Removed from the arm-05.dat file the entries:
CMake,nrf52832-dk:buttons
CMake,nrf52832-dk:wdog
CMake,nrf52840-dk:adc
CMake,nrf52840-dk:buttons
CMake,nrf52840-dk:pwm
CMake,nrf52840-dk:qspi
CMake,nrf52840-dk:timer
present in the jumbo configuration
Signed-off-by: simbit18 <simbit18@gmail.com>
This PR adds support for the CI system for native Windows as well. It allows you to build NuttX on GitHub and test it locally for Windows users.
With these CI tools with PowerShell scripts, it is possible to build NuttX for Windows Native using (for now only) Cmake + Ninja with the same logic as the CI system with Bash scripts.
This allows the msvc job to be used not only with the simulator (currently only with Visual Studio 17 2022), but also with other architectures using the same Windows runner to get more coverage and avoid future breakage.
As with the other jobs, we use artifacts to save the compilation result at the end of the workflow execution (previously for the simulator it was not done).
The proposed solution is based on the following additions and modified:
Modified Files
buildyml -> only CI Jobs MSVC
New Files in tools/
ci/cibuild.ps1 -> Added Powershell script for Run the CI Builds
ci/platforms/windows.ps1 -> Added Powershell script for installing toolchains and tools.
testlist/windows.dat -> Target (Add sim (msvc), risc-v arm)
tools/testbuild.ps1
We tested the NuttX build on GitHub and locally.
How we build on GitHub and test locally.
Locally
cd .\nuttx\tools\ci\
.\cibuild.ps1 -n -i -A -C -N .\testlist\windows.dat
Signed-off-by: simbit18 <simbit18@gmail.com>
This prevents CI from testing esp32s3-devkit:python automatically,
avoiding it to be overloaded. Manual testing can be triggered to
test xtensa-03.dat.
Signed-off-by: Tiago Medicci Serrano <tiago.medicci@espressif.com>
Split nucleo-f4x1re into nucleo-f401re and nucleo-f411re.
These are separate boards and should be in separate directories as it's
done for all other nucleo boards in NuttX
enable various types of common applications and libraries, and enable CMake CI checks.
To avoid regression issues in the build system
Signed-off-by: xuxin19 <xuxin19@xiaomi.com>
This PR splits the CI Build Job sim-01 and adds sim-03:
Before the Split: Simulator Jobs take up to 1.5 hours to complete
- sim-01 (1 hour 31 mins): adb, citest, lvgl, matter
- sim-02 (28 mins): posix_test, sqlite
After the Split: Simulator Jobs will complete within 1 hour
- sim-01 (58 mins): adb, citest
- sim-02 (35 mins): lvgl, matter
- sim-03 (28 mins): posix_test, sqlite
This will help us comply with the ASF Policy for GitHub Actions, as explained here: https://github.com/apache/nuttx/issues/14376
nucleo-144 combines 3 different ST boards. This approach is inconsistent with
the support for the rest of the nucleo boards, where each board is in separate folders.
Also nucleo-144 is no longer reserved for STM32F7 chips but other families also use this format.
After this commit nucleo-144 is divided into 3 boards:
- nucleo-f746zg
- nucleo-f767zi
- nucleo-f722ze
This PR creates the new CI Build Jobs `arm64-01` and `x86_64-01`. The new jobs will split and offload the Arm64 and x86_64 Build Targets from `other`. This will reduce our usage of GitHub Runners, to comply with the ASF Policy for GitHub Actions. (Recently we see more PRs for Arm64 and x86_64)
Before the Split: Simple PRs (One Arch and/or One Board) for Arm64 and x86_64 require almost 1 hour for CI Build
- `other` (57 mins): AVR, SPARC, x86, PinePhone, QEMU Arm64, QEMU x86_64
After the Split: Simple PRs for Arm64 and x86_64 will complete under 30 mins
- `other` (24 mins): AVR, SPARC, x86
- `arm64-01` (29 mins): PinePhone, QEMU Arm64
- `x86_64-01` (9 mins): QEMU x86_64
To skip more unnecessary builds: Our Build Rules `arch.yml` shall ignore the label "Area: Documentation", so that a Simple PR + Docs is still a Simple PR. Previously we experienced longer CI Build Times, just because we added docs to our Simple PR. (Now our PR shall be built exactly like a Simple PR)
The updated CI code is explained here: https://github.com/apache/nuttx/issues/13775
To speed up the CI Workflow, this PR splits the CI Build Jobs for RISC-V into smaller jobs. Each job will now complete within 1 hour.
Before the PR: There are 2 jobs for RISC-V, each requiring more than 1.5 hours
- `risc-v-01` (1 hour 42 mins): BL602, Ox64, ESP32-C3 / C6 / H2
- `risc-v-02` (1 hour 41 mins): K230, Icicle, QEMU, RV32M1-Vega
After the PR: The build is spread across 6 jobs for RISC-V, each job completes within 1 hour
- `risc-v-01` (19 mins): BL602, Ox64
- `risc-v-02` (44 mins): ESP32-C3
- `risc-v-03` (45 mins): ESP32-C6, ESP32-H2
- `risc-v-04` (31 mins): K230, Icicle
- `risc-v-05` (41 mins): QEMU CITest
- `risc-v-06` (38 mins): Rest of QEMU, RV32M1-Vega
Following the same convention as the Arm32 Build Jobs, the above jobs are sorted by Target Name. Performance of the RISC-V Build Jobs is discussed in https://github.com/apache/nuttx/issues/13775
CI Build Job `arm-05` (runtime 2 hours) has become the Performance Bottleneck for CI Workflow. That's because `arm-05` builds too many targets for nRF, RP2040, SAM 3, SAM A and SAM D. This PR splits `arm-05` into multiple smaller jobs, to reduce the CI Build Duration.
Before the PR: `arm-05` is overloaded, build requires 2 hours
- `arm-05` (2 hours): nRF, RP2040, SAM 3, SAM A, SAM D
- `arm-06` (56 mins): STM32 [a-m]*
After the PR: `arm-05` is offloaded (to `arm-06` and `arm-07`), completes within 1 hour
- `arm-05` (47 mins): nRF
- `arm-06` (1 hour): Reserve for RP2040 exclusively
- `arm-07` (1 hour 15 mins): SAM 3, SAM A, SAM D, STM32 [a-m]*
Build Jobs are sorted by Target Name. So we cascade the changes and rename the Build Jobs: `arm-07` becomes `arm-08`, `arm-08` becomes `arm-09` etc. Then `arm-13` becomes a new job `arm-14`. (Which we added to `build.yml`)
Performance of `arm-05` is discussed in https://github.com/apache/nuttx/issues/13775 and https://github.com/apache/nuttx/issues/12773
Currently concerns only arm.
tools/ci/testlist/msys2.dat:
At the moment I only added the board nucleo-l152re:nsh
.github/workflows/build.yml:
Enabled cmake for msys2
find: boards/risc-v/esp32c3/esp32c3-devkit/configs/cxx: No such file or directory
find: boards/risc-v/esp32c3/esp32c3-devkit/configs/wifi: No such file or directory
There have been errors like below on macOS for sim:libcxxtest
```
Error: /Applications/Xcode.../include/mach/vm_page_size.h:59:44:
error: expected ';' after top level declarator
```
This patch tries to disable that check on macOS.
Signed-off-by: Yanfeng Liu <yfliu2008@qq.com>
in order to maintain the completeness of the matter related functional code, the build of matter is added to testlist.
Signed-off-by: zhanghongyu <zhanghongyu@xiaomi.com>
since linker can't find basic_string::~basic_string() oddly:
arm-none-eabi-ld: /home/xiaoxiang/backup/os/nuttx/nuttx/staging/libxx.a(libcxx/src/locale.o): in function `__tcf_0':
/home/xiaoxiang/backup/os/nuttx/nuttx/libs/libxx/libcxx/src/locale.cpp:4941: undefined reference to `std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >::~basic_string()'
another fix is removing the follow code from _LIBCPP_STRING_V1_EXTERN_TEMPLATE_LIST and _LIBCPP_STRING_UNSTABLE_EXTERN_TEMPLATE_LIST:
_Func(_LIBCPP_FUNC_VIS basic_string<_CharType>::~basic_string()) \
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
This PR is a modification that optimizes priority inheritance
for only one holder. After the above modifications are completed,
the mutex lock->unlock process that supports priority inheritance
can be optimized by 200 cycles.
Before modify: 2000 cycle
After modify: 1742 cycle
Signed-off-by: zhangyuan21 <zhangyuan21@xiaomi.com>
