[libc-commits] [libc] [llvm] [libc] Make totalorder functions use the emulated float128 type (PR #221085)

[Muhammad Haris Awan via libc-commits]

harisawan-bit wrote:

@Sukumarsawant Done: moved `totalorderf128` and `totalordermagf128` out of the `LIBC_TYPES_HAS_NATIVE_FLOAT128` block into `TARGET_LIBM_ENTRYPOINTS` across all 8 remaining targets (`baremetal/aarch64`, `baremetal/arm`, `baremetal/riscv`, `darwin/aarch64`, `freebsd/x86_64`, `linux/aarch64`, `linux/riscv`, `linux/x86_64`). All 13 targets now have both entrypoints unconditionally enabled in `TARGET_LIBM_ENTRYPOINTS`, matching the setup for `bf16` and `fdimf128`.

Also resolved the open review threads on entrypoints and file headers.

Regarding MPFR: `totalorder` and `totalordermag` are IEEE 754 bit-level ordering predicates, so they don't have MPFR math library references; full test coverage is provided in `libc/test/src/math/smoke/` via `TotalOrderTest.h` and `TotalOrderMagTest.h` just like `totalorderf16` and `totalorderbf16`.

Please take another look when you get a chance. Thanks!


https://github.com/llvm/llvm-project/pull/221085