[PATCH v2 0/2] ASoC: cs42xx8: Add SPI bus support for CS42448/CS42888 codec

[PATCH v2 0/2] ASoC: cs42xx8: Add SPI bus support for CS42448/CS42888 codec

[chancel.liu]

From: Chancel Liu <chancel.liu@nxp.com>

The existing cs42xx8 driver only supported I2C control interface.
Add SPI bus support for the Cirrus Logic CS42448/CS42888 Audio CODEC.

Changes in v2:
- Add SPI bus support in codec binding
- Move format endian into common regmap config
- Enable use_single_read/write for common regmap config

Chancel Liu (2):
  ASoC: dt-bindings: cirrus,cs42xx8: Add SPI bus support
  ASoC: cs42xx8: Add SPI bus support for CS42448/CS42888 codec

 .../bindings/sound/cirrus,cs42xx8.yaml        |  23 ++++
 sound/soc/codecs/Kconfig                      |   7 ++
 sound/soc/codecs/Makefile                     |   2 +
 sound/soc/codecs/cs42xx8-spi.c                | 104 ++++++++++++++++++
 sound/soc/codecs/cs42xx8.c                    |   3 +
 5 files changed, 139 insertions(+)
 create mode 100644 sound/soc/codecs/cs42xx8-spi.c

--
2.50.1

[PATCH v2 1/2] ASoC: dt-bindings: cirrus,cs42xx8: Add SPI bus support

[chancel.liu]

From: Chancel Liu <chancel.liu@nxp.com>

Codec CS42448/CS42888 supports multiple control interfaces. At present,
only the I2C interface is implemented. Adding support for the SPI
control interface, operating at up to 6MHz.

Signed-off-by: Chancel Liu <chancel.liu@nxp.com>
---
 .../bindings/sound/cirrus,cs42xx8.yaml        | 23 +++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/Documentation/devicetree/bindings/sound/cirrus,cs42xx8.yaml b/Documentation/devicetree/bindings/sound/cirrus,cs42xx8.yaml
index 7ae72bd901f4..a1ae548c4b7b 100644
--- a/Documentation/devicetree/bindings/sound/cirrus,cs42xx8.yaml
+++ b/Documentation/devicetree/bindings/sound/cirrus,cs42xx8.yaml
@@ -11,6 +11,7 @@ maintainers:
 
 allOf:
   - $ref: dai-common.yaml#
+  - $ref: /schemas/spi/spi-peripheral-props.yaml#
 
 properties:
   compatible:
@@ -21,6 +22,9 @@ properties:
   reg:
     maxItems: 1
 
+  spi-max-frequency:
+    maximum: 6000000
+
   clocks:
     minItems: 1
     maxItems: 2
@@ -86,3 +90,22 @@ examples:
           reset-gpios = <&gpio 1>;
       };
     };
+
+    spi {
+      #address-cells = <1>;
+      #size-cells = <0>;
+      cs-gpios = <&gpio 8 0>;
+
+      codec@0 {
+          compatible = "cirrus,cs42888";
+          reg = <0>;
+          spi-max-frequency = <6000000>;
+          clocks = <&codec_mclk 0>;
+          clock-names = "mclk";
+          VA-supply = <&reg_audio>;
+          VD-supply = <&reg_audio>;
+          VLS-supply = <&reg_audio>;
+          VLC-supply = <&reg_audio>;
+          reset-gpios = <&gpio 1>;
+      };
+    };
-- 
2.50.1

Re: [PATCH v2 1/2] ASoC: dt-bindings: cirrus,cs42xx8: Add SPI bus support

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] Conflicting GPIO cell counts for the '&gpio' phandle in the newly added SPI example.
--

commit f4759f9ac7c7f10e085045f8fabd1df333354abf
Author: Chancel Liu <chancel.liu@nxp.com>

ASoC: dt-bindings: cirrus,cs42xx8: Add SPI bus support

This commit adds SPI control interface support to the CS42448/CS42888 codec
bindings, which can operate at up to 6MHz. It updates the schema with SPI
properties and provides a new SPI example configuration.

> diff --git a/Documentation/devicetree/bindings/sound/cirrus,cs42xx8.yaml b/Documentation/devicetree/bindings/sound/cirrus,cs42xx8.yaml
> --- a/Documentation/devicetree/bindings/sound/cirrus,cs42xx8.yaml
> +++ b/Documentation/devicetree/bindings/sound/cirrus,cs42xx8.yaml

[ ... ]

> @@ -86,3 +90,22 @@ examples:
>            reset-gpios = <&gpio 1>;
>        };
>      };
> +
> +    spi {
> +      #address-cells = <1>;
> +      #size-cells = <0>;
> +      cs-gpios = <&gpio 8 0>;
> +
> +      codec@0 {
> +          compatible = "cirrus,cs42888";
> +          reg = <0>;
> +          spi-max-frequency = <6000000>;
> +          clocks = <&codec_mclk 0>;
> +          clock-names = "mclk";
> +          VA-supply = <&reg_audio>;
> +          VD-supply = <&reg_audio>;
> +          VLS-supply = <&reg_audio>;
> +          VLC-supply = <&reg_audio>;
> +          reset-gpios = <&gpio 1>;

[Severity: Low]
Will this cause a dt_binding_check failure due to a cell size mismatch?

The cs-gpios property uses 2 argument cells for the &gpio phandle, but
reset-gpios uses 1 argument cell for the exact same &gpio label.

A single device tree node can only have a single #gpio-cells value. When
dt_binding_check auto-generates a dummy node for the unresolved &gpio
reference, it will infer a #gpio-cells value that conflicts with one of
these references. 

Should a different label (e.g., &gpio2) be used for cs-gpios, or should
reset-gpios be updated to use 2 cells?

> +      };
> +    };

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260603095041.3906558-1-chancel.liu@oss.nxp.com?part=1

[PATCH v2 2/2] ASoC: cs42xx8: Add SPI bus support for CS42448/CS42888 codec

[chancel.liu]

From: Chancel Liu <chancel.liu@nxp.com>

The existing cs42xx8 driver only supported I2C control interface.
Add SPI bus support for the Cirrus Logic CS42448/CS42888 Audio CODEC.

Signed-off-by: Chancel Liu <chancel.liu@nxp.com>
---
 sound/soc/codecs/Kconfig       |   7 +++
 sound/soc/codecs/Makefile      |   2 +
 sound/soc/codecs/cs42xx8-spi.c | 104 +++++++++++++++++++++++++++++++++
 sound/soc/codecs/cs42xx8.c     |   3 +
 4 files changed, 116 insertions(+)
 create mode 100644 sound/soc/codecs/cs42xx8-spi.c

diff --git a/sound/soc/codecs/Kconfig b/sound/soc/codecs/Kconfig
index a7c61f7c7f4c..ae36f663a5ef 100644
--- a/sound/soc/codecs/Kconfig
+++ b/sound/soc/codecs/Kconfig
@@ -93,6 +93,7 @@ config SND_SOC_ALL_CODECS
 	imply SND_SOC_CS4271_I2C
 	imply SND_SOC_CS4271_SPI
 	imply SND_SOC_CS42XX8_I2C
+	imply SND_SOC_CS42XX8_SPI
 	imply SND_SOC_CS43130
 	imply SND_SOC_CS4341
 	imply SND_SOC_CS4349
@@ -1077,6 +1078,12 @@ config SND_SOC_CS4271_SPI
 config SND_SOC_CS42XX8
 	tristate
 
+config SND_SOC_CS42XX8_SPI
+	tristate "Cirrus Logic CS42448/CS42888 CODEC (SPI)"
+	depends on SPI_MASTER
+	select SND_SOC_CS42XX8
+	select REGMAP_SPI
+
 config SND_SOC_CS42XX8_I2C
 	tristate "Cirrus Logic CS42448/CS42888 CODEC (I2C)"
 	depends on I2C
diff --git a/sound/soc/codecs/Makefile b/sound/soc/codecs/Makefile
index 73315d017c57..aa0396e5b575 100644
--- a/sound/soc/codecs/Makefile
+++ b/sound/soc/codecs/Makefile
@@ -103,6 +103,7 @@ snd-soc-cs4271-i2c-y := cs4271-i2c.o
 snd-soc-cs4271-spi-y := cs4271-spi.o
 snd-soc-cs42xx8-y := cs42xx8.o
 snd-soc-cs42xx8-i2c-y := cs42xx8-i2c.o
+snd-soc-cs42xx8-spi-y := cs42xx8-spi.o
 snd-soc-cs43130-y := cs43130.o
 snd-soc-cs4341-y := cs4341.o
 snd-soc-cs4349-y := cs4349.o
@@ -543,6 +544,7 @@ obj-$(CONFIG_SND_SOC_CS4271_I2C)	+= snd-soc-cs4271-i2c.o
 obj-$(CONFIG_SND_SOC_CS4271_SPI)	+= snd-soc-cs4271-spi.o
 obj-$(CONFIG_SND_SOC_CS42XX8)	+= snd-soc-cs42xx8.o
 obj-$(CONFIG_SND_SOC_CS42XX8_I2C) += snd-soc-cs42xx8-i2c.o
+obj-$(CONFIG_SND_SOC_CS42XX8_SPI) += snd-soc-cs42xx8-spi.o
 obj-$(CONFIG_SND_SOC_CS43130)   += snd-soc-cs43130.o
 obj-$(CONFIG_SND_SOC_CS4341)	+= snd-soc-cs4341.o
 obj-$(CONFIG_SND_SOC_CS4349)	+= snd-soc-cs4349.o
diff --git a/sound/soc/codecs/cs42xx8-spi.c b/sound/soc/codecs/cs42xx8-spi.c
new file mode 100644
index 000000000000..b86fe2fe771e
--- /dev/null
+++ b/sound/soc/codecs/cs42xx8-spi.c
@@ -0,0 +1,104 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Cirrus Logic CS42448/CS42888 Audio CODEC DAI SPI driver
+ *
+ * Copyright 2026 NXP
+ *
+ */
+
+#include <linux/mod_devicetable.h>
+#include <linux/module.h>
+#include <linux/pm_runtime.h>
+#include <linux/regmap.h>
+#include <linux/spi/spi.h>
+#include <sound/soc.h>
+
+#include "cs42xx8.h"
+
+/*
+ * CS42448/CS42888 SPI register access (from datasheet Figure 23):
+ *
+ * The SPI frame is 3 bytes:
+ *   Byte 0: chip address [7:1] = 1001111, bit[0] = R/W (0=write, 1=read)
+ *           Write: 0x9E,  Read: 0x9F
+ *   Byte 1: MAP - Memory Address Pointer
+ *           bit[7] = INCR (auto-increment for burst), bits[6:0] = address
+ *   Byte 2: data byte
+ *
+ * We configure reg_bits=16 so that regmap treats the address field as 2 bytes
+ * (big-endian). The chip address byte (0x9E/0x9F) is placed in the high byte
+ * via write_flag_mask / read_flag_mask, and the MAP register address occupies
+ * the low byte. Currently INCR (MAP bit[7]) is not set and use_single_read/write
+ * are enabled. This produces the correct 3-byte on-wire frame without any
+ * custom bus implementation:
+ *
+ *   write: [0x9E, MAP_addr, data]
+ *   read:  [0x9F, MAP_addr] -> [data]
+ */
+
+static int cs42xx8_spi_probe(struct spi_device *spi)
+{
+	struct cs42xx8_driver_data *drvdata;
+	struct regmap_config config;
+	int ret;
+
+	drvdata = (struct cs42xx8_driver_data *)spi_get_device_match_data(spi);
+	if (!drvdata)
+		return dev_err_probe(&spi->dev, -EINVAL,
+				     "failed to find driver data\n");
+
+	config = cs42xx8_regmap_config;
+	/*
+	 * reg_bits=16 makes regmap send a 2-byte address field (big-endian).
+	 * write_flag_mask/read_flag_mask are OR'd into that address field:
+	 */
+	config.reg_bits           = 16;
+	config.write_flag_mask    = 0x9E;
+	config.read_flag_mask     = 0x9F;
+
+	ret = cs42xx8_probe(&spi->dev,
+			    devm_regmap_init_spi(spi, &config), drvdata);
+	if (ret)
+		return ret;
+
+	pm_runtime_enable(&spi->dev);
+	pm_request_idle(&spi->dev);
+
+	return 0;
+}
+
+static void cs42xx8_spi_remove(struct spi_device *spi)
+{
+	pm_runtime_disable(&spi->dev);
+}
+
+static const struct of_device_id cs42xx8_of_match[] = {
+	{ .compatible = "cirrus,cs42448", .data = &cs42448_data, },
+	{ .compatible = "cirrus,cs42888", .data = &cs42888_data, },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, cs42xx8_of_match);
+
+static const struct spi_device_id cs42xx8_spi_id[] = {
+	{ .name = "cs42448", .driver_data = (kernel_ulong_t)&cs42448_data },
+	{ .name = "cs42888", .driver_data = (kernel_ulong_t)&cs42888_data },
+	{ }
+};
+MODULE_DEVICE_TABLE(spi, cs42xx8_spi_id);
+
+static struct spi_driver cs42xx8_spi_driver = {
+	.driver = {
+		.name = "cs42xx8",
+		.pm = pm_ptr(&cs42xx8_pm),
+		.of_match_table = cs42xx8_of_match,
+	},
+	.probe = cs42xx8_spi_probe,
+	.remove = cs42xx8_spi_remove,
+	.id_table = cs42xx8_spi_id,
+};
+
+module_spi_driver(cs42xx8_spi_driver);
+
+MODULE_DESCRIPTION("Cirrus Logic CS42448/CS42888 ALSA SoC Codec SPI Driver");
+MODULE_AUTHOR("Chancel Liu <chancel.liu@nxp.com>");
+MODULE_LICENSE("GPL");
diff --git a/sound/soc/codecs/cs42xx8.c b/sound/soc/codecs/cs42xx8.c
index 12fe9b3e2525..5b689549c74e 100644
--- a/sound/soc/codecs/cs42xx8.c
+++ b/sound/soc/codecs/cs42xx8.c
@@ -478,6 +478,9 @@ const struct regmap_config cs42xx8_regmap_config = {
 	.volatile_reg = cs42xx8_volatile_register,
 	.writeable_reg = cs42xx8_writeable_register,
 	.cache_type = REGCACHE_MAPLE,
+	.reg_format_endian  = REGMAP_ENDIAN_BIG,
+	.use_single_read = true,
+	.use_single_write = true,
 };
 EXPORT_SYMBOL_GPL(cs42xx8_regmap_config);
 
-- 
2.50.1

Re: [PATCH v2 2/2] ASoC: cs42xx8: Add SPI bus support for CS42448/CS42888 codec

[sashiko-bot]

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

New issues:
- [Medium] Modifying the shared cs42xx8_regmap_config structure disables bulk I/O operations for the existing I2C driver, causing performance degradation.

Pre-existing issues:
- [High] Modifying the globally shared static cs42xx8_dai structure during probe introduces shared state corruption across multiple codecs.
--

commit e668063709eddf991f98c990850087567f6a8d52
Author: Chancel Liu <chancel.liu@nxp.com>

ASoC: cs42xx8: Add SPI bus support for CS42448/CS42888 codec
    
The existing cs42xx8 driver only supported I2C control interface.
Add SPI bus support for the Cirrus Logic CS42448/CS42888 Audio CODEC.

[ ... ]
> diff --git a/sound/soc/codecs/cs42xx8-spi.c b/sound/soc/codecs/cs42xx8-spi.c
> new file mode 100644
> index 0000000000000..b86fe2fe771ee
> --- /dev/null
> +++ b/sound/soc/codecs/cs42xx8-spi.c
> @@ -0,0 +1,104 @@
[ ... ]
> +static int cs42xx8_spi_probe(struct spi_device *spi)
> +{
> +	struct cs42xx8_driver_data *drvdata;
> +	struct regmap_config config;
> +	int ret;
> +
> +	drvdata = (struct cs42xx8_driver_data *)spi_get_device_match_data(spi);
> +	if (!drvdata)
> +		return dev_err_probe(&spi->dev, -EINVAL,
> +				     "failed to find driver data\n");
> +
> +	config = cs42xx8_regmap_config;
> +	/*
> +	 * reg_bits=16 makes regmap send a 2-byte address field (big-endian).
> +	 * write_flag_mask/read_flag_mask are OR'd into that address field:
> +	 */
> +	config.reg_bits           = 16;

[Severity: Medium]
Since cs42xx8_spi_probe() already makes a local copy of the config here, could
the single read/write flags be overridden locally instead of modifying the
shared cs42xx8_regmap_config structure?

> +	config.write_flag_mask    = 0x9E;
> +	config.read_flag_mask     = 0x9F;
[ ... ]
> diff --git a/sound/soc/codecs/cs42xx8.c b/sound/soc/codecs/cs42xx8.c
> index 12fe9b3e2525e..5b689549c74ed 100644
> --- a/sound/soc/codecs/cs42xx8.c
> +++ b/sound/soc/codecs/cs42xx8.c
> @@ -478,6 +478,9 @@ const struct regmap_config cs42xx8_regmap_config = {
>  	.volatile_reg = cs42xx8_volatile_register,
>  	.writeable_reg = cs42xx8_writeable_register,
>  	.cache_type = REGCACHE_MAPLE,
> +	.reg_format_endian  = REGMAP_ENDIAN_BIG,
> +	.use_single_read = true,
> +	.use_single_write = true,

[Severity: Medium]
Does adding these restrictions directly to the shared config disable bulk I/O
operations for the existing I2C driver?

It appears the I2C driver uses this struct directly, so it is now globally
restricted to single-register reads and writes. This might force
regcache_sync() to perform a full I2C transaction for every single register
during device resume.

[Severity: High]
This isn't a bug introduced by this patch, but does modifying the globally
shared static cs42xx8_dai structure during probe introduce shared state
corruption across multiple codecs?

Looking at cs42xx8_probe():

    cs42xx8_dai.name = cs42xx8->drvdata->name;

    /* Each adc supports stereo input */
    cs42xx8_dai.capture.channels_max = cs42xx8->drvdata->num_adcs * 2;

If multiple codecs of this family are present on the same system, the probe
function of the second codec will overwrite the name and capture.channels_max
fields of the global cs42xx8_dai structure.

Because all instances share this single static struct via the pointer passed
to devm_snd_soc_register_component(), the first codec's DAI state will be
corrupted.

>  };
>  EXPORT_SYMBOL_GPL(cs42xx8_regmap_config);

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260603095041.3906558-1-chancel.liu@oss.nxp.com?part=2

Re: [PATCH v2 0/2] ASoC: cs42xx8: Add SPI bus support for CS42448/CS42888 codec

[Charles Keepax]

On Wed, Jun 03, 2026 at 06:50:39PM +0900, chancel.liu@oss.nxp.com wrote:
> From: Chancel Liu <chancel.liu@nxp.com>
> 
> The existing cs42xx8 driver only supported I2C control interface.
> Add SPI bus support for the Cirrus Logic CS42448/CS42888 Audio CODEC.
> 
> Changes in v2:
> - Add SPI bus support in codec binding
> - Move format endian into common regmap config
> - Enable use_single_read/write for common regmap config
> 
> Chancel Liu (2):
>   ASoC: dt-bindings: cirrus,cs42xx8: Add SPI bus support
>   ASoC: cs42xx8: Add SPI bus support for CS42448/CS42888 codec
> 

Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>

Thanks,
Charles

Re: [PATCH v2 0/2] ASoC: cs42xx8: Add SPI bus support for CS42448/CS42888 codec

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 556 bytes --]

On Wed, Jun 03, 2026 at 06:50:39PM +0900, chancel.liu@oss.nxp.com wrote:
> From: Chancel Liu <chancel.liu@nxp.com>
> 
> The existing cs42xx8 driver only supported I2C control interface.
> Add SPI bus support for the Cirrus Logic CS42448/CS42888 Audio CODEC.

Please don't send new patches in reply to old patches or serieses, this
makes it harder for both people and tools to understand what is going
on - it can bury things in mailboxes and make it difficult to keep track
of what current patches are, both for the new patches and the old ones.

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

RE: [PATCH v2 0/2] ASoC: cs42xx8: Add SPI bus support for CS42448/CS42888 codec

[Chancel Liu (OSS)]

> > The existing cs42xx8 driver only supported I2C control interface.
> > Add SPI bus support for the Cirrus Logic CS42448/CS42888 Audio CODEC.
> 
> Please don't send new patches in reply to old patches or serieses, this
> makes it harder for both people and tools to understand what is going
> on - it can bury things in mailboxes and make it difficult to keep track
> of what current patches are, both for the new patches and the old ones.

Thanks for the reminder. Sorry about that. Understood, I won't do that
again.

Regards, 
Chancel Liu

Re: [PATCH v2 0/2] ASoC: cs42xx8: Add SPI bus support for CS42448/CS42888 codec

[Mark Brown]

On Wed, 03 Jun 2026 18:50:39 +0900, chancel.liu@oss.nxp.com wrote:
> ASoC: cs42xx8: Add SPI bus support for CS42448/CS42888 codec
> 
> From: Chancel Liu <chancel.liu@nxp.com>
> 
> The existing cs42xx8 driver only supported I2C control interface.
> Add SPI bus support for the Cirrus Logic CS42448/CS42888 Audio CODEC.
> 
> [...]

Applied to

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-7.2

Thanks!

[1/2] ASoC: dt-bindings: cirrus,cs42xx8: Add SPI bus support
      https://git.kernel.org/broonie/sound/c/596f8d649444
[2/2] ASoC: cs42xx8: Add SPI bus support for CS42448/CS42888 codec
      https://git.kernel.org/broonie/sound/c/3158f585f4f2

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark