[LTP] [PATCH v14] thermal: add new test group

[LTP] [PATCH v14] thermal: add new test group

[Piotr Kubaj]

Currently consists of only one test for the CPU package thermal sensor
interface for Intel platforms.
It works by checking the initial count of thermal interrupts. Then it
decreases the threshold for sending a thermal interrupt to just above
the current temperature and runs a workload on the CPU. Finally, it
restores the original thermal threshold and checks whether the number
of thermal interrupts increased.

Signed-off-by: Piotr Kubaj <piotr.kubaj@intel.com>
---
Addresses Andrea's feedback, adding support for testing all
the matching thermal zones.
 runtest/thermal                               |   3 +
 testcases/kernel/Makefile                     |   1 +
 testcases/kernel/thermal/.gitignore           |   1 +
 testcases/kernel/thermal/Makefile             |   9 +
 .../kernel/thermal/thermal_interrupt_events.c | 230 ++++++++++++++++++
 5 files changed, 244 insertions(+)
 create mode 100644 runtest/thermal
 create mode 100644 testcases/kernel/thermal/.gitignore
 create mode 100644 testcases/kernel/thermal/Makefile
 create mode 100644 testcases/kernel/thermal/thermal_interrupt_events.c

diff --git a/runtest/thermal b/runtest/thermal
new file mode 100644
index 000000000..57e3d29f8
--- /dev/null
+++ b/runtest/thermal
@@ -0,0 +1,3 @@
+# Thermal driver API
+# https://docs.kernel.org/driver-api/thermal/
+thermal_interrupt_events thermal_interrupt_events
diff --git a/testcases/kernel/Makefile b/testcases/kernel/Makefile
index 98fd45a9d..ac816e4e8 100644
--- a/testcases/kernel/Makefile
+++ b/testcases/kernel/Makefile
@@ -36,6 +36,7 @@ SUBDIRS			+= connectors \
 			   sched \
 			   security \
 			   sound \
+			   thermal \
 			   tracing \
 			   uevents \
 			   watchqueue \
diff --git a/testcases/kernel/thermal/.gitignore b/testcases/kernel/thermal/.gitignore
new file mode 100644
index 000000000..1090bdad8
--- /dev/null
+++ b/testcases/kernel/thermal/.gitignore
@@ -0,0 +1 @@
+thermal_interrupt_events
diff --git a/testcases/kernel/thermal/Makefile b/testcases/kernel/thermal/Makefile
new file mode 100644
index 000000000..4657c3fb3
--- /dev/null
+++ b/testcases/kernel/thermal/Makefile
@@ -0,0 +1,9 @@
+# SPDX-License-Identifier: GPL-2.0-or-later
+# Copyright (c) 2025, Intel Corporation. All rights reserved.
+# Author:Piotr Kubaj <piotr.kubaj@intel.com>
+
+top_srcdir             ?= ../../..
+
+include $(top_srcdir)/include/mk/testcases.mk
+
+include $(top_srcdir)/include/mk/generic_leaf_target.mk
diff --git a/testcases/kernel/thermal/thermal_interrupt_events.c b/testcases/kernel/thermal/thermal_interrupt_events.c
new file mode 100644
index 000000000..0317a0ff7
--- /dev/null
+++ b/testcases/kernel/thermal/thermal_interrupt_events.c
@@ -0,0 +1,230 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+/*
+ * Copyright (C) 2026 Intel - http://www.intel.com/
+ */
+
+/*\
+ * Tests the CPU package thermal sensor interface for Intel platforms.
+
+ * Works by checking the initial count of thermal interrupts. Then it
+ * decreases the threshold for sending a thermal interrupt to just above
+ * the current temperature and runs a workload on the CPU. Finally, it restores
+ * the original thermal threshold and checks whether the number of thermal
+ * interrupts increased.
+ */
+
+#include <ctype.h>
+#include "tst_safe_stdio.h"
+#include "tst_test.h"
+#include "tst_timer_test.h"
+
+#define	RUNTIME		30
+#define	SLEEP		10
+#define	TEMP_INCREMENT	10
+
+static bool x86_pkg_temp_tz_found, *x86_pkg_temp_tz;
+static char temp_path[PATH_MAX], trip_path[PATH_MAX];
+static int nproc, temp_high, temp, trip, tz_counter;
+static uint64_t *interrupt_init, *interrupt_later;
+
+static void read_interrupts(uint64_t *interrupts, const int nproc)
+{
+	bool interrupts_found = false;
+	char line[8192];
+
+	memset(interrupts, 0, nproc * sizeof(*interrupts));
+	FILE *fp = SAFE_FOPEN("/proc/interrupts", "r");
+
+	while (fgets(line, sizeof(line), fp)) {
+		if (strstr(line, "Thermal event interrupts")) {
+			interrupts_found = true;
+			char *ptr = strchr(line, ':');
+
+			for (int i = 0; i < nproc; i++) {
+				char *endptr;
+
+				while (*ptr && !isdigit(*ptr))
+					ptr++;
+
+				errno = 0;
+
+				interrupts[i] = strtoull(ptr, &endptr, 10);
+
+				if (ptr == endptr)
+					tst_brk(TBROK, "interrupt not found");
+
+				if (errno == ERANGE)
+					tst_brk(TCONF, "interrupt out of range");
+
+				ptr = endptr;
+				tst_res(TDEBUG, "interrupts[%d]: %ld", i, interrupts[i]);
+			}
+			break;
+		}
+	}
+	SAFE_FCLOSE(fp);
+	if (!interrupts_found)
+		tst_brk(TCONF, "No Thermal event interrupts line in /proc/interrupts");
+}
+
+static void setup(void)
+{
+	char line[8192];
+
+	nproc = tst_ncpus();
+	tst_res(TDEBUG, "Number of logical cores: %d", nproc);
+	interrupt_init = calloc(nproc, sizeof(uint64_t));
+	interrupt_later = calloc(nproc, sizeof(uint64_t));
+
+	DIR *dir = SAFE_OPENDIR("/sys/class/thermal/");
+	struct dirent *entry;
+
+	while ((entry = SAFE_READDIR(dir))) {
+		if ((!strncmp(entry->d_name, "thermal_zone", sizeof("thermal_zone") - 1)))
+			tz_counter++;
+	}
+	SAFE_CLOSEDIR(dir);
+	tst_res(TDEBUG, "Found %d thermal zone(s)", tz_counter);
+
+	x86_pkg_temp_tz = calloc(tz_counter, sizeof(bool));
+
+	for (int i = 0; i < tz_counter; i++) {
+		char path[PATH_MAX];
+
+		snprintf(path, PATH_MAX, "/sys/class/thermal/thermal_zone%d/type", i);
+		tst_res(TDEBUG, "Checking whether %s is x86_pkg_temp", path);
+
+		SAFE_FILE_SCANF(path, "%s", line);
+		if (strstr(line, "x86_pkg_temp")) {
+			tst_res(TDEBUG, "Thermal zone %d uses x86_pkg_temp", i);
+			x86_pkg_temp_tz[i] = true;
+			x86_pkg_temp_tz_found = true;
+		}
+	}
+
+	if (!x86_pkg_temp_tz_found)
+		tst_brk(TCONF, "No thermal zone uses x86_pkg_temp");
+}
+
+static void *cpu_workload(double run_time)
+{
+	tst_timer_start(CLOCK_MONOTONIC);
+	int num = 2;
+
+	while (!tst_timer_expired_ms(run_time * 1000)) {
+		for (int i = 2; i * i <= num; i++) {
+			if (num % i == 0)
+				break;
+		}
+		num++;
+		SAFE_FILE_SCANF(temp_path, "%d", &temp);
+
+		if (temp > temp_high)
+			break;
+	}
+	return NULL;
+}
+
+static void test_zone(int i)
+{
+	int sleep_time = SLEEP;
+	double run_time = RUNTIME;
+
+	snprintf(temp_path, PATH_MAX, "/sys/class/thermal/thermal_zone%d/temp", i);
+	tst_res(TINFO, "Testing %s", temp_path);
+	SAFE_FILE_SCANF(temp_path, "%d", &temp);
+	if (temp < 0)
+		tst_brk(TBROK, "Unexpected zone temperature value %d", temp);
+
+	tst_res(TDEBUG, "Current temperature for %s: %d", temp_path, temp);
+
+	temp_high = temp + TEMP_INCREMENT;
+
+	snprintf(trip_path, PATH_MAX, "/sys/class/thermal/thermal_zone%d/trip_point_1_temp", i);
+
+	tst_res(TDEBUG, "Setting new trip_point_1_temp value: %d", temp_high);
+	SAFE_FILE_SCANF(trip_path, "%d", &trip);
+	SAFE_FILE_PRINTF(trip_path, "%d", temp_high);
+
+	while (sleep_time > 0) {
+		tst_res(TDEBUG, "Running for %f seconds, then sleeping for %d seconds", run_time, sleep_time);
+
+		for (int j = 0; j < nproc; j++) {
+			if (!SAFE_FORK()) {
+				cpu_workload(run_time);
+				exit(0);
+			}
+		}
+
+		tst_reap_children();
+
+		SAFE_FILE_SCANF(temp_path, "%d", &temp);
+		tst_res(TDEBUG, "Temperature for %s after a test: %d", temp_path, temp);
+
+		if (temp > temp_high)
+			break;
+		sleep(sleep_time--);
+		run_time -= 3;
+	}
+
+}
+
+static void cleanup(void)
+{
+	if (x86_pkg_temp_tz_found) {
+		for (int i = 0; i < tz_counter; i++) {
+			if (x86_pkg_temp_tz[i]) {
+				snprintf(trip_path, PATH_MAX, "/sys/class/thermal/thermal_zone%d/trip_point_1_temp", i);
+				SAFE_FILE_PRINTF(trip_path, "%d", trip);
+			}
+		}
+	}
+
+	free(x86_pkg_temp_tz);
+	free(interrupt_init);
+	free(interrupt_later);
+}
+
+static void run(void)
+{
+	for (int i = 0; i < tz_counter; i++) {
+		if (x86_pkg_temp_tz[i]) {
+			read_interrupts(interrupt_init, nproc);
+			test_zone(i);
+			read_interrupts(interrupt_later, nproc);
+			for (int i = 0; i < nproc; i++) {
+				if (interrupt_later[i] < interrupt_init[i])
+					tst_res(TFAIL, "CPU %d interrupt counter: %ld (previous: %ld)",
+						i, interrupt_later[i], interrupt_init[i]);
+			}
+			if (temp <= temp_high)
+				tst_res(TFAIL, "Zone temperature is not rising as expected");
+			else
+				tst_res(TPASS, "x86 package thermal interrupt triggered");
+		}
+	}
+
+}
+
+static struct tst_test test = {
+	.cleanup = cleanup,
+	.forks_child = 1,
+	.needs_drivers = (const char *const []) {
+		"x86_pkg_temp_thermal",
+		NULL
+	},
+	.min_runtime = 180,
+	.needs_root = 1,
+	.setup = setup,
+	.supported_archs = (const char *const []) {
+		"x86",
+		"x86_64",
+		NULL
+	},
+	.tags = (const struct tst_tag[]) {
+		{"linux-git", "9635c586a559ba0e45b2bfbff79c937ddbaf1a62"},
+		{}
+	},
+	.test_all = run
+};
-- 
2.47.3

---------------------------------------------------------------------
Intel Technology Poland sp. z o.o.
ul. Slowackiego 173 | 80-298 Gdansk | Sad Rejonowy Gdansk Polnoc | VII Wydzial Gospodarczy Krajowego Rejestru Sadowego - KRS 101882 | NIP 957-07-52-316 | Kapital zakladowy 200.000 PLN.
Spolka oswiadcza, ze posiada status duzego przedsiebiorcy w rozumieniu ustawy z dnia 8 marca 2013 r. o przeciwdzialaniu nadmiernym opoznieniom w transakcjach handlowych.

Ta wiadomosc wraz z zalacznikami jest przeznaczona dla okreslonego adresata i moze zawierac informacje poufne. W razie przypadkowego otrzymania tej wiadomosci, prosimy o powiadomienie nadawcy oraz trwale jej usuniecie; jakiekolwiek przegladanie lub rozpowszechnianie jest zabronione.
This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). If you are not the intended recipient, please contact the sender and delete all copies; any review or distribution by others is strictly prohibited.


-- 
Mailing list info: https://lists.linux.it/listinfo/ltp

Re: [LTP] [PATCH v14] thermal: add new test group

[Petr Vorel]

Hi Piotr,

Minor notes:

> +static void setup(void)
> +{
> +	char line[8192];
> +
> +	nproc = tst_ncpus();
> +	tst_res(TDEBUG, "Number of logical cores: %d", nproc);
> +	interrupt_init = calloc(nproc, sizeof(uint64_t));
> +	interrupt_later = calloc(nproc, sizeof(uint64_t));
You correctly had SAFE_CALLOC() in v13, here you accident start using calloc()
again.

> +
> +	DIR *dir = SAFE_OPENDIR("/sys/class/thermal/");
> +	struct dirent *entry;
> +
> +	while ((entry = SAFE_READDIR(dir))) {
> +		if ((!strncmp(entry->d_name, "thermal_zone", sizeof("thermal_zone") - 1)))
> +			tz_counter++;
> +	}
> +	SAFE_CLOSEDIR(dir);
> +	tst_res(TDEBUG, "Found %d thermal zone(s)", tz_counter);
> +
> +	x86_pkg_temp_tz = calloc(tz_counter, sizeof(bool));
And here as well.

...
> +	memset(interrupts, 0, nproc * sizeof(*interrupts));
> +	FILE *fp = SAFE_FOPEN("/proc/interrupts", "r");
> +
> +	while (fgets(line, sizeof(line), fp)) {
> +		if (strstr(line, "Thermal event interrupts")) {
> +			interrupts_found = true;
> +			char *ptr = strchr(line, ':');
> +
> +			for (int i = 0; i < nproc; i++) {
> +				char *endptr;
> +
> +				while (*ptr && !isdigit(*ptr))
> +					ptr++;
> +
> +				errno = 0;
> +
> +				interrupts[i] = strtoull(ptr, &endptr, 10);
> +
> +				if (ptr == endptr)
> +					tst_brk(TBROK, "interrupt not found");
Printing CPU would help to debug on error, right?
					tst_brk(TBROK, "CPU %d: interrupt not found", nproc);

> +
> +				if (errno == ERANGE)
> +					tst_brk(TCONF, "interrupt out of range");
I would expect this is quite serious error (test bug), therefore TBROK should be
used, right?  TCONF is really for skipping the test due SUT not suitable for
running the test. But if the only error, it can be changed before merge.

					tst_brk(TBROK, "CPU %d: interrupt out of range", nproc);
...
> +static struct tst_test test = {
> +	.cleanup = cleanup,
> +	.forks_child = 1,
> +	.needs_drivers = (const char *const []) {
> +		"x86_pkg_temp_thermal",
> +		NULL
> +	},
> +	.min_runtime = 180,
You had .min_runtime = 5 sec in the v13, now again back 3 min.
As I wrote earlier, using tst_set_runtime() would be much better than expect 3
min run (on my laptop the test needs few sec, but it will waste time in case of
the test get stuck for whatever reason or might not enough for really big machine).

+#define	TEST_RUNTIME	3
 #define	RUNTIME		30
 #define	SLEEP		10
 #define	TEMP_INCREMENT	10
...
@@ -73,9 +74,10 @@ static void setup(void)
 	char line[8192];
 
 	nproc = tst_ncpus();
+	tst_set_runtime(nproc * TEST_RUNTIME);

This will lead to run in my machine:
tst_test.c:1887: TINFO: Overall timeout per run is 0h 00m 30s
tst_test.c:1908: TINFO: Updating runtime to 0h 00m 12s
tst_test.c:1887: TINFO: Overall timeout per run is 0h 00m 42s

I'd be ok to apply the following changes before merge. But I'd really prefer
Cyril to give ack to this before merge.
Reviewed-by: Petr Vorel <pvorel@suse.cz>

Kind regards,
Petr

> +	.needs_root = 1,
> +	.setup = setup,
> +	.supported_archs = (const char *const []) {
> +		"x86",
> +		"x86_64",
> +		NULL
> +	},
> +	.tags = (const struct tst_tag[]) {
> +		{"linux-git", "9635c586a559ba0e45b2bfbff79c937ddbaf1a62"},
> +		{}
> +	},
> +	.test_all = run
> +};

+++ testcases/kernel/thermal/thermal_interrupt_events.c
@@ -19,6 +19,7 @@
 #include "tst_test.h"
 #include "tst_timer_test.h"
 
+#define	TEST_RUNTIME	3
 #define	RUNTIME		30
 #define	SLEEP		10
 #define	TEMP_INCREMENT	10
@@ -52,10 +53,10 @@ static void read_interrupts(uint64_t *interrupts, const int nproc)
 				interrupts[i] = strtoull(ptr, &endptr, 10);
 
 				if (ptr == endptr)
-					tst_brk(TBROK, "interrupt not found");
+					tst_brk(TBROK, "CPU %d: interrupt not found", nproc);
 
 				if (errno == ERANGE)
-					tst_brk(TCONF, "interrupt out of range");
+					tst_brk(TBROK, "CPU %d: interrupt out of range", nproc);
 
 				ptr = endptr;
 				tst_res(TDEBUG, "interrupts[%d]: %ld", i, interrupts[i]);
@@ -73,9 +74,10 @@ static void setup(void)
 	char line[8192];
 
 	nproc = tst_ncpus();
+	tst_set_runtime(nproc * TEST_RUNTIME);
 	tst_res(TDEBUG, "Number of logical cores: %d", nproc);
-	interrupt_init = calloc(nproc, sizeof(uint64_t));
-	interrupt_later = calloc(nproc, sizeof(uint64_t));
+	interrupt_init = SAFE_CALLOC(nproc, sizeof(uint64_t));
+	interrupt_later = SAFE_CALLOC(nproc, sizeof(uint64_t));
 
 	DIR *dir = SAFE_OPENDIR("/sys/class/thermal/");
 	struct dirent *entry;
@@ -87,7 +89,7 @@ static void setup(void)
 	SAFE_CLOSEDIR(dir);
 	tst_res(TDEBUG, "Found %d thermal zone(s)", tz_counter);
 
-	x86_pkg_temp_tz = calloc(tz_counter, sizeof(bool));
+	x86_pkg_temp_tz = SAFE_CALLOC(tz_counter, sizeof(bool));
 
 	for (int i = 0; i < tz_counter; i++) {
 		char path[PATH_MAX];
@@ -214,7 +216,6 @@ static struct tst_test test = {
 		"x86_pkg_temp_thermal",
 		NULL
 	},
-	.min_runtime = 180,
 	.needs_root = 1,
 	.setup = setup,
 	.supported_archs = (const char *const []) {

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp

Re: [LTP] [PATCH v14] thermal: add new test group

[Andrea Cervesato via ltp]

Hi Piotr,

> +static int nproc, temp_high, temp, trip, tz_counter;

`trip` is a single int, but cleanup() uses it to restore trip_point_1_temp
for ALL x86_pkg_temp zones. If there are multiple such zones, each
test_zone() call overwrites `trip` at the SAFE_FILE_SCANF, so only the
last zone's original value survives. The earlier zones get the wrong
value restored.

This needs to be a per-zone array, e.g. `static int *trip_orig;`
allocated alongside x86_pkg_temp_tz in setup(), with each zone's
original trip value saved individually.

> +	interrupt_init = calloc(nproc, sizeof(uint64_t));
> +	interrupt_later = calloc(nproc, sizeof(uint64_t));

[...]

> +	x86_pkg_temp_tz = calloc(tz_counter, sizeof(bool));

LTP has SAFE_CALLOC() in tst_safe_macros.h. Please use it instead of
raw calloc() — it handles the NULL check and calls tst_brk(TBROK) on
failure.

> +			tst_res(TDEBUG, "interrupts[%d]: %ld", i, interrupts[i]);

interrupts[i] is uint64_t but printed with %ld. Use PRIu64:

    tst_res(TDEBUG, "interrupts[%d]: %" PRIu64, i, interrupts[i]);

Same issue below in run():

> +				tst_res(TFAIL, "CPU %d interrupt counter: %ld (previous: %ld)",
> +					i, interrupt_later[i], interrupt_init[i]);

> +		SAFE_FILE_SCANF(path, "%s", line);

%s without a width limit can overflow line[8192]. Use "%8191s".

> +static void *cpu_workload(double run_time)
> +{

[...]

> +	return NULL;
> +}

cpu_workload() returns void * as if it were a pthread start routine, but
it is only called directly from the child process. It should return void.

> +	for (int i = 0; i < tz_counter; i++) {
> +		if (x86_pkg_temp_tz[i]) {
> +			read_interrupts(interrupt_init, nproc);
> +			test_zone(i);
> +			read_interrupts(interrupt_later, nproc);
> +			for (int i = 0; i < nproc; i++) {

Inner `i` shadows the outer loop variable `i`. Please rename to `j` or
`cpu` to avoid confusion.

Regards,
--
Andrea Cervesato
SUSE QE Automation Engineer Linux
andrea.cervesato@suse.com

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp