Skip to contentSkip to conversation

ARM: tegra: lg-x3: Add newly supported devices and fixes

PATCH Superseded Svyatoslav Ryhel 2 participants 12 messages Sep 24

Subsystems (2)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

cover ARM: tegra: lg-x3: Add newly supported devices and fixes

Add nodes for recently supported device (ALS, LED controllers, haptics) along with minor tweaks and fixups.

Svyatoslav Ryhel (7):
ARM: tegra: lg-x3: Adjust WiFi node
ARM: tegra: lg-x3: Remove pinctrl-* properties duplication
ARM: tegra: lg-x3: Lower supported suspend mode
ARM: tegra: lg-x3: Add light sensor node
ARM: tegra: lg-x3: Add flash LEDs controller node
ARM: tegra: lg-x3: Add haptic engine node
ARM: tegra: lg-x3: Add backlight LED controller node

 arch/arm/boot/dts/nvidia/tegra30-lg-p880.dts |   5 +-
 arch/arm/boot/dts/nvidia/tegra30-lg-p895.dts |   5 +-
 arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi  | 106 ++++++++++++++++++-
 3 files changed, 108 insertions(+), 8 deletions(-)
Signature · 1 line
2.53.0
#1 lore raw

PATCH 1/7 ARM: tegra: lg-x3: Adjust WiFi node

Add missing BCM4330 compatible along with keeping fallback according to current schema. Add missing lpo clock used by the module.

---
 arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +12 -1
index 492fefd7c266e..5b4da6ee991ff 100644
@@ -1754,9 +1754,12 @@ sdmmc1: mmc@78000000 {
 
 		/* BCM4330B1 37.4 MHz Class 1.5 ExtLNA */
 		wifi@1 {
-			compatible = "brcm,bcm4329-fmac";
+			compatible = "brcm,bcm4330-fmac", "brcm,bcm4329-fmac";
 			reg = <1>;
 
+			clocks = <&ext_lpo_in>;
+			clock-names = "lpo";
+
 			interrupt-parent = <&gpio>;
 			interrupts = <TEGRA_GPIO(U, 6) IRQ_TYPE_LEVEL_HIGH>;
 			interrupt-names = "host-wake";
@@ -1799,6 +1802,14 @@ clk32k_in: clock-32k {
 		clock-output-names = "pmic-oscillator";
 	};
 
+	/* PMIC's 32KHz oscillator for WiFi/GPS LPO input */
+	ext_lpo_in: clock-32k-elpo {
+		compatible = "fixed-clock";
+		clock-frequency = <32768>;
+		clock-output-names = "ext-lpo-32k-in";
+		#clock-cells = <0>;
+	};
+
 	gps_refclk: clock-gps {
 		compatible = "fixed-clock";
 		clock-frequency = <26000000>;
Signature · 1 line
2.53.0
#2 lore raw

PATCH 2/7 ARM: tegra: lg-x3: Remove pinctrl-* properties duplication

In reply to Svyatoslav Ryhel

The pinctrl-names and pinctrl-0 are defined in the common device tree, having them in the individual trees is redundant. Remove them.

---
 arch/arm/boot/dts/nvidia/tegra30-lg-p880.dts | 3 ---
 arch/arm/boot/dts/nvidia/tegra30-lg-p895.dts | 3 ---
 2 files changed, 6 deletions(-)
arch/arm/boot/dts/nvidia/tegra30-lg-p880.dts +0 -3
index d4a3513a3fd42..d76f9f465ddc6 100644
@@ -13,9 +13,6 @@ aliases {
 	};
 
 	pinmux@70000868 {
-		pinctrl-names = "default";
-		pinctrl-0 = <&state_default>;
-
 		state_default: pinmux {
 			/* WLAN SDIO pinmux */
 			host-wlan-wake {
arch/arm/boot/dts/nvidia/tegra30-lg-p895.dts +0 -3
index 7a62707f9f7fa..4ee84f4bd1878 100644
@@ -8,9 +8,6 @@ / {
 	compatible = "lg,p895", "nvidia,tegra30";
 
 	pinmux@70000868 {
-		pinctrl-names = "default";
-		pinctrl-0 = <&state_default>;
-
 		state_default: pinmux {
 			/* GNSS UART-B pinmux */
 			uartb-cts-rxd {
Signature · 1 line
2.53.0
#3 lore raw

PATCH 3/7 ARM: tegra: lg-x3: Lower supported suspend mode

In reply to Svyatoslav Ryhel

LG P880 and P895 are perfectly capable of supporting the lowest suspend mode currently available for Tegra devices which is LP1.

---
 arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +1 -1
index 5b4da6ee991ff..5f3273a4dcf46 100644
@@ -1695,7 +1695,7 @@ bridge_output: endpoint {
 	pmc@7000e400 {
 		status = "okay";
 		nvidia,invert-interrupt;
-		nvidia,suspend-mode = <2>;
+		nvidia,suspend-mode = <1>;
 		nvidia,cpu-pwr-good-time = <2000>;
 		nvidia,cpu-pwr-off-time = <200>;
 		nvidia,core-pwr-good-time = <3845 3845>;
Signature · 1 line
2.53.0
#4 lore raw

PATCH 4/7 ARM: tegra: lg-x3: Add light sensor node

In reply to Svyatoslav Ryhel

Add node for Avago APDS9900 ambient light and proximity sensor used in this device.

---
 arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi | 13 +++++++++++++
 1 file changed, 13 insertions(+)
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +13 -0
index 5f3273a4dcf46..417cc1bf42ea3 100644
@@ -1142,6 +1142,19 @@ nfc@28 {
 			firmware-gpios = <&gpio TEGRA_GPIO(R, 0) GPIO_ACTIVE_HIGH>;
 		};
 
+		light-sensor@39 {
+			compatible = "avago,apds9900";
+			reg = <0x39>;
+
+			interrupt-parent = <&gpio>;
+			interrupts = <TEGRA_GPIO(K, 2) IRQ_TYPE_EDGE_RISING>;
+
+			vdd-supply = <&vdd_3v0_proxi>;
+			vddio-supply = <&vdd_1v8_sen>;
+
+			led-max-microamp = <100000>;
+		};
+
 		imu@68 {
 			compatible = "invensense,mpu6050";
 			reg = <0x68>;
Signature · 1 line
2.53.0
#5 lore raw

Re: [PATCH v1 4/7] ARM: tegra: lg-x3: Add light sensor node

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Medium] dt: incorrect interrupt polarity for apds9900 sensor --

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] dt: incorrect interrupt polarity for apds9900 sensor
--

commit 373b2db230a91414c156f1ebbe3d13fb896ef4ef
Author: Svyatoslav Ryhel <[email protected]>

ARM: tegra: lg-x3: Add light sensor node

This commit adds a device tree node for the Avago APDS9900 ambient light and proximity sensor to the Tegra30 LG X3 device tree.

Quoted 4 lines
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +0 -0
index 9e720512e7962..74583eb07aa84 100644

[ ... ]

arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi nfc@28 { · line 1114

Quoted 10 lines
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +6 -0
@@ -1106,6 +1106,19 @@ nfc@28 {
 			firmware-gpios = <&gpio TEGRA_GPIO(R, 0) GPIO_ACTIVE_HIGH>;
 		};
 
+		light-sensor@39 {
+			compatible = "avago,apds9900";
+			reg = <0x39>;
+
+			interrupt-parent = <&gpio>;
+			interrupts = <TEGRA_GPIO(K, 2) IRQ_TYPE_EDGE_RISING>;

[Severity: Medium]
Is the interrupt trigger type correct for this sensor?

The interrupt trigger type is defined as a rising edge, which appears to conflict with the active-low (falling edge) hardware signal and driver expectations.

The tsl2772 driver expects to use IRQF_TRIGGER_FALLING when it calls devm_request_threaded_irq() during probe.

Could this mismatch cause the sensor's interrupt request to fail during probe, or cause the interrupt to trigger on the wrong edge and miss events?

Quoted 6 lines
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +6 -0
+
+			vdd-supply = <&vdd_3v0_proxi>;
+			vddio-supply = <&vdd_1v8_sen>;
+
+			led-max-microamp = <100000>;
+		};
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=4
#6 lore raw

PATCH 5/7 ARM: tegra: lg-x3: Add flash LEDs controller node

In reply to Svyatoslav Ryhel

Add node for TI LM3559 LEDs controlled used with flash LEDs in this device.

---
 arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi | 30 +++++++++++++++++++++
 1 file changed, 30 insertions(+)
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +30 -0
index 417cc1bf42ea3..a59539f36a316 100644
@@ -1277,6 +1277,7 @@ rear-camera@10 {
 			rotation = <90>;
 
 			nvmem = <&m24c08>;
+			flash-leds = <&lm3559_led0>, <&lm3559_led1>;
 			lens-focus = <&dw9714>;
 
 			assigned-clocks = <&tegra_car TEGRA30_CLK_VI_SENSOR>,
@@ -1311,6 +1312,35 @@ m24c08: eeprom@50 {
 			vcc-supply = <&vio_1v8_rear>;
 		};
 
+		led-controller@53 {
+			compatible = "ti,lm3559";
+			reg = <0x53>;
+
+			enable-gpios = <&gpio TEGRA_GPIO(BB, 3) GPIO_ACTIVE_HIGH>;
+			vin-supply = <&vdd_3v3_vbat>;
+
+			flash-max-timeout-us = <1024000>;
+			ti,peak-current-microamp = <1400000>;
+
+			#address-cells = <1>;
+			#size-cells = <0>;
+
+			/* single flash led is on 2 lines */
+			lm3559_led0: led@0 {
+				reg = <0>;
+
+				flash-max-microamp = <393750>;
+				led-max-microamp = <140625>;
+			};
+
+			lm3559_led1: led@1 {
+				reg = <1>;
+
+				flash-max-microamp = <393750>;
+				led-max-microamp = <140625>;
+			};
+		};
+
 		camera-pmic@7d {
 			compatible = "ti,lp8720";
 			reg = <0x7d>;
Signature · 1 line
2.53.0
#7 lore raw

Re: [PATCH v1 5/7] ARM: tegra: lg-x3: Add flash LEDs controller node

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Medium] dts: hw misrepresentation of single flash LED as multiple nodes --

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] dts: hw misrepresentation of single flash LED as multiple nodes --

commit 1c030c2896d9a641e8c683129fc55fbf9766228a
Author: Svyatoslav Ryhel <[email protected]>
ARM: tegra: lg-x3: Add flash LEDs controller node

This commit adds a device tree node for the ti lm3559 controller used for flash leds on the tegra30 lg-x3 device. It configures the controller and sets up two separate child nodes to represent the flash.

Quoted 4 lines
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +0 -0
index 74583eb07aa84..14d4f992222c6 100644

[ ... ]

Quoted 1 line
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +0 -0
@@ -1267,6 +1268,35 @@ m24c08: eeprom@50 {

[ ... ]

Quoted 3 lines
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +3 -0
+		led-controller@53 {
+			compatible = "ti,lm3559";
+			reg = <0x53>;

[ ... ]

arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi m24c08: eeprom@50 { · line 1337

Quoted 10 lines
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +10 -0
+			/* single flash led is on 2 lines */
+			lm3559_led0: led@0 {
+				reg = <0>;
+
+				flash-max-microamp = <393750>;
+				led-max-microamp = <140625>;
+			};
+
+			lm3559_led1: led@1 {
+				reg = <1>;

[Severity: Medium]
Does this configuration incorrectly define a single physical led as two independent led devices?

The comment indicates the single flash led is on two lines, but splitting it into two separate nodes exposes two distinct video4linux2 subdevices. Device trees generally describe hardware topology rather than driver limitations.

Should this be modeled as a single node using a property like led-sources = <0 1>? Even if the current driver lacks support for the led-sources property, duplicating child nodes to work around driver limitations is considered a device tree anti-pattern.

Quoted 5 lines
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +5 -0
+
+				flash-max-microamp = <393750>;
+				led-max-microamp = <140625>;
+			};
+		};
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=5
#8 lore raw

PATCH 6/7 ARM: tegra: lg-x3: Add haptic engine node

In reply to Svyatoslav Ryhel

Add node for Imagis ISA1200 haptic motor driver used in this device.

---
 arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi | 24 +++++++++++++++++++++
 1 file changed, 24 insertions(+)
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +24 -0
index a59539f36a316..bd82a3f7951a4 100644
@@ -1614,6 +1614,30 @@ power-sensor@40 {
 			vs-supply = <&vdd_3v0_sen>;
 		};
 
+		haptic-engine@49 {
+			compatible = "imagis,isa1200";
+			reg = <0x49>;
+
+			control-gpios = <&gpio TEGRA_GPIO(H, 1) GPIO_ACTIVE_HIGH>;
+
+			pwms = <&pwm 3 53000>;
+
+			vdd-supply = <&vdd_3v3_vbat>;
+			vddp-supply = <&vdd_1v8_vio>;
+
+			imagis,clk-div = <128>;
+			imagis,pll-div = <1>;
+
+			imagis,mode = <0>; /* LRA_MODE */
+			imagis,duty-cycle-ns = <1250>;
+
+			ldo {
+				regulator-name = "vdd_vib";
+				regulator-min-microvolt = <2300000>;
+				regulator-max-microvolt = <2300000>;
+			};
+		};
+
 		nct72: temperature-sensor@4c {
 			compatible = "onnn,nct1008";
 			reg = <0x4c>;
Signature · 1 line
2.53.0
#9 lore raw

Re: [PATCH v1 6/7] ARM: tegra: lg-x3: Add haptic engine node

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Medium] dts: missing dt-binding schema for Imagis ISA1200 haptic engine --

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] dts: missing dt-binding schema for Imagis ISA1200 haptic engine --

commit bcc2e5a5981cf130d7fde2c4cf8724ed6f15c474
Author: Svyatoslav Ryhel <[email protected]>

ARM: tegra: lg-x3: Add haptic engine node

This commit adds a device tree node for the Imagis ISA1200 haptic motor driver used in the tegra30-lg-x3 device.

Quoted 21 lines
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +13 -0
index 14d4f992222c6..c7ad3819c21d9 100644
@@ -1568,6 +1568,30 @@ power-sensor@40 {
 			vs-supply = <&vdd_3v0_sen>;
 		};
 
+		haptic-engine@49 {
+			compatible = "imagis,isa1200";
+			reg = <0x49>;
+
+			control-gpios = <&gpio TEGRA_GPIO(H, 1) GPIO_ACTIVE_HIGH>;
+
+			pwms = <&pwm 3 53000>;
+
+			vdd-supply = <&vdd_3v3_vbat>;
+			vddp-supply = <&vdd_1v8_vio>;
+
+			imagis,clk-div = <128>;
+			imagis,pll-div = <1>;

arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi power-sensor@40 { · line 1593

Quoted 10 lines
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +10 -0
+
+			imagis,mode = <0>; /* LRA_MODE */
+			imagis,duty-cycle-ns = <1250>;
+
+			ldo {
+				regulator-name = "vdd_vib";
+				regulator-min-microvolt = <2300000>;
+				regulator-max-microvolt = <2300000>;
+			};
+		};

[Severity: Medium]
Are the dt-binding schema and the driver for the "imagis,isa1200" compatible string missing?

There appears to be no dt-binding schema (YAML or txt) documenting these custom vendor properties (e.g., "imagis,clk-div", "imagis,mode") or the "ldo" subnode, nor a driver that matches this compatible.

Adding undocumented properties and using magic values (e.g., <0> /* LRA_MODE */ instead of a dt-binding macro) violates device tree standards, causes make dtbs_check validation failures, and establishes an unreviewed ABI.

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=6
#10 lore raw

PATCH 7/7 ARM: tegra: lg-x3: Add backlight LED controller node

In reply to Svyatoslav Ryhel

Add node for TI LM3533 LED controller used in this device for panel backlight.

---
 arch/arm/boot/dts/nvidia/tegra30-lg-p880.dts |  2 ++
 arch/arm/boot/dts/nvidia/tegra30-lg-p895.dts |  2 ++
 arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi  | 24 ++++++++++++++++++++
 3 files changed, 28 insertions(+)
arch/arm/boot/dts/nvidia/tegra30-lg-p880.dts +2 -0
index d76f9f465ddc6..4560d93ef438a 100644
@@ -179,6 +179,8 @@ panel@1 {
 				vdd-supply = <&vcc_3v0_lcd>;
 				vddio-supply = <&iovcc_1v8_lcd>;
 
+				backlight = <&backlight>;
+
 				port {
 					panel_input: endpoint {
 						remote-endpoint = <&bridge_output>;
arch/arm/boot/dts/nvidia/tegra30-lg-p895.dts +2 -0
index 4ee84f4bd1878..2ace0201a9311 100644
@@ -185,6 +185,8 @@ panel@1 {
 				vcc-supply = <&vcc_3v0_lcd>;
 				iovcc-supply = <&iovcc_1v8_lcd>;
 
+				backlight = <&backlight>;
+
 				port {
 					panel_input: endpoint {
 						remote-endpoint = <&bridge_output>;
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +24 -0
index bd82a3f7951a4..c86bc4b897675 100644
@@ -1206,6 +1206,30 @@ rmi4-f11@11 {
 			};
 		};
 
+		led-controller@36 {
+			compatible = "ti,lm3533";
+			reg = <0x36>;
+
+			enable-gpios = <&gpio TEGRA_GPIO(N, 6) GPIO_ACTIVE_HIGH>;
+			vin-supply = <&vdd_3v3_vbat>;
+
+			ti,boost-ovp-microvolt = <24000000>;
+			ti,boost-freq-hz = <500000>;
+
+			#address-cells = <1>;
+			#size-cells = <0>;
+
+			backlight: backlight@0 {
+				compatible = "ti,lm3533-backlight";
+				reg = <0>;
+
+				led-max-microamp = <23400>;
+				led-sources = <0 1>;
+
+				ti,linear-mapping-mode;
+			};
+		};
+
 		max14526: muic@44 {
 			compatible = "maxim,max14526";
 			reg = <0x44>;
Signature · 1 line
2.53.0
#11 lore raw

Re: [PATCH v1 7/7] ARM: tegra: lg-x3: Add backlight LED controller node

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider: - [High] drm: display probe deferral due to missing lm3533 DT support - [Medium] mfd: lm3533: GPIO lookup…

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [High] drm: display probe deferral due to missing lm3533 DT support
- [Medium] mfd: lm3533: GPIO lookup failure due to mismatched property name
- [Low] dt-bindings: missing schemas for new compatible strings
--

commit 6d421d459ccd001174a12d39a40edbf6cba8d6fc
Author: Svyatoslav Ryhel <[email protected]>

ARM: tegra: lg-x3: Add backlight LED controller node

Add node for TI LM3533 LED controller used in this device for panel backlight.

arch/arm/boot/dts/nvidia/tegra30-lg-p880.dts panel@1 { · line 158

Quoted 9 lines
arch/arm/boot/dts/nvidia/tegra30-lg-p880.dts +1 -0
index f888987ce1340..3a5e66d869b1b 100644
@@ -155,6 +155,8 @@ panel@1 {
 				vdd-supply = <&vcc_3v0_lcd>;
 				vddio-supply = <&iovcc_1v8_lcd>;
 
+				backlight = <&backlight>;

[Severity: High]
Does this cause the display panel to permanently defer probing?

When the lm3533 driver probes, it unconditionally calls dev_get_platdata() and returns -EINVAL if pdata is missing:

drivers/mfd/lm3533-core.c:lm3533_device_init() {
    ...
	if (!pdata) {
		dev_err(lm3533->dev, "no platform data\n");
		return -EINVAL;
	}
    ...
}

Since the driver lacks Device Tree parsing logic entirely, the backlight fails to probe, which causes the panel to return -EPROBE_DEFER indefinitely.

[ ... ]

arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi rmi4-f11@11 { · line 1174

Quoted 10 lines
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +2 -0
index c7ad3819c21d9..781cf38e9f183 100644
@@ -1170,6 +1170,30 @@ rmi4-f11@11 {
 			};
 		};
 
+		led-controller@36 {
+			compatible = "ti,lm3533";

[Severity: Low]
Are we missing Device Tree schemas for this new compatible string?

Adding a new compatible string without updating or creating Device Tree bindings bypasses schema validation, leaving hardware variant capabilities undocumented. Additionally, the imagis,isa1200 compatible string lacks both a schema and kernel driver support.

arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi rmi4-f11@11 { · line 1213

Quoted 3 lines
arch/arm/boot/dts/nvidia/tegra30-lg-x3.dtsi +3 -0
+			reg = <0x36>;
+
+			enable-gpios = <&gpio TEGRA_GPIO(N, 6) GPIO_ACTIVE_HIGH>;

[Severity: Medium]
Will this GPIO lookup fail due to a mismatched property name?

The driver requests the GPIO with a NULL connection ID:

drivers/mfd/lm3533-core.c:lm3533_device_init() {
    ...
	lm3533->hwen = devm_gpiod_get(lm3533->dev, NULL, GPIOD_OUT_LOW);
    ...
}

This causes gpiolib to strictly search for the base names gpios or gpio, meaning it will receive -ENOENT instead of matching the provided enable-gpios property.

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=7