[PATCH 0/9] arm64: qcom: enable display card support on SM8150 HDK

[PATCH 0/9] arm64: qcom: enable display card support on SM8150 HDK

[Dmitry Baryshkov]

The Qualcomm SM8150 HDK can be equipped with an optional display card
having the same panel as the display card for SDM845 HDK. Reuse the
driver submitted at [1] and enable display card support on this HDK.

Unlike SDM845, this also requires a new LCDB driver, providing voltage
for the panel and several fixes in the WLED driver. The FSA4480 address
fix is independent and can be applied earlier.

[1] https://lore.kernel.org/dri-devel/20261014-sdm845-hdk-v2-0-6d0b4b42f267@oss.qualcomm.com/

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
Dmitry Baryshkov (9):
      backlight: qcom-wled: fix unbalanced OVP interrupt enable
      backlight: qcom-wled: fix string detection brightness on WLED5
      dt-bindings: regulator: add Qualcomm LCDB regulator
      dt-bindings: mfd: qcom,spmi-pmic: add LCDB
      regulator: add Qualcomm LCDB driver
      arm64: dts: qcom: pm8150l: add LCDB regulators
      arm64: dts: qcom: sm8150: add MDSS reset
      arm64: dts: qcom: sm8150-hdk: fix FSA4480 address
      arm64: dts: qcom: sm8150-hdk: add display card overlay

 .../devicetree/bindings/mfd/qcom,spmi-pmic.yaml    |   4 +
 .../bindings/regulator/qcom,pm8150l-lcdb.yaml      |  81 ++++
 arch/arm64/boot/dts/qcom/Makefile                  |   4 +
 arch/arm64/boot/dts/qcom/pm8150l.dtsi              |  18 +
 .../boot/dts/qcom/sm8150-hdk-display-card.dtso     | 103 +++++
 arch/arm64/boot/dts/qcom/sm8150-hdk.dts            |   4 +-
 arch/arm64/boot/dts/qcom/sm8150.dtsi               |   1 +
 drivers/regulator/Kconfig                          |   8 +
 drivers/regulator/Makefile                         |   1 +
 drivers/regulator/qcom-lcdb-regulator.c            | 475 +++++++++++++++++++++
 drivers/video/backlight/qcom-wled.c                |  31 +-
 11 files changed, 723 insertions(+), 7 deletions(-)
---
base-commit: 1fffefb00cc0b0da1f2ae24e3d33364530ea6515
change-id: 20261014-sm8150-display-card-e43ee5b08c61
prerequisite-change-id: 20260122-sdm845-hdk-9466530c4267:v4
prerequisite-patch-id: cb272273cdb260eb5049ba5cfffe13e7904e7a70
prerequisite-patch-id: 118ae0e92f82e1d5be7b75cd310509388e6f5606
prerequisite-patch-id: efe1cd41bd8c1ddf7c9491a6fca5aec877afce3d
prerequisite-patch-id: 8759433719a8151e1a8399fe88036844f60f2ec8
prerequisite-patch-id: e46e47c4d158362a382efc455438ef1fd5c95f42

Best regards,
--  
With best wishes
Dmitry

[PATCH 1/9] backlight: qcom-wled: fix unbalanced OVP interrupt enable

[Dmitry Baryshkov]

If the bootloader leaves the WLED module enabled, the OVP interrupt is
left enabled at probe time. The first brightness update then schedules
wled_ovp_work(), which enables the interrupt for a second time and
triggers the "Unbalanced enable for IRQ" warning.

Track whether the driver has disabled the interrupt and only enable it
back in that case.

Fixes: 8663c188beea ("backlight: qcom-wled: Add auto string detection logic")
Assisted-by: LLM
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/video/backlight/qcom-wled.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/video/backlight/qcom-wled.c b/drivers/video/backlight/qcom-wled.c
index 650dd95f06ef..f2e0430cccbc 100644
--- a/drivers/video/backlight/qcom-wled.c
+++ b/drivers/video/backlight/qcom-wled.c
@@ -199,6 +199,7 @@ struct wled {
 	bool cabc_disabled;
 	int short_irq;
 	int ovp_irq;
+	bool ovp_irq_disabled;
 
 	struct wled_config cfg;
 	struct delayed_work ovp_work;
@@ -294,7 +295,11 @@ static void wled_ovp_work(struct work_struct *work)
 {
 	struct wled *wled = container_of(work,
 					 struct wled, ovp_work.work);
-	enable_irq(wled->ovp_irq);
+
+	if (wled->ovp_irq_disabled) {
+		enable_irq(wled->ovp_irq);
+		wled->ovp_irq_disabled = false;
+	}
 }
 
 static int wled_module_enable(struct wled *wled, int val)
@@ -321,8 +326,11 @@ static int wled_module_enable(struct wled *wled, int val)
 			 */
 			schedule_delayed_work(&wled->ovp_work, HZ / 100);
 		} else {
-			if (!cancel_delayed_work_sync(&wled->ovp_work))
+			cancel_delayed_work_sync(&wled->ovp_work);
+			if (!wled->ovp_irq_disabled) {
 				disable_irq(wled->ovp_irq);
+				wled->ovp_irq_disabled = true;
+			}
 		}
 	}
 
@@ -1649,8 +1657,10 @@ static int wled_configure_ovp_irq(struct wled *wled,
 		return rc;
 
 	/* Keep OVP irq disabled until module is enabled */
-	if (!(val & WLED3_CTRL_REG_MOD_EN_MASK))
+	if (!(val & WLED3_CTRL_REG_MOD_EN_MASK)) {
 		disable_irq(wled->ovp_irq);
+		wled->ovp_irq_disabled = true;
+	}
 
 	return 0;
 }

-- 
2.47.3

[PATCH 2/9] backlight: qcom-wled: fix string detection brightness on WLED5

[Dmitry Baryshkov]

The string detection sets the test brightness through
wled4_set_brightness(). On WLED5 that writes to unrelated registers and
the detection runs with whatever brightness has been programmed before.

Use the per-version callback and latch the new value on WLED5.

Fixes: 62a1d3f623ac ("backlight: qcom-wled: Add support for WLED5 peripheral that is present on PM8150L PMICs")
Assisted-by: LLM
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/video/backlight/qcom-wled.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/video/backlight/qcom-wled.c b/drivers/video/backlight/qcom-wled.c
index f2e0430cccbc..6fbac6a4c29b 100644
--- a/drivers/video/backlight/qcom-wled.c
+++ b/drivers/video/backlight/qcom-wled.c
@@ -577,6 +577,17 @@ static irqreturn_t wled_short_irq_handler(int irq, void *_wled)
 
 #define AUTO_DETECT_BRIGHTNESS		200
 
+static int wled_auto_detect_set_brightness(struct wled *wled, u16 brightness)
+{
+	int rc;
+
+	rc = wled->wled_set_brightness(wled, brightness);
+	if (rc < 0 || wled->version < 5)
+		return rc;
+
+	return wled5_mod_sync_toggle(wled);
+}
+
 static void wled_auto_string_detection(struct wled *wled)
 {
 	int rc = 0, i, j, delay_time_us;
@@ -603,7 +614,7 @@ static void wled_auto_string_detection(struct wled *wled)
 	}
 
 	/* Set low brightness across all sinks */
-	rc = wled4_set_brightness(wled, AUTO_DETECT_BRIGHTNESS);
+	rc = wled_auto_detect_set_brightness(wled, AUTO_DETECT_BRIGHTNESS);
 	if (rc < 0) {
 		dev_err(wled->dev, "Failed to set brightness for auto detection rc=%d\n",
 			rc);
@@ -743,7 +754,7 @@ static void wled_auto_string_detection(struct wled *wled)
 	}
 
 	/* Restore brightness */
-	rc = wled4_set_brightness(wled, wled->brightness);
+	rc = wled_auto_detect_set_brightness(wled, wled->brightness);
 	if (rc < 0) {
 		dev_err(wled->dev, "Failed to set brightness after auto detection rc=%d\n",
 			rc);

-- 
2.47.3

[PATCH 3/9] dt-bindings: regulator: add Qualcomm LCDB regulator

[Dmitry Baryshkov]

Several Qualcomm PMICs have an LCDB block providing the positive and
negative bias rails for LCD panels. It is not supported by Linux, so
boards using it either model the rails as dummy fixed regulators and
depend on the state left by the bootloader.

Describe the LCDB regulator present on the PM8150L. Use it as a fallback
compatible for LCDB blocks on PMI632 and PM6150L. PM660L has a different
version of LCDB block, so it's not being covered.

Assisted-by: LLM
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 .../bindings/regulator/qcom,pm8150l-lcdb.yaml      | 81 ++++++++++++++++++++++
 1 file changed, 81 insertions(+)

diff --git a/Documentation/devicetree/bindings/regulator/qcom,pm8150l-lcdb.yaml b/Documentation/devicetree/bindings/regulator/qcom,pm8150l-lcdb.yaml
new file mode 100644
index 000000000000..eb5014f86439
--- /dev/null
+++ b/Documentation/devicetree/bindings/regulator/qcom,pm8150l-lcdb.yaml
@@ -0,0 +1,81 @@
+# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/regulator/qcom,pm8150l-lcdb.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm LCDB (LCD Bias) regulator
+
+maintainers:
+  - Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
+
+description:
+  LCDB is the LCD panel bias supply found on Qualcomm PMICs such as PMI632,
+  PM6150L and PM8150L. It consists of a boost converter feeding an LDO, which
+  provides the positive bias rail, and an inverting charge pump (NCP), which
+  provides the negative one. Both rails are switched together by a single
+  module enable.
+
+properties:
+  compatible:
+    oneOf:
+      - const: qcom,pm8150l-lcdb
+      - items:
+          - enum:
+              - qcom,pm6150l-lcdb
+              - qcom,pmi632-lcdb
+          - const: qcom,pm8150l-lcdb
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    items:
+      - description: short circuit on one of the outputs
+
+  ldo:
+    type: object
+    $ref: regulator.yaml#
+    unevaluatedProperties: false
+    description: Positive bias rail.
+
+  ncp:
+    type: object
+    $ref: regulator.yaml#
+    unevaluatedProperties: false
+    description:
+      Negative bias rail. The voltage is specified as an absolute value.
+
+required:
+  - compatible
+  - reg
+  - interrupts
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/irq.h>
+
+    pmic {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        lcdb@ec00 {
+            compatible = "qcom,pm8150l-lcdb";
+            reg = <0xec00>;
+            interrupts = <0x5 0xec 0x1 IRQ_TYPE_EDGE_RISING>;
+
+            ldo {
+                regulator-min-microvolt = <4000000>;
+                regulator-max-microvolt = <6000000>;
+            };
+
+            ncp {
+                regulator-min-microvolt = <4000000>;
+                regulator-max-microvolt = <6000000>;
+            };
+        };
+    };
+
+...

-- 
2.47.3

[PATCH 4/9] dt-bindings: mfd: qcom,spmi-pmic: add LCDB

[Dmitry Baryshkov]

Allow the LCDB regulator block as a child of the SPMI PMIC node.

Assisted-by: LLM
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 Documentation/devicetree/bindings/mfd/qcom,spmi-pmic.yaml | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/devicetree/bindings/mfd/qcom,spmi-pmic.yaml b/Documentation/devicetree/bindings/mfd/qcom,spmi-pmic.yaml
index f6534e9e9912..1894fd82cbda 100644
--- a/Documentation/devicetree/bindings/mfd/qcom,spmi-pmic.yaml
+++ b/Documentation/devicetree/bindings/mfd/qcom,spmi-pmic.yaml
@@ -171,6 +171,10 @@ patternProperties:
     type: object
     $ref: /schemas/input/qcom,spmi-haptics.yaml#
 
+  "^lcdb@[0-9a-f]+$":
+    type: object
+    $ref: /schemas/regulator/qcom,pm8150l-lcdb.yaml#
+
   "^led-controller@[0-9a-f]+$":
     type: object
     $ref: /schemas/leds/qcom,spmi-flash-led.yaml#

-- 
2.47.3

[PATCH 5/9] regulator: add Qualcomm LCDB driver

[Dmitry Baryshkov]

LCD panels on boards with PMI632, PM6150L or PM8150L take their bias
rails from the LCDB block. Without a driver they only work as long as
the bootloader has enabled it and nothing turns it off.

Both rails sit behind a single module enable, so the module is kept on
while at least one of them is in use. The boost output is not exposed,
it follows the higher of the two rails plus the headroom. Enabling
starts at a lower voltage and ramps up in steps, and the LDO pull-down
is forced while disabling, both as done by the vendor driver.

Touch-to-wake, current limit and soft start tuning and PM660L are not
supported.

Assisted-by: LLM
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/regulator/Kconfig               |   8 +
 drivers/regulator/Makefile              |   1 +
 drivers/regulator/qcom-lcdb-regulator.c | 475 ++++++++++++++++++++++++++++++++
 3 files changed, 484 insertions(+)

diff --git a/drivers/regulator/Kconfig b/drivers/regulator/Kconfig
index be6b72daa76f..9fd3f2c7e365 100644
--- a/drivers/regulator/Kconfig
+++ b/drivers/regulator/Kconfig
@@ -1915,4 +1915,12 @@ config REGULATOR_QCOM_LABIBB
 	  boost regulator and IBB can be used as a negative boost regulator
 	  for LCD display panel.
 
+config REGULATOR_QCOM_LCDB
+	tristate "Qualcomm LCDB regulator support"
+	depends on SPMI || COMPILE_TEST
+	help
+	  This driver supports the LCDB (LCD bias) block found on Qualcomm
+	  PMICs such as PMI632, PM6150L and PM8150L. It provides the positive
+	  (LDO) and negative (NCP) bias rails for LCD panels.
+
 endif
diff --git a/drivers/regulator/Makefile b/drivers/regulator/Makefile
index 6aabcc54382a..84164d2b6c45 100644
--- a/drivers/regulator/Makefile
+++ b/drivers/regulator/Makefile
@@ -123,6 +123,7 @@ obj-$(CONFIG_REGULATOR_MT6397)	+= mt6397-regulator.o
 obj-$(CONFIG_REGULATOR_MTK_DVFSRC) += mtk-dvfsrc-regulator.o
 obj-$(CONFIG_REGULATOR_NEX10000UB) += nex10000ub-regulator.o
 obj-$(CONFIG_REGULATOR_QCOM_LABIBB) += qcom-labibb-regulator.o
+obj-$(CONFIG_REGULATOR_QCOM_LCDB) += qcom-lcdb-regulator.o
 obj-$(CONFIG_REGULATOR_QCOM_PM8008) += qcom-pm8008-regulator.o
 obj-$(CONFIG_REGULATOR_QCOM_REFGEN) += qcom-refgen-regulator.o
 obj-$(CONFIG_REGULATOR_QCOM_RPM) += qcom_rpm-regulator.o
diff --git a/drivers/regulator/qcom-lcdb-regulator.c b/drivers/regulator/qcom-lcdb-regulator.c
new file mode 100644
index 000000000000..45b3b7118e31
--- /dev/null
+++ b/drivers/regulator/qcom-lcdb-regulator.c
@@ -0,0 +1,475 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries.
+ */
+
+#include <linux/bits.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/ktime.h>
+#include <linux/minmax.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/property.h>
+#include <linux/regmap.h>
+#include <linux/regulator/driver.h>
+#include <linux/regulator/of_regulator.h>
+
+#define LCDB_INT_RT_STS			0x10
+#define  LCDB_VREG_OK_RT_STS		BIT(0)
+#define  LCDB_SC_ERROR_RT_STS		BIT(1)
+
+#define LCDB_BST_OUTPUT_VOLTAGE		0x41
+#define  LCDB_BST_OUTPUT_VOLTAGE_MASK	GENMASK(5, 0)
+
+#define LCDB_MODULE_RDY			0x45
+#define  LCDB_MODULE_RDY_BIT		BIT(7)
+
+#define LCDB_ENABLE_CTL1		0x46
+#define  LCDB_MODULE_EN			BIT(7)
+
+#define LCDB_LDO_OUTPUT_VOLTAGE		0x71
+#define LCDB_LDO_FORCE_PD_CTL		0x79
+#define  LCDB_LDO_FORCE_PD_EN		BIT(0)
+#define  LCDB_LDO_FORCE_PD_MODE		BIT(7)
+
+#define LCDB_NCP_OUTPUT_VOLTAGE		0x81
+
+#define LCDB_OUTPUT_VOLTAGE_MASK	GENMASK(4, 0)
+
+#define LCDB_MIN_UV			4000000
+#define LCDB_STEP_100MV_UV		100000
+#define LCDB_50MV_MIN_UV		4950000
+#define LCDB_STEP_50MV_UV		50000
+#define LCDB_50MV_MIN_SEL		10
+#define LCDB_N_VOLTAGES			32
+
+#define LCDB_BST_MIN_UV			4700000
+#define LCDB_BST_MAX_UV			6275000
+#define LCDB_BST_STEP_UV		25000
+#define LCDB_BST_HEADROOM_UV		150000
+
+#define LCDB_RAMP_START_UV		4500000
+#define LCDB_RAMP_STEP_UV		500000
+
+#define LCDB_OFF_ON_DELAY_US		20000
+#define LCDB_VREG_OK_TIMEOUT_US		50000
+
+#define LCDB_SC_RESET_CNT_DLY_US	1000000
+#define LCDB_SC_CNT_MAX			10
+
+enum {
+	LCDB_LDO,
+	LCDB_NCP,
+	LCDB_NUM_RAILS,
+};
+
+struct qcom_lcdb {
+	struct device *dev;
+	struct regmap *regmap;
+	u32 base;
+
+	struct mutex lock;
+	struct regulator_dev *rdev[LCDB_NUM_RAILS];
+	int uV[LCDB_NUM_RAILS];
+	unsigned int enabled_mask;
+	ktime_t off_time;
+
+	bool sc_disabled;
+	int sc_count;
+	ktime_t sc_time;
+};
+
+static const u8 qcom_lcdb_vreg[LCDB_NUM_RAILS] = {
+	[LCDB_LDO] = LCDB_LDO_OUTPUT_VOLTAGE,
+	[LCDB_NCP] = LCDB_NCP_OUTPUT_VOLTAGE,
+};
+
+static unsigned int qcom_lcdb_uV_to_sel(int uV)
+{
+	if (uV < LCDB_50MV_MIN_UV)
+		return (uV - LCDB_MIN_UV) / LCDB_STEP_100MV_UV;
+
+	return LCDB_50MV_MIN_SEL + (uV - LCDB_50MV_MIN_UV) / LCDB_STEP_50MV_UV;
+}
+
+static int qcom_lcdb_sel_to_uV(unsigned int sel)
+{
+	if (sel < LCDB_50MV_MIN_SEL)
+		return LCDB_MIN_UV + sel * LCDB_STEP_100MV_UV;
+
+	return LCDB_50MV_MIN_UV + (sel - LCDB_50MV_MIN_SEL) * LCDB_STEP_50MV_UV;
+}
+
+static int qcom_lcdb_write_voltages(struct qcom_lcdb *lcdb, int ldo_uV, int ncp_uV)
+{
+	const int uV[LCDB_NUM_RAILS] = { ldo_uV, ncp_uV };
+	unsigned int bst_sel, cur_sel;
+	int bst_uV, ret, i;
+
+	bst_uV = clamp(max(ldo_uV, ncp_uV) + LCDB_BST_HEADROOM_UV,
+		       LCDB_BST_MIN_UV, LCDB_BST_MAX_UV);
+	bst_sel = DIV_ROUND_UP(bst_uV - LCDB_BST_MIN_UV, LCDB_BST_STEP_UV);
+
+	ret = regmap_read(lcdb->regmap, lcdb->base + LCDB_BST_OUTPUT_VOLTAGE, &cur_sel);
+	if (ret)
+		return ret;
+	cur_sel &= LCDB_BST_OUTPUT_VOLTAGE_MASK;
+
+	/* The boost has to stay above both outputs while they are changing */
+	if (bst_sel > cur_sel) {
+		ret = regmap_update_bits(lcdb->regmap,
+					 lcdb->base + LCDB_BST_OUTPUT_VOLTAGE,
+					 LCDB_BST_OUTPUT_VOLTAGE_MASK, bst_sel);
+		if (ret)
+			return ret;
+	}
+
+	for (i = 0; i < LCDB_NUM_RAILS; i++) {
+		ret = regmap_update_bits(lcdb->regmap,
+					 lcdb->base + qcom_lcdb_vreg[i],
+					 LCDB_OUTPUT_VOLTAGE_MASK,
+					 qcom_lcdb_uV_to_sel(uV[i]));
+		if (ret)
+			return ret;
+	}
+
+	if (bst_sel < cur_sel)
+		ret = regmap_update_bits(lcdb->regmap,
+					 lcdb->base + LCDB_BST_OUTPUT_VOLTAGE,
+					 LCDB_BST_OUTPUT_VOLTAGE_MASK, bst_sel);
+
+	return ret;
+}
+
+static int qcom_lcdb_write_limited(struct qcom_lcdb *lcdb, int limit_uV)
+{
+	return qcom_lcdb_write_voltages(lcdb,
+					min(lcdb->uV[LCDB_LDO], limit_uV),
+					min(lcdb->uV[LCDB_NCP], limit_uV));
+}
+
+static int qcom_lcdb_module_disable(struct qcom_lcdb *lcdb)
+{
+	int ret;
+
+	/* Force the LDO pull-down to discharge the output quickly */
+	ret = regmap_write(lcdb->regmap, lcdb->base + LCDB_LDO_FORCE_PD_CTL,
+			   LCDB_LDO_FORCE_PD_EN | LCDB_LDO_FORCE_PD_MODE);
+	if (ret)
+		return ret;
+
+	ret = regmap_write(lcdb->regmap, lcdb->base + LCDB_ENABLE_CTL1, 0);
+	if (ret)
+		return ret;
+
+	lcdb->off_time = ktime_get();
+
+	usleep_range(10000, 11000);
+
+	return regmap_write(lcdb->regmap, lcdb->base + LCDB_LDO_FORCE_PD_CTL, 0);
+}
+
+static int qcom_lcdb_module_enable(struct qcom_lcdb *lcdb)
+{
+	int target_uV = max(lcdb->uV[LCDB_LDO], lcdb->uV[LCDB_NCP]);
+	int uV = LCDB_RAMP_START_UV;
+	unsigned int val;
+	s64 off_us;
+	int ret;
+
+	off_us = ktime_us_delta(ktime_get(), lcdb->off_time);
+	if (off_us < LCDB_OFF_ON_DELAY_US)
+		usleep_range(LCDB_OFF_ON_DELAY_US - off_us,
+			     LCDB_OFF_ON_DELAY_US - off_us + 1000);
+
+	/* Start low and ramp the outputs up in steps to limit the inrush current */
+	ret = qcom_lcdb_write_limited(lcdb, uV);
+	if (ret)
+		return ret;
+
+	ret = regmap_write(lcdb->regmap, lcdb->base + LCDB_ENABLE_CTL1,
+			   LCDB_MODULE_EN);
+	if (ret)
+		return ret;
+
+	ret = regmap_read_poll_timeout(lcdb->regmap,
+				       lcdb->base + LCDB_INT_RT_STS, val,
+				       val & LCDB_VREG_OK_RT_STS,
+				       1000, LCDB_VREG_OK_TIMEOUT_US);
+	if (ret) {
+		dev_err(lcdb->dev, "outputs failed to come up: %d\n", ret);
+		goto err_disable;
+	}
+
+	usleep_range(10000, 11000);
+
+	while (uV < target_uV) {
+		uV += LCDB_RAMP_STEP_UV;
+
+		ret = qcom_lcdb_write_limited(lcdb, uV);
+		if (ret)
+			goto err_disable;
+
+		usleep_range(1000, 1100);
+	}
+
+	return 0;
+
+err_disable:
+	qcom_lcdb_module_disable(lcdb);
+
+	return ret;
+}
+
+static int qcom_lcdb_enable(struct regulator_dev *rdev)
+{
+	struct qcom_lcdb *lcdb = rdev_get_drvdata(rdev);
+	int ret = 0;
+
+	guard(mutex)(&lcdb->lock);
+
+	if (lcdb->sc_disabled)
+		return -EIO;
+
+	if (!lcdb->enabled_mask)
+		ret = qcom_lcdb_module_enable(lcdb);
+	if (ret)
+		return ret;
+
+	lcdb->enabled_mask |= BIT(rdev_get_id(rdev));
+
+	return 0;
+}
+
+static int qcom_lcdb_disable(struct regulator_dev *rdev)
+{
+	struct qcom_lcdb *lcdb = rdev_get_drvdata(rdev);
+
+	guard(mutex)(&lcdb->lock);
+
+	lcdb->enabled_mask &= ~BIT(rdev_get_id(rdev));
+	if (lcdb->enabled_mask)
+		return 0;
+
+	return qcom_lcdb_module_disable(lcdb);
+}
+
+static int qcom_lcdb_is_enabled(struct regulator_dev *rdev)
+{
+	struct qcom_lcdb *lcdb = rdev_get_drvdata(rdev);
+
+	guard(mutex)(&lcdb->lock);
+
+	return !!(lcdb->enabled_mask & BIT(rdev_get_id(rdev)));
+}
+
+static int qcom_lcdb_set_voltage_sel(struct regulator_dev *rdev, unsigned int sel)
+{
+	struct qcom_lcdb *lcdb = rdev_get_drvdata(rdev);
+
+	guard(mutex)(&lcdb->lock);
+
+	lcdb->uV[rdev_get_id(rdev)] = qcom_lcdb_sel_to_uV(sel);
+
+	return qcom_lcdb_write_voltages(lcdb, lcdb->uV[LCDB_LDO],
+					lcdb->uV[LCDB_NCP]);
+}
+
+static int qcom_lcdb_get_voltage_sel(struct regulator_dev *rdev)
+{
+	struct qcom_lcdb *lcdb = rdev_get_drvdata(rdev);
+
+	guard(mutex)(&lcdb->lock);
+
+	return qcom_lcdb_uV_to_sel(lcdb->uV[rdev_get_id(rdev)]);
+}
+
+static const struct regulator_ops qcom_lcdb_ops = {
+	.enable			= qcom_lcdb_enable,
+	.disable		= qcom_lcdb_disable,
+	.is_enabled		= qcom_lcdb_is_enabled,
+	.set_voltage_sel	= qcom_lcdb_set_voltage_sel,
+	.get_voltage_sel	= qcom_lcdb_get_voltage_sel,
+	.list_voltage		= regulator_list_voltage_linear_range,
+	.map_voltage		= regulator_map_voltage_linear_range,
+};
+
+static const struct linear_range qcom_lcdb_ranges[] = {
+	REGULATOR_LINEAR_RANGE(LCDB_MIN_UV, 0, LCDB_50MV_MIN_SEL - 1,
+			       LCDB_STEP_100MV_UV),
+	REGULATOR_LINEAR_RANGE(LCDB_50MV_MIN_UV, LCDB_50MV_MIN_SEL,
+			       LCDB_N_VOLTAGES - 1, LCDB_STEP_50MV_UV),
+};
+
+static const struct regulator_desc qcom_lcdb_descs[LCDB_NUM_RAILS] = {
+	[LCDB_LDO] = {
+		.name			= "lcdb_ldo",
+		.of_match		= "ldo",
+		.id			= LCDB_LDO,
+		.ops			= &qcom_lcdb_ops,
+		.type			= REGULATOR_VOLTAGE,
+		.linear_ranges		= qcom_lcdb_ranges,
+		.n_linear_ranges	= ARRAY_SIZE(qcom_lcdb_ranges),
+		.n_voltages		= LCDB_N_VOLTAGES,
+		.owner			= THIS_MODULE,
+	},
+	[LCDB_NCP] = {
+		.name			= "lcdb_ncp",
+		.of_match		= "ncp",
+		.id			= LCDB_NCP,
+		.ops			= &qcom_lcdb_ops,
+		.type			= REGULATOR_VOLTAGE,
+		.linear_ranges		= qcom_lcdb_ranges,
+		.n_linear_ranges	= ARRAY_SIZE(qcom_lcdb_ranges),
+		.n_voltages		= LCDB_N_VOLTAGES,
+		.owner			= THIS_MODULE,
+	},
+};
+
+static void qcom_lcdb_handle_sc(struct qcom_lcdb *lcdb)
+{
+	int i;
+
+	guard(mutex)(&lcdb->lock);
+
+	if (!lcdb->enabled_mask || lcdb->sc_disabled)
+		return;
+
+	qcom_lcdb_module_disable(lcdb);
+
+	if (ktime_us_delta(ktime_get(), lcdb->sc_time) > LCDB_SC_RESET_CNT_DLY_US) {
+		lcdb->sc_count = 0;
+	} else if (lcdb->sc_count > LCDB_SC_CNT_MAX) {
+		dev_err(lcdb->dev, "persistent short circuit, keeping the outputs off\n");
+		lcdb->sc_disabled = true;
+
+		for (i = 0; i < LCDB_NUM_RAILS; i++)
+			regulator_notifier_call_chain(lcdb->rdev[i],
+						      REGULATOR_EVENT_FAIL, NULL);
+		return;
+	}
+
+	lcdb->sc_count++;
+	lcdb->sc_time = ktime_get();
+
+	if (qcom_lcdb_module_enable(lcdb))
+		dev_err(lcdb->dev, "failed to recover from a short circuit\n");
+}
+
+static irqreturn_t qcom_lcdb_sc_irq(int irq, void *data)
+{
+	struct qcom_lcdb *lcdb = data;
+	unsigned int val;
+	int ret;
+
+	ret = regmap_read(lcdb->regmap, lcdb->base + LCDB_INT_RT_STS, &val);
+	if (ret || !(val & LCDB_SC_ERROR_RT_STS))
+		return IRQ_HANDLED;
+
+	/* Blanking time, a real short circuit is still reported afterwards */
+	usleep_range(2000, 2100);
+
+	ret = regmap_read(lcdb->regmap, lcdb->base + LCDB_INT_RT_STS, &val);
+	if (ret || !(val & LCDB_SC_ERROR_RT_STS))
+		return IRQ_HANDLED;
+
+	qcom_lcdb_handle_sc(lcdb);
+
+	return IRQ_HANDLED;
+}
+
+static int qcom_lcdb_hw_init(struct qcom_lcdb *lcdb)
+{
+	unsigned int val;
+	int ret, i;
+
+	for (i = 0; i < LCDB_NUM_RAILS; i++) {
+		ret = regmap_read(lcdb->regmap, lcdb->base + qcom_lcdb_vreg[i], &val);
+		if (ret)
+			return ret;
+
+		lcdb->uV[i] = qcom_lcdb_sel_to_uV(val & LCDB_OUTPUT_VOLTAGE_MASK);
+	}
+
+	ret = regmap_read(lcdb->regmap, lcdb->base + LCDB_ENABLE_CTL1, &val);
+	if (ret)
+		return ret;
+
+	if (val & LCDB_MODULE_EN) {
+		lcdb->enabled_mask = GENMASK(LCDB_NUM_RAILS - 1, 0);
+		return 0;
+	}
+
+	return regmap_set_bits(lcdb->regmap, lcdb->base + LCDB_MODULE_RDY,
+			       LCDB_MODULE_RDY_BIT);
+}
+
+static int qcom_lcdb_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct regulator_config cfg = {};
+	struct qcom_lcdb *lcdb;
+	int irq, ret, i;
+
+	lcdb = devm_kzalloc(dev, sizeof(*lcdb), GFP_KERNEL);
+	if (!lcdb)
+		return -ENOMEM;
+
+	lcdb->dev = dev;
+
+	lcdb->regmap = dev_get_regmap(dev->parent, NULL);
+	if (!lcdb->regmap)
+		return dev_err_probe(dev, -ENODEV, "failed to get the parent regmap\n");
+
+	ret = device_property_read_u32(dev, "reg", &lcdb->base);
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to get the base address\n");
+
+	ret = devm_mutex_init(dev, &lcdb->lock);
+	if (ret)
+		return ret;
+
+	ret = qcom_lcdb_hw_init(lcdb);
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to read the initial state\n");
+
+	cfg.dev = dev;
+	cfg.regmap = lcdb->regmap;
+	cfg.driver_data = lcdb;
+
+	for (i = 0; i < LCDB_NUM_RAILS; i++) {
+		lcdb->rdev[i] = devm_regulator_register(dev, &qcom_lcdb_descs[i], &cfg);
+		if (IS_ERR(lcdb->rdev[i]))
+			return dev_err_probe(dev, PTR_ERR(lcdb->rdev[i]),
+					     "failed to register %s\n",
+					     qcom_lcdb_descs[i].name);
+	}
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0)
+		return irq;
+
+	return devm_request_threaded_irq(dev, irq, NULL, qcom_lcdb_sc_irq,
+					 IRQF_ONESHOT, "lcdb-sc", lcdb);
+}
+
+static const struct of_device_id qcom_lcdb_match[] = {
+	{ .compatible = "qcom,pm8150l-lcdb" },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, qcom_lcdb_match);
+
+static struct platform_driver qcom_lcdb_driver = {
+	.driver = {
+		.name = "qcom-lcdb-regulator",
+		.of_match_table = qcom_lcdb_match,
+	},
+	.probe = qcom_lcdb_probe,
+};
+module_platform_driver(qcom_lcdb_driver);
+
+MODULE_DESCRIPTION("Qualcomm LCDB (LCD bias) regulator driver");
+MODULE_LICENSE("GPL");

-- 
2.47.3

[PATCH 6/9] arm64: dts: qcom: pm8150l: add LCDB regulators

[Dmitry Baryshkov]

Describe the LCDB block providing the LCD bias rails.

Assisted-by: LLM
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 arch/arm64/boot/dts/qcom/pm8150l.dtsi | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/pm8150l.dtsi b/arch/arm64/boot/dts/qcom/pm8150l.dtsi
index 3911d6d0d2e2..f54ddb6f7946 100644
--- a/arch/arm64/boot/dts/qcom/pm8150l.dtsi
+++ b/arch/arm64/boot/dts/qcom/pm8150l.dtsi
@@ -141,5 +141,23 @@ pm8150l_wled: leds@d800 {
 
 			status = "disabled";
 		};
+
+		pm8150l_lcdb: lcdb@ec00 {
+			compatible = "qcom,pm8150l-lcdb";
+			reg = <0xec00>;
+			interrupts = <0x5 0xec 0x1 IRQ_TYPE_EDGE_RISING>;
+
+			status = "disabled";
+
+			pm8150l_lcdb_ldo: ldo {
+				regulator-min-microvolt = <4000000>;
+				regulator-max-microvolt = <6000000>;
+			};
+
+			pm8150l_lcdb_ncp: ncp {
+				regulator-min-microvolt = <4000000>;
+				regulator-max-microvolt = <6000000>;
+			};
+		};
 	};
 };

-- 
2.47.3

[PATCH 7/9] arm64: dts: qcom: sm8150: add MDSS reset

[Dmitry Baryshkov]

The bootloader can leave the display running, as it does on the SM8150
HDK with the display card attached. Describe the MDSS reset so that the
driver can bring the hardware into a known state before using it.

Assisted-by: LLM
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 arch/arm64/boot/dts/qcom/sm8150.dtsi | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm64/boot/dts/qcom/sm8150.dtsi b/arch/arm64/boot/dts/qcom/sm8150.dtsi
index 293fcbacff34..a2d6e6947c4c 100644
--- a/arch/arm64/boot/dts/qcom/sm8150.dtsi
+++ b/arch/arm64/boot/dts/qcom/sm8150.dtsi
@@ -3817,6 +3817,7 @@ mdss: display-subsystem@ae00000 {
 
 			iommus = <&apps_smmu 0x800 0x420>;
 
+			resets = <&dispcc DISP_CC_MDSS_CORE_BCR>;
 			status = "disabled";
 
 			#address-cells = <2>;

-- 
2.47.3

[PATCH 8/9] arm64: dts: qcom: sm8150-hdk: fix FSA4480 address

[Dmitry Baryshkov]

The FSA4480 on the SM8150 HDK responds at 0x43 rather than 0x42. With
the wrong address the mux driver fails to probe, the Type-C controller
and the DisplayPort controller keep waiting for it and the display
subsystem never binds.

Fixes: ba712fd55c7b ("arm64: dts: qcom: sm8150-hdk: enable DisplayPort and USB-C altmode")
Assisted-by: LLM
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 arch/arm64/boot/dts/qcom/sm8150-hdk.dts | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm64/boot/dts/qcom/sm8150-hdk.dts b/arch/arm64/boot/dts/qcom/sm8150-hdk.dts
index cdced5c848d3..8f4ea981bb89 100644
--- a/arch/arm64/boot/dts/qcom/sm8150-hdk.dts
+++ b/arch/arm64/boot/dts/qcom/sm8150-hdk.dts
@@ -434,9 +434,9 @@ &i2c4 {
 
 	status = "okay";
 
-	typec-mux@42 {
+	typec-mux@43 {
 		compatible = "fcs,fsa4480";
-		reg = <0x42>;
+		reg = <0x43>;
 
 		interrupts-extended = <&tlmm 152 IRQ_TYPE_LEVEL_LOW>;
 

-- 
2.47.3

[PATCH 9/9] arm64: dts: qcom: sm8150-hdk: add display card overlay

[Dmitry Baryshkov]

The SM8150 HDK can be fitted with a display card carrying a Truly
NT36850 WQHD dual-DSI panel. It uses the PM8150L WLED for the backlight
and LCDB for the bias rails, and it is mutually exclusive with the
on-board LT9611 HDMI bridge.

Only two of the three WLED strings are connected on the card: the
backlight stays dark and the boost runs into the overvoltage limit if
only the third string is driven.

Assisted-by: LLM
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 arch/arm64/boot/dts/qcom/Makefile                  |   4 +
 .../boot/dts/qcom/sm8150-hdk-display-card.dtso     | 103 +++++++++++++++++++++
 2 files changed, 107 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/Makefile b/arch/arm64/boot/dts/qcom/Makefile
index 85d06a950b4a..571bda210d63 100644
--- a/arch/arm64/boot/dts/qcom/Makefile
+++ b/arch/arm64/boot/dts/qcom/Makefile
@@ -409,6 +409,10 @@ dtb-$(CONFIG_ARCH_QCOM)	+= sm7325-nothing-spacewar.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= sm7325-xiaomi-lisa.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= sm7325-xiaomi-taoyao.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= sm8150-hdk.dtb
+
+sm8150-hdk-display-card-dtbs	:= sm8150-hdk.dtb sm8150-hdk-display-card.dtbo
+
+dtb-$(CONFIG_ARCH_QCOM)	+= sm8150-hdk-display-card.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= sm8150-microsoft-surface-duo.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= sm8150-mtp.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= sm8150-sony-xperia-kumano-bahamut.dtb
diff --git a/arch/arm64/boot/dts/qcom/sm8150-hdk-display-card.dtso b/arch/arm64/boot/dts/qcom/sm8150-hdk-display-card.dtso
new file mode 100644
index 000000000000..205cd6d2cd71
--- /dev/null
+++ b/arch/arm64/boot/dts/qcom/sm8150-hdk-display-card.dtso
@@ -0,0 +1,103 @@
+// SPDX-License-Identifier: BSD-3-Clause
+/*
+ * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries.
+ */
+
+/*
+ * Display Card kit overlay.
+ */
+
+#include <dt-bindings/gpio/gpio.h>
+#include <dt-bindings/interrupt-controller/irq.h>
+
+/dts-v1/;
+/plugin/;
+
+/* Disable HDMI bridge related nodes (mutually exclusive with the display card) */
+&lt9611_codec {
+	status = "disabled";
+};
+
+&mdss_dsi0 {
+	#address-cells = <1>;
+	#size-cells = <0>;
+
+	panel@0 {
+		compatible = "truly,wqhd60-nt36850";
+		reg = <0>;
+
+		reset-gpios = <&tlmm 6 GPIO_ACTIVE_LOW>;
+
+		vddio-supply = <&vreg_l14a_1p8>;
+		avdd-supply = <&pm8150l_lcdb_ldo>;
+		avee-supply = <&pm8150l_lcdb_ncp>;
+
+		backlight = <&pm8150l_wled>;
+
+		pinctrl-0 = <&disp0_reset_n_active>, <&mdp_vsync>;
+		pinctrl-1 = <&disp0_reset_n_suspend>, <&mdp_vsync>;
+		pinctrl-names = "default", "sleep";
+
+		ports {
+			#address-cells = <1>;
+			#size-cells = <0>;
+
+			port@0 {
+				reg = <0>;
+
+				panel0_in: endpoint {
+					remote-endpoint = <&mdss_dsi0_out>;
+				};
+			};
+
+			port@1 {
+				reg = <1>;
+
+				panel1_in: endpoint {
+					remote-endpoint = <&mdss_dsi1_out>;
+				};
+			};
+		};
+	};
+};
+
+&mdss_dsi0_out {
+	remote-endpoint = <&panel0_in>;
+};
+
+&mdss_dsi1_out {
+	remote-endpoint = <&panel1_in>;
+};
+
+&pm8150l_lcdb {
+	status = "okay";
+};
+
+&pm8150l_wled {
+	qcom,num-strings = <2>;
+
+	status = "okay";
+};
+
+&tlmm {
+	disp0_reset_n_active: disp0-reset-n-active-state {
+		pins = "gpio6";
+		function = "gpio";
+		drive-strength = <8>;
+		bias-disable;
+	};
+
+	disp0_reset_n_suspend: disp0-reset-n-suspend-state {
+		pins = "gpio6";
+		function = "gpio";
+		drive-strength = <2>;
+		bias-pull-down;
+	};
+
+	mdp_vsync: mdp-vsync-state {
+		pins = "gpio8";
+		function = "mdp_vsync";
+		drive-strength = <2>;
+		bias-pull-down;
+	};
+};

-- 
2.47.3

Re: [PATCH 8/9] arm64: dts: qcom: sm8150-hdk: fix FSA4480 address

[Konrad Dybcio]

On 10/15/26 10:58 AM, Dmitry Baryshkov wrote:
> The FSA4480 on the SM8150 HDK responds at 0x43 rather than 0x42. With
> the wrong address the mux driver fails to probe, the Type-C controller
> and the DisplayPort controller keep waiting for it and the display
> subsystem never binds.
> 
> Fixes: ba712fd55c7b ("arm64: dts: qcom: sm8150-hdk: enable DisplayPort and USB-C altmode")
> Assisted-by: LLM
> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
> ---

Hm, there's an ADDR pin on the FSA. I wonder if the previous guy had
an earlier revision of the board which routed that differently or
something.

Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>

Konrad

Re: [PATCH 8/9] arm64: dts: qcom: sm8150-hdk: fix FSA4480 address

[Konrad Dybcio]

On 10/15/26 11:10 AM, Konrad Dybcio wrote:
> On 10/15/26 10:58 AM, Dmitry Baryshkov wrote:
>> The FSA4480 on the SM8150 HDK responds at 0x43 rather than 0x42. With
>> the wrong address the mux driver fails to probe, the Type-C controller
>> and the DisplayPort controller keep waiting for it and the display
>> subsystem never binds.
>>
>> Fixes: ba712fd55c7b ("arm64: dts: qcom: sm8150-hdk: enable DisplayPort and USB-C altmode")
>> Assisted-by: LLM
>> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
>> ---
> 
> Hm, there's an ADDR pin on the FSA. I wonder if the previous guy had
> an earlier revision of the board which routed that differently or
> something.
> 
> Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>

oh wait.. the previous guy (author of the Fixes commit is you.. :D

Konrad

Re: [PATCH 1/9] backlight: qcom-wled: fix unbalanced OVP interrupt enable

[Konrad Dybcio]

On 10/15/26 10:58 AM, Dmitry Baryshkov wrote:
> If the bootloader leaves the WLED module enabled, the OVP interrupt is
> left enabled at probe time. The first brightness update then schedules
> wled_ovp_work(), which enables the interrupt for a second time and
> triggers the "Unbalanced enable for IRQ" warning.
> 
> Track whether the driver has disabled the interrupt and only enable it
> back in that case.
> 
> Fixes: 8663c188beea ("backlight: qcom-wled: Add auto string detection logic")
> Assisted-by: LLM
> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
> ---

This was merged yesterday:

https://lore.kernel.org/lkml/20260921-qcom-wled-backlight-v4-0-bab8c7ef73cb@ixit.cz/

Konrad

Re: [PATCH 2/9] backlight: qcom-wled: fix string detection brightness on WLED5

[Konrad Dybcio]

On 10/15/26 10:58 AM, Dmitry Baryshkov wrote:
> The string detection sets the test brightness through
> wled4_set_brightness(). On WLED5 that writes to unrelated registers and
> the detection runs with whatever brightness has been programmed before.
> 
> Use the per-version callback and latch the new value on WLED5.
> 
> Fixes: 62a1d3f623ac ("backlight: qcom-wled: Add support for WLED5 peripheral that is present on PM8150L PMICs")
> Assisted-by: LLM
> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
> ---

Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>

Konrad

Re: [PATCH 3/9] dt-bindings: regulator: add Qualcomm LCDB regulator

[Konrad Dybcio]

On 10/15/26 10:58 AM, Dmitry Baryshkov wrote:
> Several Qualcomm PMICs have an LCDB block providing the positive and
> negative bias rails for LCD panels. It is not supported by Linux, so
> boards using it either model the rails as dummy fixed regulators and
> depend on the state left by the bootloader.
> 
> Describe the LCDB regulator present on the PM8150L. Use it as a fallback
> compatible for LCDB blocks on PMI632 and PM6150L. PM660L has a different
> version of LCDB block, so it's not being covered.
> 
> Assisted-by: LLM
> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
> ---

[...]

> +        lcdb@ec00 {
> +            compatible = "qcom,pm8150l-lcdb";
> +            reg = <0xec00>;
> +            interrupts = <0x5 0xec 0x1 IRQ_TYPE_EDGE_RISING>;
> +
> +            ldo {
> +                regulator-min-microvolt = <4000000>;
> +                regulator-max-microvolt = <6000000>;
> +            };
> +
> +            ncp {
> +                regulator-min-microvolt = <4000000>;
> +                regulator-max-microvolt = <6000000>;
> +            };

Let's tighten up the example to some specific voltage to reduce
room for poor copypastas

Konrad

Re: [PATCH 7/9] arm64: dts: qcom: sm8150: add MDSS reset

[Konrad Dybcio]

On 10/15/26 10:58 AM, Dmitry Baryshkov wrote:
> The bootloader can leave the display running, as it does on the SM8150
> HDK with the display card attached. Describe the MDSS reset so that the
> driver can bring the hardware into a known state before using it.
> 
> Assisted-by: LLM
> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
> ---
>  arch/arm64/boot/dts/qcom/sm8150.dtsi | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/arch/arm64/boot/dts/qcom/sm8150.dtsi b/arch/arm64/boot/dts/qcom/sm8150.dtsi
> index 293fcbacff34..a2d6e6947c4c 100644
> --- a/arch/arm64/boot/dts/qcom/sm8150.dtsi
> +++ b/arch/arm64/boot/dts/qcom/sm8150.dtsi
> @@ -3817,6 +3817,7 @@ mdss: display-subsystem@ae00000 {
>  
>  			iommus = <&apps_smmu 0x800 0x420>;
>  
> +			resets = <&dispcc DISP_CC_MDSS_CORE_BCR>;

please put it under clocks like in other DTs

Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>

Konrad

Re: [PATCH 1/9] backlight: qcom-wled: fix unbalanced OVP interrupt enable

[Dmitry Baryshkov]

On Thu, Oct 15, 2026 at 11:11:58AM +0200, Konrad Dybcio wrote:
> On 10/15/26 10:58 AM, Dmitry Baryshkov wrote:
> > If the bootloader leaves the WLED module enabled, the OVP interrupt is
> > left enabled at probe time. The first brightness update then schedules
> > wled_ovp_work(), which enables the interrupt for a second time and
> > triggers the "Unbalanced enable for IRQ" warning.
> > 
> > Track whether the driver has disabled the interrupt and only enable it
> > back in that case.
> > 
> > Fixes: 8663c188beea ("backlight: qcom-wled: Add auto string detection logic")
> > Assisted-by: LLM
> > Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
> > ---
> 
> This was merged yesterday:
> 
> https://lore.kernel.org/lkml/20260921-qcom-wled-backlight-v4-0-bab8c7ef73cb@ixit.cz/

Nice!

-- 
With best wishes
Dmitry

Re: [PATCH 3/9] dt-bindings: regulator: add Qualcomm LCDB regulator

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 656 bytes --]

On Thu, Oct 15, 2026 at 11:58:40AM +0300, Dmitry Baryshkov wrote:
> Several Qualcomm PMICs have an LCDB block providing the positive and
> negative bias rails for LCD panels. It is not supported by Linux, so
> boards using it either model the rails as dummy fixed regulators and
> depend on the state left by the bootloader.

Please submit patches using subject lines reflecting the style for the
subsystem, this makes it easier for people to identify relevant patches.
Look at what existing commits in the area you're changing are doing and
make sure your subject lines visually resemble what they're doing.
There's no need to resubmit to fix this alone.

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