214 | | Currently a minimum RTEMS application assumes that you want items like a reentrant C library, |
215 | | shutdown code and a minimal filesystem infrastructure. Some ideas in |
216 | | this area are captured at [wiki:Projects/TinyRTEMS TinyRTEMS]. The project would focus on |
217 | | breaking linkages between components so they can be configured out by |
218 | | the user, dropped out automatically at link time, or other mechanisms. |
219 | | The goal is to be able to provide the user with a full featured RTEMS |
220 | | library and defer as much configuration as possible to application |
221 | | configuration and link time. |
| 214 | Currently a minimum RTEMS application assumes that you want items like a shutdown code, initialization of all Classic API Managers, and support for c. Some ideas in |
| 215 | this area are captured at [wiki:Projects/TinyRTEMS TinyRTEMS]. The project would focus on breaking linkages between components so they can be configured out by |
| 216 | the user, dropped out automatically at link time, or other mechanisms. The goal is to be able to provide the user with a full featured RTEMS |
| 217 | library and defer as much configuration as possible to application configuration and link time. |
| 218 | |
| 219 | One issue that will be an ongoing concern is that the minimum footprint varies by BSP. This is because some BSPs are structured to be lean, while others may have |
| 220 | unintentional dependencies. One idea is a tool to assist in identifying RTEMS symbols which are not expected occur in some of the sample applications such minimum, |
| 221 | hello, and ticker. This utility could provide guidance on the known reasons this symbol creeps in. For example, placing printk() support in the same file as |
| 222 | the console driver will result in the entirety of the Termios subsystem being in the minimum footprint even though it is unused. |