[PATCH] Added more insertps optimizations

[Filipe Cabecinhas]

Added the patterns that Andrea mentioned.

We still need to have PerformSELECTCombine for the case when we do a
vector load and extract one element of that vector. In this case we want
to narrow it to an element load.

The other cases are simplified by just dealing with them with patterns in X86InstrSSE.td

Added test cases for SSE4.1, AVX, and AVX2.

http://reviews.llvm.org/D3581

Files:
  docs/CommandGuide/llvm-symbolizer.rst
  include/llvm/CodeGen/MachineOperand.h
  include/llvm/DebugInfo/DIContext.h
  lib/DebugInfo/DWARFDebugInfoEntry.cpp
  lib/Target/PowerPC/PPCAsmPrinter.cpp
  lib/Target/X86/X86ISelLowering.cpp
  lib/Target/X86/X86ISelLowering.h
  lib/Target/X86/X86InstrSSE.td
  lib/Transforms/InstCombine/InstructionCombining.cpp
  test/Analysis/CostModel/X86/vselect-cost.ll
  test/CodeGen/X86/avx-blend.ll
  test/CodeGen/X86/avx2.ll
  test/CodeGen/X86/blend-msb.ll
  test/CodeGen/X86/fold-load-vec.ll
  test/CodeGen/X86/sse41-blend.ll
  test/CodeGen/X86/sse41.ll
  test/DebugInfo/llvm-symbolizer.test
  test/Transforms/InstCombine/gepphigep.ll
  tools/llvm-symbolizer/LLVMSymbolize.cpp
  tools/llvm-symbolizer/LLVMSymbolize.h
  tools/llvm-symbolizer/llvm-symbolizer.cpp
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D3581.9502.patch
Type: text/x-patch
Size: 55785 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140517/789f131c/attachment.bin>