This commit refactored the hrtimer_test and provided significantly
improved test-cases for both SMP and non-SMP.
Signed-off-by: ouyangxiangzhen <ouyangxiangzhen@xiaomi.com>
For QEMU, virtual CPUs can be preempted by any high priority thread in
test environments. This can cause the timer to be triggered later than
expected. This commit increased the tolerant latency to avoid the test
failures.
Signed-off-by: ouyangxiangzhen <ouyangxiangzhen@xiaomi.com>
This commit fixed the hrtimer test.
In a QEMU environment, the hrtimer latency can be arbitrary because vCPUs can be preempted. We can not assert the latency. we can only issue alerts for excessively high latency.
Signed-off-by: ouyangxiangzhen <ouyangxiangzhen@xiaomi.com>
In an effort to avoid unexpected behaviour from users not calibrating
BOARD_LOOPSPERMSEC, a compilation error occurs when it is undefined. On
some systems, this is allowed (those that use timer implementations
instead of busy-looping for delays). In these cases, we should busy-loop
using the clock time instead of relying on a possibly
undefined/uncalibrated macro.
Signed-off-by: Matteo Golin <matteo.golin@gmail.com>
Integrate up_perf_gettime() test into ostest suite with comprehensive
test coverage including monotonicity verification, interval statistics,
and frequency validation. The test verifies performance event counter
functionality across 5 independent test cases with proper error handling.
Signed-off-by: hujun5 <hujun5@xiaomi.com>
Since the kernel spinlocks can only work in flat mode, this commit
allowed spinlock_test only in flat mode.
Signed-off-by: ouyangxiangzhen <ouyangxiangzhen@xiaomi.com>
This commit refactored the spinlock test for better accuracy and
minimized jitters introduced by scheduling.
Signed-off-by: ouyangxiangzhen <ouyangxiangzhen@xiaomi.com>
When expanding the macro VERIY, (0 == pthread_barrier_wait(¶m->pub->barrier)) < 0 always false.
Regardless of the value of its operands.
Signed-off-by: anpeiyun <anpeiyun@xiaomi.com>
If we updated `callback_cnt` before the `triggered_tick` in the wdog timer callback, the `wdtest_rand` might failed randomly. This commit fixed the synchronizing bug by swapping the execution order of updating.
Signed-off-by: ouyangxiangzhen <ouyangxiangzhen@xiaomi.com>
Initialize the flags variable to 0 to fix compiler warning about
potential use of uninitialized variable. The flags variable is
conditionally assigned in wdtest_rand() based on callback count.
Signed-off-by: hujun5 <hujun5@xiaomi.com>
Inline enter_critical_section function calls in performance-critical
paths to reduce function call overhead while maintaining consistent
semantics, improving overall system latency and responsiveness in
real-time scenarios.
Signed-off-by: hujun5 <hujun5@xiaomi.com>
fix:
Building NuttX...
/d/a/nuttx-apps/nuttx-apps/sources/apps/Application.mk:238: target 'signest.c.d.a.nuttx-apps.nuttx- apps.sources.apps.testing.ostest.o' given more than once in the same rule
Signed-off-by: Chengdong Wang <wangchengdong@lixiang.com>
Fix dependency issue when signals are partially or fully enabled
Co-authored-by: guoshichao <guoshichao@xiaomi.com>
Signed-off-by: Chengdong Wang <wangchengdong@lixiang.com>
The current code incorrectly used the CONFIG_SCHED_EVENTS
macro in ostest.h for hrtimer_test(). This has been corrected to use CONFIG_HRTIMER.
Signed-off-by: Chengdong Wang wangchengdong@lixiang.com
Add functional tests for the newly added hrtimer APIs,
including hrtimer_init(), hrtimer_start(), and hrtimer_cancel().
Signed-off-by: Chengdong Wang <wangchengdong@lixiang.com>
If we set CONFIG_SMP_DEFAULT_CPUSET=1
This will cause the value of cpuset to be 0x1,
((1 << CONFIG_SMP_NCPUS) - 1) will result in the cpuset being 0x11,
leading to test case errors.
Signed-off-by: hujun5 <hujun5@xiaomi.com>
Before patch we never detect the roundrobin fail.
After patch use a array to detect if the calculation swapped when
processing.
Signed-off-by: buxiasen <buxiasen@xiaomi.com>
https://github.com/apache/nuttx/pull/17293 fixed
wd_start bug, and this patch fixed the related ostest
accordingly
Signed-off-by: Chengdong Wang <wangchengdong@lixiang.com>
The peridiodical wdog test cases should be removed after removing of the periodical wdog APIs.
Signed-off-by: ouyangxiangzhen <ouyangxiangzhen@xiaomi.com>
- Add another signal action, "interfere_action", and signal the interfere thread
- In both signal action functions (waiter and interfere), check that they get executed
in correct thread's context
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
"kill" sends the signal to all the the threads in the group. The intention of
the test is to send signals only to the "waiter" thread.
Running signal actions in parallel is not compatible behaviour with
the signest test, which assumes that signals are being run one after another.
For example running signals 38 and 40 in parallel on two threads/two cpus
would cause the test incorrectly fail on "even signals nested".
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
Add a new test for moving mutexes to verify behavior when a mutex is relocated.
* Introduced mutex_thread_args_t structure for thread arguments
* Created moved_mutex_thread_func to handle mutex operations in threads
* Updated mutex_test to include mutex_move_test for comprehensive testing
Signed-off-by: Huang Qi <huangqi3@xiaomi.com>
Summary:
In the case of slow overall system response (such as when MM_KASAN is turned on), the nxevent case work thread will not be executed and will be switched back to the main thread, so that the event does not get the expected result and the case fails.
By adjusting the thread priority, the work thread can be scheduled to avoid the expected result failure
Signed-off-by: chenrun1 <chenrun1@xiaomi.com>
Clean up code by removing redundant and unused header file includes that were identified through static analysis.
This improves code readability and reduces unnecessary dependencies.
Signed-off-by: Huang Qi <huangqi3@xiaomi.com>
Most tools used for compliance and SBOM generation use SPDX identifiers
This change brings us a step closer to an easy SBOM generation.
Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
