You've already forked linux-packaging-mono
Imported Upstream version 5.18.0.167
Former-commit-id: 289509151e0fee68a1b591a20c9f109c3c789d3a
This commit is contained in:
parent
e19d552987
commit
b084638f15
@ -1,82 +0,0 @@
|
||||
//===-- AMDGPUAsmUtils.cpp - AsmParser/InstPrinter common -----------------===//
|
||||
//
|
||||
// The LLVM Compiler Infrastructure
|
||||
//
|
||||
// This file is distributed under the University of Illinois Open Source
|
||||
// License. See LICENSE.TXT for details.
|
||||
//
|
||||
//===----------------------------------------------------------------------===//
|
||||
#include "AMDGPUAsmUtils.h"
|
||||
|
||||
namespace llvm {
|
||||
namespace AMDGPU {
|
||||
namespace SendMsg {
|
||||
|
||||
// This must be in sync with llvm::AMDGPU::SendMsg::Id enum members, see SIDefines.h.
|
||||
const char* const IdSymbolic[] = {
|
||||
nullptr,
|
||||
"MSG_INTERRUPT",
|
||||
"MSG_GS",
|
||||
"MSG_GS_DONE",
|
||||
nullptr,
|
||||
nullptr,
|
||||
nullptr,
|
||||
nullptr,
|
||||
nullptr,
|
||||
nullptr,
|
||||
nullptr,
|
||||
nullptr,
|
||||
nullptr,
|
||||
nullptr,
|
||||
nullptr,
|
||||
"MSG_SYSMSG"
|
||||
};
|
||||
|
||||
// These two must be in sync with llvm::AMDGPU::SendMsg::Op enum members, see SIDefines.h.
|
||||
const char* const OpSysSymbolic[] = {
|
||||
nullptr,
|
||||
"SYSMSG_OP_ECC_ERR_INTERRUPT",
|
||||
"SYSMSG_OP_REG_RD",
|
||||
"SYSMSG_OP_HOST_TRAP_ACK",
|
||||
"SYSMSG_OP_TTRACE_PC"
|
||||
};
|
||||
|
||||
const char* const OpGsSymbolic[] = {
|
||||
"GS_OP_NOP",
|
||||
"GS_OP_CUT",
|
||||
"GS_OP_EMIT",
|
||||
"GS_OP_EMIT_CUT"
|
||||
};
|
||||
|
||||
} // namespace SendMsg
|
||||
|
||||
namespace Hwreg {
|
||||
|
||||
// This must be in sync with llvm::AMDGPU::Hwreg::ID_SYMBOLIC_FIRST_/LAST_, see SIDefines.h.
|
||||
const char* const IdSymbolic[] = {
|
||||
nullptr,
|
||||
"HW_REG_MODE",
|
||||
"HW_REG_STATUS",
|
||||
"HW_REG_TRAPSTS",
|
||||
"HW_REG_HW_ID",
|
||||
"HW_REG_GPR_ALLOC",
|
||||
"HW_REG_LDS_ALLOC",
|
||||
"HW_REG_IB_STS"
|
||||
};
|
||||
|
||||
} // namespace Hwreg
|
||||
|
||||
namespace Swizzle {
|
||||
|
||||
// This must be in sync with llvm::AMDGPU::Swizzle::Id enum members, see SIDefines.h.
|
||||
const char* const IdSymbolic[] = {
|
||||
"QUAD_PERM",
|
||||
"BITMASK_PERM",
|
||||
"SWAP",
|
||||
"REVERSE",
|
||||
"BROADCAST",
|
||||
};
|
||||
|
||||
} // namespace Swizzle
|
||||
} // namespace AMDGPU
|
||||
} // namespace llvm
|
@ -1,37 +0,0 @@
|
||||
//===-- AMDGPUAsmUtils.h - AsmParser/InstPrinter common ---------*- C++ -*-===//
|
||||
//
|
||||
// The LLVM Compiler Infrastructure
|
||||
//
|
||||
// This file is distributed under the University of Illinois Open Source
|
||||
// License. See LICENSE.TXT for details.
|
||||
//
|
||||
//===----------------------------------------------------------------------===//
|
||||
|
||||
#ifndef LLVM_LIB_TARGET_AMDGPU_UTILS_AMDGPUASMUTILS_H
|
||||
#define LLVM_LIB_TARGET_AMDGPU_UTILS_AMDGPUASMUTILS_H
|
||||
|
||||
namespace llvm {
|
||||
namespace AMDGPU {
|
||||
namespace SendMsg { // Symbolic names for the sendmsg(...) syntax.
|
||||
|
||||
extern const char* const IdSymbolic[];
|
||||
extern const char* const OpSysSymbolic[];
|
||||
extern const char* const OpGsSymbolic[];
|
||||
|
||||
} // namespace SendMsg
|
||||
|
||||
namespace Hwreg { // Symbolic names for the hwreg(...) syntax.
|
||||
|
||||
extern const char* const IdSymbolic[];
|
||||
|
||||
} // namespace Hwreg
|
||||
|
||||
namespace Swizzle { // Symbolic names for the swizzle(...) syntax.
|
||||
|
||||
extern const char* const IdSymbolic[];
|
||||
|
||||
} // namespace Swizzle
|
||||
} // namespace AMDGPU
|
||||
} // namespace llvm
|
||||
|
||||
#endif
|
File diff suppressed because it is too large
Load Diff
@ -1,379 +0,0 @@
|
||||
//===- AMDGPUBaseInfo.h - Top level definitions for AMDGPU ------*- C++ -*-===//
|
||||
//
|
||||
// The LLVM Compiler Infrastructure
|
||||
//
|
||||
// This file is distributed under the University of Illinois Open Source
|
||||
// License. See LICENSE.TXT for details.
|
||||
//
|
||||
//===----------------------------------------------------------------------===//
|
||||
|
||||
#ifndef LLVM_LIB_TARGET_AMDGPU_UTILS_AMDGPUBASEINFO_H
|
||||
#define LLVM_LIB_TARGET_AMDGPU_UTILS_AMDGPUBASEINFO_H
|
||||
|
||||
#include "AMDGPU.h"
|
||||
#include "AMDKernelCodeT.h"
|
||||
#include "SIDefines.h"
|
||||
#include "llvm/ADT/StringRef.h"
|
||||
#include "llvm/IR/CallingConv.h"
|
||||
#include "llvm/MC/MCInstrDesc.h"
|
||||
#include "llvm/Support/Compiler.h"
|
||||
#include "llvm/Support/ErrorHandling.h"
|
||||
#include <cstdint>
|
||||
#include <string>
|
||||
#include <utility>
|
||||
|
||||
namespace llvm {
|
||||
|
||||
class Argument;
|
||||
class FeatureBitset;
|
||||
class Function;
|
||||
class GlobalValue;
|
||||
class MachineMemOperand;
|
||||
class MCContext;
|
||||
class MCRegisterClass;
|
||||
class MCRegisterInfo;
|
||||
class MCSection;
|
||||
class MCSubtargetInfo;
|
||||
class Triple;
|
||||
|
||||
namespace AMDGPU {
|
||||
namespace IsaInfo {
|
||||
|
||||
enum {
|
||||
// The closed Vulkan driver sets 96, which limits the wave count to 8 but
|
||||
// doesn't spill SGPRs as much as when 80 is set.
|
||||
FIXED_NUM_SGPRS_FOR_INIT_BUG = 96
|
||||
};
|
||||
|
||||
/// \brief Instruction set architecture version.
|
||||
struct IsaVersion {
|
||||
unsigned Major;
|
||||
unsigned Minor;
|
||||
unsigned Stepping;
|
||||
};
|
||||
|
||||
/// \returns Isa version for given subtarget \p Features.
|
||||
IsaVersion getIsaVersion(const FeatureBitset &Features);
|
||||
|
||||
/// \brief Streams isa version string for given subtarget \p STI into \p Stream.
|
||||
void streamIsaVersion(const MCSubtargetInfo *STI, raw_ostream &Stream);
|
||||
|
||||
/// \returns True if given subtarget \p Features support code object version 3,
|
||||
/// false otherwise.
|
||||
bool hasCodeObjectV3(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Wavefront size for given subtarget \p Features.
|
||||
unsigned getWavefrontSize(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Local memory size in bytes for given subtarget \p Features.
|
||||
unsigned getLocalMemorySize(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Number of execution units per compute unit for given subtarget \p
|
||||
/// Features.
|
||||
unsigned getEUsPerCU(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Maximum number of work groups per compute unit for given subtarget
|
||||
/// \p Features and limited by given \p FlatWorkGroupSize.
|
||||
unsigned getMaxWorkGroupsPerCU(const FeatureBitset &Features,
|
||||
unsigned FlatWorkGroupSize);
|
||||
|
||||
/// \returns Maximum number of waves per compute unit for given subtarget \p
|
||||
/// Features without any kind of limitation.
|
||||
unsigned getMaxWavesPerCU(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Maximum number of waves per compute unit for given subtarget \p
|
||||
/// Features and limited by given \p FlatWorkGroupSize.
|
||||
unsigned getMaxWavesPerCU(const FeatureBitset &Features,
|
||||
unsigned FlatWorkGroupSize);
|
||||
|
||||
/// \returns Minimum number of waves per execution unit for given subtarget \p
|
||||
/// Features.
|
||||
unsigned getMinWavesPerEU(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Maximum number of waves per execution unit for given subtarget \p
|
||||
/// Features without any kind of limitation.
|
||||
unsigned getMaxWavesPerEU(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Maximum number of waves per execution unit for given subtarget \p
|
||||
/// Features and limited by given \p FlatWorkGroupSize.
|
||||
unsigned getMaxWavesPerEU(const FeatureBitset &Features,
|
||||
unsigned FlatWorkGroupSize);
|
||||
|
||||
/// \returns Minimum flat work group size for given subtarget \p Features.
|
||||
unsigned getMinFlatWorkGroupSize(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Maximum flat work group size for given subtarget \p Features.
|
||||
unsigned getMaxFlatWorkGroupSize(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Number of waves per work group for given subtarget \p Features and
|
||||
/// limited by given \p FlatWorkGroupSize.
|
||||
unsigned getWavesPerWorkGroup(const FeatureBitset &Features,
|
||||
unsigned FlatWorkGroupSize);
|
||||
|
||||
/// \returns SGPR allocation granularity for given subtarget \p Features.
|
||||
unsigned getSGPRAllocGranule(const FeatureBitset &Features);
|
||||
|
||||
/// \returns SGPR encoding granularity for given subtarget \p Features.
|
||||
unsigned getSGPREncodingGranule(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Total number of SGPRs for given subtarget \p Features.
|
||||
unsigned getTotalNumSGPRs(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Addressable number of SGPRs for given subtarget \p Features.
|
||||
unsigned getAddressableNumSGPRs(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Minimum number of SGPRs that meets the given number of waves per
|
||||
/// execution unit requirement for given subtarget \p Features.
|
||||
unsigned getMinNumSGPRs(const FeatureBitset &Features, unsigned WavesPerEU);
|
||||
|
||||
/// \returns Maximum number of SGPRs that meets the given number of waves per
|
||||
/// execution unit requirement for given subtarget \p Features.
|
||||
unsigned getMaxNumSGPRs(const FeatureBitset &Features, unsigned WavesPerEU,
|
||||
bool Addressable);
|
||||
|
||||
/// \returns VGPR allocation granularity for given subtarget \p Features.
|
||||
unsigned getVGPRAllocGranule(const FeatureBitset &Features);
|
||||
|
||||
/// \returns VGPR encoding granularity for given subtarget \p Features.
|
||||
unsigned getVGPREncodingGranule(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Total number of VGPRs for given subtarget \p Features.
|
||||
unsigned getTotalNumVGPRs(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Addressable number of VGPRs for given subtarget \p Features.
|
||||
unsigned getAddressableNumVGPRs(const FeatureBitset &Features);
|
||||
|
||||
/// \returns Minimum number of VGPRs that meets given number of waves per
|
||||
/// execution unit requirement for given subtarget \p Features.
|
||||
unsigned getMinNumVGPRs(const FeatureBitset &Features, unsigned WavesPerEU);
|
||||
|
||||
/// \returns Maximum number of VGPRs that meets given number of waves per
|
||||
/// execution unit requirement for given subtarget \p Features.
|
||||
unsigned getMaxNumVGPRs(const FeatureBitset &Features, unsigned WavesPerEU);
|
||||
|
||||
} // end namespace IsaInfo
|
||||
|
||||
LLVM_READONLY
|
||||
int16_t getNamedOperandIdx(uint16_t Opcode, uint16_t NamedIdx);
|
||||
|
||||
LLVM_READONLY
|
||||
int getMaskedMIMGOp(const MCInstrInfo &MII,
|
||||
unsigned Opc, unsigned NewChannels);
|
||||
LLVM_READONLY
|
||||
int getMCOpcode(uint16_t Opcode, unsigned Gen);
|
||||
|
||||
void initDefaultAMDKernelCodeT(amd_kernel_code_t &Header,
|
||||
const FeatureBitset &Features);
|
||||
|
||||
bool isGroupSegment(const GlobalValue *GV);
|
||||
bool isGlobalSegment(const GlobalValue *GV);
|
||||
bool isReadOnlySegment(const GlobalValue *GV);
|
||||
|
||||
/// \returns True if constants should be emitted to .text section for given
|
||||
/// target triple \p TT, false otherwise.
|
||||
bool shouldEmitConstantsToTextSection(const Triple &TT);
|
||||
|
||||
/// \returns Integer value requested using \p F's \p Name attribute.
|
||||
///
|
||||
/// \returns \p Default if attribute is not present.
|
||||
///
|
||||
/// \returns \p Default and emits error if requested value cannot be converted
|
||||
/// to integer.
|
||||
int getIntegerAttribute(const Function &F, StringRef Name, int Default);
|
||||
|
||||
/// \returns A pair of integer values requested using \p F's \p Name attribute
|
||||
/// in "first[,second]" format ("second" is optional unless \p OnlyFirstRequired
|
||||
/// is false).
|
||||
///
|
||||
/// \returns \p Default if attribute is not present.
|
||||
///
|
||||
/// \returns \p Default and emits error if one of the requested values cannot be
|
||||
/// converted to integer, or \p OnlyFirstRequired is false and "second" value is
|
||||
/// not present.
|
||||
std::pair<int, int> getIntegerPairAttribute(const Function &F,
|
||||
StringRef Name,
|
||||
std::pair<int, int> Default,
|
||||
bool OnlyFirstRequired = false);
|
||||
|
||||
/// \returns Vmcnt bit mask for given isa \p Version.
|
||||
unsigned getVmcntBitMask(const IsaInfo::IsaVersion &Version);
|
||||
|
||||
/// \returns Expcnt bit mask for given isa \p Version.
|
||||
unsigned getExpcntBitMask(const IsaInfo::IsaVersion &Version);
|
||||
|
||||
/// \returns Lgkmcnt bit mask for given isa \p Version.
|
||||
unsigned getLgkmcntBitMask(const IsaInfo::IsaVersion &Version);
|
||||
|
||||
/// \returns Waitcnt bit mask for given isa \p Version.
|
||||
unsigned getWaitcntBitMask(const IsaInfo::IsaVersion &Version);
|
||||
|
||||
/// \returns Decoded Vmcnt from given \p Waitcnt for given isa \p Version.
|
||||
unsigned decodeVmcnt(const IsaInfo::IsaVersion &Version, unsigned Waitcnt);
|
||||
|
||||
/// \returns Decoded Expcnt from given \p Waitcnt for given isa \p Version.
|
||||
unsigned decodeExpcnt(const IsaInfo::IsaVersion &Version, unsigned Waitcnt);
|
||||
|
||||
/// \returns Decoded Lgkmcnt from given \p Waitcnt for given isa \p Version.
|
||||
unsigned decodeLgkmcnt(const IsaInfo::IsaVersion &Version, unsigned Waitcnt);
|
||||
|
||||
/// \brief Decodes Vmcnt, Expcnt and Lgkmcnt from given \p Waitcnt for given isa
|
||||
/// \p Version, and writes decoded values into \p Vmcnt, \p Expcnt and
|
||||
/// \p Lgkmcnt respectively.
|
||||
///
|
||||
/// \details \p Vmcnt, \p Expcnt and \p Lgkmcnt are decoded as follows:
|
||||
/// \p Vmcnt = \p Waitcnt[3:0] (pre-gfx9 only)
|
||||
/// \p Vmcnt = \p Waitcnt[3:0] | \p Waitcnt[15:14] (gfx9+ only)
|
||||
/// \p Expcnt = \p Waitcnt[6:4]
|
||||
/// \p Lgkmcnt = \p Waitcnt[11:8]
|
||||
void decodeWaitcnt(const IsaInfo::IsaVersion &Version, unsigned Waitcnt,
|
||||
unsigned &Vmcnt, unsigned &Expcnt, unsigned &Lgkmcnt);
|
||||
|
||||
/// \returns \p Waitcnt with encoded \p Vmcnt for given isa \p Version.
|
||||
unsigned encodeVmcnt(const IsaInfo::IsaVersion &Version, unsigned Waitcnt,
|
||||
unsigned Vmcnt);
|
||||
|
||||
/// \returns \p Waitcnt with encoded \p Expcnt for given isa \p Version.
|
||||
unsigned encodeExpcnt(const IsaInfo::IsaVersion &Version, unsigned Waitcnt,
|
||||
unsigned Expcnt);
|
||||
|
||||
/// \returns \p Waitcnt with encoded \p Lgkmcnt for given isa \p Version.
|
||||
unsigned encodeLgkmcnt(const IsaInfo::IsaVersion &Version, unsigned Waitcnt,
|
||||
unsigned Lgkmcnt);
|
||||
|
||||
/// \brief Encodes \p Vmcnt, \p Expcnt and \p Lgkmcnt into Waitcnt for given isa
|
||||
/// \p Version.
|
||||
///
|
||||
/// \details \p Vmcnt, \p Expcnt and \p Lgkmcnt are encoded as follows:
|
||||
/// Waitcnt[3:0] = \p Vmcnt (pre-gfx9 only)
|
||||
/// Waitcnt[3:0] = \p Vmcnt[3:0] (gfx9+ only)
|
||||
/// Waitcnt[6:4] = \p Expcnt
|
||||
/// Waitcnt[11:8] = \p Lgkmcnt
|
||||
/// Waitcnt[15:14] = \p Vmcnt[5:4] (gfx9+ only)
|
||||
///
|
||||
/// \returns Waitcnt with encoded \p Vmcnt, \p Expcnt and \p Lgkmcnt for given
|
||||
/// isa \p Version.
|
||||
unsigned encodeWaitcnt(const IsaInfo::IsaVersion &Version,
|
||||
unsigned Vmcnt, unsigned Expcnt, unsigned Lgkmcnt);
|
||||
|
||||
unsigned getInitialPSInputAddr(const Function &F);
|
||||
|
||||
LLVM_READNONE
|
||||
bool isShader(CallingConv::ID CC);
|
||||
|
||||
LLVM_READNONE
|
||||
bool isCompute(CallingConv::ID CC);
|
||||
|
||||
LLVM_READNONE
|
||||
bool isEntryFunctionCC(CallingConv::ID CC);
|
||||
|
||||
// FIXME: Remove this when calling conventions cleaned up
|
||||
LLVM_READNONE
|
||||
inline bool isKernel(CallingConv::ID CC) {
|
||||
switch (CC) {
|
||||
case CallingConv::AMDGPU_KERNEL:
|
||||
case CallingConv::SPIR_KERNEL:
|
||||
return true;
|
||||
default:
|
||||
return false;
|
||||
}
|
||||
}
|
||||
|
||||
bool isSI(const MCSubtargetInfo &STI);
|
||||
bool isCI(const MCSubtargetInfo &STI);
|
||||
bool isVI(const MCSubtargetInfo &STI);
|
||||
bool isGFX9(const MCSubtargetInfo &STI);
|
||||
|
||||
/// \brief Is Reg - scalar register
|
||||
bool isSGPR(unsigned Reg, const MCRegisterInfo* TRI);
|
||||
|
||||
/// \brief Is there any intersection between registers
|
||||
bool isRegIntersect(unsigned Reg0, unsigned Reg1, const MCRegisterInfo* TRI);
|
||||
|
||||
/// If \p Reg is a pseudo reg, return the correct hardware register given
|
||||
/// \p STI otherwise return \p Reg.
|
||||
unsigned getMCReg(unsigned Reg, const MCSubtargetInfo &STI);
|
||||
|
||||
/// \brief Convert hardware register \p Reg to a pseudo register
|
||||
LLVM_READNONE
|
||||
unsigned mc2PseudoReg(unsigned Reg);
|
||||
|
||||
/// \brief Can this operand also contain immediate values?
|
||||
bool isSISrcOperand(const MCInstrDesc &Desc, unsigned OpNo);
|
||||
|
||||
/// \brief Is this floating-point operand?
|
||||
bool isSISrcFPOperand(const MCInstrDesc &Desc, unsigned OpNo);
|
||||
|
||||
/// \brief Does this opearnd support only inlinable literals?
|
||||
bool isSISrcInlinableOperand(const MCInstrDesc &Desc, unsigned OpNo);
|
||||
|
||||
/// \brief Get the size in bits of a register from the register class \p RC.
|
||||
unsigned getRegBitWidth(unsigned RCID);
|
||||
|
||||
/// \brief Get the size in bits of a register from the register class \p RC.
|
||||
unsigned getRegBitWidth(const MCRegisterClass &RC);
|
||||
|
||||
/// \brief Get size of register operand
|
||||
unsigned getRegOperandSize(const MCRegisterInfo *MRI, const MCInstrDesc &Desc,
|
||||
unsigned OpNo);
|
||||
|
||||
LLVM_READNONE
|
||||
inline unsigned getOperandSize(const MCOperandInfo &OpInfo) {
|
||||
switch (OpInfo.OperandType) {
|
||||
case AMDGPU::OPERAND_REG_IMM_INT32:
|
||||
case AMDGPU::OPERAND_REG_IMM_FP32:
|
||||
case AMDGPU::OPERAND_REG_INLINE_C_INT32:
|
||||
case AMDGPU::OPERAND_REG_INLINE_C_FP32:
|
||||
return 4;
|
||||
|
||||
case AMDGPU::OPERAND_REG_IMM_INT64:
|
||||
case AMDGPU::OPERAND_REG_IMM_FP64:
|
||||
case AMDGPU::OPERAND_REG_INLINE_C_INT64:
|
||||
case AMDGPU::OPERAND_REG_INLINE_C_FP64:
|
||||
return 8;
|
||||
|
||||
case AMDGPU::OPERAND_REG_IMM_INT16:
|
||||
case AMDGPU::OPERAND_REG_IMM_FP16:
|
||||
case AMDGPU::OPERAND_REG_INLINE_C_INT16:
|
||||
case AMDGPU::OPERAND_REG_INLINE_C_FP16:
|
||||
case AMDGPU::OPERAND_REG_INLINE_C_V2INT16:
|
||||
case AMDGPU::OPERAND_REG_INLINE_C_V2FP16:
|
||||
return 2;
|
||||
|
||||
default:
|
||||
llvm_unreachable("unhandled operand type");
|
||||
}
|
||||
}
|
||||
|
||||
LLVM_READNONE
|
||||
inline unsigned getOperandSize(const MCInstrDesc &Desc, unsigned OpNo) {
|
||||
return getOperandSize(Desc.OpInfo[OpNo]);
|
||||
}
|
||||
|
||||
/// \brief Is this literal inlinable
|
||||
LLVM_READNONE
|
||||
bool isInlinableLiteral64(int64_t Literal, bool HasInv2Pi);
|
||||
|
||||
LLVM_READNONE
|
||||
bool isInlinableLiteral32(int32_t Literal, bool HasInv2Pi);
|
||||
|
||||
LLVM_READNONE
|
||||
bool isInlinableLiteral16(int16_t Literal, bool HasInv2Pi);
|
||||
|
||||
LLVM_READNONE
|
||||
bool isInlinableLiteralV216(int32_t Literal, bool HasInv2Pi);
|
||||
|
||||
bool isArgPassedInSGPR(const Argument *Arg);
|
||||
|
||||
/// \returns The encoding that will be used for \p ByteOffset in the SMRD
|
||||
/// offset field.
|
||||
int64_t getSMRDEncodedOffset(const MCSubtargetInfo &ST, int64_t ByteOffset);
|
||||
|
||||
/// \returns true if this offset is small enough to fit in the SMRD
|
||||
/// offset field. \p ByteOffset should be the offset in bytes and
|
||||
/// not the encoded offset.
|
||||
bool isLegalSMRDImmOffset(const MCSubtargetInfo &ST, int64_t ByteOffset);
|
||||
|
||||
} // end namespace AMDGPU
|
||||
} // end namespace llvm
|
||||
|
||||
#endif // LLVM_LIB_TARGET_AMDGPU_UTILS_AMDGPUBASEINFO_H
|
@ -1,152 +0,0 @@
|
||||
//===--------------------- AMDKernelCodeTInfo.h ---------------------------===//
|
||||
//
|
||||
// The LLVM Compiler Infrastructure
|
||||
//
|
||||
// This file is distributed under the University of Illinois Open Source
|
||||
// License. See LICENSE.TXT for details.
|
||||
//
|
||||
//===----------------------------------------------------------------------===//
|
||||
//
|
||||
//===----------------------------------------------------------------------===//
|
||||
//
|
||||
/// \file - specifies tables for amd_kernel_code_t structure parsing/printing
|
||||
//
|
||||
//===----------------------------------------------------------------------===//
|
||||
|
||||
#define QNAME(name) amd_kernel_code_t::name
|
||||
#define FLD_T(name) decltype(QNAME(name)), &QNAME(name)
|
||||
|
||||
#define FIELD2(sname, aname, name) \
|
||||
RECORD(sname, aname, printField<FLD_T(name)>, parseField<FLD_T(name)>)
|
||||
|
||||
#define FIELD(name) FIELD2(name, name, name)
|
||||
|
||||
|
||||
#define PRINTCODEPROP(name) \
|
||||
printBitField<FLD_T(code_properties),\
|
||||
AMD_CODE_PROPERTY_##name##_SHIFT,\
|
||||
AMD_CODE_PROPERTY_##name##_WIDTH>
|
||||
|
||||
#define PARSECODEPROP(name) \
|
||||
parseBitField<FLD_T(code_properties),\
|
||||
AMD_CODE_PROPERTY_##name##_SHIFT,\
|
||||
AMD_CODE_PROPERTY_##name##_WIDTH>
|
||||
|
||||
#define CODEPROP(name, shift) \
|
||||
RECORD(name, name, PRINTCODEPROP(shift), PARSECODEPROP(shift))
|
||||
|
||||
// have to define these lambdas because of Set/GetMacro
|
||||
#define PRINTCOMP(GetMacro, Shift) \
|
||||
[](StringRef Name, const amd_kernel_code_t &C, raw_ostream &OS) { \
|
||||
printName(OS, Name) << \
|
||||
(int)GetMacro(C.compute_pgm_resource_registers >> Shift); \
|
||||
}
|
||||
#define PARSECOMP(SetMacro, Shift) \
|
||||
[](amd_kernel_code_t &C, MCAsmParser &MCParser, raw_ostream &Err) { \
|
||||
int64_t Value = 0; \
|
||||
if (!expectAbsExpression(MCParser, Value, Err)) \
|
||||
return false; \
|
||||
C.compute_pgm_resource_registers |= SetMacro(Value) << Shift; \
|
||||
return true; \
|
||||
}
|
||||
|
||||
#define COMPPGM(name, aname, GetMacro, SetMacro, Shift) \
|
||||
RECORD(name, aname, PRINTCOMP(GetMacro, Shift), PARSECOMP(SetMacro, Shift))
|
||||
|
||||
#define COMPPGM1(name, aname, AccMacro) \
|
||||
COMPPGM(name, aname, G_00B848_##AccMacro, S_00B848_##AccMacro, 0)
|
||||
|
||||
#define COMPPGM2(name, aname, AccMacro) \
|
||||
COMPPGM(name, aname, G_00B84C_##AccMacro, S_00B84C_##AccMacro, 32)
|
||||
|
||||
///////////////////////////////////////////////////////////////////////////////
|
||||
// Begin of the table
|
||||
// Define RECORD(name, print, parse) in your code to get field definitions
|
||||
// and include this file
|
||||
|
||||
FIELD2(amd_code_version_major, kernel_code_version_major, amd_kernel_code_version_major),
|
||||
FIELD2(amd_code_version_minor, kernel_code_version_minor, amd_kernel_code_version_minor),
|
||||
FIELD2(amd_machine_kind, machine_kind, amd_machine_kind),
|
||||
FIELD2(amd_machine_version_major, machine_version_major, amd_machine_version_major),
|
||||
FIELD2(amd_machine_version_minor, machine_version_minor, amd_machine_version_minor),
|
||||
FIELD2(amd_machine_version_stepping, machine_version_stepping, amd_machine_version_stepping),
|
||||
|
||||
FIELD(kernel_code_entry_byte_offset),
|
||||
FIELD(kernel_code_prefetch_byte_size),
|
||||
FIELD(max_scratch_backing_memory_byte_size),
|
||||
|
||||
COMPPGM1(granulated_workitem_vgpr_count, compute_pgm_rsrc1_vgprs, VGPRS),
|
||||
COMPPGM1(granulated_wavefront_sgpr_count, compute_pgm_rsrc1_sgprs, SGPRS),
|
||||
COMPPGM1(priority, compute_pgm_rsrc1_priority, PRIORITY),
|
||||
COMPPGM1(float_mode, compute_pgm_rsrc1_float_mode, FLOAT_MODE), // TODO: split float_mode
|
||||
COMPPGM1(priv, compute_pgm_rsrc1_priv, PRIV),
|
||||
COMPPGM1(enable_dx10_clamp, compute_pgm_rsrc1_dx10_clamp, DX10_CLAMP),
|
||||
COMPPGM1(debug_mode, compute_pgm_rsrc1_debug_mode, DEBUG_MODE),
|
||||
COMPPGM1(enable_ieee_mode, compute_pgm_rsrc1_ieee_mode, IEEE_MODE),
|
||||
// TODO: bulky
|
||||
// TODO: cdbg_user
|
||||
COMPPGM2(enable_sgpr_private_segment_wave_byte_offset, compute_pgm_rsrc2_scratch_en, SCRATCH_EN),
|
||||
COMPPGM2(user_sgpr_count, compute_pgm_rsrc2_user_sgpr, USER_SGPR),
|
||||
COMPPGM2(enable_trap_handler, compute_pgm_rsrc2_trap_handler, TRAP_HANDLER),
|
||||
COMPPGM2(enable_sgpr_workgroup_id_x, compute_pgm_rsrc2_tgid_x_en, TGID_X_EN),
|
||||
COMPPGM2(enable_sgpr_workgroup_id_y, compute_pgm_rsrc2_tgid_y_en, TGID_Y_EN),
|
||||
COMPPGM2(enable_sgpr_workgroup_id_z, compute_pgm_rsrc2_tgid_z_en, TGID_Z_EN),
|
||||
COMPPGM2(enable_sgpr_workgroup_info, compute_pgm_rsrc2_tg_size_en, TG_SIZE_EN),
|
||||
COMPPGM2(enable_vgpr_workitem_id, compute_pgm_rsrc2_tidig_comp_cnt, TIDIG_COMP_CNT),
|
||||
COMPPGM2(enable_exception_msb, compute_pgm_rsrc2_excp_en_msb, EXCP_EN_MSB), // TODO: split enable_exception_msb
|
||||
COMPPGM2(granulated_lds_size, compute_pgm_rsrc2_lds_size, LDS_SIZE),
|
||||
COMPPGM2(enable_exception, compute_pgm_rsrc2_excp_en, EXCP_EN), // TODO: split enable_exception
|
||||
|
||||
CODEPROP(enable_sgpr_private_segment_buffer, ENABLE_SGPR_PRIVATE_SEGMENT_BUFFER),
|
||||
CODEPROP(enable_sgpr_dispatch_ptr, ENABLE_SGPR_DISPATCH_PTR),
|
||||
CODEPROP(enable_sgpr_queue_ptr, ENABLE_SGPR_QUEUE_PTR),
|
||||
CODEPROP(enable_sgpr_kernarg_segment_ptr, ENABLE_SGPR_KERNARG_SEGMENT_PTR),
|
||||
CODEPROP(enable_sgpr_dispatch_id, ENABLE_SGPR_DISPATCH_ID),
|
||||
CODEPROP(enable_sgpr_flat_scratch_init, ENABLE_SGPR_FLAT_SCRATCH_INIT),
|
||||
CODEPROP(enable_sgpr_private_segment_size, ENABLE_SGPR_PRIVATE_SEGMENT_SIZE),
|
||||
CODEPROP(enable_sgpr_grid_workgroup_count_x, ENABLE_SGPR_GRID_WORKGROUP_COUNT_X),
|
||||
CODEPROP(enable_sgpr_grid_workgroup_count_y, ENABLE_SGPR_GRID_WORKGROUP_COUNT_Y),
|
||||
CODEPROP(enable_sgpr_grid_workgroup_count_z, ENABLE_SGPR_GRID_WORKGROUP_COUNT_Z),
|
||||
CODEPROP(enable_ordered_append_gds, ENABLE_ORDERED_APPEND_GDS),
|
||||
CODEPROP(private_element_size, PRIVATE_ELEMENT_SIZE),
|
||||
CODEPROP(is_ptr64, IS_PTR64),
|
||||
CODEPROP(is_dynamic_callstack, IS_DYNAMIC_CALLSTACK),
|
||||
CODEPROP(is_debug_enabled, IS_DEBUG_SUPPORTED),
|
||||
CODEPROP(is_xnack_enabled, IS_XNACK_SUPPORTED),
|
||||
|
||||
FIELD(workitem_private_segment_byte_size),
|
||||
FIELD(workgroup_group_segment_byte_size),
|
||||
FIELD(gds_segment_byte_size),
|
||||
FIELD(kernarg_segment_byte_size),
|
||||
FIELD(workgroup_fbarrier_count),
|
||||
FIELD(wavefront_sgpr_count),
|
||||
FIELD(workitem_vgpr_count),
|
||||
FIELD(reserved_vgpr_first),
|
||||
FIELD(reserved_vgpr_count),
|
||||
FIELD(reserved_sgpr_first),
|
||||
FIELD(reserved_sgpr_count),
|
||||
FIELD(debug_wavefront_private_segment_offset_sgpr),
|
||||
FIELD(debug_private_segment_buffer_sgpr),
|
||||
FIELD(kernarg_segment_alignment),
|
||||
FIELD(group_segment_alignment),
|
||||
FIELD(private_segment_alignment),
|
||||
FIELD(wavefront_size),
|
||||
FIELD(call_convention),
|
||||
FIELD(runtime_loader_kernel_symbol)
|
||||
// TODO: control_directive
|
||||
|
||||
// end of the table
|
||||
///////////////////////////////////////////////////////////////////////////////
|
||||
|
||||
#undef QNAME
|
||||
#undef FLD_T
|
||||
#undef FIELD2
|
||||
#undef FIELD
|
||||
#undef PRINTCODEPROP
|
||||
#undef PARSECODEPROP
|
||||
#undef CODEPROP
|
||||
#undef PRINTCOMP
|
||||
#undef PAPSECOMP
|
||||
#undef COMPPGM
|
||||
#undef COMPPGM1
|
||||
#undef COMPPGM2
|
@ -1,180 +0,0 @@
|
||||
//===- AMDKernelCodeTUtils.cpp --------------------------------------------===//
|
||||
//
|
||||
// The LLVM Compiler Infrastructure
|
||||
//
|
||||
// This file is distributed under the University of Illinois Open Source
|
||||
// License. See LICENSE.TXT for details.
|
||||
//
|
||||
//===----------------------------------------------------------------------===//
|
||||
//
|
||||
/// \file - utility functions to parse/print amd_kernel_code_t structure
|
||||
//
|
||||
//===----------------------------------------------------------------------===//
|
||||
|
||||
#include "AMDKernelCodeTUtils.h"
|
||||
#include "SIDefines.h"
|
||||
#include "llvm/ADT/ArrayRef.h"
|
||||
#include "llvm/ADT/StringMap.h"
|
||||
#include "llvm/ADT/StringRef.h"
|
||||
#include "llvm/MC/MCParser/MCAsmLexer.h"
|
||||
#include "llvm/MC/MCParser/MCAsmParser.h"
|
||||
#include "llvm/Support/raw_ostream.h"
|
||||
#include <cassert>
|
||||
#include <cstdint>
|
||||
#include <utility>
|
||||
|
||||
using namespace llvm;
|
||||
|
||||
static ArrayRef<StringRef> get_amd_kernel_code_t_FldNames() {
|
||||
static StringRef const Table[] = {
|
||||
"", // not found placeholder
|
||||
#define RECORD(name, altName, print, parse) #name
|
||||
#include "AMDKernelCodeTInfo.h"
|
||||
#undef RECORD
|
||||
};
|
||||
return makeArrayRef(Table);
|
||||
}
|
||||
|
||||
static ArrayRef<StringRef> get_amd_kernel_code_t_FldAltNames() {
|
||||
static StringRef const Table[] = {
|
||||
"", // not found placeholder
|
||||
#define RECORD(name, altName, print, parse) #altName
|
||||
#include "AMDKernelCodeTInfo.h"
|
||||
#undef RECORD
|
||||
};
|
||||
return makeArrayRef(Table);
|
||||
}
|
||||
|
||||
static StringMap<int> createIndexMap(const ArrayRef<StringRef> &names,
|
||||
const ArrayRef<StringRef> &altNames) {
|
||||
StringMap<int> map;
|
||||
assert(names.size() == altNames.size());
|
||||
for (unsigned i = 0; i < names.size(); ++i) {
|
||||
map.insert(std::make_pair(names[i], i));
|
||||
map.insert(std::make_pair(altNames[i], i));
|
||||
}
|
||||
return map;
|
||||
}
|
||||
|
||||
static int get_amd_kernel_code_t_FieldIndex(StringRef name) {
|
||||
static const auto map = createIndexMap(get_amd_kernel_code_t_FldNames(),
|
||||
get_amd_kernel_code_t_FldAltNames());
|
||||
return map.lookup(name) - 1; // returns -1 if not found
|
||||
}
|
||||
|
||||
static StringRef get_amd_kernel_code_t_FieldName(int index) {
|
||||
return get_amd_kernel_code_t_FldNames()[index + 1];
|
||||
}
|
||||
|
||||
// Field printing
|
||||
|
||||
static raw_ostream &printName(raw_ostream &OS, StringRef Name) {
|
||||
return OS << Name << " = ";
|
||||
}
|
||||
|
||||
template <typename T, T amd_kernel_code_t::*ptr>
|
||||
static void printField(StringRef Name, const amd_kernel_code_t &C,
|
||||
raw_ostream &OS) {
|
||||
printName(OS, Name) << (int)(C.*ptr);
|
||||
}
|
||||
|
||||
template <typename T, T amd_kernel_code_t::*ptr, int shift, int width = 1>
|
||||
static void printBitField(StringRef Name, const amd_kernel_code_t &c,
|
||||
raw_ostream &OS) {
|
||||
const auto Mask = (static_cast<T>(1) << width) - 1;
|
||||
printName(OS, Name) << (int)((c.*ptr >> shift) & Mask);
|
||||
}
|
||||
|
||||
using PrintFx = void(*)(StringRef, const amd_kernel_code_t &, raw_ostream &);
|
||||
|
||||
static ArrayRef<PrintFx> getPrinterTable() {
|
||||
static const PrintFx Table[] = {
|
||||
#define RECORD(name, altName, print, parse) print
|
||||
#include "AMDKernelCodeTInfo.h"
|
||||
#undef RECORD
|
||||
};
|
||||
return makeArrayRef(Table);
|
||||
}
|
||||
|
||||
void llvm::printAmdKernelCodeField(const amd_kernel_code_t &C,
|
||||
int FldIndex,
|
||||
raw_ostream &OS) {
|
||||
auto Printer = getPrinterTable()[FldIndex];
|
||||
if (Printer)
|
||||
Printer(get_amd_kernel_code_t_FieldName(FldIndex), C, OS);
|
||||
}
|
||||
|
||||
void llvm::dumpAmdKernelCode(const amd_kernel_code_t *C,
|
||||
raw_ostream &OS,
|
||||
const char *tab) {
|
||||
const int Size = getPrinterTable().size();
|
||||
for (int i = 0; i < Size; ++i) {
|
||||
OS << tab;
|
||||
printAmdKernelCodeField(*C, i, OS);
|
||||
OS << '\n';
|
||||
}
|
||||
}
|
||||
|
||||
// Field parsing
|
||||
|
||||
static bool expectAbsExpression(MCAsmParser &MCParser, int64_t &Value, raw_ostream& Err) {
|
||||
|
||||
if (MCParser.getLexer().isNot(AsmToken::Equal)) {
|
||||
Err << "expected '='";
|
||||
return false;
|
||||
}
|
||||
MCParser.getLexer().Lex();
|
||||
|
||||
if (MCParser.parseAbsoluteExpression(Value)) {
|
||||
Err << "integer absolute expression expected";
|
||||
return false;
|
||||
}
|
||||
return true;
|
||||
}
|
||||
|
||||
template <typename T, T amd_kernel_code_t::*ptr>
|
||||
static bool parseField(amd_kernel_code_t &C, MCAsmParser &MCParser,
|
||||
raw_ostream &Err) {
|
||||
int64_t Value = 0;
|
||||
if (!expectAbsExpression(MCParser, Value, Err))
|
||||
return false;
|
||||
C.*ptr = (T)Value;
|
||||
return true;
|
||||
}
|
||||
|
||||
template <typename T, T amd_kernel_code_t::*ptr, int shift, int width = 1>
|
||||
static bool parseBitField(amd_kernel_code_t &C, MCAsmParser &MCParser,
|
||||
raw_ostream &Err) {
|
||||
int64_t Value = 0;
|
||||
if (!expectAbsExpression(MCParser, Value, Err))
|
||||
return false;
|
||||
const uint64_t Mask = ((UINT64_C(1) << width) - 1) << shift;
|
||||
C.*ptr &= (T)~Mask;
|
||||
C.*ptr |= (T)((Value << shift) & Mask);
|
||||
return true;
|
||||
}
|
||||
|
||||
using ParseFx = bool(*)(amd_kernel_code_t &, MCAsmParser &MCParser,
|
||||
raw_ostream &Err);
|
||||
|
||||
static ArrayRef<ParseFx> getParserTable() {
|
||||
static const ParseFx Table[] = {
|
||||
#define RECORD(name, altName, print, parse) parse
|
||||
#include "AMDKernelCodeTInfo.h"
|
||||
#undef RECORD
|
||||
};
|
||||
return makeArrayRef(Table);
|
||||
}
|
||||
|
||||
bool llvm::parseAmdKernelCodeField(StringRef ID,
|
||||
MCAsmParser &MCParser,
|
||||
amd_kernel_code_t &C,
|
||||
raw_ostream &Err) {
|
||||
const int Idx = get_amd_kernel_code_t_FieldIndex(ID);
|
||||
if (Idx < 0) {
|
||||
Err << "unexpected amd_kernel_code_t field name " << ID;
|
||||
return false;
|
||||
}
|
||||
auto Parser = getParserTable()[Idx];
|
||||
return Parser ? Parser(C, MCParser, Err) : false;
|
||||
}
|
@ -1,36 +0,0 @@
|
||||
//===- AMDGPUKernelCodeTUtils.h - helpers for amd_kernel_code_t -*- C++ -*-===//
|
||||
//
|
||||
// The LLVM Compiler Infrastructure
|
||||
//
|
||||
// This file is distributed under the University of Illinois Open Source
|
||||
// License. See LICENSE.TXT for details.
|
||||
//
|
||||
//===----------------------------------------------------------------------===//
|
||||
//
|
||||
/// \file AMDKernelCodeTUtils.h
|
||||
//
|
||||
//===----------------------------------------------------------------------===//
|
||||
|
||||
#ifndef LLVM_LIB_TARGET_AMDGPU_UTILS_AMDKERNELCODETUTILS_H
|
||||
#define LLVM_LIB_TARGET_AMDGPU_UTILS_AMDKERNELCODETUTILS_H
|
||||
|
||||
#include "AMDKernelCodeT.h"
|
||||
|
||||
namespace llvm {
|
||||
|
||||
class MCAsmParser;
|
||||
class raw_ostream;
|
||||
class StringRef;
|
||||
|
||||
void printAmdKernelCodeField(const amd_kernel_code_t &C, int FldIndex,
|
||||
raw_ostream &OS);
|
||||
|
||||
void dumpAmdKernelCode(const amd_kernel_code_t *C, raw_ostream &OS,
|
||||
const char *tab);
|
||||
|
||||
bool parseAmdKernelCodeField(StringRef ID, MCAsmParser &Parser,
|
||||
amd_kernel_code_t &C, raw_ostream &Err);
|
||||
|
||||
} // end namespace llvm
|
||||
|
||||
#endif // LLVM_LIB_TARGET_AMDGPU_UTILS_AMDKERNELCODETUTILS_H
|
@ -1,5 +0,0 @@
|
||||
add_llvm_library(LLVMAMDGPUUtils
|
||||
AMDGPUBaseInfo.cpp
|
||||
AMDKernelCodeTUtils.cpp
|
||||
AMDGPUAsmUtils.cpp
|
||||
)
|
@ -1,23 +0,0 @@
|
||||
;===- ./lib/Target/AMDGPU/Utils/LLVMBuild.txt ------------------*- Conf -*--===;
|
||||
;
|
||||
; The LLVM Compiler Infrastructure
|
||||
;
|
||||
; This file is distributed under the University of Illinois Open Source
|
||||
; License. See LICENSE.TXT for details.
|
||||
;
|
||||
;===------------------------------------------------------------------------===;
|
||||
;
|
||||
; This is an LLVMBuild description file for the components in this subdirectory.
|
||||
;
|
||||
; For more information on the LLVMBuild system, please see:
|
||||
;
|
||||
; http://llvm.org/docs/LLVMBuild.html
|
||||
;
|
||||
;===------------------------------------------------------------------------===;
|
||||
|
||||
[component_0]
|
||||
type = Library
|
||||
name = AMDGPUUtils
|
||||
parent = AMDGPU
|
||||
required_libraries = Core MC Support
|
||||
add_to_library_groups = AMDGPU
|
Reference in New Issue
Block a user