llvm-6502/lib/Target/PowerPC/MCTargetDesc/PPCMCCodeEmitter.cpp

227 lines
9.0 KiB
C++
Raw Normal View History

//===-- PPCMCCodeEmitter.cpp - Convert PPC code to machine code -----------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file implements the PPCMCCodeEmitter class.
//
//===----------------------------------------------------------------------===//
#define DEBUG_TYPE "mccodeemitter"
#include "MCTargetDesc/PPCMCTargetDesc.h"
#include "MCTargetDesc/PPCFixupKinds.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/MC/MCCodeEmitter.h"
#include "llvm/MC/MCContext.h"
This patch implements the general dynamic TLS model for 64-bit PowerPC. Given a thread-local symbol x with global-dynamic access, the generated code to obtain x's address is: Instruction Relocation Symbol addis ra,r2,x@got@tlsgd@ha R_PPC64_GOT_TLSGD16_HA x addi r3,ra,x@got@tlsgd@l R_PPC64_GOT_TLSGD16_L x bl __tls_get_addr(x@tlsgd) R_PPC64_TLSGD x R_PPC64_REL24 __tls_get_addr nop <use address in r3> The implementation borrows from the medium code model work for introducing special forms of ADDIS and ADDI into the DAG representation. This is made slightly more complicated by having to introduce a call to the external function __tls_get_addr. Using the full call machinery is overkill and, more importantly, makes it difficult to add a special relocation. So I've introduced another opcode GET_TLS_ADDR to represent the function call, and surrounded it with register copies to set up the parameter and return value. Most of the code is pretty straightforward. I ran into one peculiarity when I introduced a new PPC opcode BL8_NOP_ELF_TLSGD, which is just like BL8_NOP_ELF except that it takes another parameter to represent the symbol ("x" above) that requires a relocation on the call. Something in the TblGen machinery causes BL8_NOP_ELF and BL8_NOP_ELF_TLSGD to be treated identically during the emit phase, so this second operand was never visited to generate relocations. This is the reason for the slightly messy workaround in PPCMCCodeEmitter.cpp:getDirectBrEncoding(). Two new tests are included to demonstrate correct external assembly and correct generation of relocations using the integrated assembler. Comments welcome! Thanks, Bill git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@169910 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-11 20:30:11 +00:00
#include "llvm/MC/MCExpr.h"
#include "llvm/MC/MCInst.h"
#include "llvm/MC/MCInstrInfo.h"
#include "llvm/MC/MCSubtargetInfo.h"
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/raw_ostream.h"
using namespace llvm;
STATISTIC(MCNumEmitted, "Number of MC instructions emitted");
namespace {
class PPCMCCodeEmitter : public MCCodeEmitter {
PPCMCCodeEmitter(const PPCMCCodeEmitter &) LLVM_DELETED_FUNCTION;
void operator=(const PPCMCCodeEmitter &) LLVM_DELETED_FUNCTION;
const MCSubtargetInfo &STI;
const MCContext &CTX;
Triple TT;
public:
PPCMCCodeEmitter(const MCInstrInfo &mcii, const MCSubtargetInfo &sti,
MCContext &ctx)
: STI(sti), CTX(ctx), TT(STI.getTargetTriple()) {
}
~PPCMCCodeEmitter() {}
unsigned getDirectBrEncoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const;
unsigned getCondBrEncoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const;
unsigned getS16ImmEncoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const;
unsigned getMemRIEncoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const;
unsigned getMemRIXEncoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const;
unsigned getTLSRegEncoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const;
unsigned get_crbitm_encoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const;
/// getMachineOpValue - Return binary encoding of operand. If the machine
/// operand requires relocation, record the relocation and return zero.
unsigned getMachineOpValue(const MCInst &MI,const MCOperand &MO,
SmallVectorImpl<MCFixup> &Fixups) const;
// getBinaryCodeForInstr - TableGen'erated function for getting the
// binary encoding for an instruction.
uint64_t getBinaryCodeForInstr(const MCInst &MI,
SmallVectorImpl<MCFixup> &Fixups) const;
void EncodeInstruction(const MCInst &MI, raw_ostream &OS,
SmallVectorImpl<MCFixup> &Fixups) const {
uint64_t Bits = getBinaryCodeForInstr(MI, Fixups);
// BL8_NOP etc. all have a size of 8 because of the following 'nop'.
unsigned Size = 4; // FIXME: Have Desc.getSize() return the correct value!
unsigned Opcode = MI.getOpcode();
if (Opcode == PPC::BL8_NOP || Opcode == PPC::BLA8_NOP ||
Opcode == PPC::BL8_NOP_TLSGD || Opcode == PPC::BL8_NOP_TLSLD)
Size = 8;
// Output the constant in big endian byte order.
int ShiftValue = (Size * 8) - 8;
for (unsigned i = 0; i != Size; ++i) {
OS << (char)(Bits >> ShiftValue);
Bits <<= 8;
}
++MCNumEmitted; // Keep track of the # of mi's emitted.
}
};
} // end anonymous namespace
MCCodeEmitter *llvm::createPPCMCCodeEmitter(const MCInstrInfo &MCII,
const MCRegisterInfo &MRI,
const MCSubtargetInfo &STI,
MCContext &Ctx) {
return new PPCMCCodeEmitter(MCII, STI, Ctx);
}
unsigned PPCMCCodeEmitter::
getDirectBrEncoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const {
const MCOperand &MO = MI.getOperand(OpNo);
if (MO.isReg() || MO.isImm()) return getMachineOpValue(MI, MO, Fixups);
// Add a fixup for the branch target.
Fixups.push_back(MCFixup::Create(0, MO.getExpr(),
(MCFixupKind)PPC::fixup_ppc_br24));
This patch implements the general dynamic TLS model for 64-bit PowerPC. Given a thread-local symbol x with global-dynamic access, the generated code to obtain x's address is: Instruction Relocation Symbol addis ra,r2,x@got@tlsgd@ha R_PPC64_GOT_TLSGD16_HA x addi r3,ra,x@got@tlsgd@l R_PPC64_GOT_TLSGD16_L x bl __tls_get_addr(x@tlsgd) R_PPC64_TLSGD x R_PPC64_REL24 __tls_get_addr nop <use address in r3> The implementation borrows from the medium code model work for introducing special forms of ADDIS and ADDI into the DAG representation. This is made slightly more complicated by having to introduce a call to the external function __tls_get_addr. Using the full call machinery is overkill and, more importantly, makes it difficult to add a special relocation. So I've introduced another opcode GET_TLS_ADDR to represent the function call, and surrounded it with register copies to set up the parameter and return value. Most of the code is pretty straightforward. I ran into one peculiarity when I introduced a new PPC opcode BL8_NOP_ELF_TLSGD, which is just like BL8_NOP_ELF except that it takes another parameter to represent the symbol ("x" above) that requires a relocation on the call. Something in the TblGen machinery causes BL8_NOP_ELF and BL8_NOP_ELF_TLSGD to be treated identically during the emit phase, so this second operand was never visited to generate relocations. This is the reason for the slightly messy workaround in PPCMCCodeEmitter.cpp:getDirectBrEncoding(). Two new tests are included to demonstrate correct external assembly and correct generation of relocations using the integrated assembler. Comments welcome! Thanks, Bill git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@169910 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-11 20:30:11 +00:00
// For special TLS calls, add another fixup for the symbol. Apparently
// BL8_NOP, BL8_NOP_TLSGD, and BL8_NOP_TLSLD are sufficiently
// similar that TblGen will not generate a separate case for the latter
// two, so this is the only way to get the extra fixup generated.
unsigned Opcode = MI.getOpcode();
if (Opcode == PPC::BL8_NOP_TLSGD || Opcode == PPC::BL8_NOP_TLSLD) {
This patch implements the general dynamic TLS model for 64-bit PowerPC. Given a thread-local symbol x with global-dynamic access, the generated code to obtain x's address is: Instruction Relocation Symbol addis ra,r2,x@got@tlsgd@ha R_PPC64_GOT_TLSGD16_HA x addi r3,ra,x@got@tlsgd@l R_PPC64_GOT_TLSGD16_L x bl __tls_get_addr(x@tlsgd) R_PPC64_TLSGD x R_PPC64_REL24 __tls_get_addr nop <use address in r3> The implementation borrows from the medium code model work for introducing special forms of ADDIS and ADDI into the DAG representation. This is made slightly more complicated by having to introduce a call to the external function __tls_get_addr. Using the full call machinery is overkill and, more importantly, makes it difficult to add a special relocation. So I've introduced another opcode GET_TLS_ADDR to represent the function call, and surrounded it with register copies to set up the parameter and return value. Most of the code is pretty straightforward. I ran into one peculiarity when I introduced a new PPC opcode BL8_NOP_ELF_TLSGD, which is just like BL8_NOP_ELF except that it takes another parameter to represent the symbol ("x" above) that requires a relocation on the call. Something in the TblGen machinery causes BL8_NOP_ELF and BL8_NOP_ELF_TLSGD to be treated identically during the emit phase, so this second operand was never visited to generate relocations. This is the reason for the slightly messy workaround in PPCMCCodeEmitter.cpp:getDirectBrEncoding(). Two new tests are included to demonstrate correct external assembly and correct generation of relocations using the integrated assembler. Comments welcome! Thanks, Bill git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@169910 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-11 20:30:11 +00:00
const MCOperand &MO2 = MI.getOperand(OpNo+1);
Fixups.push_back(MCFixup::Create(0, MO2.getExpr(),
(MCFixupKind)PPC::fixup_ppc_nofixup));
This patch implements the general dynamic TLS model for 64-bit PowerPC. Given a thread-local symbol x with global-dynamic access, the generated code to obtain x's address is: Instruction Relocation Symbol addis ra,r2,x@got@tlsgd@ha R_PPC64_GOT_TLSGD16_HA x addi r3,ra,x@got@tlsgd@l R_PPC64_GOT_TLSGD16_L x bl __tls_get_addr(x@tlsgd) R_PPC64_TLSGD x R_PPC64_REL24 __tls_get_addr nop <use address in r3> The implementation borrows from the medium code model work for introducing special forms of ADDIS and ADDI into the DAG representation. This is made slightly more complicated by having to introduce a call to the external function __tls_get_addr. Using the full call machinery is overkill and, more importantly, makes it difficult to add a special relocation. So I've introduced another opcode GET_TLS_ADDR to represent the function call, and surrounded it with register copies to set up the parameter and return value. Most of the code is pretty straightforward. I ran into one peculiarity when I introduced a new PPC opcode BL8_NOP_ELF_TLSGD, which is just like BL8_NOP_ELF except that it takes another parameter to represent the symbol ("x" above) that requires a relocation on the call. Something in the TblGen machinery causes BL8_NOP_ELF and BL8_NOP_ELF_TLSGD to be treated identically during the emit phase, so this second operand was never visited to generate relocations. This is the reason for the slightly messy workaround in PPCMCCodeEmitter.cpp:getDirectBrEncoding(). Two new tests are included to demonstrate correct external assembly and correct generation of relocations using the integrated assembler. Comments welcome! Thanks, Bill git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@169910 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-11 20:30:11 +00:00
}
return 0;
}
unsigned PPCMCCodeEmitter::getCondBrEncoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const {
const MCOperand &MO = MI.getOperand(OpNo);
if (MO.isReg() || MO.isImm()) return getMachineOpValue(MI, MO, Fixups);
// Add a fixup for the branch target.
Fixups.push_back(MCFixup::Create(0, MO.getExpr(),
(MCFixupKind)PPC::fixup_ppc_brcond14));
return 0;
}
unsigned PPCMCCodeEmitter::getS16ImmEncoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const {
const MCOperand &MO = MI.getOperand(OpNo);
if (MO.isReg() || MO.isImm()) return getMachineOpValue(MI, MO, Fixups);
// Add a fixup for the branch target.
Fixups.push_back(MCFixup::Create(2, MO.getExpr(),
(MCFixupKind)PPC::fixup_ppc_half16));
return 0;
}
unsigned PPCMCCodeEmitter::getMemRIEncoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const {
// Encode (imm, reg) as a memri, which has the low 16-bits as the
// displacement and the next 5 bits as the register #.
assert(MI.getOperand(OpNo+1).isReg());
unsigned RegBits = getMachineOpValue(MI, MI.getOperand(OpNo+1), Fixups) << 16;
const MCOperand &MO = MI.getOperand(OpNo);
if (MO.isImm())
return (getMachineOpValue(MI, MO, Fixups) & 0xFFFF) | RegBits;
// Add a fixup for the displacement field.
Fixups.push_back(MCFixup::Create(2, MO.getExpr(),
(MCFixupKind)PPC::fixup_ppc_half16));
return RegBits;
}
unsigned PPCMCCodeEmitter::getMemRIXEncoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const {
// Encode (imm, reg) as a memrix, which has the low 14-bits as the
// displacement and the next 5 bits as the register #.
assert(MI.getOperand(OpNo+1).isReg());
unsigned RegBits = getMachineOpValue(MI, MI.getOperand(OpNo+1), Fixups) << 14;
const MCOperand &MO = MI.getOperand(OpNo);
if (MO.isImm())
[PowerPC] Use true offset value in "memrix" machine operands This is the second part of the change to always return "true" offset values from getPreIndexedAddressParts, tackling the case of "memrix" type operands. This is about instructions like LD/STD that only have a 14-bit field to encode immediate offsets, which are implicitly extended by two zero bits by the machine, so that in effect we can access 16-bit offsets as long as they are a multiple of 4. The PowerPC back end currently handles such instructions by carrying the 14-bit value (as it will get encoded into the actual machine instructions) in the machine operand fields for such instructions. This means that those values are in fact not the true offset, but rather the offset divided by 4 (and then truncated to an unsigned 14-bit value). Like in the case fixed in r182012, this makes common code operations on such offset values not work as expected. Furthermore, there doesn't really appear to be any strong reason why we should encode machine operands this way. This patch therefore changes the encoding of "memrix" type machine operands to simply contain the "true" offset value as a signed immediate value, while enforcing the rules that it must fit in a 16-bit signed value and must also be a multiple of 4. This change must be made simultaneously in all places that access machine operands of this type. However, just about all those changes make the code simpler; in many cases we can now just share the same code for memri and memrix operands. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182032 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-16 17:58:02 +00:00
return ((getMachineOpValue(MI, MO, Fixups) >> 2) & 0x3FFF) | RegBits;
// Add a fixup for the displacement field.
Fixups.push_back(MCFixup::Create(2, MO.getExpr(),
(MCFixupKind)PPC::fixup_ppc_half16ds));
return RegBits;
}
unsigned PPCMCCodeEmitter::getTLSRegEncoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const {
const MCOperand &MO = MI.getOperand(OpNo);
if (MO.isReg()) return getMachineOpValue(MI, MO, Fixups);
// Add a fixup for the TLS register, which simply provides a relocation
// hint to the linker that this statement is part of a relocation sequence.
// Return the thread-pointer register's encoding.
Fixups.push_back(MCFixup::Create(0, MO.getExpr(),
(MCFixupKind)PPC::fixup_ppc_tlsreg));
return CTX.getRegisterInfo()->getEncodingValue(PPC::X13);
}
unsigned PPCMCCodeEmitter::
get_crbitm_encoding(const MCInst &MI, unsigned OpNo,
SmallVectorImpl<MCFixup> &Fixups) const {
const MCOperand &MO = MI.getOperand(OpNo);
assert((MI.getOpcode() == PPC::MTCRF ||
MI.getOpcode() == PPC::MFOCRF ||
MI.getOpcode() == PPC::MTCRF8) &&
(MO.getReg() >= PPC::CR0 && MO.getReg() <= PPC::CR7));
return 0x80 >> CTX.getRegisterInfo()->getEncodingValue(MO.getReg());
}
unsigned PPCMCCodeEmitter::
getMachineOpValue(const MCInst &MI, const MCOperand &MO,
SmallVectorImpl<MCFixup> &Fixups) const {
if (MO.isReg()) {
// MTCRF/MFOCRF should go through get_crbitm_encoding for the CR operand.
// The GPR operand should come through here though.
assert((MI.getOpcode() != PPC::MTCRF && MI.getOpcode() != PPC::MFOCRF) ||
MO.getReg() < PPC::CR0 || MO.getReg() > PPC::CR7);
return CTX.getRegisterInfo()->getEncodingValue(MO.getReg());
}
assert(MO.isImm() &&
"Relocation required in an instruction that we cannot encode!");
return MO.getImm();
}
#include "PPCGenMCCodeEmitter.inc"