We would like to form a working group that focuses on facilitating and accelerating standard C/C++ language development in Clang/LLVM. This working group would meet at a regular cadence (weekly or bi-weekly depending on people’s schedules) and discuss topics relevant to C/C++ language development in Clang/LLVM. These topics would include (but not necessarily be limited to):
Implementation of current C/C++ language features and defects
Clarification of C/C++ language specification
In person design review and code reviews of C/C++ language features
Review and coordination of standard C/C++ implementation status in Clang
Any other topics that the group would like to discuss
We would like to invite anyone to participate in the working group, especially those that have expertise in Clang internals relating to C and C++, or who are current or were past members of WG21 or WG14. In addition to facilitating and accelerating C/C++ language development in Clang, our goal is to build skills and knowledge in these respective areas. I’ll follow up to gauge preferred meetup times if there’s sufficient interest.
Erich Keane (Intel), Aaron Ballman (Intel), Bob Monteleone (Intel), Jeff Heath (Intel)
Thanks all for the feedback, it looks like there is good support for a Clang C/C++ language working group. I’ll create a new thread in the next day or two to vote on an initial meeting time, and will create a skeleton agenda/calendar.
@jrheng99, do you know if there would be interest in driving a GSoC project through this group on improving timing traces for Clang compile-time performance? In particular, I believe more information about the “user source” whose processing is contributing to the time spent would be useful. At this time, I think the crash diagnostic stack trace contexts may serve as a model for attributing time spent to the user source.
Hi All - I haven’t been very active in the past four years or so - but this group caught my eye (exciting!) and with an in-person standards meeting on the horizon, i’m feeling a little nostalgic and was going to test the llvm-waters again - maybe try and dust some of the rust off to see if I can still swim and keep up …
With that disclosure, I was thinking of tackling the ‘deducing this’ proposal - is anyone working on it?