[PATCH] Lower certain build_vectors to insertps instructions

Filipe Cabecinhas filcab+llvm.phabricator at gmail.com
Mon Apr 28 20:20:28 PDT 2014


Perform the optimization when we're moving a single element to a different
place in the vector, while zeroing out some other elements.

http://reviews.llvm.org/D3521

Files:
  cmake/modules/AddSphinxTarget.cmake
  include/llvm/ADT/PointerUnion.h
  include/llvm/ADT/iterator.h
  include/llvm/Analysis/BlockFrequencyInfoImpl.h
  include/llvm/Analysis/LazyCallGraph.h
  include/llvm/IR/User.h
  include/llvm/MC/MCExpr.h
  lib/Analysis/BlockFrequencyInfoImpl.cpp
  lib/CodeGen/AsmPrinter/AddressPool.h
  lib/CodeGen/AsmPrinter/DIE.h
  lib/CodeGen/AsmPrinter/DwarfDebug.cpp
  lib/CodeGen/AsmPrinter/DwarfDebug.h
  lib/CodeGen/AsmPrinter/DwarfUnit.cpp
  lib/CodeGen/AsmPrinter/DwarfUnit.h
  lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
  lib/DebugInfo/DWARFDebugArangeSet.h
  lib/DebugInfo/DWARFDebugAranges.cpp
  lib/DebugInfo/DWARFDebugFrame.cpp
  lib/DebugInfo/DWARFDebugFrame.h
  lib/MC/ELFObjectWriter.cpp
  lib/MC/MCExpr.cpp
  lib/Target/ARM64/ARM64ISelLowering.cpp
  lib/Target/PowerPC/PPCFrameLowering.cpp
  lib/Target/Sparc/SparcFrameLowering.cpp
  lib/Target/X86/X86FrameLowering.cpp
  lib/Target/X86/X86ISelLowering.cpp
  lib/Target/XCore/XCoreFrameLowering.cpp
  lib/Transforms/Scalar/LoopStrengthReduce.cpp
  test/Analysis/BlockFrequencyInfo/irreducible.ll
  test/CodeGen/ARM64/ext.ll
  test/CodeGen/X86/sse41.ll
  test/DebugInfo/X86/inline-member-function.ll
  test/MC/ELF/offset.s
  test/Transforms/LoopStrengthReduce/ARM64/lit.local.cfg
  test/Transforms/LoopStrengthReduce/ARM64/req-regs.c
  unittests/ADT/PointerUnionTest.cpp
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D3521.8901.patch
Type: text/x-patch
Size: 156640 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140429/210292ef/attachment.bin>


More information about the llvm-commits mailing list