Program names containing '-' (for example, renaming hello to hello-world via PROGNAME) previously generated an invalid identifier <NAME>_main when constructing the main= compiler definition and the APP_MAIN target property used during builtin list generation. This caused the CMake build to fail because '-' is not a valid character in a C identifier. This mirrors the Make-based fix (Application.mk's PROGSYM) for the traditional build. Introduce NAME_SYM, a sanitized copy of NAME with '-' replaced by '_', and use it only where an internal C identifier is required: the main= COMPILE_DEFINITIONS property and the APP_MAIN target property. Leave NAME unchanged everywhere else, including CMake target/output names and the APP_NAME property, where hyphens are valid. The standalone/loadable executable path (MODULE/DYNLIB/kernel build) does not rename main() and therefore requires no sanitization because each executable is linked independently rather than merged into a shared builtin image. Testing (WSL2 Ubuntu, x86_64): - BOARD_CONFIG=sim/nsh, CONFIG_EXAMPLES_HELLO_PROGNAME="hello-world": clean CMake configure/build; 'hello-world' runs and prints 'Hello, World!!' - Reverted to CONFIG_EXAMPLES_HELLO_PROGNAME="hello": reconfigured and rebuilt; 'hello' runs and prints 'Hello, World!!' (no regression) Fixes #19447 Signed-off-by: Ansh Rai <anshrai331@gmail.com> |
||
|---|---|---|
| .github | ||
| arch | ||
| audio | ||
| binfmt | ||
| boards | ||
| cmake | ||
| crypto | ||
| Documentation | ||
| drivers | ||
| dummy | ||
| fs | ||
| graphics | ||
| include | ||
| libs | ||
| mm | ||
| net | ||
| openamp | ||
| pass1 | ||
| sched | ||
| syscall | ||
| tools | ||
| video | ||
| wireless | ||
| .asf.yaml | ||
| .codespell-ignore-lines | ||
| .codespellrc | ||
| .editorconfig | ||
| .gitignore | ||
| .gitmessage | ||
| .pre-commit-config.yaml | ||
| .yamllint | ||
| AUTHORS | ||
| CMakeLists.txt | ||
| CONTRIBUTING.md | ||
| INVIOLABLES.md | ||
| Kconfig | ||
| LICENSE | ||
| Makefile | ||
| NOTICE | ||
| README.md | ||
| ReleaseNotes | ||
Apache NuttX is a real-time operating system (RTOS) with an emphasis on standards compliance and small footprint. Scalable from 8-bit to 64-bit microcontroller environments, the primary governing standards in NuttX are POSIX and ANSI standards. Additional standard APIs from Unix and other common RTOSs (such as VxWorks) are adopted for functionality not available under these standards, or for functionality that is not appropriate for deeply-embedded environments (such as fork()).
For brevity, many parts of the documentation will refer to Apache NuttX as simply NuttX.
Getting Started
First time on NuttX? Read the Getting Started guide! If you don't have a board available, NuttX has its own simulator that you can run on terminal.
Documentation
You can find the current NuttX documentation on the Documentation Page.
Alternatively, you can build the documentation yourself by following the Documentation Build Instructions.
The old NuttX documentation is still available in the Apache wiki.
Supported Boards
NuttX supports a wide variety of platforms. See the full list on the Supported Platforms page.
Contributing
If you wish to contribute to the NuttX project, read the Contributing guidelines for information on Git usage, coding standard, workflow and the NuttX principles.
License
The code in this repository is under either the Apache 2 license, or a license compatible with the Apache 2 license. See the License Page for more information.