llvm-project/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp

394 lines
14 KiB
C++

//===-- AMDGPUISelLowering.cpp - AMDGPU Common DAG lowering functions -----===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This is the parent TargetLowering class for hardware code gen targets.
//
//===----------------------------------------------------------------------===//
#include "AMDGPUISelLowering.h"
#include "AMDILIntrinsicInfo.h"
#include "AMDGPUUtil.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
using namespace llvm;
AMDGPUTargetLowering::AMDGPUTargetLowering(TargetMachine &TM) :
AMDILTargetLowering(TM)
{
// We need to custom lower some of the intrinsics
setOperationAction(ISD::INTRINSIC_WO_CHAIN, MVT::Other, Custom);
setOperationAction(ISD::SELECT_CC, MVT::f32, Custom);
setOperationAction(ISD::SELECT_CC, MVT::i32, Custom);
// Library functions. These default to Expand, but we have instructions
// for them.
setOperationAction(ISD::FCEIL, MVT::f32, Legal);
setOperationAction(ISD::FEXP2, MVT::f32, Legal);
setOperationAction(ISD::FRINT, MVT::f32, Legal);
setOperationAction(ISD::UDIV, MVT::i32, Expand);
setOperationAction(ISD::UDIVREM, MVT::i32, Custom);
setOperationAction(ISD::UREM, MVT::i32, Expand);
}
SDValue AMDGPUTargetLowering::LowerOperation(SDValue Op, SelectionDAG &DAG)
const
{
switch (Op.getOpcode()) {
default: return AMDILTargetLowering::LowerOperation(Op, DAG);
case ISD::INTRINSIC_WO_CHAIN: return LowerINTRINSIC_WO_CHAIN(Op, DAG);
case ISD::SELECT_CC: return LowerSELECT_CC(Op, DAG);
case ISD::UDIVREM: return LowerUDIVREM(Op, DAG);
}
}
SDValue AMDGPUTargetLowering::LowerINTRINSIC_WO_CHAIN(SDValue Op,
SelectionDAG &DAG) const
{
unsigned IntrinsicID = cast<ConstantSDNode>(Op.getOperand(0))->getZExtValue();
DebugLoc DL = Op.getDebugLoc();
EVT VT = Op.getValueType();
switch (IntrinsicID) {
default: return Op;
case AMDGPUIntrinsic::AMDIL_abs:
return LowerIntrinsicIABS(Op, DAG);
case AMDGPUIntrinsic::AMDIL_exp:
return DAG.getNode(ISD::FEXP2, DL, VT, Op.getOperand(1));
case AMDGPUIntrinsic::AMDIL_fabs:
return DAG.getNode(ISD::FABS, DL, VT, Op.getOperand(1));
case AMDGPUIntrinsic::AMDGPU_lrp:
return LowerIntrinsicLRP(Op, DAG);
case AMDGPUIntrinsic::AMDIL_fraction:
return DAG.getNode(AMDGPUISD::FRACT, DL, VT, Op.getOperand(1));
case AMDGPUIntrinsic::AMDIL_mad:
return DAG.getNode(AMDILISD::MAD, DL, VT, Op.getOperand(1),
Op.getOperand(2), Op.getOperand(3));
case AMDGPUIntrinsic::AMDIL_max:
return DAG.getNode(AMDGPUISD::FMAX, DL, VT, Op.getOperand(1),
Op.getOperand(2));
case AMDGPUIntrinsic::AMDGPU_imax:
return DAG.getNode(AMDGPUISD::SMAX, DL, VT, Op.getOperand(1),
Op.getOperand(2));
case AMDGPUIntrinsic::AMDGPU_umax:
return DAG.getNode(AMDGPUISD::UMAX, DL, VT, Op.getOperand(1),
Op.getOperand(2));
case AMDGPUIntrinsic::AMDIL_min:
return DAG.getNode(AMDGPUISD::FMIN, DL, VT, Op.getOperand(1),
Op.getOperand(2));
case AMDGPUIntrinsic::AMDGPU_imin:
return DAG.getNode(AMDGPUISD::SMIN, DL, VT, Op.getOperand(1),
Op.getOperand(2));
case AMDGPUIntrinsic::AMDGPU_umin:
return DAG.getNode(AMDGPUISD::UMIN, DL, VT, Op.getOperand(1),
Op.getOperand(2));
case AMDGPUIntrinsic::AMDIL_round_nearest:
return DAG.getNode(ISD::FRINT, DL, VT, Op.getOperand(1));
case AMDGPUIntrinsic::AMDIL_round_posinf:
return DAG.getNode(ISD::FCEIL, DL, VT, Op.getOperand(1));
}
}
///IABS(a) = SMAX(sub(0, a), a)
SDValue AMDGPUTargetLowering::LowerIntrinsicIABS(SDValue Op,
SelectionDAG &DAG) const
{
DebugLoc DL = Op.getDebugLoc();
EVT VT = Op.getValueType();
SDValue Neg = DAG.getNode(ISD::SUB, DL, VT, DAG.getConstant(0, VT),
Op.getOperand(1));
return DAG.getNode(AMDGPUISD::SMAX, DL, VT, Neg, Op.getOperand(1));
}
/// Linear Interpolation
/// LRP(a, b, c) = muladd(a, b, (1 - a) * c)
SDValue AMDGPUTargetLowering::LowerIntrinsicLRP(SDValue Op,
SelectionDAG &DAG) const
{
DebugLoc DL = Op.getDebugLoc();
EVT VT = Op.getValueType();
SDValue OneSubA = DAG.getNode(ISD::FSUB, DL, VT,
DAG.getConstantFP(1.0f, MVT::f32),
Op.getOperand(1));
SDValue OneSubAC = DAG.getNode(ISD::FMUL, DL, VT, OneSubA,
Op.getOperand(3));
return DAG.getNode(AMDILISD::MAD, DL, VT, Op.getOperand(1),
Op.getOperand(2),
OneSubAC);
}
SDValue AMDGPUTargetLowering::LowerSELECT_CC(SDValue Op,
SelectionDAG &DAG) const
{
DebugLoc DL = Op.getDebugLoc();
EVT VT = Op.getValueType();
SDValue LHS = Op.getOperand(0);
SDValue RHS = Op.getOperand(1);
SDValue True = Op.getOperand(2);
SDValue False = Op.getOperand(3);
SDValue CC = Op.getOperand(4);
ISD::CondCode CCOpcode = cast<CondCodeSDNode>(CC)->get();
SDValue Temp;
// LHS and RHS are guaranteed to be the same value type
EVT CompareVT = LHS.getValueType();
// We need all the operands of SELECT_CC to have the same value type, so if
// necessary we need to convert LHS and RHS to be the same type True and
// False. True and False are guaranteed to have the same type as this
// SELECT_CC node.
if (CompareVT != VT) {
ISD::NodeType ConversionOp = ISD::DELETED_NODE;
if (VT == MVT::f32 && CompareVT == MVT::i32) {
if (isUnsignedIntSetCC(CCOpcode)) {
ConversionOp = ISD::UINT_TO_FP;
} else {
ConversionOp = ISD::SINT_TO_FP;
}
} else if (VT == MVT::i32 && CompareVT == MVT::f32) {
ConversionOp = ISD::FP_TO_SINT;
} else {
// I don't think there will be any other type pairings.
assert(!"Unhandled operand type parings in SELECT_CC");
}
// XXX Check the value of LHS and RHS and avoid creating sequences like
// (FTOI (ITOF))
LHS = DAG.getNode(ConversionOp, DL, VT, LHS);
RHS = DAG.getNode(ConversionOp, DL, VT, RHS);
}
// If True is a hardware TRUE value and False is a hardware FALSE value or
// vice-versa we can handle this with a native instruction (SET* instructions).
if ((isHWTrueValue(True) && isHWFalseValue(False))) {
return DAG.getNode(ISD::SELECT_CC, DL, VT, LHS, RHS, True, False, CC);
}
// XXX If True is a hardware TRUE value and False is a hardware FALSE value,
// we can handle this with a native instruction, but we need to swap true
// and false and change the conditional.
if (isHWTrueValue(False) && isHWFalseValue(True)) {
}
// XXX Check if we can lower this to a SELECT or if it is supported by a native
// operation. (The code below does this but we don't have the Instruction
// selection patterns to do this yet.
#if 0
if (isZero(LHS) || isZero(RHS)) {
SDValue Cond = (isZero(LHS) ? RHS : LHS);
bool SwapTF = false;
switch (CCOpcode) {
case ISD::SETOEQ:
case ISD::SETUEQ:
case ISD::SETEQ:
SwapTF = true;
// Fall through
case ISD::SETONE:
case ISD::SETUNE:
case ISD::SETNE:
// We can lower to select
if (SwapTF) {
Temp = True;
True = False;
False = Temp;
}
// CNDE
return DAG.getNode(ISD::SELECT, DL, VT, Cond, True, False);
default:
// Supported by a native operation (CNDGE, CNDGT)
return DAG.getNode(ISD::SELECT_CC, DL, VT, LHS, RHS, True, False, CC);
}
}
#endif
// If we make it this for it means we have no native instructions to handle
// this SELECT_CC, so we must lower it.
SDValue HWTrue, HWFalse;
if (VT == MVT::f32) {
HWTrue = DAG.getConstantFP(1.0f, VT);
HWFalse = DAG.getConstantFP(0.0f, VT);
} else if (VT == MVT::i32) {
HWTrue = DAG.getConstant(-1, VT);
HWFalse = DAG.getConstant(0, VT);
}
else {
assert(!"Unhandled value type in LowerSELECT_CC");
}
// Lower this unsupported SELECT_CC into a combination of two supported
// SELECT_CC operations.
SDValue Cond = DAG.getNode(ISD::SELECT_CC, DL, VT, LHS, RHS, HWTrue, HWFalse, CC);
return DAG.getNode(ISD::SELECT, DL, VT, Cond, True, False);
}
SDValue AMDGPUTargetLowering::LowerUDIVREM(SDValue Op,
SelectionDAG &DAG) const
{
DebugLoc DL = Op.getDebugLoc();
EVT VT = Op.getValueType();
SDValue Num = Op.getOperand(0);
SDValue Den = Op.getOperand(1);
SmallVector<SDValue, 8> Results;
// RCP = URECIP(Den) = 2^32 / Den + e
// e is rounding error.
SDValue RCP = DAG.getNode(AMDGPUISD::URECIP, DL, VT, Den);
// RCP_LO = umulo(RCP, Den) */
SDValue RCP_LO = DAG.getNode(ISD::UMULO, DL, VT, RCP, Den);
// RCP_HI = mulhu (RCP, Den) */
SDValue RCP_HI = DAG.getNode(ISD::MULHU, DL, VT, RCP, Den);
// NEG_RCP_LO = -RCP_LO
SDValue NEG_RCP_LO = DAG.getNode(ISD::SUB, DL, VT, DAG.getConstant(0, VT),
RCP_LO);
// ABS_RCP_LO = (RCP_HI == 0 ? NEG_RCP_LO : RCP_LO)
SDValue ABS_RCP_LO = DAG.getSelectCC(DL, RCP_HI, DAG.getConstant(0, VT),
NEG_RCP_LO, RCP_LO,
ISD::SETEQ);
// Calculate the rounding error from the URECIP instruction
// E = mulhu(ABS_RCP_LO, RCP)
SDValue E = DAG.getNode(ISD::MULHU, DL, VT, ABS_RCP_LO, RCP);
// RCP_A_E = RCP + E
SDValue RCP_A_E = DAG.getNode(ISD::ADD, DL, VT, RCP, E);
// RCP_S_E = RCP - E
SDValue RCP_S_E = DAG.getNode(ISD::SUB, DL, VT, RCP, E);
// Tmp0 = (RCP_HI == 0 ? RCP_A_E : RCP_SUB_E)
SDValue Tmp0 = DAG.getSelectCC(DL, RCP_HI, DAG.getConstant(0, VT),
RCP_A_E, RCP_S_E,
ISD::SETEQ);
// Quotient = mulhu(Tmp0, Num)
SDValue Quotient = DAG.getNode(ISD::MULHU, DL, VT, Tmp0, Num);
// Num_S_Remainder = Quotient * Den
SDValue Num_S_Remainder = DAG.getNode(ISD::UMULO, DL, VT, Quotient, Den);
// Remainder = Num - Num_S_Remainder
SDValue Remainder = DAG.getNode(ISD::SUB, DL, VT, Num, Num_S_Remainder);
// Remainder_GE_Den = (Remainder >= Den ? -1 : 0)
SDValue Remainder_GE_Den = DAG.getSelectCC(DL, Remainder, Den,
DAG.getConstant(-1, VT),
DAG.getConstant(0, VT),
ISD::SETGE);
// Remainder_GE_Zero = (Remainder >= 0 ? -1 : 0)
SDValue Remainder_GE_Zero = DAG.getSelectCC(DL, Remainder,
DAG.getConstant(0, VT),
DAG.getConstant(-1, VT),
DAG.getConstant(0, VT),
ISD::SETGE);
// Tmp1 = Remainder_GE_Den & Remainder_GE_Zero
SDValue Tmp1 = DAG.getNode(ISD::AND, DL, VT, Remainder_GE_Den,
Remainder_GE_Zero);
// Calculate Division result:
// Quotient_A_One = Quotient + 1
SDValue Quotient_A_One = DAG.getNode(ISD::ADD, DL, VT, Quotient,
DAG.getConstant(1, VT));
// Quotient_S_One = Quotient - 1
SDValue Quotient_S_One = DAG.getNode(ISD::SUB, DL, VT, Quotient,
DAG.getConstant(1, VT));
// Div = (Tmp1 == 0 ? Quotient : Quotient_A_One)
SDValue Div = DAG.getSelectCC(DL, Tmp1, DAG.getConstant(0, VT),
Quotient, Quotient_A_One, ISD::SETEQ);
// Div = (Remainder_GE_Zero == 0 ? Quotient_S_One : Div)
Div = DAG.getSelectCC(DL, Remainder_GE_Zero, DAG.getConstant(0, VT),
Quotient_S_One, Div, ISD::SETEQ);
// Calculate Rem result:
// Remainder_S_Den = Remainder - Den
SDValue Remainder_S_Den = DAG.getNode(ISD::SUB, DL, VT, Remainder, Den);
// Remainder_A_Den = Remainder + Den
SDValue Remainder_A_Den = DAG.getNode(ISD::ADD, DL, VT, Remainder, Den);
// Rem = (Tmp1 == 0 ? Remainder : Remainder_S_Den)
SDValue Rem = DAG.getSelectCC(DL, Tmp1, DAG.getConstant(0, VT),
Remainder, Remainder_S_Den, ISD::SETEQ);
// Rem = (Remainder_GE_Zero == 0 ? Remainder_A_Den : Rem)
Rem = DAG.getSelectCC(DL, Remainder_GE_Zero, DAG.getConstant(0, VT),
Remainder_A_Den, Rem, ISD::SETEQ);
DAG.ReplaceAllUsesWith(Op.getValue(0).getNode(), &Div);
DAG.ReplaceAllUsesWith(Op.getValue(1).getNode(), &Rem);
return Op;
}
//===----------------------------------------------------------------------===//
// Helper functions
//===----------------------------------------------------------------------===//
bool AMDGPUTargetLowering::isHWTrueValue(SDValue Op) const
{
if (ConstantFPSDNode * CFP = dyn_cast<ConstantFPSDNode>(Op)) {
return CFP->isExactlyValue(1.0);
}
if (ConstantSDNode *C = dyn_cast<ConstantSDNode>(Op)) {
return C->isAllOnesValue();
}
return false;
}
bool AMDGPUTargetLowering::isHWFalseValue(SDValue Op) const
{
if (ConstantFPSDNode * CFP = dyn_cast<ConstantFPSDNode>(Op)) {
return CFP->getValueAPF().isZero();
}
if (ConstantSDNode *C = dyn_cast<ConstantSDNode>(Op)) {
return C->isNullValue();
}
return false;
}
void AMDGPUTargetLowering::addLiveIn(MachineInstr * MI,
MachineFunction * MF, MachineRegisterInfo & MRI,
const TargetInstrInfo * TII, unsigned reg) const
{
AMDGPU::utilAddLiveIn(MF, MRI, TII, reg, MI->getOperand(0).getReg());
}
#define NODE_NAME_CASE(node) case AMDGPUISD::node: return #node;
const char* AMDGPUTargetLowering::getTargetNodeName(unsigned Opcode) const
{
switch (Opcode) {
default: return AMDILTargetLowering::getTargetNodeName(Opcode);
NODE_NAME_CASE(FRACT)
NODE_NAME_CASE(FMAX)
NODE_NAME_CASE(SMAX)
NODE_NAME_CASE(UMAX)
NODE_NAME_CASE(FMIN)
NODE_NAME_CASE(SMIN)
NODE_NAME_CASE(UMIN)
NODE_NAME_CASE(URECIP)
}
}