[PATCH] D134267: [C++] [Modules] Support one phase compilation model for named modules

[David Blaikie via Phabricator via cfe-commits]

dblaikie added a comment.

I'm getting a bit exhausted with all the words involved here & not sure how to simplify/clarify this.

If @ben.boeckel has particular use cases, it might be easier for him to be here discussing them so we can discuss the tradeoffs directly rather than through intermediaries.

I think the choices of flags, even when they represent relatively minor work on the compiler side, are important in terms of how they shape the environment the compiler exists in. I have reservations about implementing the libCody and the scanner-based solutions (let alone also caching based solutions) - but that ship's probably already sailed in terms of it's implemented in GCC and build2 is using it. (sort of like open source software - we implement things for compatibility (like LGPL) but when we're the ones innovating/creating new things we can and should be more cautious/possibly more prescriptive to avoid creating more diversity/divergence than is necessary)

Please separate this work into isolated patches & we can discuss them separately. I think this review might be best to abandon as the subject line/description's out of synch and there's been a /lot/ of discussion going in a lot of directions such that it'd be hard to understand the conclusions/focus of this review at this point.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D134267/new/

https://reviews.llvm.org/D134267