davids5 commented on PR #8992: URL: https://github.com/apache/nuttx/pull/8992#issuecomment-1512925581
> > LGTM. > > One thing that worries me is that we will stay with two versions of the pinouts and the migration process will never end. Maybe we should define some deadline when migration should be completed (in terms of release version)? Creating an issue to track progress in upstream boards migration will also be a good idea. We could follow who is willing to help in the migration process to avoid duplication of work. > > Remember there are out-of-tree boards that will need to be updated. We should allow enough time for that. How long? That should be discussed on mailing list. > > When migration begins, the first step should be to set default of STM32xx_USE_LEGACY_PINMAP to No for a while (how long? again, let's discuss on-list). After that, the legacy pinmaps can be removed. > One thing that worries me is that we will stay with two versions of the pinouts and the migration process will never end @raiden00pl Yes me too. @hartmannathan I fully agree the approach and the TBD1 & TBD2 for N should approved on the mailing list. Here is one approach I was thinking about: Once this repo has had all boards converted say in release (N). After (N+(TBD1)) releases, in a single commit, the STM32xx_USE_LEGACY_PINMAP logic would be removed from all the .h files, AND the gpio.c #"pragma message" replaced with an #error that point to the PR deleting the legacy.h files as a second commit. The legacy files are deleted (in a second single commit) Then any latecomers: to the party, can revert the second commit and have all the legacy files to run the conversion tool against. N+TBD2 remove the Kconfig entries and #error -- 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. To unsubscribe, e-mail: commits-unsubscr...@nuttx.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
