[PATCH] D87295: Prefer vpxor over vpxorps for AVX2 PR36127

[Dinar Temirbulatov via Phabricator via llvm-commits]

dtemirbulatov created this revision.
dtemirbulatov added reviewers: RKSimon, craig.topper, spatel, anton-afanasyev.
Herald added a subscriber: hiraditya.
Herald added a project: LLVM.
dtemirbulatov requested review of this revision.

As discussed in PR36127: AFAIK the throughput of the integer operations is always better or equal to the throughput of the floating-point operations in all x86/x86_64 CPU architectures to date. That is, if the default were to be changed from vxorps to vpxor things would always work out optimally (at least for now).


https://reviews.llvm.org/D87295

Files:
  llvm/lib/Target/X86/X86InstrInfo.cpp
  llvm/test/CodeGen/X86/avx-logic.ll
  llvm/test/CodeGen/X86/avx2-intrinsics-fast-isel.ll
  llvm/test/CodeGen/X86/avx512-arith.ll
  llvm/test/CodeGen/X86/bitcast-and-setcc-256.ll
  llvm/test/CodeGen/X86/fma-intrinsics-fast-isel.ll
  llvm/test/CodeGen/X86/fma_patterns.ll
  llvm/test/CodeGen/X86/fma_patterns_wide.ll
  llvm/test/CodeGen/X86/machine-combiner-int-vec.ll
  llvm/test/CodeGen/X86/promote-cmp.ll
  llvm/test/CodeGen/X86/sadd_sat_vec.ll
  llvm/test/CodeGen/X86/ssub_sat_vec.ll
  llvm/test/CodeGen/X86/stack-folding-fp-avx1.ll
  llvm/test/CodeGen/X86/v8i1-masks.ll
  llvm/test/CodeGen/X86/vec_fp_to_int.ll
  llvm/test/CodeGen/X86/vector-partial-undef.ll
  llvm/test/CodeGen/X86/vector-reduce-umax.ll
  llvm/test/CodeGen/X86/vector-reduce-umin.ll
  llvm/test/CodeGen/X86/vector-trunc-math.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D87295.290519.patch
Type: text/x-patch
Size: 27135 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200908/adc8ad51/attachment.bin>