power_management: rewrite runpwtests04.sh in C
PATCH v6 1/1 power_management: rewrite runpwtests04.sh in C
As part of the ongoing effort to reduce shell-based tests in LTP, rewrite the cpuidle sysfs smoke test in C using the modern LTP test API.
The new implementation preserves the original test semantics while removing shell dependencies.
---
Changes in v6:
- Expand test coverage to include 4 core cpuidle sysfs ABIs.
- Add verification of read() system call return values (bytes read).
- Add negative testing for 'current_governor' to verify EINVAL return value.
- Replace manual access() check with LTP .save_restore.
- Link to v5: https://lore.kernel.org/ltp/[email protected]
Changes in v5:
- Remove runpwtests04.sh
- Link to v4: https://lore.kernel.org/ltp/[email protected]
Changes in v4:
- Fix patch application failure reported by CI.
- Link to v3: https://lore.kernel.org/ltp/[email protected]
Changes in v3:
- Replace SAFE_OPEN() with open() to convert ENOENT to TCONF.
- Add a cleanup function.
- Link to v2: https://lore.kernel.org/ltp/[email protected]
Changes in v2:
- Update runtest entry
- Clarify commit message
- Link to v1: https://lore.kernel.org/ltp/[email protected]
--- runtest/power_management_tests | 2 +- testcases/kernel/power_management/.gitignore | 1 + testcases/kernel/power_management/cpuidle01.c | 90 +++++++++++++++++++ .../kernel/power_management/runpwtests04.sh | 58 ------------ 4 files changed, 92 insertions(+), 59 deletions(-) create mode 100644 testcases/kernel/power_management/cpuidle01.c delete mode 100755 testcases/kernel/power_management/runpwtests04.sh
runtest/power_management_tests +1 -1
@@ -1,5 +1,5 @@ #POWER_MANAGEMENT high_freq_hwp_cap_cppc high_freq_hwp_cap_cppc +cpuidle01 cpuidle01 runpwtests03 runpwtests03.sh -runpwtests04 runpwtests04.sh runpwtests06 runpwtests06.sh
testcases/kernel/power_management/.gitignore +1 -0
@@ -1 +1,2 @@ high_freq_hwp_cap_cppc +cpuidle01
testcases/kernel/power_management/cpuidle01.c new file +90 -0
90 lines not on this page · raw message
testcases/kernel/power_management/runpwtests04.sh deleted +0 -58
58 lines not on this page · raw message
--
2.43.0
Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
Hi Jinseok,
Quoted 4 lines · Jinseok Kim
power_management: rewriterunpwtests04.shin CThe new implementation preserves the original test semantics while removing shell dependencies.
The original shell test only checked current_governor_ro and current_driver for readability. The new test also covers available_governors, current_governor (read), and adds a write rejection check against current_governor. That is an extension, not a preservation of semantics. Please update the commit message body to reflect what is actually added.
testcases/kernel/power_management/cpuidle01.c line 42
Quoted 12 lines
testcases/kernel/power_management/cpuidle01.c new file +12
-0+static void verify_cpuidle (unsigned int i) +{ +int ret; +char path[PATH_MAX ]; +char buf[32 ]; + +snprintf (path,sizeof (path),"%s/%s" ,CPUIDLE_PATH , tcases[i].name); + +if (tcases[i].is_writable) + fd =open (path,O_RDWR ); +else + fd =open (path,O_RDONLY );
open() is a tested subject here -- its return code drives the TCONF/TFAIL decision. Bare syscalls that are the subject of the test must be wrapped in TEST(), not called directly. The same applies to read() and write() below.
testcases/kernel/power_management/cpuidle01.c line 59
Quoted 8 lines
testcases/kernel/power_management/cpuidle01.c new file +8
-0+ ret =read (fd, buf,sizeof (buf)); + +if (ret <0 ) +tst_res (TFAIL |TERRNO ,"%s read() failed" , path); +else if (ret ==0 ) +tst_res (TFAIL ,"%s read() returned 0" , path); +else +tst_res (TPASS ,"%s read() successfully" , path);
read() is the primary subject of this test and must be wrapped in TEST() so that TST_RET and TST_ERR are populated.
testcases/kernel/power_management/cpuidle01.c line 62
Quoted 2 lines
testcases/kernel/power_management/cpuidle01.c new file +2
-0+if (tcases[i].is_writable) { +if (write (fd,"invalid_governor" ,16 ) <0 ) {
Same for write().
testcases/kernel/power_management/cpuidle01.c line 83
Quoted 3 lines
testcases/kernel/power_management/cpuidle01.c new file +3
-0+static struct tst_test test = { + .cleanup = cleanup, + .needs_root =1 ,
The doc comment only mentions readability. It does not explain why root is required. Ground Rule 4 requires the reason to appear in the /*\ ... */ block. Root is needed here for writing to current_governor and for the save_restore of that path; please document that.
Quoted 1 line
+ * Verify that selected cpuidle sysfs files are readable.
The test also verifies that writing an invalid governor name to current_governor is rejected with EINVAL. The description should cover both aspects.
Verdict - Needs revision
---
Note:
The agent can sometimes produce false positives although often its findings are genuine. If you find issues with the review, please comment this email or ignore the suggestions.
Regards,
LTP AI Reviewer
Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
Hi Jinseok,
Quoted 5 lines
open()is a tested subject here -- its return code drives the TCONF/TFAIL decision. Bare syscalls that are the subject of the test must be wrapped inTEST(), not called directly. The same applies toread()andwrite()below.
This is a false positive.
Quoted 12 lines
testcases/kernel/power_management/cpuidle01.c line 59
Quoted 9 lines
testcases/kernel/power_management/cpuidle01.c new file +8
-0+ ret =read (fd, buf,sizeof (buf)); + +if (ret <0 ) +tst_res (TFAIL |TERRNO ,"%s read() failed" , path); +else if (ret ==0 ) +tst_res (TFAIL ,"%s read() returned 0" , path); +else +tst_res (TPASS ,"%s read() successfully" , path);
read()is the primary subject of this test and must be wrapped inTEST()so thatTST_RETandTST_ERRare populated.
This is false positive.
Quoted 5 lines
testcases/kernel/power_management/cpuidle01.c line 62
Quoted 3 lines
testcases/kernel/power_management/cpuidle01.c new file +2
-0+if (tcases[i].is_writable) { +if (write (fd,"invalid_governor" ,16 ) <0 ) {Same for
write().
This is false poitive.
Quoted 11 lines
testcases/kernel/power_management/cpuidle01.c line 83
Quoted 4 lines
testcases/kernel/power_management/cpuidle01.c new file +3
-0+static struct tst_test test = { + .cleanup = cleanup, + .needs_root =1 ,The doc comment only mentions readability. It does not explain why root is required. Ground Rule 4 requires the reason to appear in the /*\ ... */ block. Root is needed here for writing to
current_governorand for thesave_restoreof that path; please document that.
Quoted 7 lines
Quoted 1 line
+ * Verify that selected cpuidle sysfs files are readable.
The test also verifies that writing an invalid governor name to
current_governoris rejected with EINVAL. The description should cover both aspects.Verdict - Needs revision
The rest is correct.
--
Andrea Cervesato
SUSE QE Automation Engineer Linux
[email protected]
Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
Quoted 81 lines
runtest/power_management_tests | 2 +- testcases/kernel/power_management/.gitignore | 1 + testcases/kernel/power_management/cpuidle01.c | 90 +++++++++++++++++++ .../kernel/power_management/runpwtests04.sh | 58 ------------ 4 files changed, 92 insertions(+), 59 deletions(-) create mode 100644 testcases/kernel/power_management/cpuidle01.c delete mode 100755 testcases/kernel/power_management/runpwtests04.shruntest/power_management_tests +1
-1@@ -1,5 +1,5 @@ #POWER_MANAGEMENT high_freq_hwp_cap_cppc high_freq_hwp_cap_cppc +cpuidle01 cpuidle01 runpwtests03 runpwtests03.sh -runpwtests04 runpwtests04.sh runpwtests06 runpwtests06.shtestcases/kernel/power_management/.gitignore +1
-0@@ -1 +1,2 @@ high_freq_hwp_cap_cppc +cpuidle01testcases/kernel/power_management/cpuidle01.c new file +49
-0@@ -0,0 +1,90 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* +* Copyright (C) 2026 Jinseok Kim <[email protected]> +*/ + +/*\ +* Basic cpuidle sysfs smoke test. +* +* Verify that selected cpuidle sysfs files are readable. +*/ + +#include <fcntl.h> +#include <unistd.h> + +#include "tst_test.h" + +#define CPUIDLE_PATH "/sys/devices/system/cpu/cpuidle" + +static struct tcase { +const char *name; +bool is_writable; +} tcases[] = { + {"available_governors" , false }, + {"current_driver" , false }, + {"current_governor" , true }, + {"current_governor_ro" , false }, +}; + +static int fd =-1 ; + +static void verify_cpuidle (unsigned int i) +{ +int ret; +char path[PATH_MAX ]; +char buf[32 ]; + +snprintf (path,sizeof (path),"%s/%s" ,CPUIDLE_PATH , tcases[i].name); + +if (tcases[i].is_writable) + fd =open (path,O_RDWR ); +else + fd =open (path,O_RDONLY ); + +if (fd <0 ) { +if (errno ==ENOENT ) +tst_res (TCONF ,"%s not available" , path); +else +tst_res (TFAIL |TERRNO ,"open(%s) failed" , path); +return ;
testcases/kernel/power_management/cpuidle01.c line 59
Quoted 10 lines
testcases/kernel/power_management/cpuidle01.c new file +10
-0+ } + + ret =read (fd, buf,sizeof (buf)); + +if (ret <0 ) +tst_res (TFAIL |TERRNO ,"%s read() failed" , path); +else if (ret ==0 ) +tst_res (TFAIL ,"%s read() returned 0" , path); +else +tst_res (TPASS ,"%s read() successfully" , path);
When we actually bother writing tests maybe we should not do it with the array style, but rather add a function per sysfs file that would actually check that the data are correct.
E.g. we could write a function that would read and parse available_governors and then check that current_governor is in the set of the available_governors.
testcases/kernel/power_management/cpuidle01.c line 69
Quoted 9 lines
testcases/kernel/power_management/cpuidle01.c new file +9
-0+if (tcases[i].is_writable) { +if (write (fd,"invalid_governor" ,16 ) <0 ) { +if (errno ==EINVAL ) +tst_res (TPASS ,"%s rejected invalid input with EINVAL" , path); +else +tst_res (TFAIL |TERRNO ,"%s rejected invalid input with unexpected errno" , path); + }else { +tst_res (TFAIL ,"%s accepted invalid input successfully" , path); + }
This can be just TST_EXP_FAIL2();
Quoted 90 lines
testcases/kernel/power_management/cpuidle01.c new file +21
-0+ } + +SAFE_CLOSE (fd); +} + +static void cleanup (void ) +{ +if (fd !=-1 ) +SAFE_CLOSE (fd); +} + +static struct tst_test test = { + .cleanup = cleanup, + .needs_root =1 , + .tcnt =ARRAY_SIZE (tcases), + .test = verify_cpuidle, + .save_restore = (const struct tst_path_val []) { + {CPUIDLE_PATH "/current_governor" ,NULL ,TST_SR_TCONF }, + {} + }, +};testcases/kernel/power_management/runpwtests04.sh deleted +0
-59@@ -1,58 +0,0 @@ -#! /bin/sh -# -# Copyright (c) International Business Machines Corp., 2001 -# Author: Nageswara R Sastry <[email protected]> -# -# This program is free software; you can redistribute it and#or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation; either version 2 of the License, or -# (at your option) any later version. -# -# This program is distributed in the hope that it will be useful, but -# WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY -# or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License -# for more details. -# -# You should have received a copy of the GNU General Public License -# along with this program; if not, write to the Free Software Foundation, -# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA -# - -export TCID="Power_Management04" -export TST_TOTAL=1 - -. test.sh -. pm_include.sh - -check_cpuidle_sysfs_files() { - RC=0 - if [ -d /sys/devices/system/cpu/cpuidle ] ; then - for files in current_governor_ro current_driver - do - cat /sys/devices/system/cpu/cpuidle/${files} \ - >/dev/null 2>&1 - if [ $? -ne 0 ] ; then - echo "${0}: FAIL: cat ${files}" - RC=1 - fi - done - fi - if [ ${RC} -eq 0 ] ; then - echo "${0}: PASS: Checking cpu idle sysfs files" - else - echo "${0}: FAIL: Checking cpu idle sysfs files" - fi - return $RC -} - -# Checking test environment -check_arch - -# Checking cpuidle sysfs interface files -if check_cpuidle_sysfs_files ; then - tst_resm TPASS "CPUIDLE sysfs tests passed" -else - tst_resm TFAIL "CPUIDLE sysfs tests failed" -fi - -tst_exit --2.43.0
Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
Signature · 5 lines
Cyril Hrubis [email protected] -- Mailing list info: https://lists.linux.it/listinfo/ltp