[PATCH 0/3] selftest: rtc: rtctest: improve handling of skipped tests

[PATCH 0/3] selftest: rtc: rtctest: improve handling of skipped tests

[Wolfram Sang]

Here are some fixes for issues I found in 'rtctest' while checking RTC
drivers. Details are in the patch descriptions. Patches 1+2 have been
tested on HW. Patch 3 has not been tested, just follows the pattern of
patch 2, but may need more discussion in general. Please read there.
Looking forward to comments!


Wolfram Sang (3):
  selftest: rtc: rtctest: don't mention skipping twice
  selftest: rtc: rtctest: make missing UIEs really skip
  RFC selftest: rtc: rtctest: skip properly if setting alarm fails

 tools/testing/selftests/rtc/rtctest.c | 49 ++++++++++++---------------
 1 file changed, 21 insertions(+), 28 deletions(-)

-- 
2.53.0

[PATCH 1/3] selftest: rtc: rtctest: don't mention skipping twice

[Wolfram Sang]

The framework already annotates that we are skipping something. No need
to do in the tests themselves again.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
---
 tools/testing/selftests/rtc/rtctest.c | 31 +++++++++++++--------------
 1 file changed, 15 insertions(+), 16 deletions(-)

diff --git a/tools/testing/selftests/rtc/rtctest.c b/tools/testing/selftests/rtc/rtctest.c
index 8047d9879039..077cdc2c6f87 100644
--- a/tools/testing/selftests/rtc/rtctest.c
+++ b/tools/testing/selftests/rtc/rtctest.c
@@ -49,7 +49,7 @@ TEST_F(rtc, date_read) {
 	struct rtc_time rtc_tm;
 
 	if (self->fd == -1 && errno == ENOENT)
-		SKIP(return, "Skipping test since %s does not exist", rtc_file);
+		SKIP(return, "%s does not exist", rtc_file);
 	ASSERT_NE(-1, self->fd);
 
 	/* Read the RTC time/date */
@@ -118,7 +118,7 @@ TEST_F_TIMEOUT(rtc, date_read_loop, READ_LOOP_DURATION_SEC + 2) {
 	time_t start_rtc_read, prev_rtc_read;
 
 	if (self->fd == -1 && errno == ENOENT)
-		SKIP(return, "Skipping test since %s does not exist", rtc_file);
+		SKIP(return, "%s does not exist", rtc_file);
 	ASSERT_NE(-1, self->fd);
 
 	TH_LOG("Continuously reading RTC time for %ds (with %dms breaks after every read).",
@@ -156,7 +156,7 @@ TEST_F_TIMEOUT(rtc, uie_read, NUM_UIE + 2) {
 	unsigned long data;
 
 	if (self->fd == -1 && errno == ENOENT)
-		SKIP(return, "Skipping test since %s does not exist", rtc_file);
+		SKIP(return, "%s does not exist", rtc_file);
 	ASSERT_NE(-1, self->fd);
 
 	/* Turn on update interrupts */
@@ -185,7 +185,7 @@ TEST_F(rtc, uie_select) {
 	unsigned long data;
 
 	if (self->fd == -1 && errno == ENOENT)
-		SKIP(return, "Skipping test since %s does not exist", rtc_file);
+		SKIP(return, "%s does not exist", rtc_file);
 	ASSERT_NE(-1, self->fd);
 
 	/* Turn on update interrupts */
@@ -229,14 +229,14 @@ TEST_F(rtc, alarm_alm_set) {
 	enum rtc_alarm_state alarm_state = RTC_ALARM_UNKNOWN;
 
 	if (self->fd == -1 && errno == ENOENT)
-		SKIP(return, "Skipping test since %s does not exist", rtc_file);
+		SKIP(return, "%s does not exist", rtc_file);
 	ASSERT_NE(-1, self->fd);
 
 	alarm_state = get_rtc_alarm_state(self->fd, 1);
 	if (alarm_state == RTC_ALARM_DISABLED)
-		SKIP(return, "Skipping test since alarms are not supported.");
+		SKIP(return, "alarms are not supported.");
 	if (alarm_state == RTC_ALARM_RES_MINUTE)
-		SKIP(return, "Skipping test since alarms has only minute granularity.");
+		SKIP(return, "alarms has only minute granularity.");
 
 	rc = ioctl(self->fd, RTC_RD_TIME, &tm);
 	ASSERT_NE(-1, rc);
@@ -299,14 +299,14 @@ TEST_F(rtc, alarm_wkalm_set) {
 	enum rtc_alarm_state alarm_state = RTC_ALARM_UNKNOWN;
 
 	if (self->fd == -1 && errno == ENOENT)
-		SKIP(return, "Skipping test since %s does not exist", rtc_file);
+		SKIP(return, "%s does not exist", rtc_file);
 	ASSERT_NE(-1, self->fd);
 
 	alarm_state = get_rtc_alarm_state(self->fd, 1);
 	if (alarm_state == RTC_ALARM_DISABLED)
-		SKIP(return, "Skipping test since alarms are not supported.");
+		SKIP(return, "alarms are not supported.");
 	if (alarm_state == RTC_ALARM_RES_MINUTE)
-		SKIP(return, "Skipping test since alarms has only minute granularity.");
+		SKIP(return, "alarms has only minute granularity.");
 
 	rc = ioctl(self->fd, RTC_RD_TIME, &alarm.time);
 	ASSERT_NE(-1, rc);
@@ -363,12 +363,12 @@ TEST_F_TIMEOUT(rtc, alarm_alm_set_minute, 65) {
 	enum rtc_alarm_state alarm_state = RTC_ALARM_UNKNOWN;
 
 	if (self->fd == -1 && errno == ENOENT)
-		SKIP(return, "Skipping test since %s does not exist", rtc_file);
+		SKIP(return, "%s does not exist", rtc_file);
 	ASSERT_NE(-1, self->fd);
 
 	alarm_state = get_rtc_alarm_state(self->fd, 0);
 	if (alarm_state == RTC_ALARM_DISABLED)
-		SKIP(return, "Skipping test since alarms are not supported.");
+		SKIP(return, "alarms are not supported.");
 
 	rc = ioctl(self->fd, RTC_RD_TIME, &tm);
 	ASSERT_NE(-1, rc);
@@ -431,12 +431,12 @@ TEST_F_TIMEOUT(rtc, alarm_wkalm_set_minute, 65) {
 	enum rtc_alarm_state alarm_state = RTC_ALARM_UNKNOWN;
 
 	if (self->fd == -1 && errno == ENOENT)
-		SKIP(return, "Skipping test since %s does not exist", rtc_file);
+		SKIP(return, "%s does not exist", rtc_file);
 	ASSERT_NE(-1, self->fd);
 
 	alarm_state = get_rtc_alarm_state(self->fd, 0);
 	if (alarm_state == RTC_ALARM_DISABLED)
-		SKIP(return, "Skipping test since alarms are not supported.");
+		SKIP(return, "alarms are not supported.");
 
 	rc = ioctl(self->fd, RTC_RD_TIME, &alarm.time);
 	ASSERT_NE(-1, rc);
@@ -502,8 +502,7 @@ int main(int argc, char **argv)
 	if (access(rtc_file, R_OK) == 0)
 		ret = test_harness_run(argc, argv);
 	else
-		ksft_exit_skip("[SKIP]: Cannot access rtc file %s - Exiting\n",
-						rtc_file);
+		ksft_exit_skip("Cannot access rtc file %s - Exiting\n", rtc_file);
 
 	return ret;
 }
-- 
2.53.0

Re: [PATCH 1/3] selftest: rtc: rtctest: don't mention skipping twice

[sashiko-bot]

> The framework already annotates that we are skipping something. No need
> to do in the tests themselves again.
> 
> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>

Sashiko has reviewed this patch and found no issues. It looks great!

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260916213108.9234-5-wsa+renesas@sang-engineering.com?part=1

[PATCH 2/3] selftest: rtc: rtctest: make missing UIEs really skip

[Wolfram Sang]

Use the proper framework call to mark a test as skipped. Currently, the
test is considered good, only the printout mentions the skip.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
---
 tools/testing/selftests/rtc/rtctest.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/rtc/rtctest.c b/tools/testing/selftests/rtc/rtctest.c
index 077cdc2c6f87..962050a32985 100644
--- a/tools/testing/selftests/rtc/rtctest.c
+++ b/tools/testing/selftests/rtc/rtctest.c
@@ -163,8 +163,7 @@ TEST_F_TIMEOUT(rtc, uie_read, NUM_UIE + 2) {
 	rc = ioctl(self->fd, RTC_UIE_ON, 0);
 	if (rc == -1) {
 		ASSERT_EQ(EINVAL, errno);
-		TH_LOG("skip update IRQs not supported.");
-		return;
+		SKIP(return, "update IRQs are not supported.");
 	}
 
 	for (i = 0; i < NUM_UIE; i++) {
@@ -192,8 +191,7 @@ TEST_F(rtc, uie_select) {
 	rc = ioctl(self->fd, RTC_UIE_ON, 0);
 	if (rc == -1) {
 		ASSERT_EQ(EINVAL, errno);
-		TH_LOG("skip update IRQs not supported.");
-		return;
+		SKIP(return, "update IRQs are not supported.");
 	}
 
 	for (i = 0; i < NUM_UIE; i++) {
-- 
2.53.0

Re: [PATCH 2/3] selftest: rtc: rtctest: make missing UIEs really skip

[sashiko-bot]

> Use the proper framework call to mark a test as skipped. Currently, the
> test is considered good, only the printout mentions the skip.
> 
> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>

Sashiko has reviewed this patch and found no issues. It looks great!

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260916213108.9234-5-wsa+renesas@sang-engineering.com?part=2

[RFC PATCH 3/3] selftest: rtc: rtctest: skip properly if setting alarm fails

[Wolfram Sang]

Use the proper framework call to mark a test as skipped. Currently, the
test is considered good, only the printout mentions the skip.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
---

Maybe I am reading the logic wrong. But, the SKIPs I add here are hit
when setting an alarm fails. I wonder if this is really a SKIP and not
an error? We have already checked above if alarms are supported using a
designated check (and then skip), or?

 tools/testing/selftests/rtc/rtctest.c | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

diff --git a/tools/testing/selftests/rtc/rtctest.c b/tools/testing/selftests/rtc/rtctest.c
index 962050a32985..46fbaef9fdeb 100644
--- a/tools/testing/selftests/rtc/rtctest.c
+++ b/tools/testing/selftests/rtc/rtctest.c
@@ -250,8 +250,7 @@ TEST_F(rtc, alarm_alm_set) {
 		 */
 		ASSERT_EQ(RTC_ALARM_UNKNOWN, alarm_state);
 		ASSERT_EQ(EINVAL, errno);
-		TH_LOG("skip alarms are not supported.");
-		return;
+		SKIP(return, "alarms are not supported.");
 	}
 
 	rc = ioctl(self->fd, RTC_ALM_READ, &tm);
@@ -322,8 +321,7 @@ TEST_F(rtc, alarm_wkalm_set) {
 		 */
 		ASSERT_EQ(RTC_ALARM_UNKNOWN, alarm_state);
 		ASSERT_EQ(EINVAL, errno);
-		TH_LOG("skip alarms are not supported.");
-		return;
+		SKIP(return, "alarms are not supported.");
 	}
 
 	rc = ioctl(self->fd, RTC_WKALM_RD, &alarm);
@@ -382,8 +380,7 @@ TEST_F_TIMEOUT(rtc, alarm_alm_set_minute, 65) {
 		 */
 		ASSERT_EQ(RTC_ALARM_UNKNOWN, alarm_state);
 		ASSERT_EQ(EINVAL, errno);
-		TH_LOG("skip alarms are not supported.");
-		return;
+		SKIP(return, "alarms are not supported.");
 	}
 
 	rc = ioctl(self->fd, RTC_ALM_READ, &tm);
@@ -452,8 +449,7 @@ TEST_F_TIMEOUT(rtc, alarm_wkalm_set_minute, 65) {
 		 */
 		ASSERT_EQ(RTC_ALARM_UNKNOWN, alarm_state);
 		ASSERT_EQ(EINVAL, errno);
-		TH_LOG("skip alarms are not supported.");
-		return;
+		SKIP(return, "alarms are not supported.");
 	}
 
 	rc = ioctl(self->fd, RTC_WKALM_RD, &alarm);
-- 
2.53.0

Re: [RFC PATCH 3/3] selftest: rtc: rtctest: skip properly if setting alarm fails

[sashiko-bot]

> Use the proper framework call to mark a test as skipped. Currently, the
> test is considered good, only the printout mentions the skip.
> 
> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>

Sashiko has reviewed this patch and found no issues. It looks great!

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260916213108.9234-5-wsa+renesas@sang-engineering.com?part=3

Re: [PATCH 0/3] selftest: rtc: rtctest: improve handling of skipped tests

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 470 bytes --]

On Wed, Sep 16, 2026 at 11:31:08PM +0200, Wolfram Sang wrote:
> Here are some fixes for issues I found in 'rtctest' while checking RTC
> drivers. Details are in the patch descriptions. Patches 1+2 have been
> tested on HW. Patch 3 has not been tested, just follows the pattern of
> patch 2, but may need more discussion in general. Please read there.
> Looking forward to comments!

I guess at least patches 1+2 are good to go? Who would pick them,
Alexandre or Shuah?


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

Re: [PATCH 0/3] selftest: rtc: rtctest: improve handling of skipped tests

[Alexandre Belloni]

On 08/10/2026 21:48:16+0200, Wolfram Sang wrote:
> On Wed, Sep 16, 2026 at 11:31:08PM +0200, Wolfram Sang wrote:
> > Here are some fixes for issues I found in 'rtctest' while checking RTC
> > drivers. Details are in the patch descriptions. Patches 1+2 have been
> > tested on HW. Patch 3 has not been tested, just follows the pattern of
> > patch 2, but may need more discussion in general. Please read there.
> > Looking forward to comments!
> 
> I guess at least patches 1+2 are good to go? Who would pick them,
> Alexandre or Shuah?
> 

We discussed it at LPC, I'll take the patches.

-- 
Alexandre Belloni, co-owner and COO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com