[all-commits] [llvm/llvm-project] 7d6c2e: [clang] Use llvm.is_fpclass to implement FP classi...

[Serge Pavlov via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 7d6c2e18114de9900d1b012cf9c219803b183f63
      https://github.com/llvm/llvm-project/commit/7d6c2e18114de9900d1b012cf9c219803b183f63
  Author: Serge Pavlov <sepavloff at gmail.com>
  Date:   2023-07-11 (Tue, 11 Jul 2023)

  Changed paths:
    M clang/lib/CodeGen/CGBuiltin.cpp
    M clang/test/CodeGen/X86/strictfp_builtins.c
    M clang/test/CodeGen/aarch64-strictfp-builtins.c
    M clang/test/CodeGen/builtin_float.c
    M clang/test/CodeGen/builtin_float_strictfp.c
    M clang/test/CodeGen/builtins.c
    M clang/test/CodeGen/isfpclass.c
    M clang/test/CodeGen/strictfp_builtins.c
    M clang/test/Headers/__clang_hip_math.hip
    M clang/test/Headers/hip-header.hip
    M clang/test/Headers/openmp_device_math_isnan.cpp

  Log Message:
  -----------
  [clang] Use llvm.is_fpclass to implement FP classification functions

Builtin floating-point number classification functions:

    - __builtin_isnan,
    - __builtin_isinf,
    - __builtin_finite, and
    - __builtin_isnormal

now are implemented using `llvm.is_fpclass`.

This change makes the target callback `TargetCodeGenInfo::testFPKind`
unneeded. It is preserved in this change and should be removed later.

Differential Revision: https://reviews.llvm.org/D112932