mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-08-29 13:29:28 +00:00
When unsafe FP math operations are enabled, we can use the fre[s] and frsqrte[s] instructions, which generate reciprocal (sqrt) estimates, together with some Newton iteration, in order to quickly generate floating-point division and sqrt results. All of these instructions are separately optional, and so each has its own feature flag (except for the Altivec instructions, which are covered under the existing Altivec flag). Doing this is not only faster than using the IEEE-compliant fdiv/fsqrt instructions, but allows these computations to be pipelined with other computations in order to hide their overall latency. I've also added a couple of missing fnmsub patterns which turned out to be missing (but are necessary for good code generation of the Newton iterations). Altivec needs a similar fix, but that will probably be more complicated because fneg is expanded for Altivec's v4f32. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@178617 91177308-0d34-0410-b5e6-96231b3b80d8
1718 lines
76 KiB
TableGen
1718 lines
76 KiB
TableGen
//===-- PPCInstrInfo.td - The PowerPC Instruction Set ------*- tablegen -*-===//
|
|
//
|
|
// The LLVM Compiler Infrastructure
|
|
//
|
|
// This file is distributed under the University of Illinois Open Source
|
|
// License. See LICENSE.TXT for details.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
//
|
|
// This file describes the subset of the 32-bit PowerPC instruction set, as used
|
|
// by the PowerPC instruction selector.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
include "PPCInstrFormats.td"
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// PowerPC specific type constraints.
|
|
//
|
|
def SDT_PPCstfiwx : SDTypeProfile<0, 2, [ // stfiwx
|
|
SDTCisVT<0, f64>, SDTCisPtrTy<1>
|
|
]>;
|
|
def SDT_PPClfiwx : SDTypeProfile<1, 1, [ // lfiw[az]x
|
|
SDTCisVT<0, f64>, SDTCisPtrTy<1>
|
|
]>;
|
|
|
|
def SDT_PPCCallSeqStart : SDCallSeqStart<[ SDTCisVT<0, i32> ]>;
|
|
def SDT_PPCCallSeqEnd : SDCallSeqEnd<[ SDTCisVT<0, i32>,
|
|
SDTCisVT<1, i32> ]>;
|
|
def SDT_PPCvperm : SDTypeProfile<1, 3, [
|
|
SDTCisVT<3, v16i8>, SDTCisSameAs<0, 1>, SDTCisSameAs<0, 2>
|
|
]>;
|
|
|
|
def SDT_PPCvcmp : SDTypeProfile<1, 3, [
|
|
SDTCisSameAs<0, 1>, SDTCisSameAs<1, 2>, SDTCisVT<3, i32>
|
|
]>;
|
|
|
|
def SDT_PPCcondbr : SDTypeProfile<0, 3, [
|
|
SDTCisVT<0, i32>, SDTCisVT<2, OtherVT>
|
|
]>;
|
|
|
|
def SDT_PPClbrx : SDTypeProfile<1, 2, [
|
|
SDTCisInt<0>, SDTCisPtrTy<1>, SDTCisVT<2, OtherVT>
|
|
]>;
|
|
def SDT_PPCstbrx : SDTypeProfile<0, 3, [
|
|
SDTCisInt<0>, SDTCisPtrTy<1>, SDTCisVT<2, OtherVT>
|
|
]>;
|
|
|
|
def SDT_PPClarx : SDTypeProfile<1, 1, [
|
|
SDTCisInt<0>, SDTCisPtrTy<1>
|
|
]>;
|
|
def SDT_PPCstcx : SDTypeProfile<0, 2, [
|
|
SDTCisInt<0>, SDTCisPtrTy<1>
|
|
]>;
|
|
|
|
def SDT_PPCTC_ret : SDTypeProfile<0, 2, [
|
|
SDTCisPtrTy<0>, SDTCisVT<1, i32>
|
|
]>;
|
|
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// PowerPC specific DAG Nodes.
|
|
//
|
|
|
|
def PPCfre : SDNode<"PPCISD::FRE", SDTFPUnaryOp, []>;
|
|
def PPCfrsqrte: SDNode<"PPCISD::FRSQRTE", SDTFPUnaryOp, []>;
|
|
|
|
def PPCfcfid : SDNode<"PPCISD::FCFID", SDTFPUnaryOp, []>;
|
|
def PPCfcfidu : SDNode<"PPCISD::FCFIDU", SDTFPUnaryOp, []>;
|
|
def PPCfcfids : SDNode<"PPCISD::FCFIDS", SDTFPRoundOp, []>;
|
|
def PPCfcfidus: SDNode<"PPCISD::FCFIDUS", SDTFPRoundOp, []>;
|
|
def PPCfctidz : SDNode<"PPCISD::FCTIDZ", SDTFPUnaryOp, []>;
|
|
def PPCfctiwz : SDNode<"PPCISD::FCTIWZ", SDTFPUnaryOp, []>;
|
|
def PPCfctiduz: SDNode<"PPCISD::FCTIDUZ",SDTFPUnaryOp, []>;
|
|
def PPCfctiwuz: SDNode<"PPCISD::FCTIWUZ",SDTFPUnaryOp, []>;
|
|
def PPCstfiwx : SDNode<"PPCISD::STFIWX", SDT_PPCstfiwx,
|
|
[SDNPHasChain, SDNPMayStore]>;
|
|
def PPClfiwax : SDNode<"PPCISD::LFIWAX", SDT_PPClfiwx,
|
|
[SDNPHasChain, SDNPMayLoad]>;
|
|
def PPClfiwzx : SDNode<"PPCISD::LFIWZX", SDT_PPClfiwx,
|
|
[SDNPHasChain, SDNPMayLoad]>;
|
|
|
|
// Extract FPSCR (not modeled at the DAG level).
|
|
def PPCmffs : SDNode<"PPCISD::MFFS",
|
|
SDTypeProfile<1, 0, [SDTCisVT<0, f64>]>, []>;
|
|
|
|
// Perform FADD in round-to-zero mode.
|
|
def PPCfaddrtz: SDNode<"PPCISD::FADDRTZ", SDTFPBinOp, []>;
|
|
|
|
|
|
def PPCfsel : SDNode<"PPCISD::FSEL",
|
|
// Type constraint for fsel.
|
|
SDTypeProfile<1, 3, [SDTCisSameAs<0, 2>, SDTCisSameAs<0, 3>,
|
|
SDTCisFP<0>, SDTCisVT<1, f64>]>, []>;
|
|
|
|
def PPChi : SDNode<"PPCISD::Hi", SDTIntBinOp, []>;
|
|
def PPClo : SDNode<"PPCISD::Lo", SDTIntBinOp, []>;
|
|
def PPCtoc_entry: SDNode<"PPCISD::TOC_ENTRY", SDTIntBinOp, [SDNPMayLoad]>;
|
|
def PPCvmaddfp : SDNode<"PPCISD::VMADDFP", SDTFPTernaryOp, []>;
|
|
def PPCvnmsubfp : SDNode<"PPCISD::VNMSUBFP", SDTFPTernaryOp, []>;
|
|
|
|
def PPCaddisGotTprelHA : SDNode<"PPCISD::ADDIS_GOT_TPREL_HA", SDTIntBinOp>;
|
|
def PPCldGotTprelL : SDNode<"PPCISD::LD_GOT_TPREL_L", SDTIntBinOp,
|
|
[SDNPMayLoad]>;
|
|
def PPCaddTls : SDNode<"PPCISD::ADD_TLS", SDTIntBinOp, []>;
|
|
def PPCaddisTlsgdHA : SDNode<"PPCISD::ADDIS_TLSGD_HA", SDTIntBinOp>;
|
|
def PPCaddiTlsgdL : SDNode<"PPCISD::ADDI_TLSGD_L", SDTIntBinOp>;
|
|
def PPCgetTlsAddr : SDNode<"PPCISD::GET_TLS_ADDR", SDTIntBinOp>;
|
|
def PPCaddisTlsldHA : SDNode<"PPCISD::ADDIS_TLSLD_HA", SDTIntBinOp>;
|
|
def PPCaddiTlsldL : SDNode<"PPCISD::ADDI_TLSLD_L", SDTIntBinOp>;
|
|
def PPCgetTlsldAddr : SDNode<"PPCISD::GET_TLSLD_ADDR", SDTIntBinOp>;
|
|
def PPCaddisDtprelHA : SDNode<"PPCISD::ADDIS_DTPREL_HA", SDTIntBinOp,
|
|
[SDNPHasChain]>;
|
|
def PPCaddiDtprelL : SDNode<"PPCISD::ADDI_DTPREL_L", SDTIntBinOp>;
|
|
|
|
def PPCvperm : SDNode<"PPCISD::VPERM", SDT_PPCvperm, []>;
|
|
|
|
// These nodes represent the 32-bit PPC shifts that operate on 6-bit shift
|
|
// amounts. These nodes are generated by the multi-precision shift code.
|
|
def PPCsrl : SDNode<"PPCISD::SRL" , SDTIntShiftOp>;
|
|
def PPCsra : SDNode<"PPCISD::SRA" , SDTIntShiftOp>;
|
|
def PPCshl : SDNode<"PPCISD::SHL" , SDTIntShiftOp>;
|
|
|
|
// These are target-independent nodes, but have target-specific formats.
|
|
def callseq_start : SDNode<"ISD::CALLSEQ_START", SDT_PPCCallSeqStart,
|
|
[SDNPHasChain, SDNPOutGlue]>;
|
|
def callseq_end : SDNode<"ISD::CALLSEQ_END", SDT_PPCCallSeqEnd,
|
|
[SDNPHasChain, SDNPOptInGlue, SDNPOutGlue]>;
|
|
|
|
def SDT_PPCCall : SDTypeProfile<0, -1, [SDTCisInt<0>]>;
|
|
def PPCcall : SDNode<"PPCISD::CALL", SDT_PPCCall,
|
|
[SDNPHasChain, SDNPOptInGlue, SDNPOutGlue,
|
|
SDNPVariadic]>;
|
|
def PPCcall_nop : SDNode<"PPCISD::CALL_NOP", SDT_PPCCall,
|
|
[SDNPHasChain, SDNPOptInGlue, SDNPOutGlue,
|
|
SDNPVariadic]>;
|
|
def PPCload : SDNode<"PPCISD::LOAD", SDTypeProfile<1, 1, []>,
|
|
[SDNPHasChain, SDNPOptInGlue, SDNPOutGlue]>;
|
|
def PPCload_toc : SDNode<"PPCISD::LOAD_TOC", SDTypeProfile<0, 1, []>,
|
|
[SDNPHasChain, SDNPSideEffect,
|
|
SDNPInGlue, SDNPOutGlue]>;
|
|
def PPCtoc_restore : SDNode<"PPCISD::TOC_RESTORE", SDTypeProfile<0, 0, []>,
|
|
[SDNPHasChain, SDNPSideEffect,
|
|
SDNPInGlue, SDNPOutGlue]>;
|
|
def PPCmtctr : SDNode<"PPCISD::MTCTR", SDT_PPCCall,
|
|
[SDNPHasChain, SDNPOptInGlue, SDNPOutGlue]>;
|
|
def PPCbctrl : SDNode<"PPCISD::BCTRL", SDTNone,
|
|
[SDNPHasChain, SDNPOptInGlue, SDNPOutGlue,
|
|
SDNPVariadic]>;
|
|
|
|
def retflag : SDNode<"PPCISD::RET_FLAG", SDTNone,
|
|
[SDNPHasChain, SDNPOptInGlue, SDNPVariadic]>;
|
|
|
|
def PPCtc_return : SDNode<"PPCISD::TC_RETURN", SDT_PPCTC_ret,
|
|
[SDNPHasChain, SDNPOptInGlue, SDNPVariadic]>;
|
|
|
|
def PPCeh_sjlj_setjmp : SDNode<"PPCISD::EH_SJLJ_SETJMP",
|
|
SDTypeProfile<1, 1, [SDTCisInt<0>,
|
|
SDTCisPtrTy<1>]>,
|
|
[SDNPHasChain, SDNPSideEffect]>;
|
|
def PPCeh_sjlj_longjmp : SDNode<"PPCISD::EH_SJLJ_LONGJMP",
|
|
SDTypeProfile<0, 1, [SDTCisPtrTy<0>]>,
|
|
[SDNPHasChain, SDNPSideEffect]>;
|
|
|
|
def PPCvcmp : SDNode<"PPCISD::VCMP" , SDT_PPCvcmp, []>;
|
|
def PPCvcmp_o : SDNode<"PPCISD::VCMPo", SDT_PPCvcmp, [SDNPOutGlue]>;
|
|
|
|
def PPCcondbranch : SDNode<"PPCISD::COND_BRANCH", SDT_PPCcondbr,
|
|
[SDNPHasChain, SDNPOptInGlue]>;
|
|
|
|
def PPClbrx : SDNode<"PPCISD::LBRX", SDT_PPClbrx,
|
|
[SDNPHasChain, SDNPMayLoad]>;
|
|
def PPCstbrx : SDNode<"PPCISD::STBRX", SDT_PPCstbrx,
|
|
[SDNPHasChain, SDNPMayStore]>;
|
|
|
|
// Instructions to set/unset CR bit 6 for SVR4 vararg calls
|
|
def PPCcr6set : SDNode<"PPCISD::CR6SET", SDTNone,
|
|
[SDNPHasChain, SDNPOptInGlue, SDNPOutGlue]>;
|
|
def PPCcr6unset : SDNode<"PPCISD::CR6UNSET", SDTNone,
|
|
[SDNPHasChain, SDNPOptInGlue, SDNPOutGlue]>;
|
|
|
|
// Instructions to support atomic operations
|
|
def PPClarx : SDNode<"PPCISD::LARX", SDT_PPClarx,
|
|
[SDNPHasChain, SDNPMayLoad]>;
|
|
def PPCstcx : SDNode<"PPCISD::STCX", SDT_PPCstcx,
|
|
[SDNPHasChain, SDNPMayStore]>;
|
|
|
|
// Instructions to support medium and large code model
|
|
def PPCaddisTocHA : SDNode<"PPCISD::ADDIS_TOC_HA", SDTIntBinOp, []>;
|
|
def PPCldTocL : SDNode<"PPCISD::LD_TOC_L", SDTIntBinOp, [SDNPMayLoad]>;
|
|
def PPCaddiTocL : SDNode<"PPCISD::ADDI_TOC_L", SDTIntBinOp, []>;
|
|
|
|
|
|
// Instructions to support dynamic alloca.
|
|
def SDTDynOp : SDTypeProfile<1, 2, []>;
|
|
def PPCdynalloc : SDNode<"PPCISD::DYNALLOC", SDTDynOp, [SDNPHasChain]>;
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// PowerPC specific transformation functions and pattern fragments.
|
|
//
|
|
|
|
def SHL32 : SDNodeXForm<imm, [{
|
|
// Transformation function: 31 - imm
|
|
return getI32Imm(31 - N->getZExtValue());
|
|
}]>;
|
|
|
|
def SRL32 : SDNodeXForm<imm, [{
|
|
// Transformation function: 32 - imm
|
|
return N->getZExtValue() ? getI32Imm(32 - N->getZExtValue()) : getI32Imm(0);
|
|
}]>;
|
|
|
|
def LO16 : SDNodeXForm<imm, [{
|
|
// Transformation function: get the low 16 bits.
|
|
return getI32Imm((unsigned short)N->getZExtValue());
|
|
}]>;
|
|
|
|
def HI16 : SDNodeXForm<imm, [{
|
|
// Transformation function: shift the immediate value down into the low bits.
|
|
return getI32Imm((unsigned)N->getZExtValue() >> 16);
|
|
}]>;
|
|
|
|
def HA16 : SDNodeXForm<imm, [{
|
|
// Transformation function: shift the immediate value down into the low bits.
|
|
signed int Val = N->getZExtValue();
|
|
return getI32Imm((Val - (signed short)Val) >> 16);
|
|
}]>;
|
|
def MB : SDNodeXForm<imm, [{
|
|
// Transformation function: get the start bit of a mask
|
|
unsigned mb = 0, me;
|
|
(void)isRunOfOnes((unsigned)N->getZExtValue(), mb, me);
|
|
return getI32Imm(mb);
|
|
}]>;
|
|
|
|
def ME : SDNodeXForm<imm, [{
|
|
// Transformation function: get the end bit of a mask
|
|
unsigned mb, me = 0;
|
|
(void)isRunOfOnes((unsigned)N->getZExtValue(), mb, me);
|
|
return getI32Imm(me);
|
|
}]>;
|
|
def maskimm32 : PatLeaf<(imm), [{
|
|
// maskImm predicate - True if immediate is a run of ones.
|
|
unsigned mb, me;
|
|
if (N->getValueType(0) == MVT::i32)
|
|
return isRunOfOnes((unsigned)N->getZExtValue(), mb, me);
|
|
else
|
|
return false;
|
|
}]>;
|
|
|
|
def immSExt16 : PatLeaf<(imm), [{
|
|
// immSExt16 predicate - True if the immediate fits in a 16-bit sign extended
|
|
// field. Used by instructions like 'addi'.
|
|
if (N->getValueType(0) == MVT::i32)
|
|
return (int32_t)N->getZExtValue() == (short)N->getZExtValue();
|
|
else
|
|
return (int64_t)N->getZExtValue() == (short)N->getZExtValue();
|
|
}]>;
|
|
def immZExt16 : PatLeaf<(imm), [{
|
|
// immZExt16 predicate - True if the immediate fits in a 16-bit zero extended
|
|
// field. Used by instructions like 'ori'.
|
|
return (uint64_t)N->getZExtValue() == (unsigned short)N->getZExtValue();
|
|
}], LO16>;
|
|
|
|
// imm16Shifted* - These match immediates where the low 16-bits are zero. There
|
|
// are two forms: imm16ShiftedSExt and imm16ShiftedZExt. These two forms are
|
|
// identical in 32-bit mode, but in 64-bit mode, they return true if the
|
|
// immediate fits into a sign/zero extended 32-bit immediate (with the low bits
|
|
// clear).
|
|
def imm16ShiftedZExt : PatLeaf<(imm), [{
|
|
// imm16ShiftedZExt predicate - True if only bits in the top 16-bits of the
|
|
// immediate are set. Used by instructions like 'xoris'.
|
|
return (N->getZExtValue() & ~uint64_t(0xFFFF0000)) == 0;
|
|
}], HI16>;
|
|
|
|
def imm16ShiftedSExt : PatLeaf<(imm), [{
|
|
// imm16ShiftedSExt predicate - True if only bits in the top 16-bits of the
|
|
// immediate are set. Used by instructions like 'addis'. Identical to
|
|
// imm16ShiftedZExt in 32-bit mode.
|
|
if (N->getZExtValue() & 0xFFFF) return false;
|
|
if (N->getValueType(0) == MVT::i32)
|
|
return true;
|
|
// For 64-bit, make sure it is sext right.
|
|
return N->getZExtValue() == (uint64_t)(int)N->getZExtValue();
|
|
}], HI16>;
|
|
|
|
// Some r+i load/store instructions (such as LD, STD, LDU, etc.) that require
|
|
// restricted memrix (offset/4) constants are alignment sensitive. If these
|
|
// offsets are hidden behind TOC entries than the values of the lower-order
|
|
// bits cannot be checked directly. As a result, we need to also incorporate
|
|
// an alignment check into the relevant patterns.
|
|
|
|
def aligned4load : PatFrag<(ops node:$ptr), (load node:$ptr), [{
|
|
return cast<LoadSDNode>(N)->getAlignment() >= 4;
|
|
}]>;
|
|
def aligned4store : PatFrag<(ops node:$val, node:$ptr),
|
|
(store node:$val, node:$ptr), [{
|
|
return cast<StoreSDNode>(N)->getAlignment() >= 4;
|
|
}]>;
|
|
def aligned4sextloadi32 : PatFrag<(ops node:$ptr), (sextloadi32 node:$ptr), [{
|
|
return cast<LoadSDNode>(N)->getAlignment() >= 4;
|
|
}]>;
|
|
def aligned4pre_store : PatFrag<
|
|
(ops node:$val, node:$base, node:$offset),
|
|
(pre_store node:$val, node:$base, node:$offset), [{
|
|
return cast<StoreSDNode>(N)->getAlignment() >= 4;
|
|
}]>;
|
|
|
|
def unaligned4load : PatFrag<(ops node:$ptr), (load node:$ptr), [{
|
|
return cast<LoadSDNode>(N)->getAlignment() < 4;
|
|
}]>;
|
|
def unaligned4store : PatFrag<(ops node:$val, node:$ptr),
|
|
(store node:$val, node:$ptr), [{
|
|
return cast<StoreSDNode>(N)->getAlignment() < 4;
|
|
}]>;
|
|
def unaligned4sextloadi32 : PatFrag<(ops node:$ptr), (sextloadi32 node:$ptr), [{
|
|
return cast<LoadSDNode>(N)->getAlignment() < 4;
|
|
}]>;
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// PowerPC Flag Definitions.
|
|
|
|
class isPPC64 { bit PPC64 = 1; }
|
|
class isDOT {
|
|
list<Register> Defs = [CR0];
|
|
bit RC = 1;
|
|
}
|
|
|
|
class RegConstraint<string C> {
|
|
string Constraints = C;
|
|
}
|
|
class NoEncode<string E> {
|
|
string DisableEncoding = E;
|
|
}
|
|
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// PowerPC Operand Definitions.
|
|
|
|
def s5imm : Operand<i32> {
|
|
let PrintMethod = "printS5ImmOperand";
|
|
}
|
|
def u5imm : Operand<i32> {
|
|
let PrintMethod = "printU5ImmOperand";
|
|
}
|
|
def u6imm : Operand<i32> {
|
|
let PrintMethod = "printU6ImmOperand";
|
|
}
|
|
def s16imm : Operand<i32> {
|
|
let PrintMethod = "printS16ImmOperand";
|
|
}
|
|
def u16imm : Operand<i32> {
|
|
let PrintMethod = "printU16ImmOperand";
|
|
}
|
|
def directbrtarget : Operand<OtherVT> {
|
|
let PrintMethod = "printBranchOperand";
|
|
let EncoderMethod = "getDirectBrEncoding";
|
|
}
|
|
def condbrtarget : Operand<OtherVT> {
|
|
let PrintMethod = "printBranchOperand";
|
|
let EncoderMethod = "getCondBrEncoding";
|
|
}
|
|
def calltarget : Operand<iPTR> {
|
|
let EncoderMethod = "getDirectBrEncoding";
|
|
}
|
|
def aaddr : Operand<iPTR> {
|
|
let PrintMethod = "printAbsAddrOperand";
|
|
}
|
|
def symbolHi: Operand<i32> {
|
|
let PrintMethod = "printSymbolHi";
|
|
let EncoderMethod = "getHA16Encoding";
|
|
}
|
|
def symbolLo: Operand<i32> {
|
|
let PrintMethod = "printSymbolLo";
|
|
let EncoderMethod = "getLO16Encoding";
|
|
}
|
|
def crbitm: Operand<i8> {
|
|
let PrintMethod = "printcrbitm";
|
|
let EncoderMethod = "get_crbitm_encoding";
|
|
}
|
|
// Address operands
|
|
// A version of ptr_rc which excludes R0 (or X0 in 64-bit mode).
|
|
def ptr_rc_nor0 : PointerLikeRegClass<1>;
|
|
|
|
def dispRI : Operand<iPTR>;
|
|
def dispRIX : Operand<iPTR>;
|
|
|
|
def memri : Operand<iPTR> {
|
|
let PrintMethod = "printMemRegImm";
|
|
let MIOperandInfo = (ops dispRI:$imm, ptr_rc_nor0:$reg);
|
|
let EncoderMethod = "getMemRIEncoding";
|
|
}
|
|
def memrr : Operand<iPTR> {
|
|
let PrintMethod = "printMemRegReg";
|
|
let MIOperandInfo = (ops ptr_rc_nor0:$ptrreg, ptr_rc:$offreg);
|
|
}
|
|
def memrix : Operand<iPTR> { // memri where the imm is shifted 2 bits.
|
|
let PrintMethod = "printMemRegImmShifted";
|
|
let MIOperandInfo = (ops dispRIX:$imm, ptr_rc_nor0:$reg);
|
|
let EncoderMethod = "getMemRIXEncoding";
|
|
}
|
|
|
|
// A single-register address. This is used with the SjLj
|
|
// pseudo-instructions.
|
|
def memr : Operand<iPTR> {
|
|
let MIOperandInfo = (ops ptr_rc:$ptrreg);
|
|
}
|
|
|
|
// PowerPC Predicate operand.
|
|
def pred : Operand<OtherVT> {
|
|
let PrintMethod = "printPredicateOperand";
|
|
let MIOperandInfo = (ops i32imm:$bibo, CRRC:$reg);
|
|
}
|
|
|
|
// Define PowerPC specific addressing mode.
|
|
def iaddr : ComplexPattern<iPTR, 2, "SelectAddrImm", [], []>;
|
|
def xaddr : ComplexPattern<iPTR, 2, "SelectAddrIdx", [], []>;
|
|
def xoaddr : ComplexPattern<iPTR, 2, "SelectAddrIdxOnly",[], []>;
|
|
def ixaddr : ComplexPattern<iPTR, 2, "SelectAddrImmShift", [], []>; // "std"
|
|
|
|
// The address in a single register. This is used with the SjLj
|
|
// pseudo-instructions.
|
|
def addr : ComplexPattern<iPTR, 1, "SelectAddr",[], []>;
|
|
|
|
/// This is just the offset part of iaddr, used for preinc.
|
|
def iaddroff : ComplexPattern<iPTR, 1, "SelectAddrImmOffs", [], []>;
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// PowerPC Instruction Predicate Definitions.
|
|
def In32BitMode : Predicate<"!PPCSubTarget.isPPC64()">;
|
|
def In64BitMode : Predicate<"PPCSubTarget.isPPC64()">;
|
|
def IsBookE : Predicate<"PPCSubTarget.isBookE()">;
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// PowerPC Instruction Definitions.
|
|
|
|
// Pseudo-instructions:
|
|
|
|
let hasCtrlDep = 1 in {
|
|
let Defs = [R1], Uses = [R1] in {
|
|
def ADJCALLSTACKDOWN : Pseudo<(outs), (ins u16imm:$amt), "#ADJCALLSTACKDOWN $amt",
|
|
[(callseq_start timm:$amt)]>;
|
|
def ADJCALLSTACKUP : Pseudo<(outs), (ins u16imm:$amt1, u16imm:$amt2), "#ADJCALLSTACKUP $amt1 $amt2",
|
|
[(callseq_end timm:$amt1, timm:$amt2)]>;
|
|
}
|
|
|
|
def UPDATE_VRSAVE : Pseudo<(outs GPRC:$rD), (ins GPRC:$rS),
|
|
"UPDATE_VRSAVE $rD, $rS", []>;
|
|
}
|
|
|
|
let Defs = [R1], Uses = [R1] in
|
|
def DYNALLOC : Pseudo<(outs GPRC:$result), (ins GPRC:$negsize, memri:$fpsi), "#DYNALLOC",
|
|
[(set i32:$result,
|
|
(PPCdynalloc i32:$negsize, iaddr:$fpsi))]>;
|
|
|
|
// SELECT_CC_* - Used to implement the SELECT_CC DAG operation. Expanded after
|
|
// instruction selection into a branch sequence.
|
|
let usesCustomInserter = 1, // Expanded after instruction selection.
|
|
PPC970_Single = 1 in {
|
|
// Note that SELECT_CC_I4 and SELECT_CC_I8 use the no-r0 register classes
|
|
// because either operand might become the first operand in an isel, and
|
|
// that operand cannot be r0.
|
|
def SELECT_CC_I4 : Pseudo<(outs GPRC:$dst), (ins CRRC:$cond,
|
|
GPRC_NOR0:$T, GPRC_NOR0:$F,
|
|
i32imm:$BROPC), "#SELECT_CC_I4",
|
|
[]>;
|
|
def SELECT_CC_I8 : Pseudo<(outs G8RC:$dst), (ins CRRC:$cond,
|
|
G8RC_NOX0:$T, G8RC_NOX0:$F,
|
|
i32imm:$BROPC), "#SELECT_CC_I8",
|
|
[]>;
|
|
def SELECT_CC_F4 : Pseudo<(outs F4RC:$dst), (ins CRRC:$cond, F4RC:$T, F4RC:$F,
|
|
i32imm:$BROPC), "#SELECT_CC_F4",
|
|
[]>;
|
|
def SELECT_CC_F8 : Pseudo<(outs F8RC:$dst), (ins CRRC:$cond, F8RC:$T, F8RC:$F,
|
|
i32imm:$BROPC), "#SELECT_CC_F8",
|
|
[]>;
|
|
def SELECT_CC_VRRC: Pseudo<(outs VRRC:$dst), (ins CRRC:$cond, VRRC:$T, VRRC:$F,
|
|
i32imm:$BROPC), "#SELECT_CC_VRRC",
|
|
[]>;
|
|
}
|
|
|
|
// SPILL_CR - Indicate that we're dumping the CR register, so we'll need to
|
|
// scavenge a register for it.
|
|
let mayStore = 1 in
|
|
def SPILL_CR : Pseudo<(outs), (ins CRRC:$cond, memri:$F),
|
|
"#SPILL_CR", []>;
|
|
|
|
// RESTORE_CR - Indicate that we're restoring the CR register (previously
|
|
// spilled), so we'll need to scavenge a register for it.
|
|
let mayLoad = 1 in
|
|
def RESTORE_CR : Pseudo<(outs CRRC:$cond), (ins memri:$F),
|
|
"#RESTORE_CR", []>;
|
|
|
|
let isTerminator = 1, isBarrier = 1, PPC970_Unit = 7 in {
|
|
let isReturn = 1, Uses = [LR, RM] in
|
|
def BLR : XLForm_2_ext<19, 16, 20, 0, 0, (outs), (ins), "blr", BrB,
|
|
[(retflag)]>;
|
|
let isBranch = 1, isIndirectBranch = 1, Uses = [CTR] in
|
|
def BCTR : XLForm_2_ext<19, 528, 20, 0, 0, (outs), (ins), "bctr", BrB, []>;
|
|
}
|
|
|
|
let Defs = [LR] in
|
|
def MovePCtoLR : Pseudo<(outs), (ins), "#MovePCtoLR", []>,
|
|
PPC970_Unit_BRU;
|
|
|
|
let isBranch = 1, isTerminator = 1, hasCtrlDep = 1, PPC970_Unit = 7 in {
|
|
let isBarrier = 1 in {
|
|
def B : IForm<18, 0, 0, (outs), (ins directbrtarget:$dst),
|
|
"b $dst", BrB,
|
|
[(br bb:$dst)]>;
|
|
}
|
|
|
|
// BCC represents an arbitrary conditional branch on a predicate.
|
|
// FIXME: should be able to write a pattern for PPCcondbranch, but can't use
|
|
// a two-value operand where a dag node expects two operands. :(
|
|
let isCodeGenOnly = 1 in
|
|
def BCC : BForm<16, 0, 0, (outs), (ins pred:$cond, condbrtarget:$dst),
|
|
"b${cond:cc} ${cond:reg}, $dst"
|
|
/*[(PPCcondbranch CRRC:$crS, imm:$opc, bb:$dst)]*/>;
|
|
|
|
let Defs = [CTR], Uses = [CTR] in {
|
|
def BDZ : BForm_1<16, 18, 0, 0, (outs), (ins condbrtarget:$dst),
|
|
"bdz $dst">;
|
|
def BDNZ : BForm_1<16, 16, 0, 0, (outs), (ins condbrtarget:$dst),
|
|
"bdnz $dst">;
|
|
}
|
|
}
|
|
|
|
// The direct BCL used by the SjLj setjmp code.
|
|
let isCall = 1, hasCtrlDep = 1, isCodeGenOnly = 1, PPC970_Unit = 7 in {
|
|
let Defs = [LR], Uses = [RM] in {
|
|
def BCL : BForm_2<16, 20, 31, 0, 1, (outs), (ins condbrtarget:$dst),
|
|
"bcl 20, 31, $dst">;
|
|
}
|
|
}
|
|
|
|
let isCall = 1, PPC970_Unit = 7, Defs = [LR] in {
|
|
// Convenient aliases for call instructions
|
|
let Uses = [RM] in {
|
|
def BL : IForm<18, 0, 1, (outs), (ins calltarget:$func),
|
|
"bl $func", BrB, []>; // See Pat patterns below.
|
|
def BLA : IForm<18, 1, 1, (outs), (ins aaddr:$func),
|
|
"bla $func", BrB, [(PPCcall (i32 imm:$func))]>;
|
|
}
|
|
let Uses = [CTR, RM] in {
|
|
def BCTRL : XLForm_2_ext<19, 528, 20, 0, 1, (outs), (ins),
|
|
"bctrl", BrB, [(PPCbctrl)]>,
|
|
Requires<[In32BitMode]>;
|
|
}
|
|
}
|
|
|
|
let isCall = 1, isTerminator = 1, isReturn = 1, isBarrier = 1, Uses = [RM] in
|
|
def TCRETURNdi :Pseudo< (outs),
|
|
(ins calltarget:$dst, i32imm:$offset),
|
|
"#TC_RETURNd $dst $offset",
|
|
[]>;
|
|
|
|
|
|
let isCall = 1, isTerminator = 1, isReturn = 1, isBarrier = 1, Uses = [RM] in
|
|
def TCRETURNai :Pseudo<(outs), (ins aaddr:$func, i32imm:$offset),
|
|
"#TC_RETURNa $func $offset",
|
|
[(PPCtc_return (i32 imm:$func), imm:$offset)]>;
|
|
|
|
let isCall = 1, isTerminator = 1, isReturn = 1, isBarrier = 1, Uses = [RM] in
|
|
def TCRETURNri : Pseudo<(outs), (ins CTRRC:$dst, i32imm:$offset),
|
|
"#TC_RETURNr $dst $offset",
|
|
[]>;
|
|
|
|
|
|
let isCodeGenOnly = 1 in {
|
|
|
|
let isTerminator = 1, isBarrier = 1, PPC970_Unit = 7, isBranch = 1,
|
|
isIndirectBranch = 1, isCall = 1, isReturn = 1, Uses = [CTR, RM] in
|
|
def TAILBCTR : XLForm_2_ext<19, 528, 20, 0, 0, (outs), (ins), "bctr", BrB, []>,
|
|
Requires<[In32BitMode]>;
|
|
|
|
|
|
|
|
let isBranch = 1, isTerminator = 1, hasCtrlDep = 1, PPC970_Unit = 7,
|
|
isBarrier = 1, isCall = 1, isReturn = 1, Uses = [RM] in
|
|
def TAILB : IForm<18, 0, 0, (outs), (ins calltarget:$dst),
|
|
"b $dst", BrB,
|
|
[]>;
|
|
|
|
}
|
|
|
|
let isBranch = 1, isTerminator = 1, hasCtrlDep = 1, PPC970_Unit = 7,
|
|
isBarrier = 1, isCall = 1, isReturn = 1, Uses = [RM] in
|
|
def TAILBA : IForm<18, 0, 0, (outs), (ins aaddr:$dst),
|
|
"ba $dst", BrB,
|
|
[]>;
|
|
|
|
let hasSideEffects = 1, isBarrier = 1, usesCustomInserter = 1 in {
|
|
def EH_SjLj_SetJmp32 : Pseudo<(outs GPRC:$dst), (ins memr:$buf),
|
|
"#EH_SJLJ_SETJMP32",
|
|
[(set i32:$dst, (PPCeh_sjlj_setjmp addr:$buf))]>,
|
|
Requires<[In32BitMode]>;
|
|
let isTerminator = 1 in
|
|
def EH_SjLj_LongJmp32 : Pseudo<(outs), (ins memr:$buf),
|
|
"#EH_SJLJ_LONGJMP32",
|
|
[(PPCeh_sjlj_longjmp addr:$buf)]>,
|
|
Requires<[In32BitMode]>;
|
|
}
|
|
|
|
let isBranch = 1, isTerminator = 1 in {
|
|
def EH_SjLj_Setup : Pseudo<(outs), (ins directbrtarget:$dst),
|
|
"#EH_SjLj_Setup\t$dst", []>;
|
|
}
|
|
|
|
// DCB* instructions.
|
|
def DCBA : DCB_Form<758, 0, (outs), (ins memrr:$dst),
|
|
"dcba $dst", LdStDCBF, [(int_ppc_dcba xoaddr:$dst)]>,
|
|
PPC970_DGroup_Single;
|
|
def DCBF : DCB_Form<86, 0, (outs), (ins memrr:$dst),
|
|
"dcbf $dst", LdStDCBF, [(int_ppc_dcbf xoaddr:$dst)]>,
|
|
PPC970_DGroup_Single;
|
|
def DCBI : DCB_Form<470, 0, (outs), (ins memrr:$dst),
|
|
"dcbi $dst", LdStDCBF, [(int_ppc_dcbi xoaddr:$dst)]>,
|
|
PPC970_DGroup_Single;
|
|
def DCBST : DCB_Form<54, 0, (outs), (ins memrr:$dst),
|
|
"dcbst $dst", LdStDCBF, [(int_ppc_dcbst xoaddr:$dst)]>,
|
|
PPC970_DGroup_Single;
|
|
def DCBT : DCB_Form<278, 0, (outs), (ins memrr:$dst),
|
|
"dcbt $dst", LdStDCBF, [(int_ppc_dcbt xoaddr:$dst)]>,
|
|
PPC970_DGroup_Single;
|
|
def DCBTST : DCB_Form<246, 0, (outs), (ins memrr:$dst),
|
|
"dcbtst $dst", LdStDCBF, [(int_ppc_dcbtst xoaddr:$dst)]>,
|
|
PPC970_DGroup_Single;
|
|
def DCBZ : DCB_Form<1014, 0, (outs), (ins memrr:$dst),
|
|
"dcbz $dst", LdStDCBF, [(int_ppc_dcbz xoaddr:$dst)]>,
|
|
PPC970_DGroup_Single;
|
|
def DCBZL : DCB_Form<1014, 1, (outs), (ins memrr:$dst),
|
|
"dcbzl $dst", LdStDCBF, [(int_ppc_dcbzl xoaddr:$dst)]>,
|
|
PPC970_DGroup_Single;
|
|
|
|
def : Pat<(prefetch xoaddr:$dst, (i32 0), imm, (i32 1)),
|
|
(DCBT xoaddr:$dst)>;
|
|
|
|
// Atomic operations
|
|
let usesCustomInserter = 1 in {
|
|
let Defs = [CR0] in {
|
|
def ATOMIC_LOAD_ADD_I8 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_ADD_I8",
|
|
[(set i32:$dst, (atomic_load_add_8 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_SUB_I8 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_SUB_I8",
|
|
[(set i32:$dst, (atomic_load_sub_8 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_AND_I8 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_AND_I8",
|
|
[(set i32:$dst, (atomic_load_and_8 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_OR_I8 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_OR_I8",
|
|
[(set i32:$dst, (atomic_load_or_8 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_XOR_I8 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "ATOMIC_LOAD_XOR_I8",
|
|
[(set i32:$dst, (atomic_load_xor_8 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_NAND_I8 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_NAND_I8",
|
|
[(set i32:$dst, (atomic_load_nand_8 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_ADD_I16 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_ADD_I16",
|
|
[(set i32:$dst, (atomic_load_add_16 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_SUB_I16 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_SUB_I16",
|
|
[(set i32:$dst, (atomic_load_sub_16 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_AND_I16 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_AND_I16",
|
|
[(set i32:$dst, (atomic_load_and_16 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_OR_I16 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_OR_I16",
|
|
[(set i32:$dst, (atomic_load_or_16 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_XOR_I16 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_XOR_I16",
|
|
[(set i32:$dst, (atomic_load_xor_16 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_NAND_I16 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_NAND_I16",
|
|
[(set i32:$dst, (atomic_load_nand_16 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_ADD_I32 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_ADD_I32",
|
|
[(set i32:$dst, (atomic_load_add_32 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_SUB_I32 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_SUB_I32",
|
|
[(set i32:$dst, (atomic_load_sub_32 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_AND_I32 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_AND_I32",
|
|
[(set i32:$dst, (atomic_load_and_32 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_OR_I32 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_OR_I32",
|
|
[(set i32:$dst, (atomic_load_or_32 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_XOR_I32 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_XOR_I32",
|
|
[(set i32:$dst, (atomic_load_xor_32 xoaddr:$ptr, i32:$incr))]>;
|
|
def ATOMIC_LOAD_NAND_I32 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$incr), "#ATOMIC_LOAD_NAND_I32",
|
|
[(set i32:$dst, (atomic_load_nand_32 xoaddr:$ptr, i32:$incr))]>;
|
|
|
|
def ATOMIC_CMP_SWAP_I8 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$old, GPRC:$new), "#ATOMIC_CMP_SWAP_I8",
|
|
[(set i32:$dst, (atomic_cmp_swap_8 xoaddr:$ptr, i32:$old, i32:$new))]>;
|
|
def ATOMIC_CMP_SWAP_I16 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$old, GPRC:$new), "#ATOMIC_CMP_SWAP_I16 $dst $ptr $old $new",
|
|
[(set i32:$dst, (atomic_cmp_swap_16 xoaddr:$ptr, i32:$old, i32:$new))]>;
|
|
def ATOMIC_CMP_SWAP_I32 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$old, GPRC:$new), "#ATOMIC_CMP_SWAP_I32 $dst $ptr $old $new",
|
|
[(set i32:$dst, (atomic_cmp_swap_32 xoaddr:$ptr, i32:$old, i32:$new))]>;
|
|
|
|
def ATOMIC_SWAP_I8 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$new), "#ATOMIC_SWAP_i8",
|
|
[(set i32:$dst, (atomic_swap_8 xoaddr:$ptr, i32:$new))]>;
|
|
def ATOMIC_SWAP_I16 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$new), "#ATOMIC_SWAP_I16",
|
|
[(set i32:$dst, (atomic_swap_16 xoaddr:$ptr, i32:$new))]>;
|
|
def ATOMIC_SWAP_I32 : Pseudo<
|
|
(outs GPRC:$dst), (ins memrr:$ptr, GPRC:$new), "#ATOMIC_SWAP_I32",
|
|
[(set i32:$dst, (atomic_swap_32 xoaddr:$ptr, i32:$new))]>;
|
|
}
|
|
}
|
|
|
|
// Instructions to support atomic operations
|
|
def LWARX : XForm_1<31, 20, (outs GPRC:$rD), (ins memrr:$src),
|
|
"lwarx $rD, $src", LdStLWARX,
|
|
[(set i32:$rD, (PPClarx xoaddr:$src))]>;
|
|
|
|
let Defs = [CR0] in
|
|
def STWCX : XForm_1<31, 150, (outs), (ins GPRC:$rS, memrr:$dst),
|
|
"stwcx. $rS, $dst", LdStSTWCX,
|
|
[(PPCstcx i32:$rS, xoaddr:$dst)]>,
|
|
isDOT;
|
|
|
|
let isTerminator = 1, isBarrier = 1, hasCtrlDep = 1 in
|
|
def TRAP : XForm_24<31, 4, (outs), (ins), "trap", LdStLoad, [(trap)]>;
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// PPC32 Load Instructions.
|
|
//
|
|
|
|
// Unindexed (r+i) Loads.
|
|
let canFoldAsLoad = 1, PPC970_Unit = 2 in {
|
|
def LBZ : DForm_1<34, (outs GPRC:$rD), (ins memri:$src),
|
|
"lbz $rD, $src", LdStLoad,
|
|
[(set i32:$rD, (zextloadi8 iaddr:$src))]>;
|
|
def LHA : DForm_1<42, (outs GPRC:$rD), (ins memri:$src),
|
|
"lha $rD, $src", LdStLHA,
|
|
[(set i32:$rD, (sextloadi16 iaddr:$src))]>,
|
|
PPC970_DGroup_Cracked;
|
|
def LHZ : DForm_1<40, (outs GPRC:$rD), (ins memri:$src),
|
|
"lhz $rD, $src", LdStLoad,
|
|
[(set i32:$rD, (zextloadi16 iaddr:$src))]>;
|
|
def LWZ : DForm_1<32, (outs GPRC:$rD), (ins memri:$src),
|
|
"lwz $rD, $src", LdStLoad,
|
|
[(set i32:$rD, (load iaddr:$src))]>;
|
|
|
|
def LFS : DForm_1<48, (outs F4RC:$rD), (ins memri:$src),
|
|
"lfs $rD, $src", LdStLFD,
|
|
[(set f32:$rD, (load iaddr:$src))]>;
|
|
def LFD : DForm_1<50, (outs F8RC:$rD), (ins memri:$src),
|
|
"lfd $rD, $src", LdStLFD,
|
|
[(set f64:$rD, (load iaddr:$src))]>;
|
|
|
|
|
|
// Unindexed (r+i) Loads with Update (preinc).
|
|
let mayLoad = 1 in {
|
|
def LBZU : DForm_1<35, (outs GPRC:$rD, ptr_rc_nor0:$ea_result), (ins memri:$addr),
|
|
"lbzu $rD, $addr", LdStLoadUpd,
|
|
[]>, RegConstraint<"$addr.reg = $ea_result">,
|
|
NoEncode<"$ea_result">;
|
|
|
|
def LHAU : DForm_1<43, (outs GPRC:$rD, ptr_rc_nor0:$ea_result), (ins memri:$addr),
|
|
"lhau $rD, $addr", LdStLHAU,
|
|
[]>, RegConstraint<"$addr.reg = $ea_result">,
|
|
NoEncode<"$ea_result">;
|
|
|
|
def LHZU : DForm_1<41, (outs GPRC:$rD, ptr_rc_nor0:$ea_result), (ins memri:$addr),
|
|
"lhzu $rD, $addr", LdStLoadUpd,
|
|
[]>, RegConstraint<"$addr.reg = $ea_result">,
|
|
NoEncode<"$ea_result">;
|
|
|
|
def LWZU : DForm_1<33, (outs GPRC:$rD, ptr_rc_nor0:$ea_result), (ins memri:$addr),
|
|
"lwzu $rD, $addr", LdStLoadUpd,
|
|
[]>, RegConstraint<"$addr.reg = $ea_result">,
|
|
NoEncode<"$ea_result">;
|
|
|
|
def LFSU : DForm_1<49, (outs F4RC:$rD, ptr_rc_nor0:$ea_result), (ins memri:$addr),
|
|
"lfsu $rD, $addr", LdStLFDU,
|
|
[]>, RegConstraint<"$addr.reg = $ea_result">,
|
|
NoEncode<"$ea_result">;
|
|
|
|
def LFDU : DForm_1<51, (outs F8RC:$rD, ptr_rc_nor0:$ea_result), (ins memri:$addr),
|
|
"lfdu $rD, $addr", LdStLFDU,
|
|
[]>, RegConstraint<"$addr.reg = $ea_result">,
|
|
NoEncode<"$ea_result">;
|
|
|
|
|
|
// Indexed (r+r) Loads with Update (preinc).
|
|
def LBZUX : XForm_1<31, 119, (outs GPRC:$rD, ptr_rc_nor0:$ea_result),
|
|
(ins memrr:$addr),
|
|
"lbzux $rD, $addr", LdStLoadUpd,
|
|
[]>, RegConstraint<"$addr.ptrreg = $ea_result">,
|
|
NoEncode<"$ea_result">;
|
|
|
|
def LHAUX : XForm_1<31, 375, (outs GPRC:$rD, ptr_rc_nor0:$ea_result),
|
|
(ins memrr:$addr),
|
|
"lhaux $rD, $addr", LdStLHAU,
|
|
[]>, RegConstraint<"$addr.ptrreg = $ea_result">,
|
|
NoEncode<"$ea_result">;
|
|
|
|
def LHZUX : XForm_1<31, 311, (outs GPRC:$rD, ptr_rc_nor0:$ea_result),
|
|
(ins memrr:$addr),
|
|
"lhzux $rD, $addr", LdStLoadUpd,
|
|
[]>, RegConstraint<"$addr.ptrreg = $ea_result">,
|
|
NoEncode<"$ea_result">;
|
|
|
|
def LWZUX : XForm_1<31, 55, (outs GPRC:$rD, ptr_rc_nor0:$ea_result),
|
|
(ins memrr:$addr),
|
|
"lwzux $rD, $addr", LdStLoadUpd,
|
|
[]>, RegConstraint<"$addr.ptrreg = $ea_result">,
|
|
NoEncode<"$ea_result">;
|
|
|
|
def LFSUX : XForm_1<31, 567, (outs F4RC:$rD, ptr_rc_nor0:$ea_result),
|
|
(ins memrr:$addr),
|
|
"lfsux $rD, $addr", LdStLFDU,
|
|
[]>, RegConstraint<"$addr.ptrreg = $ea_result">,
|
|
NoEncode<"$ea_result">;
|
|
|
|
def LFDUX : XForm_1<31, 631, (outs F8RC:$rD, ptr_rc_nor0:$ea_result),
|
|
(ins memrr:$addr),
|
|
"lfdux $rD, $addr", LdStLFDU,
|
|
[]>, RegConstraint<"$addr.ptrreg = $ea_result">,
|
|
NoEncode<"$ea_result">;
|
|
}
|
|
}
|
|
|
|
// Indexed (r+r) Loads.
|
|
//
|
|
let canFoldAsLoad = 1, PPC970_Unit = 2 in {
|
|
def LBZX : XForm_1<31, 87, (outs GPRC:$rD), (ins memrr:$src),
|
|
"lbzx $rD, $src", LdStLoad,
|
|
[(set i32:$rD, (zextloadi8 xaddr:$src))]>;
|
|
def LHAX : XForm_1<31, 343, (outs GPRC:$rD), (ins memrr:$src),
|
|
"lhax $rD, $src", LdStLHA,
|
|
[(set i32:$rD, (sextloadi16 xaddr:$src))]>,
|
|
PPC970_DGroup_Cracked;
|
|
def LHZX : XForm_1<31, 279, (outs GPRC:$rD), (ins memrr:$src),
|
|
"lhzx $rD, $src", LdStLoad,
|
|
[(set i32:$rD, (zextloadi16 xaddr:$src))]>;
|
|
def LWZX : XForm_1<31, 23, (outs GPRC:$rD), (ins memrr:$src),
|
|
"lwzx $rD, $src", LdStLoad,
|
|
[(set i32:$rD, (load xaddr:$src))]>;
|
|
|
|
|
|
def LHBRX : XForm_1<31, 790, (outs GPRC:$rD), (ins memrr:$src),
|
|
"lhbrx $rD, $src", LdStLoad,
|
|
[(set i32:$rD, (PPClbrx xoaddr:$src, i16))]>;
|
|
def LWBRX : XForm_1<31, 534, (outs GPRC:$rD), (ins memrr:$src),
|
|
"lwbrx $rD, $src", LdStLoad,
|
|
[(set i32:$rD, (PPClbrx xoaddr:$src, i32))]>;
|
|
|
|
def LFSX : XForm_25<31, 535, (outs F4RC:$frD), (ins memrr:$src),
|
|
"lfsx $frD, $src", LdStLFD,
|
|
[(set f32:$frD, (load xaddr:$src))]>;
|
|
def LFDX : XForm_25<31, 599, (outs F8RC:$frD), (ins memrr:$src),
|
|
"lfdx $frD, $src", LdStLFD,
|
|
[(set f64:$frD, (load xaddr:$src))]>;
|
|
|
|
def LFIWAX : XForm_25<31, 855, (outs F8RC:$frD), (ins memrr:$src),
|
|
"lfiwax $frD, $src", LdStLFD,
|
|
[(set f64:$frD, (PPClfiwax xoaddr:$src))]>;
|
|
def LFIWZX : XForm_25<31, 887, (outs F8RC:$frD), (ins memrr:$src),
|
|
"lfiwzx $frD, $src", LdStLFD,
|
|
[(set f64:$frD, (PPClfiwzx xoaddr:$src))]>;
|
|
}
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// PPC32 Store Instructions.
|
|
//
|
|
|
|
// Unindexed (r+i) Stores.
|
|
let PPC970_Unit = 2 in {
|
|
def STB : DForm_1<38, (outs), (ins GPRC:$rS, memri:$src),
|
|
"stb $rS, $src", LdStStore,
|
|
[(truncstorei8 i32:$rS, iaddr:$src)]>;
|
|
def STH : DForm_1<44, (outs), (ins GPRC:$rS, memri:$src),
|
|
"sth $rS, $src", LdStStore,
|
|
[(truncstorei16 i32:$rS, iaddr:$src)]>;
|
|
def STW : DForm_1<36, (outs), (ins GPRC:$rS, memri:$src),
|
|
"stw $rS, $src", LdStStore,
|
|
[(store i32:$rS, iaddr:$src)]>;
|
|
def STFS : DForm_1<52, (outs), (ins F4RC:$rS, memri:$dst),
|
|
"stfs $rS, $dst", LdStSTFD,
|
|
[(store f32:$rS, iaddr:$dst)]>;
|
|
def STFD : DForm_1<54, (outs), (ins F8RC:$rS, memri:$dst),
|
|
"stfd $rS, $dst", LdStSTFD,
|
|
[(store f64:$rS, iaddr:$dst)]>;
|
|
}
|
|
|
|
// Unindexed (r+i) Stores with Update (preinc).
|
|
let PPC970_Unit = 2, mayStore = 1 in {
|
|
def STBU : DForm_1<39, (outs ptr_rc_nor0:$ea_res), (ins GPRC:$rS, memri:$dst),
|
|
"stbu $rS, $dst", LdStStoreUpd, []>,
|
|
RegConstraint<"$dst.reg = $ea_res">, NoEncode<"$ea_res">;
|
|
def STHU : DForm_1<45, (outs ptr_rc_nor0:$ea_res), (ins GPRC:$rS, memri:$dst),
|
|
"sthu $rS, $dst", LdStStoreUpd, []>,
|
|
RegConstraint<"$dst.reg = $ea_res">, NoEncode<"$ea_res">;
|
|
def STWU : DForm_1<37, (outs ptr_rc_nor0:$ea_res), (ins GPRC:$rS, memri:$dst),
|
|
"stwu $rS, $dst", LdStStoreUpd, []>,
|
|
RegConstraint<"$dst.reg = $ea_res">, NoEncode<"$ea_res">;
|
|
def STFSU : DForm_1<37, (outs ptr_rc_nor0:$ea_res), (ins F4RC:$rS, memri:$dst),
|
|
"stfsu $rS, $dst", LdStSTFDU, []>,
|
|
RegConstraint<"$dst.reg = $ea_res">, NoEncode<"$ea_res">;
|
|
def STFDU : DForm_1<37, (outs ptr_rc_nor0:$ea_res), (ins F8RC:$rS, memri:$dst),
|
|
"stfdu $rS, $dst", LdStSTFDU, []>,
|
|
RegConstraint<"$dst.reg = $ea_res">, NoEncode<"$ea_res">;
|
|
}
|
|
|
|
// Patterns to match the pre-inc stores. We can't put the patterns on
|
|
// the instruction definitions directly as ISel wants the address base
|
|
// and offset to be separate operands, not a single complex operand.
|
|
def : Pat<(pre_truncsti8 i32:$rS, iPTR:$ptrreg, iaddroff:$ptroff),
|
|
(STBU $rS, iaddroff:$ptroff, $ptrreg)>;
|
|
def : Pat<(pre_truncsti16 i32:$rS, iPTR:$ptrreg, iaddroff:$ptroff),
|
|
(STHU $rS, iaddroff:$ptroff, $ptrreg)>;
|
|
def : Pat<(pre_store i32:$rS, iPTR:$ptrreg, iaddroff:$ptroff),
|
|
(STWU $rS, iaddroff:$ptroff, $ptrreg)>;
|
|
def : Pat<(pre_store f32:$rS, iPTR:$ptrreg, iaddroff:$ptroff),
|
|
(STFSU $rS, iaddroff:$ptroff, $ptrreg)>;
|
|
def : Pat<(pre_store f64:$rS, iPTR:$ptrreg, iaddroff:$ptroff),
|
|
(STFDU $rS, iaddroff:$ptroff, $ptrreg)>;
|
|
|
|
// Indexed (r+r) Stores.
|
|
let PPC970_Unit = 2 in {
|
|
def STBX : XForm_8<31, 215, (outs), (ins GPRC:$rS, memrr:$dst),
|
|
"stbx $rS, $dst", LdStStore,
|
|
[(truncstorei8 i32:$rS, xaddr:$dst)]>,
|
|
PPC970_DGroup_Cracked;
|
|
def STHX : XForm_8<31, 407, (outs), (ins GPRC:$rS, memrr:$dst),
|
|
"sthx $rS, $dst", LdStStore,
|
|
[(truncstorei16 i32:$rS, xaddr:$dst)]>,
|
|
PPC970_DGroup_Cracked;
|
|
def STWX : XForm_8<31, 151, (outs), (ins GPRC:$rS, memrr:$dst),
|
|
"stwx $rS, $dst", LdStStore,
|
|
[(store i32:$rS, xaddr:$dst)]>,
|
|
PPC970_DGroup_Cracked;
|
|
|
|
def STHBRX: XForm_8<31, 918, (outs), (ins GPRC:$rS, memrr:$dst),
|
|
"sthbrx $rS, $dst", LdStStore,
|
|
[(PPCstbrx i32:$rS, xoaddr:$dst, i16)]>,
|
|
PPC970_DGroup_Cracked;
|
|
def STWBRX: XForm_8<31, 662, (outs), (ins GPRC:$rS, memrr:$dst),
|
|
"stwbrx $rS, $dst", LdStStore,
|
|
[(PPCstbrx i32:$rS, xoaddr:$dst, i32)]>,
|
|
PPC970_DGroup_Cracked;
|
|
|
|
def STFIWX: XForm_28<31, 983, (outs), (ins F8RC:$frS, memrr:$dst),
|
|
"stfiwx $frS, $dst", LdStSTFD,
|
|
[(PPCstfiwx f64:$frS, xoaddr:$dst)]>;
|
|
|
|
def STFSX : XForm_28<31, 663, (outs), (ins F4RC:$frS, memrr:$dst),
|
|
"stfsx $frS, $dst", LdStSTFD,
|
|
[(store f32:$frS, xaddr:$dst)]>;
|
|
def STFDX : XForm_28<31, 727, (outs), (ins F8RC:$frS, memrr:$dst),
|
|
"stfdx $frS, $dst", LdStSTFD,
|
|
[(store f64:$frS, xaddr:$dst)]>;
|
|
}
|
|
|
|
// Indexed (r+r) Stores with Update (preinc).
|
|
let PPC970_Unit = 2, mayStore = 1 in {
|
|
def STBUX : XForm_8<31, 247, (outs ptr_rc_nor0:$ea_res), (ins GPRC:$rS, memrr:$dst),
|
|
"stbux $rS, $dst", LdStStoreUpd, []>,
|
|
RegConstraint<"$dst.ptrreg = $ea_res">, NoEncode<"$ea_res">,
|
|
PPC970_DGroup_Cracked;
|
|
def STHUX : XForm_8<31, 439, (outs ptr_rc_nor0:$ea_res), (ins GPRC:$rS, memrr:$dst),
|
|
"sthux $rS, $dst", LdStStoreUpd, []>,
|
|
RegConstraint<"$dst.ptrreg = $ea_res">, NoEncode<"$ea_res">,
|
|
PPC970_DGroup_Cracked;
|
|
def STWUX : XForm_8<31, 183, (outs ptr_rc_nor0:$ea_res), (ins GPRC:$rS, memrr:$dst),
|
|
"stwux $rS, $dst", LdStStoreUpd, []>,
|
|
RegConstraint<"$dst.ptrreg = $ea_res">, NoEncode<"$ea_res">,
|
|
PPC970_DGroup_Cracked;
|
|
def STFSUX: XForm_8<31, 695, (outs ptr_rc_nor0:$ea_res), (ins F4RC:$rS, memrr:$dst),
|
|
"stfsux $rS, $dst", LdStSTFDU, []>,
|
|
RegConstraint<"$dst.ptrreg = $ea_res">, NoEncode<"$ea_res">,
|
|
PPC970_DGroup_Cracked;
|
|
def STFDUX: XForm_8<31, 759, (outs ptr_rc_nor0:$ea_res), (ins F8RC:$rS, memrr:$dst),
|
|
"stfdux $rS, $dst", LdStSTFDU, []>,
|
|
RegConstraint<"$dst.ptrreg = $ea_res">, NoEncode<"$ea_res">,
|
|
PPC970_DGroup_Cracked;
|
|
}
|
|
|
|
// Patterns to match the pre-inc stores. We can't put the patterns on
|
|
// the instruction definitions directly as ISel wants the address base
|
|
// and offset to be separate operands, not a single complex operand.
|
|
def : Pat<(pre_truncsti8 i32:$rS, iPTR:$ptrreg, iPTR:$ptroff),
|
|
(STBUX $rS, $ptrreg, $ptroff)>;
|
|
def : Pat<(pre_truncsti16 i32:$rS, iPTR:$ptrreg, iPTR:$ptroff),
|
|
(STHUX $rS, $ptrreg, $ptroff)>;
|
|
def : Pat<(pre_store i32:$rS, iPTR:$ptrreg, iPTR:$ptroff),
|
|
(STWUX $rS, $ptrreg, $ptroff)>;
|
|
def : Pat<(pre_store f32:$rS, iPTR:$ptrreg, iPTR:$ptroff),
|
|
(STFSUX $rS, $ptrreg, $ptroff)>;
|
|
def : Pat<(pre_store f64:$rS, iPTR:$ptrreg, iPTR:$ptroff),
|
|
(STFDUX $rS, $ptrreg, $ptroff)>;
|
|
|
|
def SYNC : XForm_24_sync<31, 598, (outs), (ins),
|
|
"sync", LdStSync,
|
|
[(int_ppc_sync)]>;
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// PPC32 Arithmetic Instructions.
|
|
//
|
|
|
|
let PPC970_Unit = 1 in { // FXU Operations.
|
|
def ADDI : DForm_2<14, (outs GPRC:$rD), (ins GPRC_NOR0:$rA, symbolLo:$imm),
|
|
"addi $rD, $rA, $imm", IntSimple,
|
|
[(set i32:$rD, (add i32:$rA, immSExt16:$imm))]>;
|
|
let Defs = [CARRY] in {
|
|
def ADDIC : DForm_2<12, (outs GPRC:$rD), (ins GPRC:$rA, s16imm:$imm),
|
|
"addic $rD, $rA, $imm", IntGeneral,
|
|
[(set i32:$rD, (addc i32:$rA, immSExt16:$imm))]>,
|
|
PPC970_DGroup_Cracked;
|
|
def ADDICo : DForm_2<13, (outs GPRC:$rD), (ins GPRC:$rA, s16imm:$imm),
|
|
"addic. $rD, $rA, $imm", IntGeneral,
|
|
[]>;
|
|
}
|
|
def ADDIS : DForm_2<15, (outs GPRC:$rD), (ins GPRC_NOR0:$rA, symbolHi:$imm),
|
|
"addis $rD, $rA, $imm", IntSimple,
|
|
[(set i32:$rD, (add i32:$rA, imm16ShiftedSExt:$imm))]>;
|
|
let isCodeGenOnly = 1 in
|
|
def LA : DForm_2<14, (outs GPRC:$rD), (ins GPRC_NOR0:$rA, symbolLo:$sym),
|
|
"la $rD, $sym($rA)", IntGeneral,
|
|
[(set i32:$rD, (add i32:$rA,
|
|
(PPClo tglobaladdr:$sym, 0)))]>;
|
|
def MULLI : DForm_2< 7, (outs GPRC:$rD), (ins GPRC:$rA, s16imm:$imm),
|
|
"mulli $rD, $rA, $imm", IntMulLI,
|
|
[(set i32:$rD, (mul i32:$rA, immSExt16:$imm))]>;
|
|
let Defs = [CARRY] in {
|
|
def SUBFIC : DForm_2< 8, (outs GPRC:$rD), (ins GPRC:$rA, s16imm:$imm),
|
|
"subfic $rD, $rA, $imm", IntGeneral,
|
|
[(set i32:$rD, (subc immSExt16:$imm, i32:$rA))]>;
|
|
}
|
|
|
|
let isReMaterializable = 1, isAsCheapAsAMove = 1, isMoveImm = 1 in {
|
|
def LI : DForm_2_r0<14, (outs GPRC:$rD), (ins symbolLo:$imm),
|
|
"li $rD, $imm", IntSimple,
|
|
[(set i32:$rD, immSExt16:$imm)]>;
|
|
def LIS : DForm_2_r0<15, (outs GPRC:$rD), (ins symbolHi:$imm),
|
|
"lis $rD, $imm", IntSimple,
|
|
[(set i32:$rD, imm16ShiftedSExt:$imm)]>;
|
|
}
|
|
}
|
|
|
|
let PPC970_Unit = 1 in { // FXU Operations.
|
|
def ANDIo : DForm_4<28, (outs GPRC:$dst), (ins GPRC:$src1, u16imm:$src2),
|
|
"andi. $dst, $src1, $src2", IntGeneral,
|
|
[(set i32:$dst, (and i32:$src1, immZExt16:$src2))]>,
|
|
isDOT;
|
|
def ANDISo : DForm_4<29, (outs GPRC:$dst), (ins GPRC:$src1, u16imm:$src2),
|
|
"andis. $dst, $src1, $src2", IntGeneral,
|
|
[(set i32:$dst, (and i32:$src1, imm16ShiftedZExt:$src2))]>,
|
|
isDOT;
|
|
def ORI : DForm_4<24, (outs GPRC:$dst), (ins GPRC:$src1, u16imm:$src2),
|
|
"ori $dst, $src1, $src2", IntSimple,
|
|
[(set i32:$dst, (or i32:$src1, immZExt16:$src2))]>;
|
|
def ORIS : DForm_4<25, (outs GPRC:$dst), (ins GPRC:$src1, u16imm:$src2),
|
|
"oris $dst, $src1, $src2", IntSimple,
|
|
[(set i32:$dst, (or i32:$src1, imm16ShiftedZExt:$src2))]>;
|
|
def XORI : DForm_4<26, (outs GPRC:$dst), (ins GPRC:$src1, u16imm:$src2),
|
|
"xori $dst, $src1, $src2", IntSimple,
|
|
[(set i32:$dst, (xor i32:$src1, immZExt16:$src2))]>;
|
|
def XORIS : DForm_4<27, (outs GPRC:$dst), (ins GPRC:$src1, u16imm:$src2),
|
|
"xoris $dst, $src1, $src2", IntSimple,
|
|
[(set i32:$dst, (xor i32:$src1, imm16ShiftedZExt:$src2))]>;
|
|
def NOP : DForm_4_zero<24, (outs), (ins), "nop", IntSimple,
|
|
[]>;
|
|
def CMPWI : DForm_5_ext<11, (outs CRRC:$crD), (ins GPRC:$rA, s16imm:$imm),
|
|
"cmpwi $crD, $rA, $imm", IntCompare>;
|
|
def CMPLWI : DForm_6_ext<10, (outs CRRC:$dst), (ins GPRC:$src1, u16imm:$src2),
|
|
"cmplwi $dst, $src1, $src2", IntCompare>;
|
|
}
|
|
|
|
|
|
let PPC970_Unit = 1 in { // FXU Operations.
|
|
def NAND : XForm_6<31, 476, (outs GPRC:$rA), (ins GPRC:$rS, GPRC:$rB),
|
|
"nand $rA, $rS, $rB", IntSimple,
|
|
[(set i32:$rA, (not (and i32:$rS, i32:$rB)))]>;
|
|
def AND : XForm_6<31, 28, (outs GPRC:$rA), (ins GPRC:$rS, GPRC:$rB),
|
|
"and $rA, $rS, $rB", IntSimple,
|
|
[(set i32:$rA, (and i32:$rS, i32:$rB))]>;
|
|
def ANDC : XForm_6<31, 60, (outs GPRC:$rA), (ins GPRC:$rS, GPRC:$rB),
|
|
"andc $rA, $rS, $rB", IntSimple,
|
|
[(set i32:$rA, (and i32:$rS, (not i32:$rB)))]>;
|
|
def OR : XForm_6<31, 444, (outs GPRC:$rA), (ins GPRC:$rS, GPRC:$rB),
|
|
"or $rA, $rS, $rB", IntSimple,
|
|
[(set i32:$rA, (or i32:$rS, i32:$rB))]>;
|
|
def NOR : XForm_6<31, 124, (outs GPRC:$rA), (ins GPRC:$rS, GPRC:$rB),
|
|
"nor $rA, $rS, $rB", IntSimple,
|
|
[(set i32:$rA, (not (or i32:$rS, i32:$rB)))]>;
|
|
def ORC : XForm_6<31, 412, (outs GPRC:$rA), (ins GPRC:$rS, GPRC:$rB),
|
|
"orc $rA, $rS, $rB", IntSimple,
|
|
[(set i32:$rA, (or i32:$rS, (not i32:$rB)))]>;
|
|
def EQV : XForm_6<31, 284, (outs GPRC:$rA), (ins GPRC:$rS, GPRC:$rB),
|
|
"eqv $rA, $rS, $rB", IntSimple,
|
|
[(set i32:$rA, (not (xor i32:$rS, i32:$rB)))]>;
|
|
def XOR : XForm_6<31, 316, (outs GPRC:$rA), (ins GPRC:$rS, GPRC:$rB),
|
|
"xor $rA, $rS, $rB", IntSimple,
|
|
[(set i32:$rA, (xor i32:$rS, i32:$rB))]>;
|
|
def SLW : XForm_6<31, 24, (outs GPRC:$rA), (ins GPRC:$rS, GPRC:$rB),
|
|
"slw $rA, $rS, $rB", IntGeneral,
|
|
[(set i32:$rA, (PPCshl i32:$rS, i32:$rB))]>;
|
|
def SRW : XForm_6<31, 536, (outs GPRC:$rA), (ins GPRC:$rS, GPRC:$rB),
|
|
"srw $rA, $rS, $rB", IntGeneral,
|
|
[(set i32:$rA, (PPCsrl i32:$rS, i32:$rB))]>;
|
|
let Defs = [CARRY] in {
|
|
def SRAW : XForm_6<31, 792, (outs GPRC:$rA), (ins GPRC:$rS, GPRC:$rB),
|
|
"sraw $rA, $rS, $rB", IntShift,
|
|
[(set i32:$rA, (PPCsra i32:$rS, i32:$rB))]>;
|
|
}
|
|
}
|
|
|
|
let PPC970_Unit = 1 in { // FXU Operations.
|
|
let Defs = [CARRY] in {
|
|
def SRAWI : XForm_10<31, 824, (outs GPRC:$rA), (ins GPRC:$rS, u5imm:$SH),
|
|
"srawi $rA, $rS, $SH", IntShift,
|
|
[(set i32:$rA, (sra i32:$rS, (i32 imm:$SH)))]>;
|
|
}
|
|
def CNTLZW : XForm_11<31, 26, (outs GPRC:$rA), (ins GPRC:$rS),
|
|
"cntlzw $rA, $rS", IntGeneral,
|
|
[(set i32:$rA, (ctlz i32:$rS))]>;
|
|
def EXTSB : XForm_11<31, 954, (outs GPRC:$rA), (ins GPRC:$rS),
|
|
"extsb $rA, $rS", IntSimple,
|
|
[(set i32:$rA, (sext_inreg i32:$rS, i8))]>;
|
|
def EXTSH : XForm_11<31, 922, (outs GPRC:$rA), (ins GPRC:$rS),
|
|
"extsh $rA, $rS", IntSimple,
|
|
[(set i32:$rA, (sext_inreg i32:$rS, i16))]>;
|
|
|
|
def CMPW : XForm_16_ext<31, 0, (outs CRRC:$crD), (ins GPRC:$rA, GPRC:$rB),
|
|
"cmpw $crD, $rA, $rB", IntCompare>;
|
|
def CMPLW : XForm_16_ext<31, 32, (outs CRRC:$crD), (ins GPRC:$rA, GPRC:$rB),
|
|
"cmplw $crD, $rA, $rB", IntCompare>;
|
|
}
|
|
let PPC970_Unit = 3 in { // FPU Operations.
|
|
//def FCMPO : XForm_17<63, 32, (outs CRRC:$crD), (ins FPRC:$fA, FPRC:$fB),
|
|
// "fcmpo $crD, $fA, $fB", FPCompare>;
|
|
def FCMPUS : XForm_17<63, 0, (outs CRRC:$crD), (ins F4RC:$fA, F4RC:$fB),
|
|
"fcmpu $crD, $fA, $fB", FPCompare>;
|
|
def FCMPUD : XForm_17<63, 0, (outs CRRC:$crD), (ins F8RC:$fA, F8RC:$fB),
|
|
"fcmpu $crD, $fA, $fB", FPCompare>;
|
|
|
|
let Uses = [RM] in {
|
|
def FCTIWZ : XForm_26<63, 15, (outs F8RC:$frD), (ins F8RC:$frB),
|
|
"fctiwz $frD, $frB", FPGeneral,
|
|
[(set f64:$frD, (PPCfctiwz f64:$frB))]>;
|
|
|
|
def FRSP : XForm_26<63, 12, (outs F4RC:$frD), (ins F8RC:$frB),
|
|
"frsp $frD, $frB", FPGeneral,
|
|
[(set f32:$frD, (fround f64:$frB))]>;
|
|
|
|
// The frin -> nearbyint mapping is valid only in fast-math mode.
|
|
def FRIND : XForm_26<63, 392, (outs F8RC:$frD), (ins F8RC:$frB),
|
|
"frin $frD, $frB", FPGeneral,
|
|
[(set f64:$frD, (fnearbyint f64:$frB))]>;
|
|
def FRINS : XForm_26<63, 392, (outs F4RC:$frD), (ins F4RC:$frB),
|
|
"frin $frD, $frB", FPGeneral,
|
|
[(set f32:$frD, (fnearbyint f32:$frB))]>;
|
|
|
|
// These pseudos expand to rint but also set FE_INEXACT when the result does
|
|
// not equal the argument.
|
|
let usesCustomInserter = 1, Defs = [RM] in { // FIXME: Model FPSCR!
|
|
def FRINDrint : Pseudo<(outs F8RC:$frD), (ins F8RC:$frB),
|
|
"#FRINDrint", [(set f64:$frD, (frint f64:$frB))]>;
|
|
def FRINSrint : Pseudo<(outs F4RC:$frD), (ins F4RC:$frB),
|
|
"#FRINSrint", [(set f32:$frD, (frint f32:$frB))]>;
|
|
}
|
|
|
|
def FRIPD : XForm_26<63, 456, (outs F8RC:$frD), (ins F8RC:$frB),
|
|
"frip $frD, $frB", FPGeneral,
|
|
[(set f64:$frD, (fceil f64:$frB))]>;
|
|
def FRIPS : XForm_26<63, 456, (outs F4RC:$frD), (ins F4RC:$frB),
|
|
"frip $frD, $frB", FPGeneral,
|
|
[(set f32:$frD, (fceil f32:$frB))]>;
|
|
def FRIZD : XForm_26<63, 424, (outs F8RC:$frD), (ins F8RC:$frB),
|
|
"friz $frD, $frB", FPGeneral,
|
|
[(set f64:$frD, (ftrunc f64:$frB))]>;
|
|
def FRIZS : XForm_26<63, 424, (outs F4RC:$frD), (ins F4RC:$frB),
|
|
"friz $frD, $frB", FPGeneral,
|
|
[(set f32:$frD, (ftrunc f32:$frB))]>;
|
|
def FRIMD : XForm_26<63, 488, (outs F8RC:$frD), (ins F8RC:$frB),
|
|
"frim $frD, $frB", FPGeneral,
|
|
[(set f64:$frD, (ffloor f64:$frB))]>;
|
|
def FRIMS : XForm_26<63, 488, (outs F4RC:$frD), (ins F4RC:$frB),
|
|
"frim $frD, $frB", FPGeneral,
|
|
[(set f32:$frD, (ffloor f32:$frB))]>;
|
|
|
|
def FSQRT : XForm_26<63, 22, (outs F8RC:$frD), (ins F8RC:$frB),
|
|
"fsqrt $frD, $frB", FPSqrt,
|
|
[(set f64:$frD, (fsqrt f64:$frB))]>;
|
|
def FSQRTS : XForm_26<59, 22, (outs F4RC:$frD), (ins F4RC:$frB),
|
|
"fsqrts $frD, $frB", FPSqrt,
|
|
[(set f32:$frD, (fsqrt f32:$frB))]>;
|
|
}
|
|
}
|
|
|
|
/// Note that FMR is defined as pseudo-ops on the PPC970 because they are
|
|
/// often coalesced away and we don't want the dispatch group builder to think
|
|
/// that they will fill slots (which could cause the load of a LSU reject to
|
|
/// sneak into a d-group with a store).
|
|
def FMR : XForm_26<63, 72, (outs F4RC:$frD), (ins F4RC:$frB),
|
|
"fmr $frD, $frB", FPGeneral,
|
|
[]>, // (set f32:$frD, f32:$frB)
|
|
PPC970_Unit_Pseudo;
|
|
|
|
let PPC970_Unit = 3 in { // FPU Operations.
|
|
// These are artificially split into two different forms, for 4/8 byte FP.
|
|
def FABSS : XForm_26<63, 264, (outs F4RC:$frD), (ins F4RC:$frB),
|
|
"fabs $frD, $frB", FPGeneral,
|
|
[(set f32:$frD, (fabs f32:$frB))]>;
|
|
def FABSD : XForm_26<63, 264, (outs F8RC:$frD), (ins F8RC:$frB),
|
|
"fabs $frD, $frB", FPGeneral,
|
|
[(set f64:$frD, (fabs f64:$frB))]>;
|
|
def FNABSS : XForm_26<63, 136, (outs F4RC:$frD), (ins F4RC:$frB),
|
|
"fnabs $frD, $frB", FPGeneral,
|
|
[(set f32:$frD, (fneg (fabs f32:$frB)))]>;
|
|
def FNABSD : XForm_26<63, 136, (outs F8RC:$frD), (ins F8RC:$frB),
|
|
"fnabs $frD, $frB", FPGeneral,
|
|
[(set f64:$frD, (fneg (fabs f64:$frB)))]>;
|
|
def FNEGS : XForm_26<63, 40, (outs F4RC:$frD), (ins F4RC:$frB),
|
|
"fneg $frD, $frB", FPGeneral,
|
|
[(set f32:$frD, (fneg f32:$frB))]>;
|
|
def FNEGD : XForm_26<63, 40, (outs F8RC:$frD), (ins F8RC:$frB),
|
|
"fneg $frD, $frB", FPGeneral,
|
|
[(set f64:$frD, (fneg f64:$frB))]>;
|
|
|
|
// Reciprocal estimates.
|
|
def FRE : XForm_26<63, 24, (outs F8RC:$frD), (ins F8RC:$frB),
|
|
"fre $frD, $frB", FPGeneral,
|
|
[(set f64:$frD, (PPCfre f64:$frB))]>;
|
|
def FRES : XForm_26<59, 24, (outs F4RC:$frD), (ins F4RC:$frB),
|
|
"fres $frD, $frB", FPGeneral,
|
|
[(set f32:$frD, (PPCfre f32:$frB))]>;
|
|
def FRSQRTE : XForm_26<63, 26, (outs F8RC:$frD), (ins F8RC:$frB),
|
|
"frsqrte $frD, $frB", FPGeneral,
|
|
[(set f64:$frD, (PPCfrsqrte f64:$frB))]>;
|
|
def FRSQRTES : XForm_26<59, 26, (outs F4RC:$frD), (ins F4RC:$frB),
|
|
"frsqrtes $frD, $frB", FPGeneral,
|
|
[(set f32:$frD, (PPCfrsqrte f32:$frB))]>;
|
|
}
|
|
|
|
// XL-Form instructions. condition register logical ops.
|
|
//
|
|
def MCRF : XLForm_3<19, 0, (outs CRRC:$BF), (ins CRRC:$BFA),
|
|
"mcrf $BF, $BFA", BrMCR>,
|
|
PPC970_DGroup_First, PPC970_Unit_CRU;
|
|
|
|
def CREQV : XLForm_1<19, 289, (outs CRBITRC:$CRD),
|
|
(ins CRBITRC:$CRA, CRBITRC:$CRB),
|
|
"creqv $CRD, $CRA, $CRB", BrCR,
|
|
[]>;
|
|
|
|
def CROR : XLForm_1<19, 449, (outs CRBITRC:$CRD),
|
|
(ins CRBITRC:$CRA, CRBITRC:$CRB),
|
|
"cror $CRD, $CRA, $CRB", BrCR,
|
|
[]>;
|
|
|
|
let isCodeGenOnly = 1 in {
|
|
def CRSET : XLForm_1_ext<19, 289, (outs CRBITRC:$dst), (ins),
|
|
"creqv $dst, $dst, $dst", BrCR,
|
|
[]>;
|
|
|
|
def CRUNSET: XLForm_1_ext<19, 193, (outs CRBITRC:$dst), (ins),
|
|
"crxor $dst, $dst, $dst", BrCR,
|
|
[]>;
|
|
|
|
let Defs = [CR1EQ], CRD = 6 in {
|
|
def CR6SET : XLForm_1_ext<19, 289, (outs), (ins),
|
|
"creqv 6, 6, 6", BrCR,
|
|
[(PPCcr6set)]>;
|
|
|
|
def CR6UNSET: XLForm_1_ext<19, 193, (outs), (ins),
|
|
"crxor 6, 6, 6", BrCR,
|
|
[(PPCcr6unset)]>;
|
|
}
|
|
}
|
|
|
|
// XFX-Form instructions. Instructions that deal with SPRs.
|
|
//
|
|
let Uses = [CTR] in {
|
|
def MFCTR : XFXForm_1_ext<31, 339, 9, (outs GPRC:$rT), (ins),
|
|
"mfctr $rT", SprMFSPR>,
|
|
PPC970_DGroup_First, PPC970_Unit_FXU;
|
|
}
|
|
let Defs = [CTR], Pattern = [(PPCmtctr i32:$rS)] in {
|
|
def MTCTR : XFXForm_7_ext<31, 467, 9, (outs), (ins GPRC:$rS),
|
|
"mtctr $rS", SprMTSPR>,
|
|
PPC970_DGroup_First, PPC970_Unit_FXU;
|
|
}
|
|
|
|
let Defs = [LR] in {
|
|
def MTLR : XFXForm_7_ext<31, 467, 8, (outs), (ins GPRC:$rS),
|
|
"mtlr $rS", SprMTSPR>,
|
|
PPC970_DGroup_First, PPC970_Unit_FXU;
|
|
}
|
|
let Uses = [LR] in {
|
|
def MFLR : XFXForm_1_ext<31, 339, 8, (outs GPRC:$rT), (ins),
|
|
"mflr $rT", SprMFSPR>,
|
|
PPC970_DGroup_First, PPC970_Unit_FXU;
|
|
}
|
|
|
|
// Move to/from VRSAVE: despite being a SPR, the VRSAVE register is renamed like
|
|
// a GPR on the PPC970. As such, copies in and out have the same performance
|
|
// characteristics as an OR instruction.
|
|
def MTVRSAVE : XFXForm_7_ext<31, 467, 256, (outs), (ins GPRC:$rS),
|
|
"mtspr 256, $rS", IntGeneral>,
|
|
PPC970_DGroup_Single, PPC970_Unit_FXU;
|
|
def MFVRSAVE : XFXForm_1_ext<31, 339, 256, (outs GPRC:$rT), (ins),
|
|
"mfspr $rT, 256", IntGeneral>,
|
|
PPC970_DGroup_First, PPC970_Unit_FXU;
|
|
|
|
let isCodeGenOnly = 1 in {
|
|
def MTVRSAVEv : XFXForm_7_ext<31, 467, 256,
|
|
(outs VRSAVERC:$reg), (ins GPRC:$rS),
|
|
"mtspr 256, $rS", IntGeneral>,
|
|
PPC970_DGroup_Single, PPC970_Unit_FXU;
|
|
def MFVRSAVEv : XFXForm_1_ext<31, 339, 256, (outs GPRC:$rT),
|
|
(ins VRSAVERC:$reg),
|
|
"mfspr $rT, 256", IntGeneral>,
|
|
PPC970_DGroup_First, PPC970_Unit_FXU;
|
|
}
|
|
|
|
// SPILL_VRSAVE - Indicate that we're dumping the VRSAVE register,
|
|
// so we'll need to scavenge a register for it.
|
|
let mayStore = 1 in
|
|
def SPILL_VRSAVE : Pseudo<(outs), (ins VRSAVERC:$vrsave, memri:$F),
|
|
"#SPILL_VRSAVE", []>;
|
|
|
|
// RESTORE_VRSAVE - Indicate that we're restoring the VRSAVE register (previously
|
|
// spilled), so we'll need to scavenge a register for it.
|
|
let mayLoad = 1 in
|
|
def RESTORE_VRSAVE : Pseudo<(outs VRSAVERC:$vrsave), (ins memri:$F),
|
|
"#RESTORE_VRSAVE", []>;
|
|
|
|
def MTCRF : XFXForm_5<31, 144, (outs crbitm:$FXM), (ins GPRC:$rS),
|
|
"mtcrf $FXM, $rS", BrMCRX>,
|
|
PPC970_MicroCode, PPC970_Unit_CRU;
|
|
|
|
// This is a pseudo for MFCR, which implicitly uses all 8 of its subregisters;
|
|
// declaring that here gives the local register allocator problems with this:
|
|
// vreg = MCRF CR0
|
|
// MFCR <kill of whatever preg got assigned to vreg>
|
|
// while not declaring it breaks DeadMachineInstructionElimination.
|
|
// As it turns out, in all cases where we currently use this,
|
|
// we're only interested in one subregister of it. Represent this in the
|
|
// instruction to keep the register allocator from becoming confused.
|
|
//
|
|
// FIXME: Make this a real Pseudo instruction when the JIT switches to MC.
|
|
let isCodeGenOnly = 1 in
|
|
def MFCRpseud: XFXForm_3<31, 19, (outs GPRC:$rT), (ins crbitm:$FXM),
|
|
"#MFCRpseud", SprMFCR>,
|
|
PPC970_MicroCode, PPC970_Unit_CRU;
|
|
|
|
def MFCR : XFXForm_3<31, 19, (outs GPRC:$rT), (ins),
|
|
"mfcr $rT", SprMFCR>,
|
|
PPC970_MicroCode, PPC970_Unit_CRU;
|
|
|
|
def MFOCRF: XFXForm_5a<31, 19, (outs GPRC:$rT), (ins crbitm:$FXM),
|
|
"mfocrf $rT, $FXM", SprMFCR>,
|
|
PPC970_DGroup_First, PPC970_Unit_CRU;
|
|
|
|
// Pseudo instruction to perform FADD in round-to-zero mode.
|
|
let usesCustomInserter = 1, Uses = [RM] in {
|
|
def FADDrtz: Pseudo<(outs F8RC:$FRT), (ins F8RC:$FRA, F8RC:$FRB), "",
|
|
[(set f64:$FRT, (PPCfaddrtz f64:$FRA, f64:$FRB))]>;
|
|
}
|
|
|
|
// The above pseudo gets expanded to make use of the following instructions
|
|
// to manipulate FPSCR. Note that FPSCR is not modeled at the DAG level.
|
|
let Uses = [RM], Defs = [RM] in {
|
|
def MTFSB0 : XForm_43<63, 70, (outs), (ins u5imm:$FM),
|
|
"mtfsb0 $FM", IntMTFSB0, []>,
|
|
PPC970_DGroup_Single, PPC970_Unit_FPU;
|
|
def MTFSB1 : XForm_43<63, 38, (outs), (ins u5imm:$FM),
|
|
"mtfsb1 $FM", IntMTFSB0, []>,
|
|
PPC970_DGroup_Single, PPC970_Unit_FPU;
|
|
def MTFSF : XFLForm<63, 711, (outs), (ins i32imm:$FM, F8RC:$rT),
|
|
"mtfsf $FM, $rT", IntMTFSB0, []>,
|
|
PPC970_DGroup_Single, PPC970_Unit_FPU;
|
|
}
|
|
let Uses = [RM] in {
|
|
def MFFS : XForm_42<63, 583, (outs F8RC:$rT), (ins),
|
|
"mffs $rT", IntMFFS,
|
|
[(set f64:$rT, (PPCmffs))]>,
|
|
PPC970_DGroup_Single, PPC970_Unit_FPU;
|
|
}
|
|
|
|
|
|
let PPC970_Unit = 1 in { // FXU Operations.
|
|
|
|
// XO-Form instructions. Arithmetic instructions that can set overflow bit
|
|
//
|
|
def ADD4 : XOForm_1<31, 266, 0, (outs GPRC:$rT), (ins GPRC:$rA, GPRC:$rB),
|
|
"add $rT, $rA, $rB", IntSimple,
|
|
[(set i32:$rT, (add i32:$rA, i32:$rB))]>;
|
|
let Defs = [CARRY] in {
|
|
def ADDC : XOForm_1<31, 10, 0, (outs GPRC:$rT), (ins GPRC:$rA, GPRC:$rB),
|
|
"addc $rT, $rA, $rB", IntGeneral,
|
|
[(set i32:$rT, (addc i32:$rA, i32:$rB))]>,
|
|
PPC970_DGroup_Cracked;
|
|
}
|
|
def DIVW : XOForm_1<31, 491, 0, (outs GPRC:$rT), (ins GPRC:$rA, GPRC:$rB),
|
|
"divw $rT, $rA, $rB", IntDivW,
|
|
[(set i32:$rT, (sdiv i32:$rA, i32:$rB))]>,
|
|
PPC970_DGroup_First, PPC970_DGroup_Cracked;
|
|
def DIVWU : XOForm_1<31, 459, 0, (outs GPRC:$rT), (ins GPRC:$rA, GPRC:$rB),
|
|
"divwu $rT, $rA, $rB", IntDivW,
|
|
[(set i32:$rT, (udiv i32:$rA, i32:$rB))]>,
|
|
PPC970_DGroup_First, PPC970_DGroup_Cracked;
|
|
def MULHW : XOForm_1<31, 75, 0, (outs GPRC:$rT), (ins GPRC:$rA, GPRC:$rB),
|
|
"mulhw $rT, $rA, $rB", IntMulHW,
|
|
[(set i32:$rT, (mulhs i32:$rA, i32:$rB))]>;
|
|
def MULHWU : XOForm_1<31, 11, 0, (outs GPRC:$rT), (ins GPRC:$rA, GPRC:$rB),
|
|
"mulhwu $rT, $rA, $rB", IntMulHWU,
|
|
[(set i32:$rT, (mulhu i32:$rA, i32:$rB))]>;
|
|
def MULLW : XOForm_1<31, 235, 0, (outs GPRC:$rT), (ins GPRC:$rA, GPRC:$rB),
|
|
"mullw $rT, $rA, $rB", IntMulHW,
|
|
[(set i32:$rT, (mul i32:$rA, i32:$rB))]>;
|
|
def SUBF : XOForm_1<31, 40, 0, (outs GPRC:$rT), (ins GPRC:$rA, GPRC:$rB),
|
|
"subf $rT, $rA, $rB", IntGeneral,
|
|
[(set i32:$rT, (sub i32:$rB, i32:$rA))]>;
|
|
let Defs = [CARRY] in {
|
|
def SUBFC : XOForm_1<31, 8, 0, (outs GPRC:$rT), (ins GPRC:$rA, GPRC:$rB),
|
|
"subfc $rT, $rA, $rB", IntGeneral,
|
|
[(set i32:$rT, (subc i32:$rB, i32:$rA))]>,
|
|
PPC970_DGroup_Cracked;
|
|
}
|
|
def NEG : XOForm_3<31, 104, 0, (outs GPRC:$rT), (ins GPRC:$rA),
|
|
"neg $rT, $rA", IntSimple,
|
|
[(set i32:$rT, (ineg i32:$rA))]>;
|
|
let Uses = [CARRY], Defs = [CARRY] in {
|
|
def ADDE : XOForm_1<31, 138, 0, (outs GPRC:$rT), (ins GPRC:$rA, GPRC:$rB),
|
|
"adde $rT, $rA, $rB", IntGeneral,
|
|
[(set i32:$rT, (adde i32:$rA, i32:$rB))]>;
|
|
def ADDME : XOForm_3<31, 234, 0, (outs GPRC:$rT), (ins GPRC:$rA),
|
|
"addme $rT, $rA", IntGeneral,
|
|
[(set i32:$rT, (adde i32:$rA, -1))]>;
|
|
def ADDZE : XOForm_3<31, 202, 0, (outs GPRC:$rT), (ins GPRC:$rA),
|
|
"addze $rT, $rA", IntGeneral,
|
|
[(set i32:$rT, (adde i32:$rA, 0))]>;
|
|
def SUBFE : XOForm_1<31, 136, 0, (outs GPRC:$rT), (ins GPRC:$rA, GPRC:$rB),
|
|
"subfe $rT, $rA, $rB", IntGeneral,
|
|
[(set i32:$rT, (sube i32:$rB, i32:$rA))]>;
|
|
def SUBFME : XOForm_3<31, 232, 0, (outs GPRC:$rT), (ins GPRC:$rA),
|
|
"subfme $rT, $rA", IntGeneral,
|
|
[(set i32:$rT, (sube -1, i32:$rA))]>;
|
|
def SUBFZE : XOForm_3<31, 200, 0, (outs GPRC:$rT), (ins GPRC:$rA),
|
|
"subfze $rT, $rA", IntGeneral,
|
|
[(set i32:$rT, (sube 0, i32:$rA))]>;
|
|
}
|
|
}
|
|
|
|
// A-Form instructions. Most of the instructions executed in the FPU are of
|
|
// this type.
|
|
//
|
|
let PPC970_Unit = 3 in { // FPU Operations.
|
|
let Uses = [RM] in {
|
|
def FMADD : AForm_1<63, 29,
|
|
(outs F8RC:$FRT), (ins F8RC:$FRA, F8RC:$FRC, F8RC:$FRB),
|
|
"fmadd $FRT, $FRA, $FRC, $FRB", FPFused,
|
|
[(set f64:$FRT, (fma f64:$FRA, f64:$FRC, f64:$FRB))]>;
|
|
def FMADDS : AForm_1<59, 29,
|
|
(outs F4RC:$FRT), (ins F4RC:$FRA, F4RC:$FRC, F4RC:$FRB),
|
|
"fmadds $FRT, $FRA, $FRC, $FRB", FPGeneral,
|
|
[(set f32:$FRT, (fma f32:$FRA, f32:$FRC, f32:$FRB))]>;
|
|
def FMSUB : AForm_1<63, 28,
|
|
(outs F8RC:$FRT), (ins F8RC:$FRA, F8RC:$FRC, F8RC:$FRB),
|
|
"fmsub $FRT, $FRA, $FRC, $FRB", FPFused,
|
|
[(set f64:$FRT,
|
|
(fma f64:$FRA, f64:$FRC, (fneg f64:$FRB)))]>;
|
|
def FMSUBS : AForm_1<59, 28,
|
|
(outs F4RC:$FRT), (ins F4RC:$FRA, F4RC:$FRC, F4RC:$FRB),
|
|
"fmsubs $FRT, $FRA, $FRC, $FRB", FPGeneral,
|
|
[(set f32:$FRT,
|
|
(fma f32:$FRA, f32:$FRC, (fneg f32:$FRB)))]>;
|
|
def FNMADD : AForm_1<63, 31,
|
|
(outs F8RC:$FRT), (ins F8RC:$FRA, F8RC:$FRC, F8RC:$FRB),
|
|
"fnmadd $FRT, $FRA, $FRC, $FRB", FPFused,
|
|
[(set f64:$FRT,
|
|
(fneg (fma f64:$FRA, f64:$FRC, f64:$FRB)))]>;
|
|
def FNMADDS : AForm_1<59, 31,
|
|
(outs F4RC:$FRT), (ins F4RC:$FRA, F4RC:$FRC, F4RC:$FRB),
|
|
"fnmadds $FRT, $FRA, $FRC, $FRB", FPGeneral,
|
|
[(set f32:$FRT,
|
|
(fneg (fma f32:$FRA, f32:$FRC, f32:$FRB)))]>;
|
|
def FNMSUB : AForm_1<63, 30,
|
|
(outs F8RC:$FRT), (ins F8RC:$FRA, F8RC:$FRC, F8RC:$FRB),
|
|
"fnmsub $FRT, $FRA, $FRC, $FRB", FPFused,
|
|
[(set f64:$FRT, (fneg (fma f64:$FRA, f64:$FRC,
|
|
(fneg f64:$FRB))))]>;
|
|
def FNMSUBS : AForm_1<59, 30,
|
|
(outs F4RC:$FRT), (ins F4RC:$FRA, F4RC:$FRC, F4RC:$FRB),
|
|
"fnmsubs $FRT, $FRA, $FRC, $FRB", FPGeneral,
|
|
[(set f32:$FRT, (fneg (fma f32:$FRA, f32:$FRC,
|
|
(fneg f32:$FRB))))]>;
|
|
}
|
|
// FSEL is artificially split into 4 and 8-byte forms for the result. To avoid
|
|
// having 4 of these, force the comparison to always be an 8-byte double (code
|
|
// should use an FMRSD if the input comparison value really wants to be a float)
|
|
// and 4/8 byte forms for the result and operand type..
|
|
def FSELD : AForm_1<63, 23,
|
|
(outs F8RC:$FRT), (ins F8RC:$FRA, F8RC:$FRC, F8RC:$FRB),
|
|
"fsel $FRT, $FRA, $FRC, $FRB", FPGeneral,
|
|
[(set f64:$FRT, (PPCfsel f64:$FRA, f64:$FRC, f64:$FRB))]>;
|
|
def FSELS : AForm_1<63, 23,
|
|
(outs F4RC:$FRT), (ins F8RC:$FRA, F4RC:$FRC, F4RC:$FRB),
|
|
"fsel $FRT, $FRA, $FRC, $FRB", FPGeneral,
|
|
[(set f32:$FRT, (PPCfsel f64:$FRA, f32:$FRC, f32:$FRB))]>;
|
|
let Uses = [RM] in {
|
|
def FADD : AForm_2<63, 21,
|
|
(outs F8RC:$FRT), (ins F8RC:$FRA, F8RC:$FRB),
|
|
"fadd $FRT, $FRA, $FRB", FPAddSub,
|
|
[(set f64:$FRT, (fadd f64:$FRA, f64:$FRB))]>;
|
|
def FADDS : AForm_2<59, 21,
|
|
(outs F4RC:$FRT), (ins F4RC:$FRA, F4RC:$FRB),
|
|
"fadds $FRT, $FRA, $FRB", FPGeneral,
|
|
[(set f32:$FRT, (fadd f32:$FRA, f32:$FRB))]>;
|
|
def FDIV : AForm_2<63, 18,
|
|
(outs F8RC:$FRT), (ins F8RC:$FRA, F8RC:$FRB),
|
|
"fdiv $FRT, $FRA, $FRB", FPDivD,
|
|
[(set f64:$FRT, (fdiv f64:$FRA, f64:$FRB))]>;
|
|
def FDIVS : AForm_2<59, 18,
|
|
(outs F4RC:$FRT), (ins F4RC:$FRA, F4RC:$FRB),
|
|
"fdivs $FRT, $FRA, $FRB", FPDivS,
|
|
[(set f32:$FRT, (fdiv f32:$FRA, f32:$FRB))]>;
|
|
def FMUL : AForm_3<63, 25,
|
|
(outs F8RC:$FRT), (ins F8RC:$FRA, F8RC:$FRC),
|
|
"fmul $FRT, $FRA, $FRC", FPFused,
|
|
[(set f64:$FRT, (fmul f64:$FRA, f64:$FRC))]>;
|
|
def FMULS : AForm_3<59, 25,
|
|
(outs F4RC:$FRT), (ins F4RC:$FRA, F4RC:$FRC),
|
|
"fmuls $FRT, $FRA, $FRC", FPGeneral,
|
|
[(set f32:$FRT, (fmul f32:$FRA, f32:$FRC))]>;
|
|
def FSUB : AForm_2<63, 20,
|
|
(outs F8RC:$FRT), (ins F8RC:$FRA, F8RC:$FRB),
|
|
"fsub $FRT, $FRA, $FRB", FPAddSub,
|
|
[(set f64:$FRT, (fsub f64:$FRA, f64:$FRB))]>;
|
|
def FSUBS : AForm_2<59, 20,
|
|
(outs F4RC:$FRT), (ins F4RC:$FRA, F4RC:$FRB),
|
|
"fsubs $FRT, $FRA, $FRB", FPGeneral,
|
|
[(set f32:$FRT, (fsub f32:$FRA, f32:$FRB))]>;
|
|
}
|
|
}
|
|
|
|
let PPC970_Unit = 1 in { // FXU Operations.
|
|
def ISEL : AForm_4<31, 15,
|
|
(outs GPRC:$rT), (ins GPRC_NOR0:$rA, GPRC:$rB, CRBITRC:$cond),
|
|
"isel $rT, $rA, $rB, $cond", IntGeneral,
|
|
[]>;
|
|
}
|
|
|
|
let PPC970_Unit = 1 in { // FXU Operations.
|
|
// M-Form instructions. rotate and mask instructions.
|
|
//
|
|
let isCommutable = 1 in {
|
|
// RLWIMI can be commuted if the rotate amount is zero.
|
|
def RLWIMI : MForm_2<20,
|
|
(outs GPRC:$rA), (ins GPRC:$rSi, GPRC:$rS, u5imm:$SH, u5imm:$MB,
|
|
u5imm:$ME), "rlwimi $rA, $rS, $SH, $MB, $ME", IntRotate,
|
|
[]>, PPC970_DGroup_Cracked, RegConstraint<"$rSi = $rA">,
|
|
NoEncode<"$rSi">;
|
|
}
|
|
def RLWINM : MForm_2<21,
|
|
(outs GPRC:$rA), (ins GPRC:$rS, u5imm:$SH, u5imm:$MB, u5imm:$ME),
|
|
"rlwinm $rA, $rS, $SH, $MB, $ME", IntGeneral,
|
|
[]>;
|
|
def RLWINMo : MForm_2<21,
|
|
(outs GPRC:$rA), (ins GPRC:$rS, u5imm:$SH, u5imm:$MB, u5imm:$ME),
|
|
"rlwinm. $rA, $rS, $SH, $MB, $ME", IntGeneral,
|
|
[]>, isDOT, PPC970_DGroup_Cracked;
|
|
def RLWNM : MForm_2<23,
|
|
(outs GPRC:$rA), (ins GPRC:$rS, GPRC:$rB, u5imm:$MB, u5imm:$ME),
|
|
"rlwnm $rA, $rS, $rB, $MB, $ME", IntGeneral,
|
|
[]>;
|
|
}
|
|
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// PowerPC Instruction Patterns
|
|
//
|
|
|
|
// Arbitrary immediate support. Implement in terms of LIS/ORI.
|
|
def : Pat<(i32 imm:$imm),
|
|
(ORI (LIS (HI16 imm:$imm)), (LO16 imm:$imm))>;
|
|
|
|
// Implement the 'not' operation with the NOR instruction.
|
|
def NOT : Pat<(not i32:$in),
|
|
(NOR $in, $in)>;
|
|
|
|
// ADD an arbitrary immediate.
|
|
def : Pat<(add i32:$in, imm:$imm),
|
|
(ADDIS (ADDI $in, (LO16 imm:$imm)), (HA16 imm:$imm))>;
|
|
// OR an arbitrary immediate.
|
|
def : Pat<(or i32:$in, imm:$imm),
|
|
(ORIS (ORI $in, (LO16 imm:$imm)), (HI16 imm:$imm))>;
|
|
// XOR an arbitrary immediate.
|
|
def : Pat<(xor i32:$in, imm:$imm),
|
|
(XORIS (XORI $in, (LO16 imm:$imm)), (HI16 imm:$imm))>;
|
|
// SUBFIC
|
|
def : Pat<(sub immSExt16:$imm, i32:$in),
|
|
(SUBFIC $in, imm:$imm)>;
|
|
|
|
// SHL/SRL
|
|
def : Pat<(shl i32:$in, (i32 imm:$imm)),
|
|
(RLWINM $in, imm:$imm, 0, (SHL32 imm:$imm))>;
|
|
def : Pat<(srl i32:$in, (i32 imm:$imm)),
|
|
(RLWINM $in, (SRL32 imm:$imm), imm:$imm, 31)>;
|
|
|
|
// ROTL
|
|
def : Pat<(rotl i32:$in, i32:$sh),
|
|
(RLWNM $in, $sh, 0, 31)>;
|
|
def : Pat<(rotl i32:$in, (i32 imm:$imm)),
|
|
(RLWINM $in, imm:$imm, 0, 31)>;
|
|
|
|
// RLWNM
|
|
def : Pat<(and (rotl i32:$in, i32:$sh), maskimm32:$imm),
|
|
(RLWNM $in, $sh, (MB maskimm32:$imm), (ME maskimm32:$imm))>;
|
|
|
|
// Calls
|
|
def : Pat<(PPCcall (i32 tglobaladdr:$dst)),
|
|
(BL tglobaladdr:$dst)>;
|
|
def : Pat<(PPCcall (i32 texternalsym:$dst)),
|
|
(BL texternalsym:$dst)>;
|
|
|
|
|
|
def : Pat<(PPCtc_return (i32 tglobaladdr:$dst), imm:$imm),
|
|
(TCRETURNdi tglobaladdr:$dst, imm:$imm)>;
|
|
|
|
def : Pat<(PPCtc_return (i32 texternalsym:$dst), imm:$imm),
|
|
(TCRETURNdi texternalsym:$dst, imm:$imm)>;
|
|
|
|
def : Pat<(PPCtc_return CTRRC:$dst, imm:$imm),
|
|
(TCRETURNri CTRRC:$dst, imm:$imm)>;
|
|
|
|
|
|
|
|
// Hi and Lo for Darwin Global Addresses.
|
|
def : Pat<(PPChi tglobaladdr:$in, 0), (LIS tglobaladdr:$in)>;
|
|
def : Pat<(PPClo tglobaladdr:$in, 0), (LI tglobaladdr:$in)>;
|
|
def : Pat<(PPChi tconstpool:$in, 0), (LIS tconstpool:$in)>;
|
|
def : Pat<(PPClo tconstpool:$in, 0), (LI tconstpool:$in)>;
|
|
def : Pat<(PPChi tjumptable:$in, 0), (LIS tjumptable:$in)>;
|
|
def : Pat<(PPClo tjumptable:$in, 0), (LI tjumptable:$in)>;
|
|
def : Pat<(PPChi tblockaddress:$in, 0), (LIS tblockaddress:$in)>;
|
|
def : Pat<(PPClo tblockaddress:$in, 0), (LI tblockaddress:$in)>;
|
|
def : Pat<(PPChi tglobaltlsaddr:$g, i32:$in),
|
|
(ADDIS $in, tglobaltlsaddr:$g)>;
|
|
def : Pat<(PPClo tglobaltlsaddr:$g, i32:$in),
|
|
(ADDI $in, tglobaltlsaddr:$g)>;
|
|
def : Pat<(add i32:$in, (PPChi tglobaladdr:$g, 0)),
|
|
(ADDIS $in, tglobaladdr:$g)>;
|
|
def : Pat<(add i32:$in, (PPChi tconstpool:$g, 0)),
|
|
(ADDIS $in, tconstpool:$g)>;
|
|
def : Pat<(add i32:$in, (PPChi tjumptable:$g, 0)),
|
|
(ADDIS $in, tjumptable:$g)>;
|
|
def : Pat<(add i32:$in, (PPChi tblockaddress:$g, 0)),
|
|
(ADDIS $in, tblockaddress:$g)>;
|
|
|
|
// Standard shifts. These are represented separately from the real shifts above
|
|
// so that we can distinguish between shifts that allow 5-bit and 6-bit shift
|
|
// amounts.
|
|
def : Pat<(sra i32:$rS, i32:$rB),
|
|
(SRAW $rS, $rB)>;
|
|
def : Pat<(srl i32:$rS, i32:$rB),
|
|
(SRW $rS, $rB)>;
|
|
def : Pat<(shl i32:$rS, i32:$rB),
|
|
(SLW $rS, $rB)>;
|
|
|
|
def : Pat<(zextloadi1 iaddr:$src),
|
|
(LBZ iaddr:$src)>;
|
|
def : Pat<(zextloadi1 xaddr:$src),
|
|
(LBZX xaddr:$src)>;
|
|
def : Pat<(extloadi1 iaddr:$src),
|
|
(LBZ iaddr:$src)>;
|
|
def : Pat<(extloadi1 xaddr:$src),
|
|
(LBZX xaddr:$src)>;
|
|
def : Pat<(extloadi8 iaddr:$src),
|
|
(LBZ iaddr:$src)>;
|
|
def : Pat<(extloadi8 xaddr:$src),
|
|
(LBZX xaddr:$src)>;
|
|
def : Pat<(extloadi16 iaddr:$src),
|
|
(LHZ iaddr:$src)>;
|
|
def : Pat<(extloadi16 xaddr:$src),
|
|
(LHZX xaddr:$src)>;
|
|
def : Pat<(f64 (extloadf32 iaddr:$src)),
|
|
(COPY_TO_REGCLASS (LFS iaddr:$src), F8RC)>;
|
|
def : Pat<(f64 (extloadf32 xaddr:$src)),
|
|
(COPY_TO_REGCLASS (LFSX xaddr:$src), F8RC)>;
|
|
|
|
def : Pat<(f64 (fextend f32:$src)),
|
|
(COPY_TO_REGCLASS $src, F8RC)>;
|
|
|
|
// Memory barriers
|
|
def : Pat<(membarrier (i32 imm /*ll*/),
|
|
(i32 imm /*ls*/),
|
|
(i32 imm /*sl*/),
|
|
(i32 imm /*ss*/),
|
|
(i32 imm /*device*/)),
|
|
(SYNC)>;
|
|
|
|
def : Pat<(atomic_fence (imm), (imm)), (SYNC)>;
|
|
|
|
// Additional FNMSUB patterns: -a*c + b == -(a*c - b)
|
|
def : Pat<(fma (fneg f64:$A), f64:$C, f64:$B),
|
|
(FNMSUB $A, $C, $B)>;
|
|
def : Pat<(fma f64:$A, (fneg f64:$C), f64:$B),
|
|
(FNMSUB $A, $C, $B)>;
|
|
def : Pat<(fma (fneg f32:$A), f32:$C, f32:$B),
|
|
(FNMSUBS $A, $C, $B)>;
|
|
def : Pat<(fma f32:$A, (fneg f32:$C), f32:$B),
|
|
(FNMSUBS $A, $C, $B)>;
|
|
|
|
include "PPCInstrAltivec.td"
|
|
include "PPCInstr64Bit.td"
|