[all-commits] [llvm/llvm-project] 90c31b: [X86] Custom lower ISD::FROUND with SSE4.1 to avoi...

[topperc via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 90c31b0f428fe911255277a60782ea9114700475
      https://github.com/llvm/llvm-project/commit/90c31b0f428fe911255277a60782ea9114700475
  Author: Craig Topper <craig.topper at intel.com>
  Date:   2020-01-29 (Wed, 29 Jan 2020)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/extractelement-fp.ll
    M llvm/test/CodeGen/X86/vec-libcalls.ll
    R llvm/test/CodeGen/X86/vec_round.ll

  Log Message:
  -----------
  [X86] Custom lower ISD::FROUND with SSE4.1 to avoid a libcall.

ISD::FROUND is defined to round to nearest with ties rounding
away from 0. This mode isn't supported in hardware on X86.

But as long as we aren't compiling with trapping math, we can
emulate this with floor(X + copysign(nextafter(0.5, 0.0), X)).

We have to use nextafter to avoid some corner cases that adding
0.5 would have. For example, if X is nextafter(0.5, 0.0) it should
round to 0.0, but adding 0.5 would need one extra bit of mantissa
than can be stored so it rounds to 1.0. Adding nextafter(0.5, 0.0)
instead will just increase the exponent by 1 and leave the mantissa
as all 1s. This would be nextafter(1.0, 0.0) which will floor to 0.0.

Techically this requires -fno-trapping-math which isn't our default.
But if we care about exceptions we should be using constrained
intrinsics. Constrained intrinsics would use STRICT_FROUND which
won't go through this code.

Fixes PR42195.

Differential Revision: https://reviews.llvm.org/D73607