20 @@ static const unsigned int es8389_pga_values[] = {
 	1, 5, 6
 };
 
+static const struct soc_enum es8389_adcl_mux_enum =
+	SOC_ENUM_SINGLE(ES8389_ADC_MODE, 5,
+			ARRAY_SIZE(es8389_adcl_mux_txt), es8389_adcl_mux_txt);
+
+static const struct snd_kcontrol_new es8389_adcl_mux_controls =
+	SOC_DAPM_ENUM("INPUTL MUX", es8389_adcl_mux_enum);
+
+static const struct soc_enum es8389_adcr_mux_enum =
+	SOC_ENUM_SINGLE(ES8389_ADC_MODE, 4,
+			ARRAY_SIZE(es8389_adcr_mux_txt), es8389_adcr_mux_txt);
+
+static const struct snd_kcontrol_new es8389_adcr_mux_controls =
+	SOC_DAPM_ENUM("INPUTR MUX", es8389_adcr_mux_enum);
+
 static const struct soc_enum es8389_outl_mux_enum =
 	SOC_ENUM_SINGLE(ES8389_DAC_MIX, 5,
 			ARRAY_SIZE(es8389_outl_mux_txt), es8389_outl_mux_txt);
@@ -319,6 +343,8 @@ static const struct snd_soc_dapm_widget es8389_dapm_widgets[] = {
 			   &es8389_adc_mixer_controls[0],
 			   ARRAY_SIZE(es8389_adc_mixer_controls)),
 	SND_SOC_DAPM_MUX("ADC MUX", SND_SOC_NOPM, 0, 0, &es8389_dmic_mux_controls),
+	SND_SOC_DAPM_MUX("INPUTL MUX", SND_SOC_NOPM, 0, 0, &es8389_adcl_mux_controls),
+	SND_SOC_DAPM_MUX("INPUTR MUX", SND_SOC_NOPM, 0, 0, &es8389_adcr_mux_controls),
 
 	SND_SOC_DAPM_MUX("OUTL MUX", SND_SOC_NOPM, 0, 0, &es8389_outl_mux_controls),
 	SND_SOC_DAPM_MUX("OUTR MUX", SND_SOC_NOPM, 0, 0, &es8389_outr_mux_controls),
@@ -332,10 +358,15 @@ static const struct snd_soc_dapm_route es8389_dapm_routes[] = {
 	{"ADCL", NULL, "PGAL"},
 	{"ADCR", NULL, "PGAR"},
 
+	{"INPUTL MUX", "Normal", "ADCL"},
+	{"INPUTL MUX", "ADC2 channel to ADC1 channel", "ADCR"},
+	{"INPUTR MUX", "Normal", "ADCR"},
+	{"INPUTR MUX", "ADC1 channel to ADC2 channel", "ADCL"},
+
 	{"ADC Mixer", "DACL ADCL Mixer", "DACL"},
 	{"ADC Mixer", "DACR ADCR Mixer", "DACR"},
-	{"ADC Mixer", NULL, "ADCL"},
-	{"ADC Mixer", NULL, "ADCR"},
+	{"ADC Mixer", NULL, "INPUTL MUX"},
+	{"ADC Mixer", NULL, "INPUTR MUX"},
 
 	{"ADC MUX", "AMIC", "ADC Mixer"},
 	{"ADC MUX", "DMIC", "DMIC"},
-- 
2.17.1

[PATCH 7/7] ASoC: dt-bindings: ES8389: Add members about HPF and clock

[Zhang Yi]

Add members related to HPF and mclk_source

Signed-off-by: Zhang Yi <zhangyi@everest-semi.com>
---
 .../bindings/sound/everest,es8389.yaml        | 23 +++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/Documentation/devicetree/bindings/sound/everest,es8389.yaml b/Documentation/devicetree/bindings/sound/everest,es8389.yaml
index 75ce0bc48..be92014c0 100644
--- a/Documentation/devicetree/bindings/sound/everest,es8389.yaml
+++ b/Documentation/devicetree/bindings/sound/everest,es8389.yaml
@@ -38,6 +38,27 @@ properties:
     description:
       Interface power supply.
 
+  everest,mclk-from-sclk:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description:
+      Indicates that SCLK is used as the internal clock.
+
+  everest,hpfl:
+    $ref: /schemas/types.yaml#/definitions/uint8
+    description:
+      the HPF value of ADCL.
+    minimum: 0x00
+    maximum: 0x0f
+    default: 0x0a
+
+  everest,hpfr:
+    $ref: /schemas/types.yaml#/definitions/uint8
+    description:
+      the HPF value of ADCR.
+    minimum: 0x00
+    maximum: 0x0f
+    default: 0x0a
+
 required:
   - compatible
   - reg
@@ -58,5 +79,7 @@ examples:
         #sound-dai-cells = <0>;
         vddd-supply = <&vdd3v3>;
         vdda-supply = <&vdd3v3>;
+        everest,hpfl = [0a];
+        everest,hpfr = [0a];
       };
     };
-- 
2.17.1

[PATCH 0/7] ASoC: codecs: ES8389: Modify the ES8389 driver

[Zhang Yi]

v0:
  - Modify volatile_register
  - Add MCLK_MASK to resolve the issue related to mclk_src
  - Modify configuration about clock
  - Add private members about HPF at driver and devicetree
  - Add INPUT MUX at route

Zhang Yi (7):
  ASoC: codecs: ES8389: Modify volatile_register
  ASoC: codecs: ES8389: Fix the issue about mclk_src
  ASoC: codecs: ES8389: Modify the clock table
  ASoC: codecs: ES8389: Modify the initial configuration
  ASoC: codecs: ES8389: Add private members related to HPF
  ASoC: codecs: ES8389: Add INPUTL MUX and INPUTR MUX
  ASoC: dt-bindings: ES8389: Add members about HPF and clock

 .../bindings/sound/everest,es8389.yaml        |  23 +++
 sound/soc/codecs/es8389.c                     | 136 +++++++++++++-----
 sound/soc/codecs/es8389.h                     |   6 +-
 3 files changed, 125 insertions(+), 40 deletions(-)

-- 
2.17.1

[PATCH 1/7] ASoC: codecs: ES8389: Modify volatile_register

[Zhang Yi]

Mark some registers that are not volatile as false

Signed-off-by: Zhang Yi <zhangyi@everest-semi.com>
---
 sound/soc/codecs/es8389.c | 25 ++++++++++++++++++++++---
 1 file changed, 22 insertions(+), 3 deletions(-)

diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
index 449d9574b..9c8164916 100644
--- a/sound/soc/codecs/es8389.c
+++ b/sound/soc/codecs/es8389.c
@@ -50,10 +50,29 @@ static const char * const es8389_core_supplies[] = {
 static bool es8389_volatile_register(struct device *dev,
 			unsigned int reg)
 {
-	if ((reg  <= 0xff))
-		return true;
-	else
+	switch (reg) {
+	case ES8389_ADC_HPF1:
+	case ES8389_ADC_HPF2:
+	case ES8389_ADCL_VOL:
+	case ES8389_ADCR_VOL:
+	case ES8389_MIC1_GAIN:
+	case ES8389_MIC2_GAIN:
+	case ES8389_DACL_VOL:
+	case ES8389_DACR_VOL:
+	case ES8389_ALC_ON:
+	case ES8389_ALC_CTL:
+	case ES8389_ALC_TARGET:
+	case ES8389_ALC_GAIN:
+	case ES8389_ADC_MUTE:
+	case ES8389_OSR_VOL:
+	case ES8389_DAC_INV:
+	case ES8389_MIX_VOL:
+	case ES8389_DAC_MIX:
+	case ES8389_ADC_RESET:
 		return false;
+	default:
+		return true;
+	}
 }
 
 static const DECLARE_TLV_DB_SCALE(dac_vol_tlv, -9550, 50, 0);
-- 
2.17.1

Re: [PATCH 1/7] ASoC: codecs: ES8389: Modify volatile_register

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:

New issues:
- [High] The regmap cache for non-volatile registers is overwritten with hardcoded default values during resume if the device lost power, neutralizing the patch's fix.

Pre-existing issues:
- [High] Critical configuration registers modified during stream setup are marked as volatile, causing them to be lost upon suspend/resume.
- [High] The variable `regv` is evaluated uninitialized if `regmap_read` fails during resume.
--

commit 146c46405cf59e3e0362628efb42fd4f52cff71a
Author: Zhang Yi <zhangyi@everest-semi.com>

ASoC: codecs: ES8389: Modify volatile_register

This patch modifies es8389_volatile_register() to mark a specific set of 
registers as non-volatile, enabling regcache for them. All other registers 
now default to being volatile.

> diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
> index 449d9574b03a8..9c816491648c7 100644
> --- a/sound/soc/codecs/es8389.c
> +++ b/sound/soc/codecs/es8389.c
> @@ -50,10 +50,29 @@ static const char * const es8389_core_supplies[] = {
>  static bool es8389_volatile_register(struct device *dev,
>  			unsigned int reg)
>  {
> -	if ((reg  <= 0xff))
> -		return true;
> -	else
> +	switch (reg) {
> +	case ES8389_ADC_HPF1:
> +	case ES8389_ADC_HPF2:

[ ... ]

> +	case ES8389_ADC_RESET:
>  		return false;
> +	default:
> +		return true;

[Severity: High]
This is a pre-existing issue, but does this fall-through prevent critical
configuration registers from being restored by regcache_sync()?

Critical configuration registers modified during stream setup like
ES8389_DMIC_EN, ES8389_ADC_MODE, ES8389_ADC_FORMAT_MUTE, and ES8389_ADC_OSR
remain uncached.

Since the ALSA framework does not re-invoke hw_params on resume, these
registers must be cached and restored by regcache_sync(). Because they are
volatile, active audio streams will fail or lose routing after a
suspend/resume cycle.

> +	}
>  }

[Severity: High]
By newly marking these registers as non-volatile, does this cause the
regcache to be overwritten with defaults during a power-loss resume?

In sound/soc/codecs/es8389.c:es8389_resume(), if the device lost power
(regv == 0xff), es8389_init() is called without regcache_cache_bypass being
enabled:

    regmap_read(es8389->regmap, ES8389_RESET, &regv);
    regcache_cache_bypass(es8389->regmap, false);

    if (regv == 0xff)
        es8389_init(component);

Inside es8389_init(), direct regmap_write() calls are made:

    regmap_write(es8389->regmap, ES8389_ADC_HPF1, 0x64);

Because bypass is disabled, these writes clobber the newly added regmap cache
with hardcoded default values. The subsequent regcache_sync() then syncs these
defaults instead of the user's prior settings, breaking state restoration.

[Severity: High]
This is a pre-existing issue, but can regv be evaluated uninitialized if the
regmap_read fails during resume?

In sound/soc/codecs/es8389.c:es8389_resume():

    regmap_read(es8389->regmap, ES8389_RESET, &regv);
    regcache_cache_bypass(es8389->regmap, false);

    if (regv == 0xff)

If regmap_read() fails (which can occur during resume due to bus readiness
issues), regv contains uninitialized stack memory. The subsequent evaluation
against 0xff produces undefined behavior, potentially skipping necessary
device initialization and leading to silent failure.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260609030623.17404-1-zhangyi@everest-semi.com?part=1

[PATCH 0/7] ASoC: codecs: ES8389: Modify the ES8389 driver

[Zhang Yi]

v0:
  - Modify volatile_register
  - Add MCLK_MASK to resolve the issue related to mclk_src
  - Modify configuration about clock
  - Add private members about HPF at driver and devicetree
  - Add INPUT MUX at route

Zhang Yi (7):
  ASoC: codecs: ES8389: Modify volatile_register
  ASoC: codecs: ES8389: Fix the issue about mclk_src
  ASoC: codecs: ES8389: Modify the clock table
  ASoC: codecs: ES8389: Modify the initial configuration
  ASoC: codecs: ES8389: Add private members related to HPF
  ASoC: codecs: ES8389: Add INPUTL MUX and INPUTR MUX
  ASoC: dt-bindings: ES8389: Add members about HPF

 .../bindings/sound/everest,es8389.yaml        |  23 +++
 sound/soc/codecs/es8389.c                     | 136 +++++++++++++-----
 sound/soc/codecs/es8389.h                     |   6 +-
 3 files changed, 125 insertions(+), 40 deletions(-)

-- 
2.17.1

[PATCH 1/7] ASoC: codecs: ES8389: Modify volatile_register

[Zhang Yi]

Mark some registers that are not volatile as false

Signed-off-by: Zhang Yi <zhangyi@everest-semi.com>
---
 sound/soc/codecs/es8389.c | 25 ++++++++++++++++++++++---
 1 file changed, 22 insertions(+), 3 deletions(-)

diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
index 449d9574b..9c8164916 100644
--- a/sound/soc/codecs/es8389.c
+++ b/sound/soc/codecs/es8389.c
@@ -50,10 +50,29 @@ static const char * const es8389_core_supplies[] = {
 static bool es8389_volatile_register(struct device *dev,
 			unsigned int reg)
 {
-	if ((reg  <= 0xff))
-		return true;
-	else
+	switch (reg) {
+	case ES8389_ADC_HPF1:
+	case ES8389_ADC_HPF2:
+	case ES8389_ADCL_VOL:
+	case ES8389_ADCR_VOL:
+	case ES8389_MIC1_GAIN:
+	case ES8389_MIC2_GAIN:
+	case ES8389_DACL_VOL:
+	case ES8389_DACR_VOL:
+	case ES8389_ALC_ON:
+	case ES8389_ALC_CTL:
+	case ES8389_ALC_TARGET:
+	case ES8389_ALC_GAIN:
+	case ES8389_ADC_MUTE:
+	case ES8389_OSR_VOL:
+	case ES8389_DAC_INV:
+	case ES8389_MIX_VOL:
+	case ES8389_DAC_MIX:
+	case ES8389_ADC_RESET:
 		return false;
+	default:
+		return true;
+	}
 }
 
 static const DECLARE_TLV_DB_SCALE(dac_vol_tlv, -9550, 50, 0);
-- 
2.17.1

Re: [PATCH 1/7] ASoC: codecs: ES8389: Modify volatile_register

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:

Pre-existing issues:
- [High] Incomplete volatile register list leaves critical configuration un-cached, causing state loss on resume.
- [High] regmap cache is clobbered with hardware defaults during resume, nullifying the preservation of user settings.
--

commit 18a4d58b5b0816359115dead13dc5872d2f0a08a
Author: Zhang Yi <zhangyi@everest-semi.com>

ASoC: codecs: ES8389: Modify volatile_register

Mark some registers that are not volatile as false.

> diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
> index 449d9574b03a8..9c816491648c7 100644
> --- a/sound/soc/codecs/es8389.c
> +++ b/sound/soc/codecs/es8389.c
> @@ -50,10 +50,29 @@ static const char * const es8389_core_supplies[] = {
>  static bool es8389_volatile_register(struct device *dev,
>  			unsigned int reg)
>  {
> -	if ((reg  <= 0xff))
> -		return true;
> -	else
> +	switch (reg) {
> +	case ES8389_ADC_HPF1:
> +	case ES8389_ADC_HPF2:
> +	case ES8389_ADCL_VOL:
> +	case ES8389_ADCR_VOL:
> +	case ES8389_MIC1_GAIN:
> +	case ES8389_MIC2_GAIN:

[Severity: High]
This is a pre-existing issue, but does making these registers non-volatile
cause their cached user settings to be clobbered during resume?

In es8389_resume(), regcache_cache_bypass(es8389->regmap, false) is called
before invoking es8389_init():

sound/soc/codecs/es8389.c:es8389_resume() {
    ...
	regcache_cache_bypass(es8389->regmap, false);

	if (regv == 0xff)
		es8389_init(component);
    ...
}

Because caching is enabled during es8389_init(), its explicit writes to these
registers overwrite the user's settings in the regmap cache with hardware
defaults:

sound/soc/codecs/es8389.c:es8389_init() {
    ...
	regmap_write(es8389->regmap, ES8389_ADC_MUTE, 0xC0);
    ...
}

When regcache_sync() subsequently runs, it syncs these defaults instead of the
user's configuration, which silently mutes the microphone and resets its gain
upon every resume.

> +	case ES8389_DACL_VOL:
> +	case ES8389_DACR_VOL:
> +	case ES8389_ALC_ON:
> +	case ES8389_ALC_CTL:
> +	case ES8389_ALC_TARGET:
> +	case ES8389_ALC_GAIN:
> +	case ES8389_ADC_MUTE:
> +	case ES8389_OSR_VOL:
> +	case ES8389_DAC_INV:
> +	case ES8389_MIX_VOL:
> +	case ES8389_DAC_MIX:
> +	case ES8389_ADC_RESET:
>  		return false;
> +	default:
> +		return true;

[Severity: High]
This is a pre-existing issue, but does this default case incorrectly omit
essential configuration registers?

It leaves important configuration registers like ES8389_MASTER_MODE,
ES8389_ADC_MODE, ES8389_DMIC_EN, ES8389_ADC_FORMAT_MUTE, and
ES8389_DAC_FORMAT_MUTE as volatile.

Because volatile registers bypass the regmap cache, any user changes, such as
DAI formats and DAPM MUX selections, are not cached.

When es8389_init() writes default values to these registers, regcache_sync()
skips them, so the user's MUX and format configurations are never restored,
breaking audio routing.

> +	}
>  }

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260609025605.16945-1-zhangyi@everest-semi.com?part=1