[PATCH] ASoC: rt722: reset codec to fix abnormal sound

[PATCH] ASoC: rt722: reset codec to fix abnormal sound

[shumingf]

From: Shuming Fan <shumingf@realtek.com>

The audio output may become abnormal after a warm reboot from Windows.
Reset the codec once during hardware initialization to restore it to a known
state and prevent the issue.
BugLink: https://github.com/thesofproject/linux/issues/5845

Signed-off-by: Shuming Fan <shumingf@realtek.com>
---
 sound/soc/codecs/rt722-sdca-sdw.c |  4 ++++
 sound/soc/codecs/rt722-sdca.c     | 14 ++++++++++++++
 2 files changed, 18 insertions(+)

diff --git a/sound/soc/codecs/rt722-sdca-sdw.c b/sound/soc/codecs/rt722-sdca-sdw.c
index 05e8e23f1f43..5bc9aed58d14 100644
--- a/sound/soc/codecs/rt722-sdca-sdw.c
+++ b/sound/soc/codecs/rt722-sdca-sdw.c
@@ -114,6 +114,7 @@ static int rt722_sdca_mbq_size(struct device *dev, unsigned int reg)
 	case 0x2000090 ... 0x2000094:
 	case 0x20000b1:
 	case 0x20000b4:
+	case 0x3010000:
 	case 0x3110000:
 	case 0x5300000 ... 0x5300300:
 	case 0x5400002:
@@ -197,15 +198,18 @@ static bool rt722_sdca_volatile_register(struct device *dev, unsigned int reg)
 	case 0x2000007:
 	case 0x200000d:
 	case 0x2000019:
+	case 0x200001a:
 	case 0x2000020:
 	case 0x2000030:
 	case 0x2000046:
 	case 0x2000067:
 	case 0x2000084:
 	case 0x2000086:
+	case 0x3010000:
 	case 0x3110000:
 	case 0x5800003:
 	case 0x5810000:
+	case 0x6100008:
 	case 0x44011000 ... 0x440115ff:
 	case 0x44012000:
 	case 0x44012021:
diff --git a/sound/soc/codecs/rt722-sdca.c b/sound/soc/codecs/rt722-sdca.c
index decf9407ab6d..a26253864d48 100644
--- a/sound/soc/codecs/rt722-sdca.c
+++ b/sound/soc/codecs/rt722-sdca.c
@@ -1851,6 +1851,17 @@ static void rt722_sdca_jack_preset(struct rt722_sdca_priv *rt722)
 	}
 }
 
+static void rt722_sdca_reset(struct rt722_sdca_priv *rt722)
+{
+	rt722_sdca_index_update_bits(rt722, RT722_VENDOR_REG,
+		RT722_LDO1_CTL, RT722_HIDDEN_REG_SW_RESET,
+		RT722_HIDDEN_REG_SW_RESET);
+	rt722_sdca_index_update_bits(rt722, RT722_VENDOR_HDA_CTL,
+		RT722_HDA_LEGACY_RESET_CTL, 0x1, 0x1);
+	if (rt722->hw_vid == RT722_VA)
+		rt722_sdca_index_write(rt722, RT722_VENDOR_REG, RT722_LDO1_CTL, 0xb091);
+}
+
 int rt722_sdca_io_init(struct device *dev, struct sdw_slave *slave)
 {
 	struct rt722_sdca_priv *rt722 = dev_get_drvdata(dev);
@@ -1888,6 +1899,9 @@ int rt722_sdca_io_init(struct device *dev, struct sdw_slave *slave)
 	rt722->hw_vid = (val & 0x0f00) >> 8;
 	dev_dbg(&slave->dev, "%s hw_vid=0x%x\n", __func__, rt722->hw_vid);
 
+	if (!rt722->first_hw_init)
+		rt722_sdca_reset(rt722);
+
 	rt722_sdca_dmic_preset(rt722);
 	rt722_sdca_amp_preset(rt722);
 	rt722_sdca_jack_preset(rt722);
-- 
2.54.0

Re: [PATCH] ASoC: rt722: reset codec to fix abnormal sound

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 793 bytes --]

On Tue, Jul 28, 2026 at 05:01:11PM +0800, shumingf@realtek.com wrote:

> The audio output may become abnormal after a warm reboot from Windows.
> Reset the codec once during hardware initialization to restore it to a known
> state and prevent the issue.

> @@ -1888,6 +1899,9 @@ int rt722_sdca_io_init(struct device *dev, struct sdw_slave *slave)
>  	rt722->hw_vid = (val & 0x0f00) >> 8;
>  	dev_dbg(&slave->dev, "%s hw_vid=0x%x\n", __func__, rt722->hw_vid);
>  
> +	if (!rt722->first_hw_init)
> +		rt722_sdca_reset(rt722);
> +
>  	rt722_sdca_dmic_preset(rt722);
>  	rt722_sdca_amp_preset(rt722);
>  	rt722_sdca_jack_preset(rt722);

Slirhgly separate issue but you probably need a _jack_init() after the
reset - your other devices do that, I think there's some races there.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

RE: [PATCH] ASoC: rt722: reset codec to fix abnormal sound

[Shuming [范書銘]]

> > The audio output may become abnormal after a warm reboot from
> Windows.
> > Reset the codec once during hardware initialization to restore it to a
> > known state and prevent the issue.
> 
> > @@ -1888,6 +1899,9 @@ int rt722_sdca_io_init(struct device *dev, struct
> sdw_slave *slave)
> >  	rt722->hw_vid = (val & 0x0f00) >> 8;
> >  	dev_dbg(&slave->dev, "%s hw_vid=0x%x\n", __func__, rt722->hw_vid);
> >
> > +	if (!rt722->first_hw_init)
> > +		rt722_sdca_reset(rt722);
> > +
> >  	rt722_sdca_dmic_preset(rt722);
> >  	rt722_sdca_amp_preset(rt722);
> >  	rt722_sdca_jack_preset(rt722);
> 
> Slirhgly separate issue but you probably need a _jack_init() after the reset -
> your other devices do that, I think there's some races there.

Understood. I'll add _jack_init() after the reset to avoid its settings being reset by the reset.