[X86] Remove the multiply by 8 that goes into the shift constant for X86ISD::VSHLDQ...
authorCraig Topper <craig.topper@gmail.com>
Mon, 16 Feb 2015 20:52:07 +0000 (20:52 +0000)
committerCraig Topper <craig.topper@gmail.com>
Mon, 16 Feb 2015 20:52:07 +0000 (20:52 +0000)
commit49df44e2e2f12526a48f2d4dcffd975a25b49558
tree50dc6dbd14c944a2883c793be30badb157180cd1
parent44026efa8879e5ec0f91cea42d98f0b45e9076dd
[X86] Remove the multiply by 8 that goes into the shift constant for X86ISD::VSHLDQ and X86ISD::VSRLDQ. This simplifies the pattern matching in isel and allows these nodes to become the patterns embedded in the instruction.

llvm-svn: 229431
llvm/lib/IR/AutoUpgrade.cpp
llvm/lib/Target/X86/X86ISelLowering.cpp
llvm/lib/Target/X86/X86InstrSSE.td
llvm/test/CodeGen/X86/avx2-intrinsics-x86-upgrade.ll
llvm/test/CodeGen/X86/avx2-intrinsics-x86.ll