kernel/linux.git/tools/testing/selftests/cgroup/test_cpu.c, branch v6.18.22

selftests: cgroup: Use values_close_report in test_cpu

2025-10-15T15:00:59+00:00

Convert test_cpu to use the newly added values_close_report() helper to print detailed diagnostics when a tolerance check fails. This provides clearer insight into deviations while run in the CI. Signed-off-by: Sebastian Chlad Signed-off-by: Tejun Heo

selftests/cgroup: fix cpu.max tests

2025-07-17T18:12:19+00:00

Current cpu.max tests (both the normal one and the nested one) are broken. They setup cpu.max with 1000 us quota and the default period (100,000 us). A cpu hog is run for a duration of 1s as per wall clock time. This corresponds to 10 periods, hence an expected usage of 10,000 us. We want the measured usage (as per cpu.stat) to be close to 10,000 us. Previously, this approximate equality test was done by `!values_close(usage_usec, expected_usage_usec, 95)`: if the absolute difference between usage_usec and expected_usage_usec is greater than 95% of their sum, then we pass. And expected_usage_usec was set to 1,000,000 us. Mathematically, this translates to the following being true for pass: |usage - expected_usage| > (usage + expected_usage)*0.95 If usage > expected_usage: usage - expected_usage > (usage + expected_usage)*0.95 0.05*usage > 1.95*expected_usage usage > 39*expected_usage = 39s If usage < expected_usage: expected_usage - usage > (usage + expected_usage)*0.95 0.05*expected_usage > 1.95*usage usage < 0.0256*expected_usage = 25,600 us Combined, Pass if usage < 25,600 us or > 39 s, which makes no sense given that all we need is for usage_usec to be close to 10,000 us. Fix this by explicitly calcuating the expected usage duration based on the configured quota, default period, and the duration, and compare usage_usec and expected_usage_usec using values_close() with a 10% error margin. Also, use snprintf to get the quota string to write to cpu.max instead of hardcoding the quota, ensuring a single source of truth. Remove the check comparing user_usec and expected_usage_usec, since on running this test modified with printfs, it's seen that user_usec and usage_usec can regularly exceed the theoretical expected_usage_usec: $ sudo ./test_cpu user: 10485, usage: 10485, expected: 10000 ok 1 test_cpucg_max user: 11127, usage: 11127, expected: 10000 ok 2 test_cpucg_max_nested $ sudo ./test_cpu user: 10286, usage: 10286, expected: 10000 ok 1 test_cpucg_max user: 10404, usage: 11271, expected: 10000 ok 2 test_cpucg_max_nested Hence, a values_close() check of usage_usec and expected_usage_usec is sufficient. Fixes: a79906570f9646ae17 ("cgroup: Add test_cpucg_max_nested() testcase") Fixes: 889ab8113ef1386c57 ("cgroup: Add test_cpucg_max() testcase") Acked-by: Michal Koutný Signed-off-by: Shashank Balaji Signed-off-by: Tejun Heo

selftests/cgroup: Fix compile error in test_cpu.c

2024-10-15T01:05:39+00:00

When compiling the cgroup selftests with the following command: make -C tools/testing/selftests/cgroup/ the compiler complains as below: test_cpu.c: In function ‘test_cpucg_nice’: test_cpu.c:284:39: error: incompatible type for argument 2 of ‘hog_cpus_timed’ 284 | hog_cpus_timed(cpucg, param); | ^~~~~ | | | struct cpu_hog_func_param test_cpu.c:132:53: note: expected ‘void *’ but argument is of type ‘struct cpu_hog_func_param’ 132 | static int hog_cpus_timed(const char *cgroup, void *arg) | ~~~~~~^~~ Fix it by passing the address of param to hog_cpus_timed(). Fixes: 2e82c0d4562a ("cgroup/rstat: Selftests for niced CPU statistics") Signed-off-by: Xiu Jianfeng Signed-off-by: Tejun Heo

cgroup/rstat: Selftests for niced CPU statistics

2024-10-08T18:50:54+00:00

Creates a cgroup with a single nice CPU hog process running. fork() is called to generate the nice process because un-nicing is not possible (see man nice(3)). If fork() was not used to generate the CPU hog, we would run the rest of the cgroup selftest suite as a nice process. Signed-off-by: Joshua Hahn Signed-off-by: Tejun Heo

Revert "selftests/cgroup: Drop define _GNU_SOURCE"

2024-05-20T15:00:15+00:00

This reverts commit c1457d9aad5ee2feafcf85aa9a58ab50500159d2. The framework change to add D_GNU_SOURCE to KHDR_INCLUDES to Makefile, lib.mk, and kselftest_harness.h is reverted as it is causing build failures and warnings. Revert this change as this change depends on the framework change. Reported-by: Mark Brown Signed-off-by: Shuah Khan

selftests/cgroup: Drop define _GNU_SOURCE

2024-05-13T17:02:36+00:00

_GNU_SOURCE is provided by lib.mk, so it should be dropped to prevent redefinition warnings. Signed-off-by: Edward Liaw Signed-off-by: Tejun Heo

selftests/cgroup: cpu_hogger init: use {} instead of {NULL}

2024-05-03T19:06:09+00:00

First of all, in order to build with clang at all, one must first apply Valentin Obst's build fix for LLVM [1]. Once that is done, then when building with clang, via: make LLVM=1 -C tools/testing/selftests ...clang generates warning here, because struct cpu_hogger has multiple fields, and the code is initializing an array of these structs, and it is incorrect to specify a single NULL value as the initializer. Fix this by initializing with {}, so that the compiler knows to use default initializer values for all fields in each array entry. [1] https://lore.kernel.org/all/20240329-selftests-libmk-llvm-rfc-v1-1-2f9ed7d1c49f@valentinobst.de/ Signed-off-by: John Hubbard Reviewed-by: Roman Gushchin Signed-off-by: Tejun Heo

selftests: cgroup: skip test_cgcore_lesser_ns_open when cgroup2 mounted without nsdelegate

2024-04-03T19:42:13+00:00

The test case test_cgcore_lesser_ns_open only tasks effect when cgroup2 is mounted with "nsdelegate" mount option. If it misses this option, or is remounted without "nsdelegate", the test case will fail. For example, running bpf/test_cgroup_storage first, and then run cgroup/test_core will fail on test_cgcore_lesser_ns_open. Skip it if "nsdelegate" is not detected in cgroup2 mount options. Fixes: bf35a7879f1d ("selftests: cgroup: Test open-time cgroup namespace usage for migration checks") Signed-off-by: Tianchen Ding Reviewed-by: Muhammad Usama Anjum Signed-off-by: Tejun Heo

cgroup: Add test_cpucg_max_nested() testcase

2022-04-25T17:27:31+00:00

The cgroup cpu controller selftests have a test_cpucg_max() testcase that validates the behavior of the cpu.max knob. Let's also add a testcase that verifies that the behavior works correctly when set on a nested cgroup. Signed-off-by: David Vernet Signed-off-by: Tejun Heo

cgroup: Add test_cpucg_max() testcase

2022-04-25T17:27:31+00:00

The cgroup cpu controller test suite has a number of testcases that validate the expected behavior of the cpu.weight knob, but none for cpu.max. This testcase fixes that by adding a testcase for cpu.max as well. Signed-off-by: David Vernet Signed-off-by: Tejun Heo