[all-commits] [llvm/llvm-project] 90ff34: [X86] Add custom widening for v2i32->v2f64 strict_...

[topperc via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 90ff34e6ab38a333b16890c0d4b7ca891d213652
      https://github.com/llvm/llvm-project/commit/90ff34e6ab38a333b16890c0d4b7ca891d213652
  Author: Craig Topper <craig.topper at intel.com>
  Date:   2019-12-26 (Thu, 26 Dec 2019)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/lib/Target/X86/X86InstrAVX512.td
    M llvm/test/CodeGen/X86/vec-strict-inttofp-128.ll
    M llvm/test/CodeGen/X86/vec_int_to_fp.ll
    M llvm/test/CodeGen/X86/vector-constrained-fp-intrinsics.ll

  Log Message:
  -----------
  [X86] Add custom widening for v2i32->v2f64 strict_uint_to_fp with AVX512F, but not AVX512VL.

Previously we were widening with isel patterns, but that wasn't
exception safe for strict FP. So now we widen to v4i32->v4f64
during type legalization. And then let op legalization further
widen to v8i32->v8f64.

The vec_int_to_fp.ll changes are caused by us no longer narrowing
extracts of strict_uint_to_fp to the v4i32->v2f64 instruction
without AVX512VL only to have isel rewiden it. Now we just keep
it wide throughout. So we don't have an opportunity to narrow
the load.