v01d commented on issue #1639: URL: https://github.com/apache/incubator-nuttx/issues/1639#issuecomment-680106002
I also do not like the idea of basing any improvements of the NuttX BLE stack by depending on external actively developed project such as Zephyr, which moves fast and its purpose is not to provide others with a BLE stack. This means that any internal reorganization of Zephyr stack will negatively impact NuttX (even if it is just adapting the shims). For something as critical and internal as a networking stack, I agree with Greg that this should be a fully contained and native solution. If it really feels like reinventing the wheel and there's a good generic BLE stack that can provide API stability over a long time (MyNewt?), I would consider using that, but only if the shim is clean. Since we already have a good foundation for a 4.2 BLE stack in NuttX, maybe we should try to improve it with what was recently discussed in the list and what this issue is about first. If we want to then move to 5.0 I would try to improve the existing stack unless it proves to be a complete rewrite (I wouldn't think so since there's some degree of compatibility between spec versions). In any case, I would prioritize consolidating the current stack to upgrade to newer spec version. But I understand that companies with BLE products might want to prioritize 5.0 at some point. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
