The test consists of two parts:
- A tester that tries to trigger wrong states of work queue
- A verifier that checks whether the wqueue is still working properly
The tester is trying to queue and cancel work several times with
priority lower/same/higher than the work queue.
Most wrong cases are likely to happen with high priority like:
- If `cancel` never decreases semcount, the count may keep growing
and finally overflow
- If `cancel` is decreasing semcount too much, the `work_thread` may
be waken up less times than expected
The lower/same priority testers are just added for covering other
unexpected situations.
The verifier is trying to queue some works and check they are called as
expected:
- Frist queue a 'sleep' worker, to let a work queue thread be in busy
status and not waiting on sem, while other work queue thread(s) (if
any) still waiting for sem. If sem is in wrong state, it may cause
wrong behavior in either thread waiting/not waiting on the sem.
- Then queue a few count works, if the work queue(s) are still working
properly, these works should finally be all called once.
Signed-off-by: Zhe Weng <wengzhe@xiaomi.com>
Summary:
- This commit shows errno in message queue test
Impact:
- None
Testing:
- Tested with esp32-devkitc:smp on QEMU-7.1
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Summary:
- This commit fixes the task_restart test for the 2nd time execution
Impact:
- None
Testing:
- sim:ostest
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Summary:
- This commit adds ASSERT in case of errors to stop the
execution of ostest
- Also, add some 'ERROR' messages
Impact:
- None
Testing:
- Run ostest with several configs
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
When application is compiled without optimisation the stack usage
is higher. 1K stack of adversary task is getting overflowed and that
leads to crash. Set stack size to CONFIG_DEFAULT_TASK_STACKSIZE for
adversary task
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
Gregory Nutt has submitted the SGA
Haltian Ltd has submitted the SGA
as a result we can migrate the licenses to Apache.
Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
task is blocked by semphore1, signal handler is blocked by semphore2,
after post semphore2, the task must get -EINTR.
Signed-off-by: Zeng Zhaoxiu <zhaoxiu.zeng@gmail.com>
aio.c: In function ‘check_done’:
aio.c:158:41: warning: format ‘%d’ expects argument of type ‘int’, but argument 3 has type ‘ssize_t’ {aka ‘volatile long int’} [-Wformat=]
158 | printf(" list[%d]. result = %d\n", i, aiocbp->aio_result);
| ~^ ~~~~~~~~~~~~~~~~~~
| | |
| int ssize_t {aka volatile long int}
| %ld
aio.c: In function ‘remove_done’:
aio.c:222:41: warning: format ‘%d’ expects argument of type ‘int’, but argument 3 has type ‘ssize_t’ {aka ‘volatile long int’} [-Wformat=]
222 | printf(" list[%d]. result = %d\n", i, aiocbp->aio_result);
| ~^ ~~~~~~~~~~~~~~~~~~
| | |
| int ssize_t {aka volatile long int}
| %ld
sporadic2.c: In function ‘sporadic2_test’:
sporadic2.c:313:10: warning: format ‘%i’ expects argument of type ‘int’, but argument 4 has type ‘long int’ [-Wformat=]
313 | printf("Sporadic 1: prio high %d, low %d, repl %" PRIi32 " ns\n",
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from /home/xiaoxiang/backup/os/nuttx/nuttx/include/stdint.h:35,
from /home/xiaoxiang/backup/os/nuttx/nuttx/include/sys/types.h:32,
from /home/xiaoxiang/backup/os/nuttx/nuttx/include/stdio.h:30,
from sporadic2.c:27:
/home/xiaoxiang/backup/os/nuttx/nuttx/include/arch/inttypes.h:51:24: note: format string is defined here
51 | # define PRIi32 "i"
sporadic2.c:315:10: warning: format ‘%i’ expects argument of type ‘int’, but argument 4 has type ‘long int’ [-Wformat=]
315 | printf("Sporadic 2: prio high %d, low %d, repl %" PRIi32 " ns\n",
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Gregory Nutt is has submitted the SGA
Mark Schulte has submitted the ICLA
Ken Pettit has submitted the ICLA
as a result we can migrate the licenses to Apache.
Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
Gregory Nutt is the copyright holder for those files and he has submitted the
SGA as a result we can migrate the licenses to Apache.
Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
The Linux man page requires that the getopt_long() and getopt_long_only() functions accept arguments to options in a form like:
--option=argument
This PR adds a test that missing functionality that was recently added to NuttX.
This change also fixes an error in string comparison that was working before only because of the way that strings are stored by in linker ELF. The address of the strings were being compared, not the value of the string.
This change effects only the getopt() tests of the OS test.
Tested on a simulator NSH configuration and used to verify the NuttX change.
Fix optind range checking. optind may index through argc (to the NULL argv entry) on the last option since optind is required to always point to the next command line argument.
Add two more test cases were the final thing on the command line is an invalid long option.
Fix a check that used an older, obsoleted hard-coded value that was not updated.
This should have no impact other than to the getopt() test cases of the OS test.
Tested on the simulator using a modified NSH configuration.
Add logic to assure that the getopt() functions parse exactly the correct number of parameters. Previously if the test terminated early, the error would be undetected. Also, prevents indexing past the end of the results array.
Fixes a coding error that was not causing a test failure.
Add some minimal level of testing for invalid options (this found a good bug in getopt_long()).
Affects only the getopt() test cases of the OS test.
Verified using an NSH simulator configuration set up to add apps/testing/ostest as a built-in command.
This new test was used to verify the new implementations of getopt_long() and getopt_long_only() and to verify that there are no regressions to the existing getopts().
Only the OS test is affect by this change.
Verified with the getopt() modifications on a modified version of the sim:nsh configuration.
Results were difficult to interpret because the counts during the high priority, budget interval were included with other counts during the low priority interval. This corrects that reporting by using two counts: One for the low and one for the high priority interval. This makes the results much easier to interpret.
There should be no impact to anything other that the sporadic scheduler case of the OS test.
Tested using the the stm32f4discovery:sporadic configration. That configuration has no yet been merged but is available from incubator-apps PR 3097.
