| Message ID | 20240930075451.14196-1-macpaul.lin@mediatek.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | [v2,1/2] ASoC: dt-bindings: mt6359: Update generic node name and dmic-mode | expand |
Il 30/09/24 09:54, Macpaul Lin ha scritto: > Some fix and updates in the following items: > 1. examples: > Update generic node name to 'audio-codec' to comply with the > coming change in 'mt6359.dtsi'. This change is necessary to fix the > dtbs_check error: > pmic: 'mt6359codec' does not match any of the regexes: 'pinctrl-[0-9]+' > > 2. mediatek,dmic-mode: > After inspecting the .dts and .dtsi files using 'mt6359-codec', it was > discovered that the definitions of 'two wires' and 'one wire' are > inverted compared to the DT schema. > For example, the following boards using MT6359 PMIC: > - mt8192-asurada.dtsi > - mt8195-cherry.dtsi > These boards use the same definitions of 'dmic-mode' as other boards > using MT6358 PMIC. The meaning of '0' or '1' has been noted as comments > in the device trees. > > Upon examining the code in [1] and [2], it was confirmed that the > definitions of 'dmic-mode' are consistent between "MT6359 PMIC" and > "MT6358 PMIC". Therefore, the DT Schema should be correct as is. > > References: > [1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/sound/soc/codecs/mt6358.c#n1875 > [2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/sound/soc/codecs/mt6359.c#L1515 > > Fixes: 539237d1c609 ("dt-bindings: mediatek: mt6359: add codec document") > Signed-off-by: Jiaxin Yu <jiaxin.yu@mediatek.com> > Signed-off-by: Macpaul Lin <macpaul.lin@mediatek.com> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
diff --git a/Documentation/devicetree/bindings/sound/mt6359.yaml b/Documentation/devicetree/bindings/sound/mt6359.yaml index 23d411fc4200..128698630c86 100644 --- a/Documentation/devicetree/bindings/sound/mt6359.yaml +++ b/Documentation/devicetree/bindings/sound/mt6359.yaml @@ -23,8 +23,8 @@ properties: Indicates how many data pins are used to transmit two channels of PDM signal. 0 means two wires, 1 means one wire. Default value is 0. enum: - - 0 # one wire - - 1 # two wires + - 0 # two wires + - 1 # one wire mediatek,mic-type-0: $ref: /schemas/types.yaml#/definitions/uint32 @@ -53,9 +53,9 @@ additionalProperties: false examples: - | - mt6359codec: mt6359codec { - mediatek,dmic-mode = <0>; - mediatek,mic-type-0 = <2>; + mt6359codec: audio-codec { + mediatek,dmic-mode = <0>; + mediatek,mic-type-0 = <2>; }; ...
Some fix and updates in the following items: 1. examples: Update generic node name to 'audio-codec' to comply with the coming change in 'mt6359.dtsi'. This change is necessary to fix the dtbs_check error: pmic: 'mt6359codec' does not match any of the regexes: 'pinctrl-[0-9]+' 2. mediatek,dmic-mode: After inspecting the .dts and .dtsi files using 'mt6359-codec', it was discovered that the definitions of 'two wires' and 'one wire' are inverted compared to the DT schema. For example, the following boards using MT6359 PMIC: - mt8192-asurada.dtsi - mt8195-cherry.dtsi These boards use the same definitions of 'dmic-mode' as other boards using MT6358 PMIC. The meaning of '0' or '1' has been noted as comments in the device trees. Upon examining the code in [1] and [2], it was confirmed that the definitions of 'dmic-mode' are consistent between "MT6359 PMIC" and "MT6358 PMIC". Therefore, the DT Schema should be correct as is. References: [1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/sound/soc/codecs/mt6358.c#n1875 [2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/sound/soc/codecs/mt6359.c#L1515 Fixes: 539237d1c609 ("dt-bindings: mediatek: mt6359: add codec document") Signed-off-by: Jiaxin Yu <jiaxin.yu@mediatek.com> Signed-off-by: Macpaul Lin <macpaul.lin@mediatek.com> --- Documentation/devicetree/bindings/sound/mt6359.yaml | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) Changes for v2: - No change. Because the 2/3 patch has been dropped from the v1 patch. Hence the new patch set v2 needs to be updated.