in function `mp_mul':
apps/interpreters/quickjs/quickjs/libbf.c:1179: multiple definition of `mp_mul';
nuttx/staging/libapps.a:apps/math/libtommath/libtommath/bn_mp_mul.c:8: first defined here
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
make[2]: *** No rule to make target '/home/nuttx/libs/libxx/uClibc++/tests/testframework.cpp',
needed by '/home/nuttx/libs/libxx/uClibc++/tests/testframework.cpp.home.apps.testing.uclibcxx_test.o'. Stop.
Signed-off-by: haopengxiang <haopengxiang@xiaomi.com>
`posix_spawn` depends on these configurations:
- !BINFMT_DISABLE
- !DISABLE_MOUNTPOINT
- BOARDCTL
- BOARDCTL_APP_SYMTAB
- ELF
- FS_ROMFS
Check them in compile time may waste some time if the dependencies are not met,
move them to Kconfig to avoid this.
Signed-off-by: Huang Qi <huangqi3@xiaomi.com>
libuv:
CC: pthread/pthread_mutexattr_setrobust.c libuv/src/unix/thread.c: In function ‘uv_thread_create_ex’:
libuv/src/unix/thread.c:174:24: error: storage size of ‘param’ isn’t known
174 | struct sched_param param;
| ^~~~~
btsak_main:
In file included from btsak_main.c:39:
btsak.h:149:1: error: unknown type name ‘bool’
149 | bool btsak_str2bool(FAR const char *str);
pipe_main.c:
CC: sim/sim_registerdump.c pipe_main.c:44:30: error: unknown type name ‘pthread_addr_t’
44 | static void *open_write_only(pthread_addr_t pvarg)
| ^~~~~~~~~~~~~~
pipe_main.c: In function ‘pipe_main’:
pipe_main.c:81:3: error: unknown type name ‘pthread_t’
81 | pthread_t writeonly;
redirect_test.c: In function ‘redirection_test’:
redirect_test.c:205:3: error: unknown type name ‘pthread_t’
205 | pthread_t readerid;
| ^~~~~~~~~
redirect_test.c:206:3: error: unknown type name ‘pthread_t’
206 | pthread_t writerid;
| ^~~~~~~~~
drivertest_posix_timer.c:48:29: error: ‘optarg’ undeclared (first use in this function)
48 | value = (type)strtoul(optarg, &ptr, base); \
drivertest_posix_timer.c:208:3: warning: implicit declaration of function ‘sleep’ [-Wimplicit-function-declaration]
208 | sleep(SLEEPSECONDS);
| ^~~~~
drivertest_uart.c:92:13: warning: implicit declaration of function ‘read’; did you mean ‘fread’? [-Wimplicit-function-declaration]
92 | len = read(fd, &tmp_char, 1);
| ^~~~
| fread
drivertest_uart.c:323:44: error: ‘optopt’ undeclared (first use in this function)
323 | printf("Unknown option: %c\n", optopt);
kernel/fs/cases/fs_getfilep_test.c:71:9: warning: implicit declaration of function ‘fs_getfilep’ [-Wimplicit-function-declaration]
71 | ret = fs_getfilep(fileno(fp), &filep);
| ^~~~~~~~~~~
kernel/mm/cases/mm_test_008.c:148:9: warning: implicit declaration of function ‘task_create’; did you mean ‘timer_create’? [-Wimplicit-function-declaration]
148 | pid = task_create("TestNuttx08_routine_1",
| ^~~~~~~~~~~
| timer_create
/home/ajh/work/nuttxwork/apps/testing/cmocka/cmocka_main.c:171:11: warning: implicit declaration of function ‘setenv’ [-Wimplicit-function-declaration]
171 | setenv("CMOCKA_XML_FILE", xml_path, 1);
| ^~~~~~
drivertest_pm_runtime.c:123:7: warning: implicit declaration of function ‘usleep’ [-Wimplicit-function-declaration]
123 | usleep(210 * 1000);
| ^~~~~~
Signed-off-by: anjiahao <anjiahao@xiaomi.com>
stdio.h remove <nuttx/xxxx.h> Some .c previously depended on
<nuttx/xxx.h> and needed to be added to .c instead of stdio.h
Signed-off-by: anjiahao <anjiahao@xiaomi.com>
1. Input redirect flags currently is hardcode, passing by arguments maybe better.
2. Only support redirect to path_name, redirect to fd is needed for pipeline.
Test
1. Redirect in
cat < /etc/init.d/rc.sysinit
2. Redirect with FIFO
mkfifo /dev/testfifo
cat /dev/testfifo &
ls > /dev/testfifo
3. NSH Params
set name `uname`
echo $name
Signed-off-by: wangjianyu3 <wangjianyu3@xiaomi.com>
Without this patch
nsh> ls /etc/group | dd | dd
sh [13:100]
sh [14:100]
nsh: dd: read failed: 4
nsh>
Signed-off-by: wangjianyu3 <wangjianyu3@xiaomi.com>
LVGL 9.2.1 archive is ~70MB. Over poor connections, it may take
longer to download it. This renders CMake build unusable in such
environments.
Signed-off-by: Daniel Jasinski <jasinskidaniel95szcz@gmail.com>
write returns in case some bytes were written but not everything can
fit.
This wasn't the case in NuttX but commit
d0680fd1bc51b7ead0b068fb24a31a144d22dc6c introduced this standard
behavior.
- Ensure `AUDIO_STOP` via `ioctrl` is followed by `mq_send STOP` without race conditions.
- Modify loop condition to correctly handle `running = false` upon receiving STOP signal.
- Resolve potential issue where `AUDIO_MSG_DEQUEUE` could still accept buffers after STOP signal due to timing.
Before:
- `nxaudio_stop` would call `ioctrl AUDIO_STOP` followed by `mq_send STOP`, which might lead to `AUDIO_MSG_DEQUEUE` accepting buffers after STOP.
After:
- Synchronized the sequence of `ioctrl AUDIO_STOP` and `mq_send STOP` to prevent buffer addition after STOP.
- Enhanced loop condition to accurately reflect the STOP state.
Signed-off-by: xinbingnan <xinbingnan@xiaomi.com>
Error: smp_call.c:36:31: error: 'g_call_data' defined but not used [-Werror=unused-variable]
36 | static struct smp_call_data_s g_call_data;
| ^~~~~~~~~~~
cc1: all warnings being treated as errors
make[2]: *** [/github/workspace/sources/apps/Application.mk:237: smp_call.c.github.workspace.sources.apps.testing.ostest.o] Error 1
make[2]: Target 'all' not remade because of errors.
make[1]: *** [Makefile:52: /github/workspace/sources/apps/testing/ostest_all] Error 2
make[1]: Target 'all' not remade because of errors.
make: *** [tools/LibTargets.mk:248: /github/workspace/sources/apps/libapps.a] Error 2
make: Target 'all' not remade because of errors.
/github/workspace/sources/nuttx/tools/testbuild.sh: line 385: /github/workspace/sources/nuttx/../nuttx/nuttx.manifest: No such file or directory
Signed-off-by: hujun5 <hujun5@xiaomi.com>
reason:
The old implementation of the SMP call, even when using the "no wait" parameter,
could still result in waiting, if invoking it within a critical section
may lead to deadlocks. Therefore, in order to implement a truly asynchronous SMP
call strategy, we have added nxsched_smp_call_async.
Signed-off-by: hujun5 <hujun5@xiaomi.com>
If the experimental feature Wasm build enabled with multi thread
compiling, the COPYFILE (cp) maybe raise the error below:
```
cp: cannot create regular file 'math.h': file exists
```
This error cause by that the wasi-sdk don't provide the math.h for
non-wasi envrioment, for this case, provide a math.h for it instead of copy
file each time is better.
Signed-off-by: Huang Qi <huangqi3@xiaomi.com>
API function nxboot_get_confirm was returning incorrect value if
primary image was flashed directly into the embedded flash (this
image does not have a tail, but is automatically considered as valid
and stable based on the header magic value).
Signed-off-by: Michal Lenc <michallenc@seznam.cz>
There is no corresponding sched_lock() for this and therefore this causes
random crashes.
Found when running a long init script which utilizes shell variables in
SMP mode.
When compiling the Release Branch of `nuttx-apps`, our CI Workflow `build.yml` checks out the Master Branch of `nuttx` repo, which is incorrect. This happens due to a typo in `build.yml`: https://github.com/apache/nuttx/issues/14513
This PR fixes the typo in `build.yml` to checkout the correct branch of `nuttx` repo.
Due to the [recent cost-cutting](https://github.com/apache/nuttx/issues/14376), we are no longer running PR Merge Jobs in the `nuttx` and `nuttx-apps` repos. For this to happen, I am now running a script on my computer that will cancel any PR Merge Jobs that appear: [kill-push-master.sh](https://github.com/lupyuen/nuttx-release/blob/main/kill-push-master.sh)
This PR disables PR Merge Jobs permanently, so that we no longer need to run the script. This prevents our CI Charges from over-running, in case the script fails to operate properly.
Summary
---------------------------------------------------------------------------------------------------------
Created a minimal demo application for Bluetooth. It works by displaying some data over Bluetooth and the
user can connect to the device running it on NuttX and read this data.
This can work in 2 scenarios: a real scenario, in which data from the BME680 (temperature, humidity, etc..)
is used and the user will be able to read it. The second scenario involves displaying some dummy data
(hardcoded values) for testing purposes.
This application can serve as an example for users implementing more complex applications using Bluetooth on NuttX.
Testing
---------------------------------------------------------------------------------------------------------
I used the ESP32-Sparrow (which includes the BME680 sensor) and the ESP32-devkitc boards for testing.
The Bluetooth stack starts by default with advertising when it initializes, however I noticed that enabling
advertising in the stack does not work (at least on ESP32). However, when scanning is also enabled, the
device starts advertising. Therefore, in our application, during the BLE services initialization part, I also
enable scanning as a temporary workaround in order to make sure the device advertises and the user can see it
and connect to it.
This patch allow from other module that outside
WAMR directory include `wasm_export.h` to use WAMR as library.
Signed-off-by: Huang Qi <huangqi3@xiaomi.com>
