beacause _SCHED_GETTID is in nuttx/sched.h, when use dump_stack()
compile error log:
unqlite.c:51256: undefined reference to `_SCHED_GETTID'
Signed-off-by: wangbowen6 <wangbowen6@xiaomi.com>
When using stm32, the starting address of the function parsed by mkallsyms.py is an odd number, one large than the actual address
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
when using the clock_gettime() function to print the timestamp, the sinfo() function is called, as same as syslog(). But syslog() itself has the ability to print a timestamp, and called clock_gettime(). Thus, the clock_gettime() and syslog() have a recursive call problem. In order to solve this problem, it is necessary to ensure that SYSLOG_TIMESTAMP is close when DEBUG_SCHED is open.
Signed-off-by: cuiziwei <cuiziwei@xiaomi.com>
SAMv7 PWM driver supports complementary PWM output if both pins (H and L)
are defined and configured. This commit adds a configuration option to
configure the complementary L pin.
The pin definition has to be provided by board level support.
Signed-off-by: Michal Lenc <michallenc@seznam.cz>
There might be some cases that we receive an event after socket is closed because of out-of-ordered rpmsg, since it may not cause problem, we may omit the error.
Also change log level to err when returning negated errno, because if we return negated errno, we'll directly trigger RPMSG_ASSERT outside, so we need at least an error message to indicate the reason.
Signed-off-by: Zhe Weng <wengzhe@xiaomi.com>
As far as I can interpret how signal delivery should work when the signal
is blocked, it should still be sent to the pending queue even if the signal
is masked. When the sigmask changes it will be delivered.
The original implementation did not add the pending signal action, if
stcb->task_state == TSTATE_WAIT_SIG is true.
An attempt to patch this was made in #8563 but it is insufficient as it
creates an issue when the task is not waiting for a signal, but is in
syscall, in this case the signal is incorrectly queued twice.
Summary:
- I noticed that there are two kinds of descriptors for imxrt_enet.c
- The first one is the legacy descriptor and its size is 8bytes.
- The second one is the enhanced descriptor and its size is 32bytes.
- In both cases, we can not use a descriptor chain like stm32.
- Considering cache line alignment, the second one is perfect because
one descriptor fits the Cortex-M7 cache line which would fix networking
stability issues in d-cache write-back mode.
Impact:
- imxrt ethernet in d-cache write-back mode
Testing:
- Tested with ixmrt1060-evk:netnsh_dcache_wb (will be added later)
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Summary:
- I noticed that CONFIG_IMXRT_ENET_ENHANCEDBD is not correctly
used though it is defined in Kconfig.
- This commit fixes this issue.
Impact:
- None
Testing:
- Tested with imxrt1060-evk:netnsh_dcache_wb (will be added later)
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
The POSIX standard dictates that during abnormal termination the functions
registered by atexit() are _not_ called, also flushing the streams is
optional. So in this case, it is perfectly legal / better to call the
kernel system call _exit() instead.
This fixes regression issues caused by removal exit() from the kernel.
