diff options
author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2016-04-21 18:03:06 +0000 |
---|---|---|
committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2016-04-21 18:03:06 +0000 |
commit | 8d1052f55cf807afc4fb21734d36f166b83ae4ea (patch) | |
tree | 0617fb1baba1f867f534338ec0c90ed46d02c1e5 /llvm/lib/CodeGen/SelectionDAG | |
parent | a98c7ead306647142fd954f64cc9d5184a17fd28 (diff) | |
download | bcm5719-llvm-8d1052f55cf807afc4fb21734d36f166b83ae4ea.tar.gz bcm5719-llvm-8d1052f55cf807afc4fb21734d36f166b83ae4ea.zip |
DAGCombiner: Reduce 64-bit BFE pattern to pattern on 32-bit component
If the extracted bits are restricted to the upper half or lower half,
this can be truncated.
llvm-svn: 267024
Diffstat (limited to 'llvm/lib/CodeGen/SelectionDAG')
-rw-r--r-- | llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp | 44 |
1 files changed, 44 insertions, 0 deletions
diff --git a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp index d1ad438a1bc..6c9800824ef 100644 --- a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp +++ b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp @@ -2969,6 +2969,50 @@ SDValue DAGCombiner::visitANDLike(SDValue N0, SDValue N1, } } + // Reduce bit extract of low half of an integer to the narrower type. + // (and (srl i64:x, K), KMask) -> + // (i64 zero_extend (and (srl (i32 (trunc i64:x)), K)), KMask) + if (N0.getOpcode() == ISD::SRL && N0.hasOneUse()) { + if (ConstantSDNode *CAnd = dyn_cast<ConstantSDNode>(N1)) { + if (ConstantSDNode *CShift = dyn_cast<ConstantSDNode>(N0.getOperand(1))) { + unsigned Size = VT.getSizeInBits(); + const APInt &AndMask = CAnd->getAPIntValue(); + unsigned ShiftBits = CShift->getZExtValue(); + unsigned MaskBits = AndMask.countTrailingOnes(); + EVT HalfVT = EVT::getIntegerVT(*DAG.getContext(), Size / 2); + + if (APIntOps::isMask(AndMask) && + // Required bits must not span the two halves of the integer and + // must fit in the half size type. + (ShiftBits + MaskBits <= Size / 2) && + TLI.isNarrowingProfitable(VT, HalfVT) && + TLI.isTypeDesirableForOp(ISD::AND, HalfVT) && + TLI.isTypeDesirableForOp(ISD::SRL, HalfVT) && + TLI.isTruncateFree(VT, HalfVT) && + TLI.isZExtFree(HalfVT, VT)) { + // The isNarrowingProfitable is to avoid regressions on PPC and + // AArch64 which match a few 64-bit bit insert / bit extract patterns + // on downstream users of this. Those patterns could probably be + // extended to handle extensions mixed in. + + SDValue SL(N0); + assert(ShiftBits != 0 && MaskBits <= Size); + + // Extracting the highest bit of the low half. + EVT ShiftVT = TLI.getShiftAmountTy(HalfVT, DAG.getDataLayout()); + SDValue Trunc = DAG.getNode(ISD::TRUNCATE, SL, HalfVT, + N0.getOperand(0)); + + SDValue NewMask = DAG.getConstant(AndMask.trunc(Size / 2), SL, HalfVT); + SDValue ShiftK = DAG.getConstant(ShiftBits, SL, ShiftVT); + SDValue Shift = DAG.getNode(ISD::SRL, SL, HalfVT, Trunc, ShiftK); + SDValue And = DAG.getNode(ISD::AND, SL, HalfVT, Shift, NewMask); + return DAG.getNode(ISD::ZERO_EXTEND, SL, VT, And); + } + } + } + } + return SDValue(); } |