setOperationAction(ISD::STACKRESTORE, MVT::Other, Expand);
setOperationAction(ISD::DYNAMIC_STACKALLOC, MVT::i64, Expand);
- // Exception handling.
- // FIXME: These are guesses. Has this been defined yet?
- setExceptionPointerRegister(AArch64::X0);
- setExceptionSelectorRegister(AArch64::X1);
-
// Constant pool entries
setOperationAction(ISD::ConstantPool, MVT::i64, Custom);
// AArch64 lacks both left-rotate and popcount instructions.
setOperationAction(ISD::ROTL, MVT::i32, Expand);
setOperationAction(ISD::ROTL, MVT::i64, Expand);
+ for (MVT VT : MVT::vector_valuetypes()) {
+ setOperationAction(ISD::ROTL, VT, Expand);
+ setOperationAction(ISD::ROTR, VT, Expand);
+ }
// AArch64 doesn't have {U|S}MUL_LOHI.
setOperationAction(ISD::UMUL_LOHI, MVT::i64, Expand);
setTargetDAGCombine(ISD::BITCAST);
setTargetDAGCombine(ISD::CONCAT_VECTORS);
setTargetDAGCombine(ISD::STORE);
+ if (Subtarget->supportsAddressTopByteIgnored())
+ setTargetDAGCombine(ISD::LOAD);
setTargetDAGCombine(ISD::MUL);
SDValue AArch64TargetLowering::LowerF128Call(SDValue Op, SelectionDAG &DAG,
RTLIB::Libcall Call) const {
SmallVector<SDValue, 2> Ops(Op->op_begin(), Op->op_end());
- return makeLibCall(DAG, Call, MVT::f128, &Ops[0], Ops.size(), false,
- SDLoc(Op)).first;
+ return makeLibCall(DAG, Call, MVT::f128, Ops, false, SDLoc(Op)).first;
}
static SDValue LowerXOR(SDValue Op, SelectionDAG &DAG) {
// precise. That doesn't take part in the LibCall so we can't directly use
// LowerF128Call.
SDValue SrcVal = Op.getOperand(0);
- return makeLibCall(DAG, LC, Op.getValueType(), &SrcVal, 1,
- /*isSigned*/ false, SDLoc(Op)).first;
+ return makeLibCall(DAG, LC, Op.getValueType(), SrcVal, /*isSigned*/ false,
+ SDLoc(Op)).first;
}
static SDValue LowerVectorFP_TO_INT(SDValue Op, SelectionDAG &DAG) {
LC = RTLIB::getFPTOUINT(Op.getOperand(0).getValueType(), Op.getValueType());
SmallVector<SDValue, 2> Ops(Op->op_begin(), Op->op_end());
- return makeLibCall(DAG, LC, Op.getValueType(), &Ops[0], Ops.size(), false,
- SDLoc(Op)).first;
+ return makeLibCall(DAG, LC, Op.getValueType(), Ops, false, SDLoc(Op)).first;
}
static SDValue LowerVectorINT_TO_FP(SDValue Op, SelectionDAG &DAG) {
unsigned Val = Cst->getZExtValue();
unsigned Size = Op.getValueType().getSizeInBits();
- if (Val == 0) {
- switch (Size) {
- case 8:
- return DAG.getTargetExtractSubreg(AArch64::bsub, dl, Op.getValueType(),
- Op.getOperand(0));
- case 16:
- return DAG.getTargetExtractSubreg(AArch64::hsub, dl, Op.getValueType(),
- Op.getOperand(0));
- case 32:
- return DAG.getTargetExtractSubreg(AArch64::ssub, dl, Op.getValueType(),
- Op.getOperand(0));
- case 64:
- return DAG.getTargetExtractSubreg(AArch64::dsub, dl, Op.getValueType(),
- Op.getOperand(0));
- default:
- llvm_unreachable("Unexpected vector type in extract_subvector!");
- }
- }
+
+ // This will get lowered to an appropriate EXTRACT_SUBREG in ISel.
+ if (Val == 0)
+ return Op;
+
// If this is extracting the upper 64-bits of a 128-bit vector, we match
// that directly.
if (Size == 64 && Val * VT.getVectorElementType().getSizeInBits() == 64)
return NewST1;
}
-static SDValue performSTORECombine(SDNode *N,
- TargetLowering::DAGCombinerInfo &DCI,
- SelectionDAG &DAG,
- const AArch64Subtarget *Subtarget) {
+static SDValue split16BStores(SDNode *N, TargetLowering::DAGCombinerInfo &DCI,
+ SelectionDAG &DAG,
+ const AArch64Subtarget *Subtarget) {
if (!DCI.isBeforeLegalize())
return SDValue();
return SDValue();
}
-/// This function handles the log2-shuffle pattern produced by the
+/// Simplify \Addr given that the top byte of it is ignored by HW during
+/// address translation.
+static bool performTBISimplification(SDValue Addr,
+ TargetLowering::DAGCombinerInfo &DCI,
+ SelectionDAG &DAG) {
+ APInt DemandedMask = APInt::getLowBitsSet(64, 56);
+ APInt KnownZero, KnownOne;
+ TargetLowering::TargetLoweringOpt TLO(DAG, DCI.isBeforeLegalize(),
+ DCI.isBeforeLegalizeOps());
+ const TargetLowering &TLI = DAG.getTargetLoweringInfo();
+ if (TLI.SimplifyDemandedBits(Addr, DemandedMask, KnownZero, KnownOne, TLO)) {
+ DCI.CommitTargetLoweringOpt(TLO);
+ return true;
+ }
+ return false;
+}
+
+static SDValue performSTORECombine(SDNode *N,
+ TargetLowering::DAGCombinerInfo &DCI,
+ SelectionDAG &DAG,
+ const AArch64Subtarget *Subtarget) {
+ SDValue Split = split16BStores(N, DCI, DAG, Subtarget);
+ if (Split.getNode())
+ return Split;
+
+ if (Subtarget->supportsAddressTopByteIgnored() &&
+ performTBISimplification(N->getOperand(2), DCI, DAG))
+ return SDValue(N, 0);
+
+ return SDValue();
+}
+
+ /// This function handles the log2-shuffle pattern produced by the
/// LoopVectorizer for the across vector reduction. It consists of
/// log2(NumVectorElements) steps and, in each step, 2^(s) elements
/// are reduced, where s is an induction variable from 0 to
}
case ISD::VSELECT:
return performVSelectCombine(N, DCI.DAG);
+ case ISD::LOAD:
+ if (performTBISimplification(N->getOperand(1), DCI, DAG))
+ return SDValue(N, 0);
+ break;
case ISD::STORE:
return performSTORECombine(N, DCI, DAG, Subtarget);
case AArch64ISD::BRCOND:
Results.push_back(DAG.getNode(ISD::TRUNCATE, DL, MVT::i16, Op));
}
+static void ReplaceReductionResults(SDNode *N,
+ SmallVectorImpl<SDValue> &Results,
+ SelectionDAG &DAG, unsigned InterOp,
+ unsigned AcrossOp) {
+ EVT LoVT, HiVT;
+ SDValue Lo, Hi;
+ SDLoc dl(N);
+ std::tie(LoVT, HiVT) = DAG.GetSplitDestVTs(N->getValueType(0));
+ std::tie(Lo, Hi) = DAG.SplitVectorOperand(N, 0);
+ SDValue InterVal = DAG.getNode(InterOp, dl, LoVT, Lo, Hi);
+ SDValue SplitVal = DAG.getNode(AcrossOp, dl, LoVT, InterVal);
+ Results.push_back(SplitVal);
+}
+
void AArch64TargetLowering::ReplaceNodeResults(
SDNode *N, SmallVectorImpl<SDValue> &Results, SelectionDAG &DAG) const {
switch (N->getOpcode()) {
case ISD::BITCAST:
ReplaceBITCASTResults(N, Results, DAG);
return;
+ case AArch64ISD::SADDV:
+ ReplaceReductionResults(N, Results, DAG, ISD::ADD, AArch64ISD::SADDV);
+ return;
+ case AArch64ISD::UADDV:
+ ReplaceReductionResults(N, Results, DAG, ISD::ADD, AArch64ISD::UADDV);
+ return;
+ case AArch64ISD::SMINV:
+ ReplaceReductionResults(N, Results, DAG, ISD::SMIN, AArch64ISD::SMINV);
+ return;
+ case AArch64ISD::UMINV:
+ ReplaceReductionResults(N, Results, DAG, ISD::UMIN, AArch64ISD::UMINV);
+ return;
+ case AArch64ISD::SMAXV:
+ ReplaceReductionResults(N, Results, DAG, ISD::SMAX, AArch64ISD::SMAXV);
+ return;
+ case AArch64ISD::UMAXV:
+ ReplaceReductionResults(N, Results, DAG, ISD::UMAX, AArch64ISD::UMAXV);
+ return;
case ISD::FP_TO_UINT:
case ISD::FP_TO_SINT:
assert(N->getValueType(0) == MVT::i128 && "unexpected illegal conversion");
EVT) const {
return false;
}
+
+Value *AArch64TargetLowering::getSafeStackPointerLocation(IRBuilder<> &IRB) const {
+ if (!Subtarget->isTargetAndroid())
+ return TargetLowering::getSafeStackPointerLocation(IRB);
+
+ // Android provides a fixed TLS slot for the SafeStack pointer. See the
+ // definition of TLS_SLOT_SAFESTACK in
+ // https://android.googlesource.com/platform/bionic/+/master/libc/private/bionic_tls.h
+ const unsigned TlsOffset = 0x48;
+ Module *M = IRB.GetInsertBlock()->getParent()->getParent();
+ Function *ThreadPointerFunc =
+ Intrinsic::getDeclaration(M, Intrinsic::aarch64_thread_pointer);
+ return IRB.CreatePointerCast(
+ IRB.CreateConstGEP1_32(IRB.CreateCall(ThreadPointerFunc), TlsOffset),
+ Type::getInt8PtrTy(IRB.getContext())->getPointerTo(0));
+}