[PATCH v4 0/8] power: supply: Add several features support in qcom-battmgr driver

[PATCH v4 0/8] power: supply: Add several features support in qcom-battmgr driver

[Fenglin Wu via B4 Relay]

Add following features in qcom-battmgr drivers as the battery management
firmware has provided such capabilities:
 - Add resistance power supply property in core driver and qcom-battmgr
   driver to get battery resistance
 - Add state_of_health power supply property in core driver and
   qcom-battmgr driver to get battery health percentage
 - Add charge control start/end threshold control by using
   charge_control_start_threshold and charge_control_end_threshold power
   supply properties

The changes have been tested on QRD8650 and X1E80100-CRD devices based on
qcom/linux.git for-next commit a679f3f6931cdb0c2ef5dc0c26f895ae3f6c1ddc.

Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
---
Changes in v4:
- Address review comments from Maud and Konrad in [PATCH v3 8/8]:
  - Fix the address typo in nvmem cell node name "charge-limit-delta@75"
  - Add a blank line after the nvmem cell DT assignments
- Retrieve code-review trailers
- Link to v3: https://lore.kernel.org/r/20250826-qcom_battmgr_update-v3-0-74ea410ef146@oss.qualcomm.com

Changes in v3:
- Change power supply property "RESISTANCE" to "INTERNAL_RESISTANCE" and
  update the ABI document accordingly.
- Update the ABI document for "STATE_OF_HEALTH" property to explain it
  better.
- Drop following patches, as they were made due to my misunderstanding of
  the compats fallback behavior. Thank you to Krzysztof for the correction
  with patience.
    [PATCH v2 6/8] dt-bindings: soc: qcom: pmic-glink: Move X1E80100 out of fallbacks
    [PATCH v2 7/8] usb: typec: ucsi_glink: Add UCSI quirk for X1E80100 platform
    [PATCH v2 8/8] arm64: dts: qcom: x1*: Remove qcom,sm8550-pmic-glink fallback
- Addressed several comments in [PATCH v2 5/8]:
   - Separated the compat string addition change
   - Fixed the coding style issues in several places to address this
     checkpatch error:
       "CHECK: Alignment should match open parenthesis"
   - Add logic to read charge control thresholds from SDAM registers in driver
     probe. It addresses the issue on X1E80100, where there is no interface
     to retrieve the thresholds from the battery management firmware after
     boot-up.
- Add a DT binding change for charge_limit_xx "nvmem" DT properties.
- Add a DT change to specifiy charge_limit_xx "nvmem" properties for X1E80100
  devices.
- Link to v2: https://lore.kernel.org/r/20250530-qcom_battmgr_update-v2-0-9e377193a656@oss.qualcomm.com

Changes in v2:
- Corrected "qcom-battmgr" to "qcom_battmgr" in the commit subject of
  patch 4/5.
- Added charge control support for X1E80100 platform in patch 5.
- X1E80100 is no longer a fallback of SM8550 in pmic-glink battmgr support,
  hence added patch 6 in the pmic-glink binding to move X1E80100 out of the
  fallbacks.
- Added patch 7 in glink-ucsi driver to include UCSI quirk for X1E80100
  platform
- Added patch 8 to remove "qcom,sm8550-pmic-glink" compatible string in
  x1* board files.
- Rebased the changes on qcom/linux.git for-next commit 44ef9ab4baaf496d227ab98d368016700f0b9300.
- Link to v1: https://lore.kernel.org/r/20250523-qcom_battmgr_update-v1-0-2bb6d4e0a56e@oss.qualcomm.com

---
Fenglin Wu (8):
      power: supply: core: Add resistance power supply property
      power: supply: core: Add state_of_health power supply property
      power: supply: qcom_battmgr: Add resistance power supply property
      power: supply: qcom_battmgr: Add state_of_health property
      power: supply: qcom_battmgr: update compats for SM8550 and X1E80100
      dt-bindings: soc: qcom,pmic-glink: Add charge limit nvmem properties
      power: supply: qcom_battmgr: Add charge control support
      arm64: dts: qcom: x1e80100-crd: Add charge limit nvmem

 Documentation/ABI/testing/sysfs-class-power        |  31 ++
 .../bindings/soc/qcom/qcom,pmic-glink.yaml         |  14 +
 arch/arm64/boot/dts/qcom/x1-crd.dtsi               |   3 +
 arch/arm64/boot/dts/qcom/x1e80100-pmics.dtsi       |  20 ++
 drivers/power/supply/power_supply_sysfs.c          |   2 +
 drivers/power/supply/qcom_battmgr.c                | 316 ++++++++++++++++++++-
 include/linux/power_supply.h                       |   2 +
 7 files changed, 379 insertions(+), 9 deletions(-)
---
base-commit: abbf1025002e4966bfcbf8a069234e485d49edf1
change-id: 20250520-qcom_battmgr_update-3561dc526c05

Best regards,
-- 
Fenglin Wu <fenglin.wu@oss.qualcomm.com>

[PATCH v4 1/8] power: supply: core: Add resistance power supply property

[Fenglin Wu via B4 Relay]

From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>

Some battery drivers provide the ability to export internal resistance
as a parameter. Add internal_resistance power supply property for that
purpose.

Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
---
 Documentation/ABI/testing/sysfs-class-power | 16 ++++++++++++++++
 drivers/power/supply/power_supply_sysfs.c   |  1 +
 include/linux/power_supply.h                |  1 +
 3 files changed, 18 insertions(+)

diff --git a/Documentation/ABI/testing/sysfs-class-power b/Documentation/ABI/testing/sysfs-class-power
index 560124cc31770cde03bcdbbba0d85a5bd78b15a0..cea1a38f5a8fb754d4e6323967ef6cf2e20a68ce 100644
--- a/Documentation/ABI/testing/sysfs-class-power
+++ b/Documentation/ABI/testing/sysfs-class-power
@@ -552,6 +552,22 @@ Description:
 			Integer > 0: representing full cycles
 			Integer = 0: cycle_count info is not available
 
+What:		/sys/class/power_supply/<supply_name>/internal_resistance
+Date:		August 2025
+Contact:	linux-arm-msm@vger.kernel.org
+Description:
+		Represent the battery's internal resistance, often referred
+		to as Equivalent Series Resistance (ESR). It is a dynamic
+		parameter that reflects the opposition to current flow within
+		the cell. It is not a fixed value but varies significantly
+		based on several operational conditions, including battery
+		state of charge (SoC), temperature, and whether the battery
+		is in a charging or discharging state.
+
+		Access: Read
+
+		Valid values: Represented in microohms
+
 **USB Properties**
 
 What:		/sys/class/power_supply/<supply_name>/input_current_limit
diff --git a/drivers/power/supply/power_supply_sysfs.c b/drivers/power/supply/power_supply_sysfs.c
index a438f7983d4f6a832e9d479184c7c35453e1757c..cfa8f90a88ebc8fc1c7447198f138e5d2e699e5a 100644
--- a/drivers/power/supply/power_supply_sysfs.c
+++ b/drivers/power/supply/power_supply_sysfs.c
@@ -220,6 +220,7 @@ static struct power_supply_attr power_supply_attrs[] __ro_after_init = {
 	POWER_SUPPLY_ATTR(MANUFACTURE_YEAR),
 	POWER_SUPPLY_ATTR(MANUFACTURE_MONTH),
 	POWER_SUPPLY_ATTR(MANUFACTURE_DAY),
+	POWER_SUPPLY_ATTR(INTERNAL_RESISTANCE),
 	/* Properties of type `const char *' */
 	POWER_SUPPLY_ATTR(MODEL_NAME),
 	POWER_SUPPLY_ATTR(MANUFACTURER),
diff --git a/include/linux/power_supply.h b/include/linux/power_supply.h
index c4cb854971f53a244ba7742a15ce7a5515da6199..8bc3b7a67eb5693a16db9b7d123e7881711c6bf4 100644
--- a/include/linux/power_supply.h
+++ b/include/linux/power_supply.h
@@ -174,6 +174,7 @@ enum power_supply_property {
 	POWER_SUPPLY_PROP_MANUFACTURE_YEAR,
 	POWER_SUPPLY_PROP_MANUFACTURE_MONTH,
 	POWER_SUPPLY_PROP_MANUFACTURE_DAY,
+	POWER_SUPPLY_PROP_INTERNAL_RESISTANCE,
 	/* Properties of type `const char *' */
 	POWER_SUPPLY_PROP_MODEL_NAME,
 	POWER_SUPPLY_PROP_MANUFACTURER,

-- 
2.34.1

[PATCH v4 2/8] power: supply: core: Add state_of_health power supply property

[Fenglin Wu via B4 Relay]

From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>

Add state_of_health power supply property to represent battery
health percentage.

Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
---
 Documentation/ABI/testing/sysfs-class-power | 15 +++++++++++++++
 drivers/power/supply/power_supply_sysfs.c   |  1 +
 include/linux/power_supply.h                |  1 +
 3 files changed, 17 insertions(+)

diff --git a/Documentation/ABI/testing/sysfs-class-power b/Documentation/ABI/testing/sysfs-class-power
index cea1a38f5a8fb754d4e6323967ef6cf2e20a68ce..04f82e3e33aad6e16dc4fbace066b5d26069bf44 100644
--- a/Documentation/ABI/testing/sysfs-class-power
+++ b/Documentation/ABI/testing/sysfs-class-power
@@ -568,6 +568,21 @@ Description:
 
 		Valid values: Represented in microohms
 
+What:		/sys/class/power_supply/<supply_name>/state_of_health
+Date:		August 2025
+Contact:	linux-arm-msm@vger.kernel.org
+Description:
+		The state_of_health parameter quantifies the overall condition
+		of a battery as a percentage, reflecting its ability to deliver
+		rated performance relative to its original specifications. It is
+		dynamically computed using a combination of learned capacity
+		and impedance-based degradation indicators, both of which evolve
+		over the battery's lifecycle.
+
+		Access: Read
+
+		Valid values: 0 - 100 (percent)
+
 **USB Properties**
 
 What:		/sys/class/power_supply/<supply_name>/input_current_limit
diff --git a/drivers/power/supply/power_supply_sysfs.c b/drivers/power/supply/power_supply_sysfs.c
index cfa8f90a88ebc8fc1c7447198f138e5d2e699e5a..d96a8578308e3af60cc1a352845662aa922c29b3 100644
--- a/drivers/power/supply/power_supply_sysfs.c
+++ b/drivers/power/supply/power_supply_sysfs.c
@@ -221,6 +221,7 @@ static struct power_supply_attr power_supply_attrs[] __ro_after_init = {
 	POWER_SUPPLY_ATTR(MANUFACTURE_MONTH),
 	POWER_SUPPLY_ATTR(MANUFACTURE_DAY),
 	POWER_SUPPLY_ATTR(INTERNAL_RESISTANCE),
+	POWER_SUPPLY_ATTR(STATE_OF_HEALTH),
 	/* Properties of type `const char *' */
 	POWER_SUPPLY_ATTR(MODEL_NAME),
 	POWER_SUPPLY_ATTR(MANUFACTURER),
diff --git a/include/linux/power_supply.h b/include/linux/power_supply.h
index 8bc3b7a67eb5693a16db9b7d123e7881711c6bf4..ccb43fe44381965069dc3bd9505d45050b9b1bd8 100644
--- a/include/linux/power_supply.h
+++ b/include/linux/power_supply.h
@@ -175,6 +175,7 @@ enum power_supply_property {
 	POWER_SUPPLY_PROP_MANUFACTURE_MONTH,
 	POWER_SUPPLY_PROP_MANUFACTURE_DAY,
 	POWER_SUPPLY_PROP_INTERNAL_RESISTANCE,
+	POWER_SUPPLY_PROP_STATE_OF_HEALTH,
 	/* Properties of type `const char *' */
 	POWER_SUPPLY_PROP_MODEL_NAME,
 	POWER_SUPPLY_PROP_MANUFACTURER,

-- 
2.34.1

[PATCH v4 3/8] power: supply: qcom_battmgr: Add resistance power supply property

[Fenglin Wu via B4 Relay]

From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>

Add power supply property to get battery internal resistance from
the battery management firmware.

Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
---
 drivers/power/supply/qcom_battmgr.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/power/supply/qcom_battmgr.c b/drivers/power/supply/qcom_battmgr.c
index fe27676fbc7cd12292caa6fb3b5b46a18c426e6d..55477ae92fd56ede465b32d6f7ed9da78ebd869c 100644
--- a/drivers/power/supply/qcom_battmgr.c
+++ b/drivers/power/supply/qcom_battmgr.c
@@ -2,6 +2,7 @@
 /*
  * Copyright (c) 2019-2020, The Linux Foundation. All rights reserved.
  * Copyright (c) 2022, Linaro Ltd
+ * Copyright (c) 2025 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 #include <linux/auxiliary_bus.h>
 #include <linux/module.h>
@@ -254,6 +255,7 @@ struct qcom_battmgr_status {
 	unsigned int voltage_now;
 	unsigned int voltage_ocv;
 	unsigned int temperature;
+	unsigned int resistance;
 
 	unsigned int discharge_time;
 	unsigned int charge_time;
@@ -418,6 +420,7 @@ static const u8 sm8350_bat_prop_map[] = {
 	[POWER_SUPPLY_PROP_MODEL_NAME] = BATT_MODEL_NAME,
 	[POWER_SUPPLY_PROP_TIME_TO_FULL_AVG] = BATT_TTF_AVG,
 	[POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG] = BATT_TTE_AVG,
+	[POWER_SUPPLY_PROP_INTERNAL_RESISTANCE] = BATT_RESISTANCE,
 	[POWER_SUPPLY_PROP_POWER_NOW] = BATT_POWER_NOW,
 };
 
@@ -582,6 +585,9 @@ static int qcom_battmgr_bat_get_property(struct power_supply *psy,
 	case POWER_SUPPLY_PROP_TEMP:
 		val->intval = battmgr->status.temperature;
 		break;
+	case POWER_SUPPLY_PROP_INTERNAL_RESISTANCE:
+		val->intval = battmgr->status.resistance;
+		break;
 	case POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG:
 		val->intval = battmgr->status.discharge_time;
 		break;
@@ -665,6 +671,7 @@ static const enum power_supply_property sm8350_bat_props[] = {
 	POWER_SUPPLY_PROP_MODEL_NAME,
 	POWER_SUPPLY_PROP_TIME_TO_FULL_AVG,
 	POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG,
+	POWER_SUPPLY_PROP_INTERNAL_RESISTANCE,
 	POWER_SUPPLY_PROP_POWER_NOW,
 };
 
@@ -1174,6 +1181,9 @@ static void qcom_battmgr_sm8350_callback(struct qcom_battmgr *battmgr,
 		case BATT_TTE_AVG:
 			battmgr->status.discharge_time = le32_to_cpu(resp->intval.value);
 			break;
+		case BATT_RESISTANCE:
+			battmgr->status.resistance = le32_to_cpu(resp->intval.value);
+			break;
 		case BATT_POWER_NOW:
 			battmgr->status.power_now = le32_to_cpu(resp->intval.value);
 			break;

-- 
2.34.1

[PATCH v4 4/8] power: supply: qcom_battmgr: Add state_of_health property

[Fenglin Wu via B4 Relay]

From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>

Add state_of_health property to read battery health percentage from
battery management firmware.

Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
---
 drivers/power/supply/qcom_battmgr.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/power/supply/qcom_battmgr.c b/drivers/power/supply/qcom_battmgr.c
index 55477ae92fd56ede465b32d6f7ed9da78ebd869c..008e241e3eac3574a78459a2256e006e48c9f508 100644
--- a/drivers/power/supply/qcom_battmgr.c
+++ b/drivers/power/supply/qcom_battmgr.c
@@ -256,6 +256,7 @@ struct qcom_battmgr_status {
 	unsigned int voltage_ocv;
 	unsigned int temperature;
 	unsigned int resistance;
+	unsigned int soh_percent;
 
 	unsigned int discharge_time;
 	unsigned int charge_time;
@@ -421,6 +422,7 @@ static const u8 sm8350_bat_prop_map[] = {
 	[POWER_SUPPLY_PROP_TIME_TO_FULL_AVG] = BATT_TTF_AVG,
 	[POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG] = BATT_TTE_AVG,
 	[POWER_SUPPLY_PROP_INTERNAL_RESISTANCE] = BATT_RESISTANCE,
+	[POWER_SUPPLY_PROP_STATE_OF_HEALTH] = BATT_SOH,
 	[POWER_SUPPLY_PROP_POWER_NOW] = BATT_POWER_NOW,
 };
 
@@ -588,6 +590,9 @@ static int qcom_battmgr_bat_get_property(struct power_supply *psy,
 	case POWER_SUPPLY_PROP_INTERNAL_RESISTANCE:
 		val->intval = battmgr->status.resistance;
 		break;
+	case POWER_SUPPLY_PROP_STATE_OF_HEALTH:
+		val->intval = battmgr->status.soh_percent;
+		break;
 	case POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG:
 		val->intval = battmgr->status.discharge_time;
 		break;
@@ -672,6 +677,7 @@ static const enum power_supply_property sm8350_bat_props[] = {
 	POWER_SUPPLY_PROP_TIME_TO_FULL_AVG,
 	POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG,
 	POWER_SUPPLY_PROP_INTERNAL_RESISTANCE,
+	POWER_SUPPLY_PROP_STATE_OF_HEALTH,
 	POWER_SUPPLY_PROP_POWER_NOW,
 };
 
@@ -1141,6 +1147,9 @@ static void qcom_battmgr_sm8350_callback(struct qcom_battmgr *battmgr,
 		case BATT_CAPACITY:
 			battmgr->status.percent = le32_to_cpu(resp->intval.value) / 100;
 			break;
+		case BATT_SOH:
+			battmgr->status.soh_percent = le32_to_cpu(resp->intval.value);
+			break;
 		case BATT_VOLT_OCV:
 			battmgr->status.voltage_ocv = le32_to_cpu(resp->intval.value);
 			break;

-- 
2.34.1

[PATCH v4 5/8] power: supply: qcom_battmgr: update compats for SM8550 and X1E80100

[Fenglin Wu via B4 Relay]

From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>

Add variant definitions for SM8550 and X1E80100 platforms. Add a compat
for SM8550 and update match data for X1E80100 specifically so that they
could be handled differently in supporting charge control functionality.

Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
---
 drivers/power/supply/qcom_battmgr.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/power/supply/qcom_battmgr.c b/drivers/power/supply/qcom_battmgr.c
index 008e241e3eac3574a78459a2256e006e48c9f508..174d3f83ac2b070bb90c21a498686e91cc629ebe 100644
--- a/drivers/power/supply/qcom_battmgr.c
+++ b/drivers/power/supply/qcom_battmgr.c
@@ -19,8 +19,10 @@
 #define BATTMGR_STRING_LEN	128
 
 enum qcom_battmgr_variant {
-	QCOM_BATTMGR_SM8350,
 	QCOM_BATTMGR_SC8280XP,
+	QCOM_BATTMGR_SM8350,
+	QCOM_BATTMGR_SM8550,
+	QCOM_BATTMGR_X1E80100,
 };
 
 #define BATTMGR_BAT_STATUS		0x1
@@ -1333,7 +1335,8 @@ static void qcom_battmgr_pdr_notify(void *priv, int state)
 static const struct of_device_id qcom_battmgr_of_variants[] = {
 	{ .compatible = "qcom,sc8180x-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
 	{ .compatible = "qcom,sc8280xp-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
-	{ .compatible = "qcom,x1e80100-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
+	{ .compatible = "qcom,sm8550-pmic-glink", .data = (void *)QCOM_BATTMGR_SM8550 },
+	{ .compatible = "qcom,x1e80100-pmic-glink", .data = (void *)QCOM_BATTMGR_X1E80100 },
 	/* Unmatched devices falls back to QCOM_BATTMGR_SM8350 */
 	{}
 };

-- 
2.34.1

[PATCH v4 6/8] dt-bindings: soc: qcom,pmic-glink: Add charge limit nvmem properties

[Fenglin Wu via B4 Relay]

From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>

Add nvmem properties to retrieve charge control configurations
from the PMIC SDAM registers.

Acked-by: Rob Herring (Arm) <robh@kernel.org>
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
---
 .../devicetree/bindings/soc/qcom/qcom,pmic-glink.yaml      | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/Documentation/devicetree/bindings/soc/qcom/qcom,pmic-glink.yaml b/Documentation/devicetree/bindings/soc/qcom/qcom,pmic-glink.yaml
index 4c9e78f29523e3d77aacb4299f64ab96f9b1a831..9d6db4825da389aa14d77f653d2f8a3442e22162 100644
--- a/Documentation/devicetree/bindings/soc/qcom/qcom,pmic-glink.yaml
+++ b/Documentation/devicetree/bindings/soc/qcom/qcom,pmic-glink.yaml
@@ -55,6 +55,20 @@ properties:
       The array should contain a gpio entry for each PMIC Glink connector, in reg order.
       It is defined that GPIO active level means "CC2" or Reversed/Flipped orientation.
 
+  nvmem-cells:
+    minItems: 3
+    maxItems: 3
+    description:
+      The nvmem cells contain the charge control settings, including the charge control
+      enable status, the battery state of charge (SoC) threshold for stopping charging,
+      and the battery SoC delta required to restart charging.
+
+  nvmem-cell-names:
+    items:
+      - const: charge_limit_en
+      - const: charge_limit_end
+      - const: charge_limit_delta
+
 patternProperties:
   '^connector@\d$':
     $ref: /schemas/connector/usb-connector.yaml#

-- 
2.34.1

[PATCH v4 7/8] power: supply: qcom_battmgr: Add charge control support

[Fenglin Wu via B4 Relay]

From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>

Add charge control support for SM8550 and X1E80100. It's supported
with below two power supply properties:

charge_control_end_threshold: The battery SoC (State of Charge)
threshold at which the charging should be terminated.

charge_control_start_threshold: The battery SoC threshold at
which the charging should be resumed.

Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
---
 drivers/power/supply/qcom_battmgr.c | 290 +++++++++++++++++++++++++++++++++++-
 1 file changed, 283 insertions(+), 7 deletions(-)

diff --git a/drivers/power/supply/qcom_battmgr.c b/drivers/power/supply/qcom_battmgr.c
index 174d3f83ac2b070bb90c21a498686e91cc629ebe..23c68eb942f1b213f634e31a45420b3e113e2764 100644
--- a/drivers/power/supply/qcom_battmgr.c
+++ b/drivers/power/supply/qcom_battmgr.c
@@ -7,6 +7,7 @@
 #include <linux/auxiliary_bus.h>
 #include <linux/module.h>
 #include <linux/mutex.h>
+#include <linux/nvmem-consumer.h>
 #include <linux/of_device.h>
 #include <linux/power_supply.h>
 #include <linux/property.h>
@@ -68,6 +69,9 @@ enum qcom_battmgr_variant {
 #define BATT_RESISTANCE			21
 #define BATT_POWER_NOW			22
 #define BATT_POWER_AVG			23
+#define BATT_CHG_CTRL_EN		24
+#define BATT_CHG_CTRL_START_THR		25
+#define BATT_CHG_CTRL_END_THR		26
 
 #define BATTMGR_USB_PROPERTY_GET	0x32
 #define BATTMGR_USB_PROPERTY_SET	0x33
@@ -92,6 +96,13 @@ enum qcom_battmgr_variant {
 #define WLS_TYPE			5
 #define WLS_BOOST_EN			6
 
+#define BATTMGR_CHG_CTRL_LIMIT_EN	0x48
+#define CHARGE_CTRL_START_THR_MIN	50
+#define CHARGE_CTRL_START_THR_MAX	95
+#define CHARGE_CTRL_END_THR_MIN		55
+#define CHARGE_CTRL_END_THR_MAX		100
+#define CHARGE_CTRL_DELTA_SOC		5
+
 struct qcom_battmgr_enable_request {
 	struct pmic_glink_hdr hdr;
 	__le32 battery_id;
@@ -126,6 +137,13 @@ struct qcom_battmgr_discharge_time_request {
 	__le32 reserved;
 };
 
+struct qcom_battmgr_charge_ctrl_request {
+	struct pmic_glink_hdr hdr;
+	__le32 enable;
+	__le32 target_soc;
+	__le32 delta_soc;
+};
+
 struct qcom_battmgr_message {
 	struct pmic_glink_hdr hdr;
 	union {
@@ -238,6 +256,8 @@ struct qcom_battmgr_info {
 	unsigned int capacity_warning;
 	unsigned int cycle_count;
 	unsigned int charge_count;
+	unsigned int charge_ctrl_start;
+	unsigned int charge_ctrl_end;
 	char model_number[BATTMGR_STRING_LEN];
 	char serial_number[BATTMGR_STRING_LEN];
 	char oem_info[BATTMGR_STRING_LEN];
@@ -426,6 +446,8 @@ static const u8 sm8350_bat_prop_map[] = {
 	[POWER_SUPPLY_PROP_INTERNAL_RESISTANCE] = BATT_RESISTANCE,
 	[POWER_SUPPLY_PROP_STATE_OF_HEALTH] = BATT_SOH,
 	[POWER_SUPPLY_PROP_POWER_NOW] = BATT_POWER_NOW,
+	[POWER_SUPPLY_PROP_CHARGE_CONTROL_START_THRESHOLD] = BATT_CHG_CTRL_START_THR,
+	[POWER_SUPPLY_PROP_CHARGE_CONTROL_END_THRESHOLD] = BATT_CHG_CTRL_END_THR,
 };
 
 static int qcom_battmgr_bat_sm8350_update(struct qcom_battmgr *battmgr,
@@ -496,7 +518,8 @@ static int qcom_battmgr_bat_get_property(struct power_supply *psy,
 	if (!battmgr->service_up)
 		return -EAGAIN;
 
-	if (battmgr->variant == QCOM_BATTMGR_SC8280XP)
+	if (battmgr->variant == QCOM_BATTMGR_SC8280XP ||
+	    battmgr->variant == QCOM_BATTMGR_X1E80100)
 		ret = qcom_battmgr_bat_sc8280xp_update(battmgr, psp);
 	else
 		ret = qcom_battmgr_bat_sm8350_update(battmgr, psp);
@@ -601,6 +624,12 @@ static int qcom_battmgr_bat_get_property(struct power_supply *psy,
 	case POWER_SUPPLY_PROP_TIME_TO_FULL_AVG:
 		val->intval = battmgr->status.charge_time;
 		break;
+	case POWER_SUPPLY_PROP_CHARGE_CONTROL_START_THRESHOLD:
+		val->intval = battmgr->info.charge_ctrl_start;
+		break;
+	case POWER_SUPPLY_PROP_CHARGE_CONTROL_END_THRESHOLD:
+		val->intval = battmgr->info.charge_ctrl_end;
+		break;
 	case POWER_SUPPLY_PROP_MANUFACTURE_YEAR:
 		val->intval = battmgr->info.year;
 		break;
@@ -626,6 +655,149 @@ static int qcom_battmgr_bat_get_property(struct power_supply *psy,
 	return 0;
 }
 
+static int qcom_battmgr_set_charge_control(struct qcom_battmgr *battmgr,
+					   u32 target_soc, u32 delta_soc)
+{
+	struct qcom_battmgr_charge_ctrl_request request = {
+		.hdr.owner = cpu_to_le32(PMIC_GLINK_OWNER_BATTMGR),
+		.hdr.type = cpu_to_le32(PMIC_GLINK_REQ_RESP),
+		.hdr.opcode = cpu_to_le32(BATTMGR_CHG_CTRL_LIMIT_EN),
+		.enable = cpu_to_le32(1),
+		.target_soc = cpu_to_le32(target_soc),
+		.delta_soc = cpu_to_le32(delta_soc),
+	};
+
+	return qcom_battmgr_request(battmgr, &request, sizeof(request));
+}
+
+static int qcom_battmgr_set_charge_start_threshold(struct qcom_battmgr *battmgr, int start_soc)
+{
+	u32 target_soc, delta_soc;
+	int ret;
+
+	if (start_soc < CHARGE_CTRL_START_THR_MIN ||
+	    start_soc > CHARGE_CTRL_START_THR_MAX) {
+		dev_err(battmgr->dev, "charge control start threshold exceed range: [%u - %u]\n",
+			CHARGE_CTRL_START_THR_MIN, CHARGE_CTRL_START_THR_MAX);
+		return -EINVAL;
+	}
+
+	/*
+	 * If the new start threshold is larger than the old end threshold,
+	 * move the end threshold one step (DELTA_SOC) after the new start
+	 * threshold.
+	 */
+	if (start_soc > battmgr->info.charge_ctrl_end) {
+		target_soc = start_soc + CHARGE_CTRL_DELTA_SOC;
+		target_soc = min_t(u32, target_soc, CHARGE_CTRL_END_THR_MAX);
+		delta_soc = target_soc - start_soc;
+		delta_soc = min_t(u32, delta_soc, CHARGE_CTRL_DELTA_SOC);
+	} else {
+		target_soc =  battmgr->info.charge_ctrl_end;
+		delta_soc = battmgr->info.charge_ctrl_end - start_soc;
+	}
+
+	mutex_lock(&battmgr->lock);
+	ret = qcom_battmgr_set_charge_control(battmgr, target_soc, delta_soc);
+	mutex_unlock(&battmgr->lock);
+	if (!ret) {
+		battmgr->info.charge_ctrl_start = start_soc;
+		battmgr->info.charge_ctrl_end = target_soc;
+	}
+
+	return 0;
+}
+
+static int qcom_battmgr_set_charge_end_threshold(struct qcom_battmgr *battmgr, int end_soc)
+{
+	u32 delta_soc = CHARGE_CTRL_DELTA_SOC;
+	int ret;
+
+	if (end_soc < CHARGE_CTRL_END_THR_MIN ||
+	    end_soc > CHARGE_CTRL_END_THR_MAX) {
+		dev_err(battmgr->dev, "charge control end threshold exceed range: [%u - %u]\n",
+			CHARGE_CTRL_END_THR_MIN, CHARGE_CTRL_END_THR_MAX);
+		return -EINVAL;
+	}
+
+	if (battmgr->info.charge_ctrl_start && end_soc > battmgr->info.charge_ctrl_start)
+		delta_soc = end_soc - battmgr->info.charge_ctrl_start;
+
+	mutex_lock(&battmgr->lock);
+	ret = qcom_battmgr_set_charge_control(battmgr, end_soc, delta_soc);
+	mutex_unlock(&battmgr->lock);
+	if (!ret) {
+		battmgr->info.charge_ctrl_start = end_soc - delta_soc;
+		battmgr->info.charge_ctrl_end = end_soc;
+	}
+
+	return 0;
+}
+
+static int qcom_battmgr_charge_control_thresholds_init(struct qcom_battmgr *battmgr)
+{
+	int ret;
+	u8 en, end_soc, start_soc, delta_soc;
+
+	ret = nvmem_cell_read_u8(battmgr->dev->parent, "charge_limit_en", &en);
+	if (!ret && en != 0) {
+		ret = nvmem_cell_read_u8(battmgr->dev->parent, "charge_limit_end", &end_soc);
+		if (ret < 0)
+			return ret;
+
+		ret = nvmem_cell_read_u8(battmgr->dev->parent, "charge_limit_delta", &delta_soc);
+		if (ret < 0)
+			return ret;
+
+		if (delta_soc >= end_soc)
+			return -EINVAL;
+
+		start_soc = end_soc - delta_soc;
+		end_soc = clamp(end_soc, CHARGE_CTRL_END_THR_MIN, CHARGE_CTRL_END_THR_MAX);
+		start_soc = clamp(start_soc, CHARGE_CTRL_START_THR_MIN, CHARGE_CTRL_START_THR_MAX);
+
+		battmgr->info.charge_ctrl_start = start_soc;
+		battmgr->info.charge_ctrl_end = end_soc;
+	}
+
+	return 0;
+}
+
+static int qcom_battmgr_bat_is_writeable(struct power_supply *psy,
+					 enum power_supply_property psp)
+{
+	switch (psp) {
+	case POWER_SUPPLY_PROP_CHARGE_CONTROL_START_THRESHOLD:
+	case POWER_SUPPLY_PROP_CHARGE_CONTROL_END_THRESHOLD:
+		return 1;
+	default:
+		return 0;
+	}
+
+	return 0;
+}
+
+static int qcom_battmgr_bat_set_property(struct power_supply *psy,
+					 enum power_supply_property psp,
+					 const union power_supply_propval *pval)
+{
+	struct qcom_battmgr *battmgr = power_supply_get_drvdata(psy);
+
+	if (!battmgr->service_up)
+		return -EAGAIN;
+
+	switch (psp) {
+	case POWER_SUPPLY_PROP_CHARGE_CONTROL_START_THRESHOLD:
+		return qcom_battmgr_set_charge_start_threshold(battmgr, pval->intval);
+	case POWER_SUPPLY_PROP_CHARGE_CONTROL_END_THRESHOLD:
+		return qcom_battmgr_set_charge_end_threshold(battmgr, pval->intval);
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static const enum power_supply_property sc8280xp_bat_props[] = {
 	POWER_SUPPLY_PROP_STATUS,
 	POWER_SUPPLY_PROP_PRESENT,
@@ -659,6 +831,43 @@ static const struct power_supply_desc sc8280xp_bat_psy_desc = {
 	.get_property = qcom_battmgr_bat_get_property,
 };
 
+static const enum power_supply_property x1e80100_bat_props[] = {
+	POWER_SUPPLY_PROP_STATUS,
+	POWER_SUPPLY_PROP_PRESENT,
+	POWER_SUPPLY_PROP_TECHNOLOGY,
+	POWER_SUPPLY_PROP_CYCLE_COUNT,
+	POWER_SUPPLY_PROP_VOLTAGE_MAX_DESIGN,
+	POWER_SUPPLY_PROP_VOLTAGE_NOW,
+	POWER_SUPPLY_PROP_POWER_NOW,
+	POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN,
+	POWER_SUPPLY_PROP_CHARGE_FULL,
+	POWER_SUPPLY_PROP_CHARGE_EMPTY,
+	POWER_SUPPLY_PROP_CHARGE_NOW,
+	POWER_SUPPLY_PROP_ENERGY_FULL_DESIGN,
+	POWER_SUPPLY_PROP_ENERGY_FULL,
+	POWER_SUPPLY_PROP_ENERGY_EMPTY,
+	POWER_SUPPLY_PROP_ENERGY_NOW,
+	POWER_SUPPLY_PROP_TEMP,
+	POWER_SUPPLY_PROP_MANUFACTURE_YEAR,
+	POWER_SUPPLY_PROP_MANUFACTURE_MONTH,
+	POWER_SUPPLY_PROP_MANUFACTURE_DAY,
+	POWER_SUPPLY_PROP_MODEL_NAME,
+	POWER_SUPPLY_PROP_MANUFACTURER,
+	POWER_SUPPLY_PROP_SERIAL_NUMBER,
+	POWER_SUPPLY_PROP_CHARGE_CONTROL_START_THRESHOLD,
+	POWER_SUPPLY_PROP_CHARGE_CONTROL_END_THRESHOLD,
+};
+
+static const struct power_supply_desc x1e80100_bat_psy_desc = {
+	.name = "qcom-battmgr-bat",
+	.type = POWER_SUPPLY_TYPE_BATTERY,
+	.properties = x1e80100_bat_props,
+	.num_properties = ARRAY_SIZE(x1e80100_bat_props),
+	.get_property = qcom_battmgr_bat_get_property,
+	.set_property = qcom_battmgr_bat_set_property,
+	.property_is_writeable = qcom_battmgr_bat_is_writeable,
+};
+
 static const enum power_supply_property sm8350_bat_props[] = {
 	POWER_SUPPLY_PROP_STATUS,
 	POWER_SUPPLY_PROP_HEALTH,
@@ -691,6 +900,42 @@ static const struct power_supply_desc sm8350_bat_psy_desc = {
 	.get_property = qcom_battmgr_bat_get_property,
 };
 
+static const enum power_supply_property sm8550_bat_props[] = {
+	POWER_SUPPLY_PROP_STATUS,
+	POWER_SUPPLY_PROP_HEALTH,
+	POWER_SUPPLY_PROP_PRESENT,
+	POWER_SUPPLY_PROP_CHARGE_TYPE,
+	POWER_SUPPLY_PROP_CAPACITY,
+	POWER_SUPPLY_PROP_VOLTAGE_OCV,
+	POWER_SUPPLY_PROP_VOLTAGE_NOW,
+	POWER_SUPPLY_PROP_VOLTAGE_MAX,
+	POWER_SUPPLY_PROP_CURRENT_NOW,
+	POWER_SUPPLY_PROP_TEMP,
+	POWER_SUPPLY_PROP_TECHNOLOGY,
+	POWER_SUPPLY_PROP_CHARGE_COUNTER,
+	POWER_SUPPLY_PROP_CYCLE_COUNT,
+	POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN,
+	POWER_SUPPLY_PROP_CHARGE_FULL,
+	POWER_SUPPLY_PROP_MODEL_NAME,
+	POWER_SUPPLY_PROP_TIME_TO_FULL_AVG,
+	POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG,
+	POWER_SUPPLY_PROP_INTERNAL_RESISTANCE,
+	POWER_SUPPLY_PROP_STATE_OF_HEALTH,
+	POWER_SUPPLY_PROP_POWER_NOW,
+	POWER_SUPPLY_PROP_CHARGE_CONTROL_START_THRESHOLD,
+	POWER_SUPPLY_PROP_CHARGE_CONTROL_END_THRESHOLD,
+};
+
+static const struct power_supply_desc sm8550_bat_psy_desc = {
+	.name = "qcom-battmgr-bat",
+	.type = POWER_SUPPLY_TYPE_BATTERY,
+	.properties = sm8550_bat_props,
+	.num_properties = ARRAY_SIZE(sm8550_bat_props),
+	.get_property = qcom_battmgr_bat_get_property,
+	.set_property = qcom_battmgr_bat_set_property,
+	.property_is_writeable = qcom_battmgr_bat_is_writeable,
+};
+
 static int qcom_battmgr_ac_get_property(struct power_supply *psy,
 					enum power_supply_property psp,
 					union power_supply_propval *val)
@@ -766,7 +1011,8 @@ static int qcom_battmgr_usb_get_property(struct power_supply *psy,
 	if (!battmgr->service_up)
 		return -EAGAIN;
 
-	if (battmgr->variant == QCOM_BATTMGR_SC8280XP)
+	if (battmgr->variant == QCOM_BATTMGR_SC8280XP ||
+	    battmgr->variant == QCOM_BATTMGR_X1E80100)
 		ret = qcom_battmgr_bat_sc8280xp_update(battmgr, psp);
 	else
 		ret = qcom_battmgr_usb_sm8350_update(battmgr, psp);
@@ -888,7 +1134,8 @@ static int qcom_battmgr_wls_get_property(struct power_supply *psy,
 	if (!battmgr->service_up)
 		return -EAGAIN;
 
-	if (battmgr->variant == QCOM_BATTMGR_SC8280XP)
+	if (battmgr->variant == QCOM_BATTMGR_SC8280XP ||
+	    battmgr->variant == QCOM_BATTMGR_X1E80100)
 		ret = qcom_battmgr_bat_sc8280xp_update(battmgr, psp);
 	else
 		ret = qcom_battmgr_wls_sm8350_update(battmgr, psp);
@@ -1085,6 +1332,9 @@ static void qcom_battmgr_sc8280xp_callback(struct qcom_battmgr *battmgr,
 	case BATTMGR_BAT_CHARGE_TIME:
 		battmgr->status.charge_time = le32_to_cpu(resp->time);
 		break;
+	case BATTMGR_CHG_CTRL_LIMIT_EN:
+		battmgr->error = 0;
+		break;
 	default:
 		dev_warn(battmgr->dev, "unknown message %#x\n", opcode);
 		break;
@@ -1198,6 +1448,12 @@ static void qcom_battmgr_sm8350_callback(struct qcom_battmgr *battmgr,
 		case BATT_POWER_NOW:
 			battmgr->status.power_now = le32_to_cpu(resp->intval.value);
 			break;
+		case BATT_CHG_CTRL_START_THR:
+			battmgr->info.charge_ctrl_start = le32_to_cpu(resp->intval.value);
+			break;
+		case BATT_CHG_CTRL_END_THR:
+			battmgr->info.charge_ctrl_end = le32_to_cpu(resp->intval.value);
+			break;
 		default:
 			dev_warn(battmgr->dev, "unknown property %#x\n", property);
 			break;
@@ -1280,6 +1536,7 @@ static void qcom_battmgr_sm8350_callback(struct qcom_battmgr *battmgr,
 		}
 		break;
 	case BATTMGR_REQUEST_NOTIFICATION:
+	case BATTMGR_CHG_CTRL_LIMIT_EN:
 		battmgr->error = 0;
 		break;
 	default:
@@ -1299,7 +1556,8 @@ static void qcom_battmgr_callback(const void *data, size_t len, void *priv)
 
 	if (opcode == BATTMGR_NOTIFICATION)
 		qcom_battmgr_notification(battmgr, data, len);
-	else if (battmgr->variant == QCOM_BATTMGR_SC8280XP)
+	else if (battmgr->variant == QCOM_BATTMGR_SC8280XP ||
+		 battmgr->variant == QCOM_BATTMGR_X1E80100)
 		qcom_battmgr_sc8280xp_callback(battmgr, data, len);
 	else
 		qcom_battmgr_sm8350_callback(battmgr, data, len);
@@ -1346,11 +1604,13 @@ static char *qcom_battmgr_battery[] = { "battery" };
 static int qcom_battmgr_probe(struct auxiliary_device *adev,
 			      const struct auxiliary_device_id *id)
 {
+	const struct power_supply_desc *psy_desc;
 	struct power_supply_config psy_cfg_supply = {};
 	struct power_supply_config psy_cfg = {};
 	const struct of_device_id *match;
 	struct qcom_battmgr *battmgr;
 	struct device *dev = &adev->dev;
+	int ret;
 
 	battmgr = devm_kzalloc(dev, sizeof(*battmgr), GFP_KERNEL);
 	if (!battmgr)
@@ -1376,8 +1636,19 @@ static int qcom_battmgr_probe(struct auxiliary_device *adev,
 	else
 		battmgr->variant = QCOM_BATTMGR_SM8350;
 
-	if (battmgr->variant == QCOM_BATTMGR_SC8280XP) {
-		battmgr->bat_psy = devm_power_supply_register(dev, &sc8280xp_bat_psy_desc, &psy_cfg);
+	ret = qcom_battmgr_charge_control_thresholds_init(battmgr);
+	if (ret < 0)
+		return dev_err_probe(dev, ret,
+				     "failed to init battery charge control thresholds\n");
+
+	if (battmgr->variant == QCOM_BATTMGR_SC8280XP ||
+	    battmgr->variant == QCOM_BATTMGR_X1E80100) {
+		if (battmgr->variant == QCOM_BATTMGR_X1E80100)
+			psy_desc = &x1e80100_bat_psy_desc;
+		else
+			psy_desc = &sc8280xp_bat_psy_desc;
+
+		battmgr->bat_psy = devm_power_supply_register(dev, psy_desc, &psy_cfg);
 		if (IS_ERR(battmgr->bat_psy))
 			return dev_err_probe(dev, PTR_ERR(battmgr->bat_psy),
 					     "failed to register battery power supply\n");
@@ -1397,7 +1668,12 @@ static int qcom_battmgr_probe(struct auxiliary_device *adev,
 			return dev_err_probe(dev, PTR_ERR(battmgr->wls_psy),
 					     "failed to register wireless charing power supply\n");
 	} else {
-		battmgr->bat_psy = devm_power_supply_register(dev, &sm8350_bat_psy_desc, &psy_cfg);
+		if (battmgr->variant == QCOM_BATTMGR_SM8550)
+			psy_desc = &sm8550_bat_psy_desc;
+		else
+			psy_desc = &sm8350_bat_psy_desc;
+
+		battmgr->bat_psy = devm_power_supply_register(dev, psy_desc, &psy_cfg);
 		if (IS_ERR(battmgr->bat_psy))
 			return dev_err_probe(dev, PTR_ERR(battmgr->bat_psy),
 					     "failed to register battery power supply\n");

-- 
2.34.1

[PATCH v4 8/8] arm64: dts: qcom: x1e80100-crd: Add charge limit nvmem

[Fenglin Wu via B4 Relay]

From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>

Add nvmem cells for getting charge control thresholds if they have
been set previously.

Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
---
 arch/arm64/boot/dts/qcom/x1-crd.dtsi         |  3 +++
 arch/arm64/boot/dts/qcom/x1e80100-pmics.dtsi | 20 ++++++++++++++++++++
 2 files changed, 23 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/x1-crd.dtsi b/arch/arm64/boot/dts/qcom/x1-crd.dtsi
index c9f0d505267081af66b0973fe6c1e33832a2c86b..fee65391653ae9c2ee23f9f3954d9ed018c9aecd 100644
--- a/arch/arm64/boot/dts/qcom/x1-crd.dtsi
+++ b/arch/arm64/boot/dts/qcom/x1-crd.dtsi
@@ -82,6 +82,9 @@ pmic-glink {
 				    <&tlmm 123 GPIO_ACTIVE_HIGH>,
 				    <&tlmm 125 GPIO_ACTIVE_HIGH>;
 
+		nvmem-cells = <&charge_limit_en>, <&charge_limit_end>, <&charge_limit_delta>;
+		nvmem-cell-names = "charge_limit_en", "charge_limit_end", "charge_limit_delta";
+
 		/* Left-side rear port */
 		connector@0 {
 			compatible = "usb-c-connector";
diff --git a/arch/arm64/boot/dts/qcom/x1e80100-pmics.dtsi b/arch/arm64/boot/dts/qcom/x1e80100-pmics.dtsi
index c02fd4d15c9649c222caaafa5ed2c777a10fb4f5..eb5562e4393c88faa16d9172ee2a1ceabef076ff 100644
--- a/arch/arm64/boot/dts/qcom/x1e80100-pmics.dtsi
+++ b/arch/arm64/boot/dts/qcom/x1e80100-pmics.dtsi
@@ -239,6 +239,26 @@ reboot_reason: reboot-reason@48 {
 			};
 		};
 
+		pmk8550_sdam_15: nvram@7e00 {
+			compatible = "qcom,spmi-sdam";
+			reg = <0x7e00>;
+			#address-cells = <1>;
+			#size-cells = <1>;
+			ranges = <0 0x7e00 0x100>;
+
+			charge_limit_en: charge-limit-en@73 {
+				reg = <0x73 0x1>;
+			};
+
+			charge_limit_end: charge-limit-end@75 {
+				reg = <0x75 0x1>;
+			};
+
+			charge_limit_delta: charge-limit-delta@76 {
+				reg = <0x76 0x1>;
+			};
+		};
+
 		pmk8550_gpios: gpio@8800 {
 			compatible = "qcom,pmk8550-gpio", "qcom,spmi-gpio";
 			reg = <0xb800>;

-- 
2.34.1

Re: [PATCH v4 5/8] power: supply: qcom_battmgr: update compats for SM8550 and X1E80100

[Stephan Gerhold]

On Mon, Sep 15, 2025 at 04:49:57PM +0800, Fenglin Wu via B4 Relay wrote:
> From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> 
> Add variant definitions for SM8550 and X1E80100 platforms. Add a compat
> for SM8550 and update match data for X1E80100 specifically so that they
> could be handled differently in supporting charge control functionality.
> 
> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
> Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> ---
>  drivers/power/supply/qcom_battmgr.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/power/supply/qcom_battmgr.c b/drivers/power/supply/qcom_battmgr.c
> index 008e241e3eac3574a78459a2256e006e48c9f508..174d3f83ac2b070bb90c21a498686e91cc629ebe 100644
> --- a/drivers/power/supply/qcom_battmgr.c
> +++ b/drivers/power/supply/qcom_battmgr.c
> @@ -19,8 +19,10 @@
>  #define BATTMGR_STRING_LEN	128
>  
>  enum qcom_battmgr_variant {
> -	QCOM_BATTMGR_SM8350,
>  	QCOM_BATTMGR_SC8280XP,
> +	QCOM_BATTMGR_SM8350,
> +	QCOM_BATTMGR_SM8550,
> +	QCOM_BATTMGR_X1E80100,
>  };
>  
>  #define BATTMGR_BAT_STATUS		0x1
> @@ -1333,7 +1335,8 @@ static void qcom_battmgr_pdr_notify(void *priv, int state)
>  static const struct of_device_id qcom_battmgr_of_variants[] = {
>  	{ .compatible = "qcom,sc8180x-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
>  	{ .compatible = "qcom,sc8280xp-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
> -	{ .compatible = "qcom,x1e80100-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
> +	{ .compatible = "qcom,sm8550-pmic-glink", .data = (void *)QCOM_BATTMGR_SM8550 },
> +	{ .compatible = "qcom,x1e80100-pmic-glink", .data = (void *)QCOM_BATTMGR_X1E80100 },
>  	/* Unmatched devices falls back to QCOM_BATTMGR_SM8350 */
>  	{}
>  };

I think you need to squash this with "[PATCH 7/8] power: supply:
qcom_battmgr: Add charge control support", or move the modified checks
for

	if (battmgr->variant == QCOM_BATTMGR_SC8280XP ||
	    battmgr->variant == QCOM_BATTMGR_X1E80100) {

into this patch.

With this patch right now, I would expect that your series is not
bisectable: The wrong code paths are chosen if you only apply this patch
because e.g. X1E doesn't use the QCOM_BATTMGR_SC8280XP code anymore.

Thanks,
Stephan

Re: [PATCH v4 5/8] power: supply: qcom_battmgr: update compats for SM8550 and X1E80100

[Fenglin Wu]

On 9/15/2025 5:02 PM, Stephan Gerhold wrote:
> On Mon, Sep 15, 2025 at 04:49:57PM +0800, Fenglin Wu via B4 Relay wrote:
>> From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
>>
>> Add variant definitions for SM8550 and X1E80100 platforms. Add a compat
>> for SM8550 and update match data for X1E80100 specifically so that they
>> could be handled differently in supporting charge control functionality.
>>
>> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
>> Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
>> ---
>>   drivers/power/supply/qcom_battmgr.c | 7 +++++--
>>   1 file changed, 5 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/power/supply/qcom_battmgr.c b/drivers/power/supply/qcom_battmgr.c
>> index 008e241e3eac3574a78459a2256e006e48c9f508..174d3f83ac2b070bb90c21a498686e91cc629ebe 100644
>> --- a/drivers/power/supply/qcom_battmgr.c
>> +++ b/drivers/power/supply/qcom_battmgr.c
>> @@ -19,8 +19,10 @@
>>   #define BATTMGR_STRING_LEN	128
>>   
>>   enum qcom_battmgr_variant {
>> -	QCOM_BATTMGR_SM8350,
>>   	QCOM_BATTMGR_SC8280XP,
>> +	QCOM_BATTMGR_SM8350,
>> +	QCOM_BATTMGR_SM8550,
>> +	QCOM_BATTMGR_X1E80100,
>>   };
>>   
>>   #define BATTMGR_BAT_STATUS		0x1
>> @@ -1333,7 +1335,8 @@ static void qcom_battmgr_pdr_notify(void *priv, int state)
>>   static const struct of_device_id qcom_battmgr_of_variants[] = {
>>   	{ .compatible = "qcom,sc8180x-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
>>   	{ .compatible = "qcom,sc8280xp-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
>> -	{ .compatible = "qcom,x1e80100-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
>> +	{ .compatible = "qcom,sm8550-pmic-glink", .data = (void *)QCOM_BATTMGR_SM8550 },
>> +	{ .compatible = "qcom,x1e80100-pmic-glink", .data = (void *)QCOM_BATTMGR_X1E80100 },
>>   	/* Unmatched devices falls back to QCOM_BATTMGR_SM8350 */
>>   	{}
>>   };
> I think you need to squash this with "[PATCH 7/8] power: supply:
> qcom_battmgr: Add charge control support", or move the modified checks
> for
>
> 	if (battmgr->variant == QCOM_BATTMGR_SC8280XP ||
> 	    battmgr->variant == QCOM_BATTMGR_X1E80100) {
>
> into this patch.
>
> With this patch right now, I would expect that your series is not
> bisectable: The wrong code paths are chosen if you only apply this patch
> because e.g. X1E doesn't use the QCOM_BATTMGR_SC8280XP code anymore.
>
> Thanks,
> Stephan

I see.

I was making it this way to address the review comment from Bryan in 
patch v2 about separating the compats change. See here: 
https://lore.kernel.org/all/7f001134-e099-492d-8ce5-4122d83a3de3@linaro.org/

If I revise it according to your 2nd suggestion, would it conflict with 
Bryan's feedback?

Re: [PATCH v4 5/8] power: supply: qcom_battmgr: update compats for SM8550 and X1E80100

[Stephan Gerhold]

On Mon, Sep 15, 2025 at 05:44:40PM +0800, Fenglin Wu wrote:
> 
> On 9/15/2025 5:02 PM, Stephan Gerhold wrote:
> > On Mon, Sep 15, 2025 at 04:49:57PM +0800, Fenglin Wu via B4 Relay wrote:
> > > From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> > > 
> > > Add variant definitions for SM8550 and X1E80100 platforms. Add a compat
> > > for SM8550 and update match data for X1E80100 specifically so that they
> > > could be handled differently in supporting charge control functionality.
> > > 
> > > Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
> > > Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> > > ---
> > >   drivers/power/supply/qcom_battmgr.c | 7 +++++--
> > >   1 file changed, 5 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/drivers/power/supply/qcom_battmgr.c b/drivers/power/supply/qcom_battmgr.c
> > > index 008e241e3eac3574a78459a2256e006e48c9f508..174d3f83ac2b070bb90c21a498686e91cc629ebe 100644
> > > --- a/drivers/power/supply/qcom_battmgr.c
> > > +++ b/drivers/power/supply/qcom_battmgr.c
> > > @@ -19,8 +19,10 @@
> > >   #define BATTMGR_STRING_LEN	128
> > >   enum qcom_battmgr_variant {
> > > -	QCOM_BATTMGR_SM8350,
> > >   	QCOM_BATTMGR_SC8280XP,
> > > +	QCOM_BATTMGR_SM8350,
> > > +	QCOM_BATTMGR_SM8550,
> > > +	QCOM_BATTMGR_X1E80100,
> > >   };
> > >   #define BATTMGR_BAT_STATUS		0x1
> > > @@ -1333,7 +1335,8 @@ static void qcom_battmgr_pdr_notify(void *priv, int state)
> > >   static const struct of_device_id qcom_battmgr_of_variants[] = {
> > >   	{ .compatible = "qcom,sc8180x-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
> > >   	{ .compatible = "qcom,sc8280xp-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
> > > -	{ .compatible = "qcom,x1e80100-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
> > > +	{ .compatible = "qcom,sm8550-pmic-glink", .data = (void *)QCOM_BATTMGR_SM8550 },
> > > +	{ .compatible = "qcom,x1e80100-pmic-glink", .data = (void *)QCOM_BATTMGR_X1E80100 },
> > >   	/* Unmatched devices falls back to QCOM_BATTMGR_SM8350 */
> > >   	{}
> > >   };
> > I think you need to squash this with "[PATCH 7/8] power: supply:
> > qcom_battmgr: Add charge control support", or move the modified checks
> > for
> > 
> > 	if (battmgr->variant == QCOM_BATTMGR_SC8280XP ||
> > 	    battmgr->variant == QCOM_BATTMGR_X1E80100) {
> > 
> > into this patch.
> > 
> > With this patch right now, I would expect that your series is not
> > bisectable: The wrong code paths are chosen if you only apply this patch
> > because e.g. X1E doesn't use the QCOM_BATTMGR_SC8280XP code anymore.
> > 
> > Thanks,
> > Stephan
> 
> I see.
> 
> I was making it this way to address the review comment from Bryan in patch
> v2 about separating the compats change. See here:
> https://lore.kernel.org/all/7f001134-e099-492d-8ce5-4122d83a3de3@linaro.org/
> 
> If I revise it according to your 2nd suggestion, would it conflict with
> Bryan's feedback?
> 

I would expect that Bryan had my second suggestion in mind - separating
the refactoring (without functional change) from the new feature
addition.

You need to add the new cases to the if statements in this patch, or you
will (temporarily) change and break functionality.

Thanks,
Stephan

Re: [PATCH v4 3/8] power: supply: qcom_battmgr: Add resistance power supply property

[Dmitry Baryshkov]

On Mon, Sep 15, 2025 at 04:49:55PM +0800, Fenglin Wu via B4 Relay wrote:
> From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> 
> Add power supply property to get battery internal resistance from
> the battery management firmware.
> 
> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED

T14S is X1E80100, which uses SC8280XP-specific sets of properties. This
patch changes only SM8350-related data. How was it tested?

> Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> ---
>  drivers/power/supply/qcom_battmgr.c | 10 ++++++++++
>  1 file changed, 10 insertions(+)
> 
> diff --git a/drivers/power/supply/qcom_battmgr.c b/drivers/power/supply/qcom_battmgr.c
> index fe27676fbc7cd12292caa6fb3b5b46a18c426e6d..55477ae92fd56ede465b32d6f7ed9da78ebd869c 100644
> --- a/drivers/power/supply/qcom_battmgr.c
> +++ b/drivers/power/supply/qcom_battmgr.c
> @@ -2,6 +2,7 @@
>  /*
>   * Copyright (c) 2019-2020, The Linux Foundation. All rights reserved.
>   * Copyright (c) 2022, Linaro Ltd
> + * Copyright (c) 2025 Qualcomm Innovation Center, Inc. All rights reserved.

Please follow marketing guidelines here.

>   */
>  #include <linux/auxiliary_bus.h>
>  #include <linux/module.h>
> @@ -254,6 +255,7 @@ struct qcom_battmgr_status {
>  	unsigned int voltage_now;
>  	unsigned int voltage_ocv;
>  	unsigned int temperature;
> +	unsigned int resistance;
>  
>  	unsigned int discharge_time;
>  	unsigned int charge_time;
> @@ -418,6 +420,7 @@ static const u8 sm8350_bat_prop_map[] = {
>  	[POWER_SUPPLY_PROP_MODEL_NAME] = BATT_MODEL_NAME,
>  	[POWER_SUPPLY_PROP_TIME_TO_FULL_AVG] = BATT_TTF_AVG,
>  	[POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG] = BATT_TTE_AVG,
> +	[POWER_SUPPLY_PROP_INTERNAL_RESISTANCE] = BATT_RESISTANCE,
>  	[POWER_SUPPLY_PROP_POWER_NOW] = BATT_POWER_NOW,
>  };
>  
> @@ -582,6 +585,9 @@ static int qcom_battmgr_bat_get_property(struct power_supply *psy,
>  	case POWER_SUPPLY_PROP_TEMP:
>  		val->intval = battmgr->status.temperature;
>  		break;
> +	case POWER_SUPPLY_PROP_INTERNAL_RESISTANCE:
> +		val->intval = battmgr->status.resistance;
> +		break;
>  	case POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG:
>  		val->intval = battmgr->status.discharge_time;
>  		break;
> @@ -665,6 +671,7 @@ static const enum power_supply_property sm8350_bat_props[] = {
>  	POWER_SUPPLY_PROP_MODEL_NAME,
>  	POWER_SUPPLY_PROP_TIME_TO_FULL_AVG,
>  	POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG,
> +	POWER_SUPPLY_PROP_INTERNAL_RESISTANCE,
>  	POWER_SUPPLY_PROP_POWER_NOW,
>  };
>  
> @@ -1174,6 +1181,9 @@ static void qcom_battmgr_sm8350_callback(struct qcom_battmgr *battmgr,
>  		case BATT_TTE_AVG:
>  			battmgr->status.discharge_time = le32_to_cpu(resp->intval.value);
>  			break;
> +		case BATT_RESISTANCE:
> +			battmgr->status.resistance = le32_to_cpu(resp->intval.value);
> +			break;
>  		case BATT_POWER_NOW:
>  			battmgr->status.power_now = le32_to_cpu(resp->intval.value);
>  			break;
> 
> -- 
> 2.34.1
> 
> 

-- 
With best wishes
Dmitry

Re: [PATCH v4 5/8] power: supply: qcom_battmgr: update compats for SM8550 and X1E80100

[Dmitry Baryshkov]

On Mon, Sep 15, 2025 at 04:49:57PM +0800, Fenglin Wu via B4 Relay wrote:
> From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> 
> Add variant definitions for SM8550 and X1E80100 platforms. Add a compat
> for SM8550 and update match data for X1E80100 specifically so that they
> could be handled differently in supporting charge control functionality.

Why?

> 
> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
> Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> ---
>  drivers/power/supply/qcom_battmgr.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/power/supply/qcom_battmgr.c b/drivers/power/supply/qcom_battmgr.c
> index 008e241e3eac3574a78459a2256e006e48c9f508..174d3f83ac2b070bb90c21a498686e91cc629ebe 100644
> --- a/drivers/power/supply/qcom_battmgr.c
> +++ b/drivers/power/supply/qcom_battmgr.c
> @@ -19,8 +19,10 @@
>  #define BATTMGR_STRING_LEN	128
>  
>  enum qcom_battmgr_variant {
> -	QCOM_BATTMGR_SM8350,
>  	QCOM_BATTMGR_SC8280XP,
> +	QCOM_BATTMGR_SM8350,
> +	QCOM_BATTMGR_SM8550,
> +	QCOM_BATTMGR_X1E80100,
>  };
>  
>  #define BATTMGR_BAT_STATUS		0x1
> @@ -1333,7 +1335,8 @@ static void qcom_battmgr_pdr_notify(void *priv, int state)
>  static const struct of_device_id qcom_battmgr_of_variants[] = {
>  	{ .compatible = "qcom,sc8180x-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
>  	{ .compatible = "qcom,sc8280xp-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
> -	{ .compatible = "qcom,x1e80100-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
> +	{ .compatible = "qcom,sm8550-pmic-glink", .data = (void *)QCOM_BATTMGR_SM8550 },
> +	{ .compatible = "qcom,x1e80100-pmic-glink", .data = (void *)QCOM_BATTMGR_X1E80100 },
>  	/* Unmatched devices falls back to QCOM_BATTMGR_SM8350 */
>  	{}
>  };
> 
> -- 
> 2.34.1
> 
> 

-- 
With best wishes
Dmitry

Re: [PATCH v4 8/8] arm64: dts: qcom: x1e80100-crd: Add charge limit nvmem

[Dmitry Baryshkov]

On Mon, Sep 15, 2025 at 04:50:00PM +0800, Fenglin Wu via B4 Relay wrote:
> From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> 
> Add nvmem cells for getting charge control thresholds if they have
> been set previously.
> 
> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
> Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> ---
>  arch/arm64/boot/dts/qcom/x1-crd.dtsi         |  3 +++
>  arch/arm64/boot/dts/qcom/x1e80100-pmics.dtsi | 20 ++++++++++++++++++++
>  2 files changed, 23 insertions(+)
> 
> diff --git a/arch/arm64/boot/dts/qcom/x1-crd.dtsi b/arch/arm64/boot/dts/qcom/x1-crd.dtsi
> index c9f0d505267081af66b0973fe6c1e33832a2c86b..fee65391653ae9c2ee23f9f3954d9ed018c9aecd 100644
> --- a/arch/arm64/boot/dts/qcom/x1-crd.dtsi
> +++ b/arch/arm64/boot/dts/qcom/x1-crd.dtsi
> @@ -82,6 +82,9 @@ pmic-glink {
>  				    <&tlmm 123 GPIO_ACTIVE_HIGH>,
>  				    <&tlmm 125 GPIO_ACTIVE_HIGH>;
>  
> +		nvmem-cells = <&charge_limit_en>, <&charge_limit_end>, <&charge_limit_delta>;

One item per line, please.

> +		nvmem-cell-names = "charge_limit_en", "charge_limit_end", "charge_limit_delta";

And here too.

> +
>  		/* Left-side rear port */
>  		connector@0 {
>  			compatible = "usb-c-connector";

-- 
With best wishes
Dmitry

Re: [PATCH v4 3/8] power: supply: qcom_battmgr: Add resistance power supply property

[Fenglin Wu]

On 9/15/2025 6:18 PM, Dmitry Baryshkov wrote:
> On Mon, Sep 15, 2025 at 04:49:55PM +0800, Fenglin Wu via B4 Relay wrote:
>> From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
>>
>> Add power supply property to get battery internal resistance from
>> the battery management firmware.
>>
>> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
> T14S is X1E80100, which uses SC8280XP-specific sets of properties. This
> patch changes only SM8350-related data. How was it tested?

I assumed that Neil has picked the series of the changes and tested the 
charge control limit functionality on his T14S device.

When I run "b4 trailers -u", the tag was added on all patches. I will 
remove the "Tested-by" trailer for the patches with functionality not 
applicable for X1E80100 platform.

Re: [PATCH v4 5/8] power: supply: qcom_battmgr: update compats for SM8550 and X1E80100

[Fenglin Wu]

On 9/15/2025 6:19 PM, Dmitry Baryshkov wrote:
> On Mon, Sep 15, 2025 at 04:49:57PM +0800, Fenglin Wu via B4 Relay wrote:
>> From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
>>
>> Add variant definitions for SM8550 and X1E80100 platforms. Add a compat
>> for SM8550 and update match data for X1E80100 specifically so that they
>> could be handled differently in supporting charge control functionality.
> Why?
Is the question about why this was submitted as a separate patch, or 
about the need for the change itself? The reason for the change is 
explained in the commit text. As for submitting it separately, that was 
done to address Bryan's comments to split out the compats changes. 
Anyway, I will address the further comments from Stephen to make the 
change bisectable.
>
>> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
>> Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
>> ---
>>   drivers/power/supply/qcom_battmgr.c | 7 +++++--
>>   1 file changed, 5 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/power/supply/qcom_battmgr.c b/drivers/power/supply/qcom_battmgr.c
>> index 008e241e3eac3574a78459a2256e006e48c9f508..174d3f83ac2b070bb90c21a498686e91cc629ebe 100644
>> --- a/drivers/power/supply/qcom_battmgr.c
>> +++ b/drivers/power/supply/qcom_battmgr.c
>> @@ -19,8 +19,10 @@
>>   #define BATTMGR_STRING_LEN	128
>>   
>>   enum qcom_battmgr_variant {
>> -	QCOM_BATTMGR_SM8350,
>>   	QCOM_BATTMGR_SC8280XP,
>> +	QCOM_BATTMGR_SM8350,
>> +	QCOM_BATTMGR_SM8550,
>> +	QCOM_BATTMGR_X1E80100,
>>   };
>>   
>>   #define BATTMGR_BAT_STATUS		0x1
>> @@ -1333,7 +1335,8 @@ static void qcom_battmgr_pdr_notify(void *priv, int state)
>>   static const struct of_device_id qcom_battmgr_of_variants[] = {
>>   	{ .compatible = "qcom,sc8180x-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
>>   	{ .compatible = "qcom,sc8280xp-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
>> -	{ .compatible = "qcom,x1e80100-pmic-glink", .data = (void *)QCOM_BATTMGR_SC8280XP },
>> +	{ .compatible = "qcom,sm8550-pmic-glink", .data = (void *)QCOM_BATTMGR_SM8550 },
>> +	{ .compatible = "qcom,x1e80100-pmic-glink", .data = (void *)QCOM_BATTMGR_X1E80100 },
>>   	/* Unmatched devices falls back to QCOM_BATTMGR_SM8350 */
>>   	{}
>>   };
>>
>> -- 
>> 2.34.1
>>
>>

Re: [PATCH v4 3/8] power: supply: qcom_battmgr: Add resistance power supply property

[Konrad Dybcio]

On 9/16/25 4:31 AM, Fenglin Wu wrote:
> 
> On 9/15/2025 6:18 PM, Dmitry Baryshkov wrote:
>> On Mon, Sep 15, 2025 at 04:49:55PM +0800, Fenglin Wu via B4 Relay wrote:
>>> From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
>>>
>>> Add power supply property to get battery internal resistance from
>>> the battery management firmware.
>>>
>>> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
>> T14S is X1E80100, which uses SC8280XP-specific sets of properties. This
>> patch changes only SM8350-related data. How was it tested?
> 
> I assumed that Neil has picked the series of the changes and tested the charge control limit functionality on his T14S device.
> 
> When I run "b4 trailers -u", the tag was added on all patches. I will remove the "Tested-by" trailer for the patches with functionality not applicable for X1E80100 platform.

+ Konstantin

It's quite common to see someone leaving a T-b on the cover letter,
trying to say "I gave this series a spin" and then seeing the tag
appear on unrelated commits within the series (e.g. bindings or some
cosmetic fixes". Maybe some sort of an interactive (opt-in is fine)
dialog for "which patches to apply t-b/tags to" could be worth the
effort?

I was imagining two options:

$ b4 trailers -u --lalala
> Grabbing tags..
> Found:
> [Patch 0/n] Very Nice Changeset
>   Tested-by: Foo Bar <foo@bar.com>
>
> Which patches do you want the Tested-by tags to apply to? [all]: 2-5

or:

$ b4 trailers -u --lalala2
> Grabbing tagsd..
> Found:
> [Patch 0/n] Very Nice Changeset
>   Apply to Patch 1 ("soc: qcom: Fix all bugs")? [Y/n/a] y
>   Apply to Patch 2 ("dt-bindings: foobarbaz")? [Y/n/a] n
>   Apply to Patch 3 ("clk: qcom: Fix ABCD")? [Y/n/a] a
>   Applying to Patch 4 ("clk: qcom: Fix DEFG")
>   . . .
>   Applying to Patch n ("clk: qcom: Fix XYZ")
> Tags applied!

As I'm writing this, I'm thinking option 2 offers much more
fine-grained control, which is always nice to see..

Konrad

Re: [PATCH v4 5/8] power: supply: qcom_battmgr: update compats for SM8550 and X1E80100

[Dmitry Baryshkov]

On Tue, Sep 16, 2025 at 10:40:03AM +0800, Fenglin Wu wrote:
> 
> On 9/15/2025 6:19 PM, Dmitry Baryshkov wrote:
> > On Mon, Sep 15, 2025 at 04:49:57PM +0800, Fenglin Wu via B4 Relay wrote:
> > > From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> > > 
> > > Add variant definitions for SM8550 and X1E80100 platforms. Add a compat
> > > for SM8550 and update match data for X1E80100 specifically so that they
> > > could be handled differently in supporting charge control functionality.
> > Why?
> Is the question about why this was submitted as a separate patch, or about
> the need for the change itself? The reason for the change is explained in
> the commit text.

It's not, and that was my question. Why do you need to handle them
differently?

Please always start your commit message with the description of the
issue that you are facing.


> As for submitting it separately, that was done to address
> Bryan's comments to split out the compats changes. Anyway, I will address
> the further comments from Stephen to make the change bisectable.
> > 
> > > Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
> > > Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> > > ---
> > >   drivers/power/supply/qcom_battmgr.c | 7 +++++--
> > >   1 file changed, 5 insertions(+), 2 deletions(-)
> > > 

-- 
With best wishes
Dmitry

Re: [PATCH v4 8/8] arm64: dts: qcom: x1e80100-crd: Add charge limit nvmem

[Konrad Dybcio]

On 9/15/25 12:20 PM, Dmitry Baryshkov wrote:
> On Mon, Sep 15, 2025 at 04:50:00PM +0800, Fenglin Wu via B4 Relay wrote:
>> From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
>>
>> Add nvmem cells for getting charge control thresholds if they have
>> been set previously.
>>
>> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
>> Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
>> ---
>>  arch/arm64/boot/dts/qcom/x1-crd.dtsi         |  3 +++
>>  arch/arm64/boot/dts/qcom/x1e80100-pmics.dtsi | 20 ++++++++++++++++++++
>>  2 files changed, 23 insertions(+)
>>
>> diff --git a/arch/arm64/boot/dts/qcom/x1-crd.dtsi b/arch/arm64/boot/dts/qcom/x1-crd.dtsi
>> index c9f0d505267081af66b0973fe6c1e33832a2c86b..fee65391653ae9c2ee23f9f3954d9ed018c9aecd 100644
>> --- a/arch/arm64/boot/dts/qcom/x1-crd.dtsi
>> +++ b/arch/arm64/boot/dts/qcom/x1-crd.dtsi
>> @@ -82,6 +82,9 @@ pmic-glink {
>>  				    <&tlmm 123 GPIO_ACTIVE_HIGH>,
>>  				    <&tlmm 125 GPIO_ACTIVE_HIGH>;
>>  
>> +		nvmem-cells = <&charge_limit_en>, <&charge_limit_end>, <&charge_limit_delta>;
> 
> One item per line, please.
> 
>> +		nvmem-cell-names = "charge_limit_en", "charge_limit_end", "charge_limit_delta";
> 
> And here too.

With that:

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

Konrad

Re: [PATCH v4 3/8] power: supply: qcom_battmgr: Add resistance power supply property

[Konstantin Ryabitsev]

On Tue, Sep 16, 2025 at 09:59:04AM +0200, Konrad Dybcio wrote:
> + Konstantin
> 
> It's quite common to see someone leaving a T-b on the cover letter,
> trying to say "I gave this series a spin" and then seeing the tag
> appear on unrelated commits within the series (e.g. bindings or some
> cosmetic fixes". Maybe some sort of an interactive (opt-in is fine)
> dialog for "which patches to apply t-b/tags to" could be worth the
> effort?

The plan is to add interactive mode to a few commands, including to the
trailers command. This will open an interface similar to interactive rebase,
where you can mark trailers as accept, skip, or ignore. That should do what
you're asking for, I believe.

Best regards,
-K

Re: [PATCH v4 3/8] power: supply: qcom_battmgr: Add resistance power supply property

[Konrad Dybcio]

On 9/16/25 3:37 PM, Konstantin Ryabitsev wrote:
> On Tue, Sep 16, 2025 at 09:59:04AM +0200, Konrad Dybcio wrote:
>> + Konstantin
>>
>> It's quite common to see someone leaving a T-b on the cover letter,
>> trying to say "I gave this series a spin" and then seeing the tag
>> appear on unrelated commits within the series (e.g. bindings or some
>> cosmetic fixes". Maybe some sort of an interactive (opt-in is fine)
>> dialog for "which patches to apply t-b/tags to" could be worth the
>> effort?
> 
> The plan is to add interactive mode to a few commands, including to the
> trailers command. This will open an interface similar to interactive rebase,
> where you can mark trailers as accept, skip, or ignore. That should do what
> you're asking for, I believe.

That is amazing to hear, thank you

Konrad

Re: [PATCH v4 2/8] power: supply: core: Add state_of_health power supply property

[Sebastian Reichel]

[-- Attachment #1: Type: text/plain, Size: 3420 bytes --]

Hi,

On Mon, Sep 15, 2025 at 04:49:54PM +0800, Fenglin Wu via B4 Relay wrote:
> From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> 
> Add state_of_health power supply property to represent battery
> health percentage.
> 
> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
> Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> ---
>  Documentation/ABI/testing/sysfs-class-power | 15 +++++++++++++++
>  drivers/power/supply/power_supply_sysfs.c   |  1 +
>  include/linux/power_supply.h                |  1 +
>  3 files changed, 17 insertions(+)
> 
> diff --git a/Documentation/ABI/testing/sysfs-class-power b/Documentation/ABI/testing/sysfs-class-power
> index cea1a38f5a8fb754d4e6323967ef6cf2e20a68ce..04f82e3e33aad6e16dc4fbace066b5d26069bf44 100644
> --- a/Documentation/ABI/testing/sysfs-class-power
> +++ b/Documentation/ABI/testing/sysfs-class-power
> @@ -568,6 +568,21 @@ Description:
>  
>  		Valid values: Represented in microohms
>  
> +What:		/sys/class/power_supply/<supply_name>/state_of_health
> +Date:		August 2025
> +Contact:	linux-arm-msm@vger.kernel.org
> +Description:
> +		The state_of_health parameter quantifies the overall condition
> +		of a battery as a percentage, reflecting its ability to deliver
> +		rated performance relative to its original specifications. It is
> +		dynamically computed using a combination of learned capacity
> +		and impedance-based degradation indicators, both of which evolve
> +		over the battery's lifecycle.

I think this should be extended by the following:

Note that the exact algorithms are kept secret by most battery
vendors and the value from different battery vendors cannot be
compared with each other as there is no vendor-agnostic definition
of "performance". Also this usually cannot be used for any
calculations (i.e. this is not the factor between charge_full and
charge_full_design).

-- Sebastian

> +
> +		Access: Read
> +
> +		Valid values: 0 - 100 (percent)
> +
>  **USB Properties**
>  
>  What:		/sys/class/power_supply/<supply_name>/input_current_limit
> diff --git a/drivers/power/supply/power_supply_sysfs.c b/drivers/power/supply/power_supply_sysfs.c
> index cfa8f90a88ebc8fc1c7447198f138e5d2e699e5a..d96a8578308e3af60cc1a352845662aa922c29b3 100644
> --- a/drivers/power/supply/power_supply_sysfs.c
> +++ b/drivers/power/supply/power_supply_sysfs.c
> @@ -221,6 +221,7 @@ static struct power_supply_attr power_supply_attrs[] __ro_after_init = {
>  	POWER_SUPPLY_ATTR(MANUFACTURE_MONTH),
>  	POWER_SUPPLY_ATTR(MANUFACTURE_DAY),
>  	POWER_SUPPLY_ATTR(INTERNAL_RESISTANCE),
> +	POWER_SUPPLY_ATTR(STATE_OF_HEALTH),
>  	/* Properties of type `const char *' */
>  	POWER_SUPPLY_ATTR(MODEL_NAME),
>  	POWER_SUPPLY_ATTR(MANUFACTURER),
> diff --git a/include/linux/power_supply.h b/include/linux/power_supply.h
> index 8bc3b7a67eb5693a16db9b7d123e7881711c6bf4..ccb43fe44381965069dc3bd9505d45050b9b1bd8 100644
> --- a/include/linux/power_supply.h
> +++ b/include/linux/power_supply.h
> @@ -175,6 +175,7 @@ enum power_supply_property {
>  	POWER_SUPPLY_PROP_MANUFACTURE_MONTH,
>  	POWER_SUPPLY_PROP_MANUFACTURE_DAY,
>  	POWER_SUPPLY_PROP_INTERNAL_RESISTANCE,
> +	POWER_SUPPLY_PROP_STATE_OF_HEALTH,
>  	/* Properties of type `const char *' */
>  	POWER_SUPPLY_PROP_MODEL_NAME,
>  	POWER_SUPPLY_PROP_MANUFACTURER,
> 
> -- 
> 2.34.1
> 
> 

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH v4 0/8] power: supply: Add several features support in qcom-battmgr driver

[Sebastian Reichel]

[-- Attachment #1: Type: text/plain, Size: 1372 bytes --]

Hi,

On Mon, Sep 15, 2025 at 04:49:52PM +0800, Fenglin Wu via B4 Relay wrote:
> Add following features in qcom-battmgr drivers as the battery management
> firmware has provided such capabilities:
>  - Add resistance power supply property in core driver and qcom-battmgr
>    driver to get battery resistance
>  - Add state_of_health power supply property in core driver and
>    qcom-battmgr driver to get battery health percentage
>  - Add charge control start/end threshold control by using
>    charge_control_start_threshold and charge_control_end_threshold power
>    supply properties
> 
> The changes have been tested on QRD8650 and X1E80100-CRD devices based on
> qcom/linux.git for-next commit a679f3f6931cdb0c2ef5dc0c26f895ae3f6c1ddc.
> 
> Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
> ---

Unrelated to this series specifically, but can you look into fixing
the following errors (appearing at least on X1E based Thinkpad T14s)?

qcom_battmgr.pmic_glink_power_supply pmic_glink.power-supply.0: unknown notification: 0x283
qcom_battmgr.pmic_glink_power_supply pmic_glink.power-supply.0: unknown notification: 0x283
qcom_battmgr.pmic_glink_power_supply pmic_glink.power-supply.0: unknown notification: 0x483
qcom_battmgr.pmic_glink_power_supply pmic_glink.power-supply.0: unknown notification: 0x83

Thanks,

-- Sebastian

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH v4 0/8] power: supply: Add several features support in qcom-battmgr driver

[Fenglin Wu]

On 9/17/2025 3:35 AM, Sebastian Reichel wrote:
> Hi,
>
> On Mon, Sep 15, 2025 at 04:49:52PM +0800, Fenglin Wu via B4 Relay wrote:
>> Add following features in qcom-battmgr drivers as the battery management
>> firmware has provided such capabilities:
>>   - Add resistance power supply property in core driver and qcom-battmgr
>>     driver to get battery resistance
>>   - Add state_of_health power supply property in core driver and
>>     qcom-battmgr driver to get battery health percentage
>>   - Add charge control start/end threshold control by using
>>     charge_control_start_threshold and charge_control_end_threshold power
>>     supply properties
>>
>> The changes have been tested on QRD8650 and X1E80100-CRD devices based on
>> qcom/linux.git for-next commit a679f3f6931cdb0c2ef5dc0c26f895ae3f6c1ddc.
>>
>> Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
>> ---
> Unrelated to this series specifically, but can you look into fixing
> the following errors (appearing at least on X1E based Thinkpad T14s)?
>
> qcom_battmgr.pmic_glink_power_supply pmic_glink.power-supply.0: unknown notification: 0x283
> qcom_battmgr.pmic_glink_power_supply pmic_glink.power-supply.0: unknown notification: 0x283
> qcom_battmgr.pmic_glink_power_supply pmic_glink.power-supply.0: unknown notification: 0x483
> qcom_battmgr.pmic_glink_power_supply pmic_glink.power-supply.0: unknown notification: 0x83
>
> Thanks,
>
> -- Sebastian

Sure.

I will look into this and fix them if I can see the same on my X1E80100 
CRD device.

Re: [PATCH v4 5/8] power: supply: qcom_battmgr: update compats for SM8550 and X1E80100

[Fenglin Wu]

On 9/16/2025 6:16 PM, Dmitry Baryshkov wrote:
> On Tue, Sep 16, 2025 at 10:40:03AM +0800, Fenglin Wu wrote:
>> On 9/15/2025 6:19 PM, Dmitry Baryshkov wrote:
>>> On Mon, Sep 15, 2025 at 04:49:57PM +0800, Fenglin Wu via B4 Relay wrote:
>>>> From: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
>>>>
>>>> Add variant definitions for SM8550 and X1E80100 platforms. Add a compat
>>>> for SM8550 and update match data for X1E80100 specifically so that they
>>>> could be handled differently in supporting charge control functionality.
>>> Why?
>> Is the question about why this was submitted as a separate patch, or about
>> the need for the change itself? The reason for the change is explained in
>> the commit text.
> It's not, and that was my question. Why do you need to handle them
> differently?
>
> Please always start your commit message with the description of the
> issue that you are facing.
>
The simple answer is, the charge control functionality is only supported 
in battery management firmware starting from SM8550 and X1E80100.

I will add this statement in the commit text.

>> As for submitting it separately, that was done to address
>> Bryan's comments to split out the compats changes. Anyway, I will address
>> the further comments from Stephen to make the change bisectable.
>>>> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Thinkpad T14S OLED
>>>> Signed-off-by: Fenglin Wu <fenglin.wu@oss.qualcomm.com>
>>>> ---
>>>>    drivers/power/supply/qcom_battmgr.c | 7 +++++--
>>>>    1 file changed, 5 insertions(+), 2 deletions(-)
>>>>