[all-commits] [llvm/llvm-project] 4745c9: Set std::numeric_limits<>::tinyness_before to true...

[Owen Anderson via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 4745c994e4a794ca177152c4c0bd0f640d0cbe8b
      https://github.com/llvm/llvm-project/commit/4745c994e4a794ca177152c4c0bd0f640d0cbe8b
  Author: Owen Anderson <resistor at mac.com>
  Date:   2022-02-22 (Tue, 22 Feb 2022)

  Changed paths:
    M libcxx/include/limits
    M libcxx/test/std/language.support/support.limits/limits/numeric.limits.members/tinyness_before.pass.cpp

  Log Message:
  -----------
  Set std::numeric_limits<>::tinyness_before to true for floating point types on ARM platforms.

Set std::numeric_limits<>::tinyness_before to true for floating point types on ARM platforms.

Section E1.3.5 in the ARMv8 Architecture Reference Manual specifies:
  Underflow. The bit is set to 1 if the absolute value of the result
  of an operation, produced before rounding, is less than the minimum
  positive normalized number for the destination precision, and the
  rounded result is inexact.

Reviewed By: #libc, majnemer, EricWF

Differential Revision: https://reviews.llvm.org/D116338