[x86] eliminate movddup of horizontal op

This pattern would show up as a regression if we more
aggressively convert vector FP ops to scalar ops.

There's still a missed optimization for the v4f64 legal
case (AVX) because we create that h-op with an undef operand.
We should probably just duplicate the operands for that
pattern to avoid trouble.

llvm-svn: 357642
2 files changed
tree: 425dd5d12ac5e3b497936692d24d63a2b3bfaa26
  1. clang/
  2. clang-tools-extra/
  3. compiler-rt/
  4. debuginfo-tests/
  5. libclc/
  6. libcxx/
  7. libcxxabi/
  8. libunwind/
  9. lld/
  10. lldb/
  11. llgo/
  12. llvm/
  13. openmp/
  14. parallel-libs/
  15. polly/
  16. pstl/
  17. .arcconfig
  18. .clang-format
  19. .clang-tidy
  20. README.md
README.md

The LLVM Compiler Infrastructure

This directory and its subdirectories contain source code for LLVM, a toolkit for the construction of highly optimized compilers, optimizers, and runtime environments.