On 16/11/2022 11:40, Krzysztof Kozlowski wrote:
On 16/11/2022 11:20, Konrad Dybcio wrote:
On 16/11/2022 11:13, Krzysztof Kozlowski wrote:
From: Srinivas Kandagatla <srinivas.kandagatla@xxxxxxxxxx>
Add Soundwire controllers, Low Power Audio SubSystem (LPASS) devices and
LPASS pin controller.
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@xxxxxxxxxx>
Co-developed-by: Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx>
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx>
---
Changes since v2:
1. Use lower-case hex.
Changes since v1:
1. Whitespace cleanups.
2. Correct include - do not use deprecated one.
---
arch/arm64/boot/dts/qcom/sm8450.dtsi | 295 +++++++++++++++++++++++++++
1 file changed, 295 insertions(+)
diff --git a/arch/arm64/boot/dts/qcom/sm8450.dtsi b/arch/arm64/boot/dts/qcom/sm8450.dtsi
index 4b0a1eee8bd9..747440d0445a 100644
--- a/arch/arm64/boot/dts/qcom/sm8450.dtsi
+++ b/arch/arm64/boot/dts/qcom/sm8450.dtsi
@@ -15,6 +15,7 @@
#include <dt-bindings/interconnect/qcom,sm8450.h>
#include <dt-bindings/soc/qcom,gpr.h>
#include <dt-bindings/soc/qcom,rpmh-rsc.h>
+#include <dt-bindings/sound/qcom,q6dsp-lpass-ports.h>
#include <dt-bindings/thermal/thermal.h>
/ {
@@ -2097,6 +2098,212 @@ compute-cb@3 {
};
};
+ wsa2macro: codec@31e0000 {
+ compatible = "qcom,sm8450-lpass-wsa-macro";
+ reg = <0 0x031e0000 0 0x1000>;
+ clocks = <&q6prmcc LPASS_CLK_ID_WSA_CORE_TX_MCLK LPASS_CLK_ATTRIBUTE_COUPLE_NO>,
+ <&q6prmcc LPASS_CLK_ID_RX_CORE_MCLK2_2X_MCLK LPASS_CLK_ATTRIBUTE_COUPLE_NO>,
+ <&q6prmcc LPASS_HW_MACRO_VOTE LPASS_CLK_ATTRIBUTE_COUPLE_NO>,
+ <&q6prmcc LPASS_HW_DCODEC_VOTE LPASS_CLK_ATTRIBUTE_COUPLE_NO>,
+ <&vamacro>;
+ clock-names = "mclk", "npl", "macro", "dcodec", "fsgen";
+ assigned-clocks = <&q6prmcc LPASS_CLK_ID_WSA_CORE_TX_MCLK LPASS_CLK_ATTRIBUTE_COUPLE_NO>,
+ <&q6prmcc LPASS_CLK_ID_RX_CORE_MCLK2_2X_MCLK LPASS_CLK_ATTRIBUTE_COUPLE_NO>;
+ assigned-clock-rates = <19200000>, <19200000>;
+
+ #clock-cells = <0>;
+ clock-output-names = "wsa2-mclk";
+ #sound-dai-cells = <1>;
I think I'm being a bit too picky, but #-cells could go as the last
bunch of properties.
I was thinking about this as well, but some of other codecs which are
very similar (also "macro") do not have pinctrls and this makes them
unified with additions at the end.
Are you sure you still prefer alphabetical order?
Thinking about it, IMO it makes sense to add things "roughly where they
belong", a.k.a if there was a codec that for whatever reason also
required a power domain, we would stick it somewhere in the middle and
not at the end..
+
+ pinctrl-names = "default";
+ pinctrl-0 = <&wsa2_swr_active>;
+ };
+
+ /* WSA2 */
+ swr4: soundwire-controller@31f0000 {
+ reg = <0 0x031f0000 0 0x2000>;
+ compatible = "qcom,soundwire-v1.7.0";
+ interrupts = <GIC_SPI 171 IRQ_TYPE_LEVEL_HIGH>;
+ clocks = <&wsa2macro>;
+ clock-names = "iface";
+
+ qcom,din-ports = <2>;
+ qcom,dout-ports = <6>;
+
+ qcom,ports-sinterval-low = /bits/ 8 <0x07 0x1f 0x3f 0x07 0x1f 0x3f 0x0f 0x0f>;
+ qcom,ports-offset1 = /bits/ 8 <0x01 0x02 0x0c 0x06 0x12 0x0d 0x07 0x0a>;
+ qcom,ports-offset2 = /bits/ 8 <0xff 0x00 0x1f 0xff 0x00 0x1f 0x00 0x00>;
+ qcom,ports-hstart = /bits/ 8 <0xff 0xff 0xff 0xff 0xff 0xff 0xff 0xff>;
+ qcom,ports-hstop = /bits/ 8 <0xff 0xff 0xff 0xff 0xff 0xff 0xff 0xff>;
+ qcom,ports-word-length = /bits/ 8 <0xff 0xff 0xff 0xff 0xff 0xff 0xff 0xff>;
+ qcom,ports-block-pack-mode = /bits/ 8 <0xff 0xff 0x01 0xff 0xff 0x01 0xff 0xff>;
+ qcom,ports-block-group-count = /bits/ 8 <0xff 0xff 0xff 0xff 0xff 0xff 0xff 0xff>;
+ qcom,ports-lane-control = /bits/ 8 <0xff 0xff 0xff 0xff 0xff 0xff 0xff 0xff>;
+
+ #sound-dai-cells = <1>;
+ #address-cells = <2>;
+ #size-cells = <0>;
+ };
+
+ rxmacro: codec@3200000 {
+ compatible = "qcom,sm8450-lpass-rx-macro";
+ reg = <0 0x3200000 0 0x1000>;
+ clocks = <&q6prmcc LPASS_CLK_ID_RX_CORE_TX_MCLK LPASS_CLK_ATTRIBUTE_COUPLE_NO>,
+ <&q6prmcc LPASS_CLK_ID_RX_CORE_MCLK2_2X_MCLK LPASS_CLK_ATTRIBUTE_COUPLE_NO>,
+ <&q6prmcc LPASS_HW_MACRO_VOTE LPASS_CLK_ATTRIBUTE_COUPLE_NO>,
+ <&q6prmcc LPASS_HW_DCODEC_VOTE LPASS_CLK_ATTRIBUTE_COUPLE_NO>,
+ <&vamacro>;
+ clock-names = "mclk", "npl", "macro", "dcodec", "fsgen";
+
+ assigned-clocks = <&q6prmcc LPASS_CLK_ID_RX_CORE_TX_MCLK LPASS_CLK_ATTRIBUTE_COUPLE_NO>,
+ <&q6prmcc LPASS_CLK_ID_RX_CORE_MCLK2_2X_MCLK LPASS_CLK_ATTRIBUTE_COUPLE_NO>;
+ assigned-clock-rates = <19200000>, <19200000>;
+
+ #clock-cells = <0>;
+ clock-output-names = "mclk";
+ #sound-dai-cells = <1>;
+
+ pinctrl-names = "default";
+ pinctrl-0 = <&rx_swr_active>;
+ };
+
+ swr1: soundwire-controller@3210000 {
+ reg = <0 0x3210000 0 0x2000>;
+ compatible = "qcom,soundwire-v1.7.0";
Some nodes have reg and compatible flipped.
Ack
+ interrupts = <GIC_SPI 155 IRQ_TYPE_LEVEL_HIGH>;
+ clocks = <&rxmacro>;
+ clock-names = "iface";
+ label = "RX";
+ qcom,din-ports = <0>;
+ qcom,dout-ports = <5>;
+
(...)
apps_smmu: iommu@15000000 {
compatible = "qcom,sm8450-smmu-500", "arm,mmu-500";
reg = <0 0x15000000 0 0x100000>;
@@ -3507,6 +3799,9 @@ lpass_ag_noc: interconnect@3c40000 {
};
};
+ sound: sound {
+ };
You asked another folk working on sa8540p to not include sound in the
SoC dtsi.
I asked not to put it in soc node. It can be in DTSI, but not under soc.
Oh okay, then I misread that, sorry.
Konrad
Best regards,
Krzysztof
