mirror of
https://github.com/intel/llvm.git
synced 2026-02-06 06:31:50 +08:00
Standard says that implementation of math functions that have floating-point-type parameter should provide an "overload for each cv-unqualified floating-point type".