os/board/bk7239n: Optimize the internal code of bk7239n by Poly-J · Pull Request #7280 · Samsung/TizenRT

Poly-J · 2026-04-16T06:34:19Z

The form of the security interface has changed, and the corresponding test code also needs to be modified accordingly.
Logic optimization related to security dump
Optimize the logic related to the system entering sleep

ewoodev

I think, this is only important agenda of this commit.

The form of the security interface has changed, and
the corresponding test code also needs to be modified accordingly.

It's optimization

Logic optimization related to security dump

it's another agenda

Optimize the logic related to the system entering sleep

Could you please seperate commit?

ewoodev · 2026-04-16T09:26:21Z


 void bk_security_to_nosecurity_dump_register_callback(void)
 {
-    if (get_security_level() >  LOW_SECURITY_LEVEL) {


is it not needed?

Yes, it should not be added here, because after adding this logic, since security_level defaults to HIGH, it will cause this callback not to be injected.

ah okay, this fuction is just register callback.

Poly-J · 2026-04-16T09:36:19Z

I think, this is only important agenda of this commit.

The form of the security interface has changed, and
the corresponding test code also needs to be modified accordingly.

It's optimization

Logic optimization related to security dump

it's another agenda

Optimize the logic related to the system entering sleep

Could you please seperate commit?

Okay, I will split it into three commits

ewoodev · 2026-04-16T09:53:16Z

 void bk_security_donmain_notifies_non_security_domain_to_dump(uint32_t *reg)
 {
    // High security level will not dump the exception information.
    if (get_security_level() > LOW_SECURITY_LEVEL) { 


if (get_security_level() != LOW_SECURITY_LEVEL)

The value can be changed, please use !=

Thank you for your feedback, I have already modified itit.

ewoodev · 2026-04-16T09:54:09Z


 void bk_security_to_nosecurity_dump_register_callback(void)
 {
-    if (get_security_level() >  LOW_SECURITY_LEVEL) {


ah okay, this fuction is just register callback.

ewoodev · 2026-04-16T09:58:02Z

+	{
+		if(bk_rtc_get_first_alarm_name() != NULL)
+		{
+			if(strcmp(bk_rtc_get_first_alarm_name(), PM_WIFI_RTC_ALARM_NAME) == 0)


Could you please use strncmp?

Dear customer
Okay, we will make the modification. Thank you

ewoodev · 2026-04-16T10:00:57Z

+#define PM_VANALDO_STEP_SIZE        (2)
+#define PM_VANALDO_STEP_UP_DELAY_US (10)
+__attribute__((section(".iram"))) int32 sys_hal_vanaldo_set(uint32_t value)
+{
+	uint32_t cur = 0;
+	cur = sys_ana_ll_get_ana_reg7_vanaldosel();
+	if (cur == value) {
+		return 0;
+	}
+
+	if (value < cur) {
+		/* Target is lower or equal: set directly */
+		sys_ana_ll_set_ana_reg7_vanaldosel(value);
+	} else {
+		/* Target is higher: step up by PM_VANALDO_STEP_SIZE, delay PM_VANALDO_STEP_UP_DELAY_US each step */
+		uint32_t v;
+		for (v = cur + PM_VANALDO_STEP_SIZE; v < value; v += PM_VANALDO_STEP_SIZE) {
+			sys_ana_ll_set_ana_reg7_vanaldosel(v);
+			pm_delay_us(PM_VANALDO_STEP_UP_DELAY_US);
+		}
+		sys_ana_ll_set_ana_reg7_vanaldosel(value);
+
+		pm_delay_us(PM_VANALDO_STEP_UP_DELAY_US);
+	}
+	return 0;
+}


Could you please explain this changes in commit description?

Dear customer
This adjustment is an optimization of the recovery sequence after low-power wake-up: first adjust the voltage, then configure other items. Compared to the previous approach of writing all at once, it is more robust. This modification does not affect functionality.

ewoodev · 2026-04-16T10:01:51Z

+	/*step 1: restore ana voltage*/
+	sys_hal_vanaldo_set(vanaldosel);
+
+	sys_ana_ll_set_ana_reg7_spi_pwd_regpow(pwd_regpow);
+	sys_ana_ll_set_ana_reg7_vbspbuf_lp(vbspbuf_lp);
+	sys_ana_ll_set_ana_reg7_envrefh1v(envrefh1v);
+	sys_ana_ll_set_ana_reg7_aldohp(aldohp);
+	sys_ana_ll_set_ana_reg7_dldohp(dldohp);


it's also, what changes?

Dear customer
The same as the above

…rface The form of the security interface has changed, and the corresponding test code also needs to be modified accordingly

…leep Optimized the logic of some wake-up source prompts after the system enters sleep

Poly-J force-pushed the bk7239n_platform branch 2 times, most recently from e7bc91f to 9da17b6 Compare April 16, 2026 08:43

ewoodev reviewed Apr 16, 2026

View reviewed changes

Poly-J force-pushed the bk7239n_platform branch from 9da17b6 to 25c4876 Compare April 16, 2026 09:49

ewoodev reviewed Apr 16, 2026

View reviewed changes

Poly-J force-pushed the bk7239n_platform branch from 25c4876 to 45037fd Compare April 16, 2026 10:00

ewoodev reviewed Apr 16, 2026

View reviewed changes

Poly-J force-pushed the bk7239n_platform branch from 45037fd to b497f9c Compare April 16, 2026 11:03

os/board/bk7239n: Internal security test code adapted to the new inte…

7825af7

…rface The form of the security interface has changed, and the corresponding test code also needs to be modified accordingly

Poly-J force-pushed the bk7239n_platform branch 3 times, most recently from 58302ac to 0ad7c04 Compare April 21, 2026 11:38

os/board/bk7239n: Optimize the logic related to the system entering s…

0ad7c04

…leep Optimized the logic of some wake-up source prompts after the system enters sleep

Conversation

Poly-J commented Apr 16, 2026

Uh oh!

ewoodev 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!

Choose a reason for hiding this comment

Uh oh!

Poly-J commented Apr 16, 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!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants