drivers: power: ltm4700: Add support for LTM4700 dual-channel PMBus regulator by cjones-adi · Pull Request #3111 · analogdevicesinc/no-OS

cjones-adi · 2026-05-28T04:35:07Z

Pull Request Description

Add comprehensive support for the LTM4700 dual-channel 50A/100A μModule regulator with complete IIO subsystem integration, multi-example project build, and unit testing.

Driver Features

PMBus 1.3 compliant I2C interface (default address 0x5A)
Dual-channel power monitoring and control (VIN, IIN, PIN, VOUT, IOUT, POUT per channel)
Temperature monitoring
Switching frequency monitoring
LINEAR11/LINEAR16 data format conversion
Device identification and verification (LTM4700 special ID check)
CRC-8 support for reliable PMBus communication
Operation control (ON/OFF per channel or both)

Files Changed

Drivers: 4 files (ltm4700.h/c, iio_ltm4700.h/c)
Project: 13 files (complete multi-platform structure)
Tests: 4 files (comprehensive unit tests)
Documentation: 4 files (driver/project README + Sphinx)

PR Type

Bug fix (change that fixes an issue)
New feature (change that adds new functionality)
Breaking change (has dependencies in other repos or will cause CI to fail)

PR Checklist

I have followed the Coding style guidelines
I have complied with the Submission Checklist
I have performed a self-review of the changes
I have commented my code, at least hard-to-understand parts
I have build all projects affected by the changes in this PR
I have tested in hardware affected projects, at the relevant boards
I have signed off all commits from this PR
I have updated the documentation (wiki pages, ReadMe etc), if applies

edelweiseescala · 2026-06-01T04:18:56Z

+		return -EINVAL;
+
+	return ltm4700_send_byte(dev, 0, LTM4700_MFR_CLEAR_PEAKS);
+}


missing new line at end, check all files

Addressed in V2

edelweiseescala · 2026-06-01T05:05:34Z

+	if (ret)
+		goto exit;
+
+	ret = ltm4700_set_operation(dev, LTM4700_CHAN_ALL,


in ltm4700_init num of channels is set to
LTM4700_CHAN_ALL is 0xFF
won't it return invalid in ltm4700_set_operation because of
if (channel >= dev->num_channels)
return -EINVAL;

Addressed in V2

edelweiseescala · 2026-06-01T05:15:33Z

+
+	ret = no_os_gpio_remove(dev->alert_desc);
+	if (ret)
+		return ret;


Maybe we should still attempt to remove i2c to prevent partial clean ups

Addressed in V2

edelweiseescala · 2026-06-01T05:21:33Z

+#define LTM4700_LIN11_EXPONENT_MSK		NO_OS_GENMASK(15,11)
+#define LTM4700_LIN11_EXPONENT(x)		((int16_t)(x) >> 11)
+#define LTM4700_LIN11_MANTISSA(x)		(((int16_t)((x & LTM4700_LIN11_MANTISSA_MSK) << 5)) >> 5)
+#define LTM4700_LIN16_EXPONENT			-12


maybe enclose negative values in ()

Addressed in V2

cjones-adi · 2026-06-01T09:22:27Z

V2:

Fix bug in sscanf parsing
Fix bug in all channel filter in set operation (and set page)
Add i2c clean up
Add newline at the end of file for affected files
Add parenthesis on negative #define values

liviutomoiaga · 2026-06-02T12:03:08Z

Xilinx build fails due to missing authorization. By default, PRs from forks cannot access repository secrets.

FYI @cjones-adi

amiclaus · 2026-06-09T11:38:33Z

+ * @param tr - Transition rate in µV/ms
+ * @return 0 in case of success, negative error code otherwise
+ */
+int ltm4700_vout_tr(struct ltm4700_dev *dev, uint8_t channel, int tr);


this seems to be redundant.

Removed in V3

amiclaus · 2026-06-09T11:39:48Z

@@ -0,0 +1,10 @@
+{
+	"maxim": {
+		"basic_example_max32665": {


the project documentation references only max32655. what platform was actually used?

max32665 is the one. addressed in V3

Changed back to max32655 in V4

amiclaus · 2026-06-09T11:40:07Z

+        $(PLATFORM_DRIVERS)/../common/maxim_dma.c \
+        $(PLATFORM_DRIVERS)/maxim_i2c.c \
+        $(PLATFORM_DRIVERS)/maxim_uart.c \
+        $(PLATFORM_DRIVERS)/maxim_uart_stdio.c


missing newline at eof. check all occurences in all files.

Added in V3

amiclaus · 2026-06-09T11:41:21Z

+    struct ltm4700_init_param init_param = {
+        .i2c_init = &i2c_init_param,
+        .crc_en = false,
+        .id = ID_LTM4700


this member is non-existent.

Removed in V3

amiclaus · 2026-06-09T11:44:40Z

+	struct no_os_gpio_init_param **run_params;
+	struct no_os_gpio_init_param **fault_params;
+
+	bool external_clk_en;


this doesn't seem to be used.

Removed in V3

cjones-adi · 2026-06-17T07:07:33Z

V3

Remove unused external_clk_en member from ltm4700_init_param
Remove redundant ltm4700_vout_tr function declaration
Fix driver README example code
Fix platform inconsistency in project documentation
Add missing newlines at end of files

edelweiseescala · 2026-06-30T07:35:40Z

+	else
+		mantissa = uval << (-exp);
+
+	return (uint16_t)mantissa;


When value overflows truncated value may give garbage values. Better to clamp

Addressed in V4

edelweiseescala · 2026-06-30T07:35:56Z

+	else
+		val = mantissa >> (-exp);
+
+	return val;


Same clamp issue

Addressed in V4

edelweiseescala · 2026-06-30T07:37:11Z

+	if (exponent >= 0)
+		val = mantissa << exponent;
+	else
+		val = mantissa >> (-exponent);


same issue with overflow

Addressed in V4

edelweiseescala · 2026-06-30T07:39:17Z

+		goto error_dev;
+
+	if (init_param->crc_en) {
+		no_os_crc8_populate_msb(ltm4700_crc_table, LTM4700_CRC_POLYNOMIAL);


crc table is in init but seems to not be used

Addressed in V4

edelweiseescala · 2026-06-30T07:41:02Z

+		ptr++;
+	}
+
+	if (sscanf(ptr, "%d.%d", &int_part, &frac_part) == 2) {


if fractional part has leading zeros like 3.05
.05 will just be parsed as 5 this in turn will give a wrong value

Addressed in V4

edelweiseescala · 2026-06-30T07:43:04Z

+
+#define LTM4700_IIO_REG_CHAN(reg, chan)			(reg | (chan << 8))
+#define LTM4700_IIO_REG(x)				(x & 0xFF)
+#define LTM4700_IIO_CHAN(x)				((x >> 8) & 0xFF)


Are these macros used?

Removed in V4 (stale debug codes)

edelweiseescala · 2026-06-30T07:45:00Z

+		return -EINVAL;
+
+	if (ltm4700_cmd_is_paged(cmd)) {
+		int ret = ltm4700_set_page(dev, page);


Inconsistent with coding style should be declared at the top

Fixed in V4

edelweiseescala · 2026-06-30T07:45:10Z

+		return -EINVAL;
+
+	if (ltm4700_cmd_is_paged(cmd)) {
+		int ret = ltm4700_set_page(dev, page);


Inconsistent with coding style should be declared at the top

Fixed in V4

edelweiseescala · 2026-06-30T07:45:18Z

+		return -EINVAL;
+
+	if (ltm4700_cmd_is_paged(cmd)) {
+		int ret = ltm4700_set_page(dev, page);


Inconsistent with coding style should be declared at the top

Fixed in V4

buha · 2026-07-13T06:56:37Z

please migrate to the cmake/kconfig build system

buha · 2026-07-13T06:57:04Z

/AzurePipelines run

azure-pipelines · 2026-07-13T06:57:17Z

Azure Pipelines successfully started running 2 pipeline(s).

Add comprehensive PMBus driver for LTM4700 dual-channel regulator. Includes support for dual-channel monitoring, and complete LINEAR format conversion. Key features: - PMBus 1.3 compliant I2C interface (400kHz) - Dual-channel 50A per channel or single 100A operation - LINEAR11/LINEAR16 data format support - Comprehensive status monitoring - Complete telemetry reading (VIN, IIN, VOUT, IOUT, temperatures) - Peak value tracking and management - NVM operations for configuration storage Signed-off-by: Carlos Jones Jr <carlosjr.jones@analog.com>

Add comprehensive Industrial I/O (IIO) framework integration for LTM4700 dual-channel regulator with real-time telemetry and control interface. Key features: - 13 IIO channels covering all telemetry (VIN, IIN, VOUT0/1, IOUT0/1, etc.) - Real-time voltage/current/power/temperature monitoring - Write support for VOUT control and operation mode - Peak value management and fault clearing - Engineering unit conversion (mV, mA, m°C, mW) - Multi-channel configuration support - Complete Linux IIO subsystem compliance IIO Channels: - Input: VIN, IIN, PIN, IC temperature - Output: VOUT0/1, IOUT0/1, POUT0/1, external temperatures - Control: Operation mode, voltage commands, margins - Frequency: Per-channel switching frequency monitoring Signed-off-by: Carlos Jones Jr <carlosjr.jones@analog.com>

Add comprehensive driver documentation covering device features, PMBus command interface, IIO integration, and usage examples. Documentation includes: - Device overview - Key features and specifications - PMBus communication interface details - Data format descriptions (LINEAR11/LINEAR16) - Complete API usage examples - IIO framework integration guide - Design considerations and best practices - Build configuration instructions The documentation provides complete guidance for driver integration and application development with the LTM4700 regulator. Signed-off-by: Carlos Jones Jr <carlosjr.jones@analog.com>

Add comprehensive project structure for LTM4700 dual-channel regulator with support for both basic telemetry monitoring and IIO integration. Includes platform-specific configuration for Maxim MAX32655 with proper I2C, UART, and GPIO setup for PMBus communication at 400kHz. Signed-off-by: Carlos Jones Jr <carlosjr.jones@analog.com>

Add comprehensive project documentation covering hardware setup, build instructions, examples usage, and troubleshooting guide. Includes detailed configuration options for both basic telemetry and IIO integration examples with performance characteristics and platform-specific setup instructions. Signed-off-by: Carlos Jones Jr <carlosjr.jones@analog.com>

Add comprehensive unit test suite for LTM4700 dual-channel regulator driver using Ceedling/Unity/CMock framework. Includes core driver tests for initialization, PMBus communication, LINEAR format conversion, voltage/current/temperature reading, and IIO integration tests. Test coverage includes: - Device initialization - PMBus read/write operations with I2C mocking - LINEAR11/LINEAR16 format conversion validation - Multi-channel operation and page selection - IIO channel mapping and attribute handling - Error handling and parameter validation - Support functions for test data generation Targets 80%+ code coverage with automated testing framework. Signed-off-by: Carlos Jones Jr <carlosjr.jones@analog.com>

cjones-adi · 2026-07-14T06:47:13Z

V4:

Migrate build system to cmake/kconfig
Handle overflows in data conversion
Address CRC and scannf issue
Clean up unused macros
Fix code style issues

cjones-adi requested review from CiprianRegus, RaduSabau1, amiclaus, buha, cristina-suteu, danmois, dbogdan, raezt, ralucabozdog, ranechita, rbolboac, rbudai98 and stefpopa as code owners May 28, 2026 04:35

github-actions Bot added drivers projects documentation labels May 28, 2026

cjones-adi force-pushed the staging/ltm4700 branch from 3d242f5 to fab2333 Compare May 29, 2026 08:52

edelweiseescala reviewed Jun 1, 2026

View reviewed changes

cjones-adi force-pushed the staging/ltm4700 branch 2 times, most recently from 20b4639 to 10ebcb5 Compare June 1, 2026 09:18

amiclaus reviewed Jun 9, 2026

View reviewed changes

cjones-adi force-pushed the staging/ltm4700 branch 3 times, most recently from 63dfde6 to 17892f4 Compare June 17, 2026 07:06

cjones-adi requested review from amiclaus and edelweiseescala June 30, 2026 02:05

edelweiseescala reviewed Jun 30, 2026

View reviewed changes

cjones-adi added 6 commits July 14, 2026 13:44

cjones-adi force-pushed the staging/ltm4700 branch from 17892f4 to 1ddba9e Compare July 14, 2026 06:40

Uh oh!

Conversation

cjones-adi commented May 28, 2026

Pull Request Description

Driver Features

Files Changed

PR Type

PR Checklist

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

cjones-adi commented Jun 1, 2026

Uh oh!

liviutomoiaga commented Jun 2, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

cjones-adi commented Jun 17, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment