MSPM0: RTC Support #90844

karthi012 · 2025-05-30T06:46:04Z

This PR add real time clock support for Texas Instruments MSPM0 Family Series

bjarki-andreasen

Please add an overlay to and test out the driver with https://docs.zephyrproject.org/latest/hardware/peripherals/rtc.html#rtc-device-driver-test-suite :)

bjarki-andreasen · 2025-05-30T08:50:53Z

dts/bindings/rtc/ti,mspm0-rtc.yaml

+compatible: "ti,mspm0-rtc"
+
+include:
+  - name: rtc.yaml


only rtc-device.yaml should be included, rtc.yaml is an old relic :)

bjarki-andreasen · 2025-05-30T08:52:13Z

drivers/rtc/rtc_ti_mspm0.c

+
+#define DT_DRV_COMPAT ti_mspm0_rtc
+
+#include <errno.h>


not needed since its part of zephyr/kernel.h :)

drivers/rtc/rtc_ti_mspm0.c

bjarki-andreasen · 2025-05-30T08:58:39Z

drivers/rtc/rtc_ti_mspm0.c

+	__disable_irq();
+	ret = alarm->is_pending ? 1 : 0;
+	alarm->is_pending = false;
+	__enable_irq();


disable IRQ is not needed here as that is what a spinlock does, so you are already in a critical section here because of line 349 :)

also, this is not tracking (they key part), so it seems it could break the kernel by not restoring interrupt mask to correct state :) I have not seen __disable_irq() before

bjarki-andreasen · 2025-05-30T09:00:33Z

drivers/rtc/rtc_ti_mspm0.c

+{
+#if defined(CONFIG_RTC_ALARM)
+	uint8_t id;
+	uint32_t key = irq_lock();


Use spinlock if needed :)

bjarki-andreasen · 2025-05-30T09:03:23Z

drivers/rtc/rtc_ti_mspm0.c

+	alarm->is_pending = true;
+	if (alarm->callback) {
+		alarm->callback(dev, id, alarm->user_data);
+	}


Its safer to store the callback and user data in a local variable, reenable irqs as critical section is no longer needed, then call the callback. This way the callback is not called with irqs disabled (or more appropriately spinlock held) :)

bjarki-andreasen · 2025-05-30T09:05:53Z

drivers/rtc/rtc_ti_mspm0.c

+	.base		 = (RTC_Regs *)DT_INST_REG_ADDR(n),		     \
+	.rtc_x		 = DT_INST_PROP(n, ti_rtc_x),			     \
+	.irq_config_func = ti_mspm0_config_irq_##n,			     \


alingment here should be

static type sym = { .member = foo, };

bjarki-andreasen · 2025-05-30T09:08:39Z

dts/arm/ti/mspm0/mspm0.dtsi

+			reg = <0x40094000 0x2000>;
+			interrupts = <30 0>;
+			prescaler = <1>;
+			clock-frequency = <32768>;


missing alarms-count = <2>; property inhereted from rtc-device.yaml :)

karthi012 · 2025-05-31T12:41:50Z

@bjarki-andreasen Updated the changes. Please have a look

tbursztyka · 2025-06-11T13:18:57Z

dts/bindings/rtc/ti,mspm0-rtc.yaml

+  ti,rtc-x:
+    type: boolean
+    description: |
+      Indicate as Real-Time-Clock_x(RTC_x) instances.


add a space between 'x('

tbursztyka · 2025-06-11T13:19:56Z

drivers/rtc/rtc_ti_mspm0.c

+#if defined(CONFIG_RTC_ALARM)
+struct rtc_ti_mspm0_alarm {
+	uint16_t mask;
+	bool is_pending;


move these 2 attributes at the end of the structure for better alignment

tbursztyka · 2025-06-11T13:20:17Z

drivers/rtc/rtc_ti_mspm0.c

+
+struct rtc_ti_mspm0_config {
+	RTC_Regs *base;
+	bool rtc_x;


move this attribute at the end of the structure for better alignment

drivers/rtc/rtc_ti_mspm0.c

tbursztyka · 2025-06-11T13:30:10Z

drivers/rtc/rtc_ti_mspm0.c

+	const struct rtc_ti_mspm0_config *cfg = dev->config;
+
+	DL_RTC_Common_disableInterrupt(cfg->base, DL_RTC_COMMON_INTERRUPT_CALENDAR_ALARM1);
+	if (mask & RTC_ALARM_TIME_MASK_MINUTE) {


add an empty line before the if ()

tbursztyka · 2025-06-11T13:30:15Z

drivers/rtc/rtc_ti_mspm0.c

+	const struct rtc_ti_mspm0_config *cfg = dev->config;
+
+	DL_RTC_Common_disableInterrupt(cfg->base, DL_RTC_COMMON_INTERRUPT_CALENDAR_ALARM2);
+	if (mask & RTC_ALARM_TIME_MASK_MINUTE) {


tbursztyka · 2025-06-11T13:33:05Z

drivers/rtc/rtc_ti_mspm0.c

+#define RTC_TI_MAX_ALARM	DT_INST_PROP(0, alarms_count)
+
+struct rtc_ti_mspm0_config {
+	RTC_Regs *base;


s/base/regs would be better semantically, like it has been done for the uart driver.

tbursztyka · 2025-06-11T13:33:30Z

drivers/rtc/rtc_ti_mspm0.c

+
+	K_SPINLOCK(&data->lock) {
+		rtc_ti_mspm0_clear_alarm(dev, id);
+		if (id == RTC_TI_ALARM_1) {


d-philpot · 2025-06-16T15:21:43Z

The RTC_Common APIs is the superset file in the SDK for all flavors of RTC that exist on MSPM0 devices. The base rtc apis (in rtc.h/c) should be used instead.

karthi012 · 2025-06-21T10:15:25Z

The RTC_Common APIs is the superset file in the SDK for all flavors of RTC that exist on MSPM0 devices. The base rtc apis (in rtc.h/c) should be used instead.

RTC common(dl_rtc_common.h) is compatible with rtc and rtc_x variant in RTC IP block from MSPM0. We don't need a separate driver each, instead we can adapt the existing driver to support both.

karthi012 · 2025-06-28T06:59:32Z

@bjarki-andreasen Ping for review. Thanks

bjarki-andreasen

I would encourage you to add a .conf overlay to the rtc api test suite to enable CONFIG_RTC_ALARM to test them as well, looks good otherwise :)

drivers/rtc/rtc_ti_mspm0.c

add Real-Time Clock binding for Texas Instruments MSPM0 Family Signed-off-by: Karthikeyan Krishnasamy <[email protected]>

added Real-Time Clock driver support for Texas Instruments MSPM0 Family, currently this driver supports rtc time keeping and calendar alarms functionality Signed-off-by: Karthikeyan Krishnasamy <[email protected]>

Add a rtc support for mspm0 Signed-off-by: Karthikeyan Krishnasamy <[email protected]>

Add rtc_api test coverage support for lp_mspm0g3507 Signed-off-by: Karthikeyan Krishnasamy <[email protected]>

sonarqubecloud · 2025-07-03T11:28:39Z

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

karthi012 · 2025-07-05T05:44:00Z

@bjarki-andreasen Ping for a review. Thanks

github-actions bot added platform: Texas Instruments MSPM0 platform: TI SimpleLink Texas Instruments SimpleLink MCU area: RTC Real Time Clock labels May 30, 2025

github-actions bot requested review from bjarki-andreasen, ssekar15, vaishnavachath and vanti May 30, 2025 06:46

github-actions bot assigned bjarki-andreasen May 30, 2025

bjarki-andreasen requested changes May 30, 2025

View reviewed changes

parthitce linked an issue May 30, 2025 that may be closed by this pull request

RTC linumiz/zephyr#87

Open

karthi012 force-pushed the upstream/ti/mspm0-rtc branch from 4e49499 to bb57eaa Compare May 31, 2025 10:14

tbursztyka requested changes Jun 11, 2025

View reviewed changes

karthi012 force-pushed the upstream/ti/mspm0-rtc branch from bb57eaa to 6dcab95 Compare June 21, 2025 11:56

parthitce removed the platform: TI SimpleLink Texas Instruments SimpleLink MCU label Jun 26, 2025

bjarki-andreasen previously approved these changes Jun 30, 2025

View reviewed changes

drivers/rtc/rtc_ti_mspm0.c Outdated Show resolved Hide resolved

drivers/rtc/rtc_ti_mspm0.c Outdated Show resolved Hide resolved

karthi012 added 4 commits July 3, 2025 16:22

dts: bindings: rtc: introduce ti mspm0 rtc binding

882aa4b

add Real-Time Clock binding for Texas Instruments MSPM0 Family Signed-off-by: Karthikeyan Krishnasamy <[email protected]>

drivers: rtc: add driver support for ti mspm0 rtc

007e081

added Real-Time Clock driver support for Texas Instruments MSPM0 Family, currently this driver supports rtc time keeping and calendar alarms functionality Signed-off-by: Karthikeyan Krishnasamy <[email protected]>

dts: arm: ti: mspm0: add support for rtc

5a28971

Add a rtc support for mspm0 Signed-off-by: Karthikeyan Krishnasamy <[email protected]>

tests: drivers: rtc: add rtc_api test suite support

5bce82b

Add rtc_api test coverage support for lp_mspm0g3507 Signed-off-by: Karthikeyan Krishnasamy <[email protected]>

karthi012 dismissed bjarki-andreasen’s stale review via 5bce82b July 3, 2025 11:11

karthi012 force-pushed the upstream/ti/mspm0-rtc branch from 6dcab95 to 5bce82b Compare July 3, 2025 11:11

github-actions bot added the platform: TI SimpleLink Texas Instruments SimpleLink MCU label Jul 3, 2025

MSPM0: RTC Support #90844

Are you sure you want to change the base?

MSPM0: RTC Support #90844

Conversation

karthi012 commented May 30, 2025

Uh oh!

bjarki-andreasen left a comment

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!

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!

karthi012 commented May 31, 2025

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!

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!

d-philpot commented Jun 16, 2025

Uh oh!

karthi012 commented Jun 21, 2025

Uh oh!

karthi012 commented Jun 28, 2025

Uh oh!

bjarki-andreasen left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

sonarqubecloud bot commented Jul 3, 2025

Quality Gate passed

Uh oh!

karthi012 commented Jul 5, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

karthi012 commented Jul 5, 2025 •

edited

Loading