https://gcc.gnu.org/bugzilla/show_bug.cgi?id=124174
--- Comment #6 from Alex Coplan <acoplan at gcc dot gnu.org> --- The ACLE does say that intrinsic implementations shouldn't introduce UB: https://github.com/ARM-software/acle/blob/main/main/acle.md#undefined-behavior > Care should be taken by compiler implementers not to introduce the concept of > undefined behavior to the semantics of an intrinsic. For example, the > vabsd_s64 intrinsic has well defined behaviour for all input values, while > the C99 llabs has undefined behaviour if the result would not be > representable in a long long type. It would thus be incorrect to implement > vabsd_s64 as a wrapper function or macro around llabs.
