[all-commits] [llvm/llvm-project] 228a2b: [X86] Teach combineCVTPH2PS to shrink v8i16 loads ...

[topperc via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 228a2bc9b70c3d93bd28f0038a8664ef8dac042e
      https://github.com/llvm/llvm-project/commit/228a2bc9b70c3d93bd28f0038a8664ef8dac042e
  Author: Craig Topper <craig.topper at intel.com>
  Date:   2020-02-21 (Fri, 21 Feb 2020)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/lib/Target/X86/X86InstrAVX512.td
    M llvm/lib/Target/X86/X86InstrSSE.td

  Log Message:
  -----------
  [X86] Teach combineCVTPH2PS to shrink v8i16 loads when the output type is v4f32. Remove extra isel patterns.

Similar to what do for other operations that use a subset of bits.
Allows us to remove a pattern that shrinks a load. Which was
incorrect if the load was volatile.