Stephan T. Lavavej

We talked about this at the weekly maintainer meeting and @barcharcraz and I agree that this LWG issue resolution is simple enough that we can just directly implement it, no...

@alekasm I can see that you've put a lot of time and effort into following the reporting guidance so I'll also make an effort to get this properly tracked. However,...

We talked about this at the weekly maintainer meeting and @barcharcraz had a suggestion - build as `Release`, but then add compiler options (e.g. `/Zi`) to generate debug info. https://github.com/microsoft/STL/blob/8e2d724cc1072b4052b14d8c5f81a830b8f1d8cb/benchmarks/CMakeLists.txt#L62-L65...

Yes, I agree. Labeled, thanks.

At this time, we *definitely* still need to support ancient CPUs, e.g. those that don't have SSE4.2. (There was a recent compiler back-end bug regarding unguarded use of the `popcnt`...

We can do this before vNext - I have the necessary changes, plus an argument for why it's reasonable to do. Stay tuned...

I've contacted the compiler team about getting support for this.

* `__reference_constructs_from_temporary` has been implemented in Clang 18. * `__reference_converts_from_temporary` is being implemented in Clang.

We've picked up Clang 19 in #5247, so I believe this is now unblocked.

We talked about this at the weekly maintainer meeting and we're in favor of implementing the `ranges::minmax` value-based family with dedicated code in the core header `` (currently only providing...