JIT compiler for x86

This commit is contained in:
tevador 2018-12-18 22:00:58 +01:00
parent ddc29cb4d3
commit ed0bc906d6
12 changed files with 917 additions and 46 deletions

View File

@ -11,7 +11,7 @@ SRCDIR=src
OBJDIR=obj
LDFLAGS=
TOBJS=$(addprefix $(OBJDIR)/,instructionsPortable.o TestAluFpu.o)
ROBJS=$(addprefix $(OBJDIR)/,argon2_core.o argon2_ref.o AssemblyGeneratorX86.o blake2b.o CompiledVirtualMachine.o dataset.o executeProgram-linux.o instructionsPortable.o Instruction.o InterpretedVirtualMachine.o main.o Program.o softAes.o VirtualMachine.o)
ROBJS=$(addprefix $(OBJDIR)/,argon2_core.o argon2_ref.o AssemblyGeneratorX86.o blake2b.o CompiledVirtualMachine.o dataset.o JitCompilerX86.o instructionsPortable.o Instruction.o InterpretedVirtualMachine.o main.o Program.o softAes.o VirtualMachine.o)
SRC1=$(addprefix $(SRCDIR)/,TestAluFpu.cpp instructions.hpp Pcg32.hpp)
all: release test
@ -55,8 +55,8 @@ $(OBJDIR)/CompiledVirtualMachine.o: $(addprefix $(SRCDIR)/,CompiledVirtualMachin
$(OBJDIR)/dataset.o: $(addprefix $(SRCDIR)/,dataset.cpp common.hpp Pcg32.hpp argon2_core.h) | $(OBJDIR)
$(CXX) $(CXXFLAGS) -c $(SRCDIR)/dataset.cpp -o $@
$(OBJDIR)/executeProgram-linux.o: $(addprefix $(SRCDIR)/,executeProgram-linux.cpp common.hpp) | $(OBJDIR)
$(CXX) $(CXXFLAGS) -c $(SRCDIR)/executeProgram-linux.cpp -o $@
$(OBJDIR)/JitCompilerX86.o: $(addprefix $(SRCDIR)/,JitCompilerX86.cpp JitCompilerX86.hpp Instruction.hpp) | $(OBJDIR)
$(CXX) $(CXXFLAGS) -c $(SRCDIR)/JitCompilerX86.cpp -o $@
$(OBJDIR)/instructionsPortable.o: $(addprefix $(SRCDIR)/,instructionsPortable.cpp instructions.hpp intrinPortable.h) | $(OBJDIR)
$(CXX) $(CXXFLAGS) -c $(SRCDIR)/instructionsPortable.cpp -o $@

View File

@ -218,10 +218,6 @@ namespace RandomX {
}
}
static inline int wrapi(int i) {
return i % RandomX::ProgramLength;
}
void AssemblyGeneratorX86::h_ADD_64(Instruction& instr, int i) {
gena(instr);
asmCode << "\tadd rax, ";
@ -468,14 +464,14 @@ namespace RandomX {
asmCode << "\tcmp " << regR32[instr.regb % RegistersCount] << ", " << instr.imm1 << std::endl;
asmCode << "\tjbe short taken_call_" << i << std::endl;
gencr(instr);
asmCode << "\tjmp rx_i_" << wrapi(i + 1) << std::endl;
asmCode << "\tjmp rx_i_" << wrapInstr(i + 1) << std::endl;
asmCode << "taken_call_" << i << ":" << std::endl;
}
if (trace) {
asmCode << "\tmov qword ptr [rsi + rdi * 8 + 262144], rax" << std::endl;
}
asmCode << "\tpush rax" << std::endl;
asmCode << "\tcall rx_i_" << wrapi(i + (instr.imm0 & 127) + 2) << std::endl;
asmCode << "\tcall rx_i_" << wrapInstr(i + (instr.imm0 & 127) + 2) << std::endl;
}
void AssemblyGeneratorX86::h_RET(Instruction& instr, int i) {

View File

@ -21,23 +21,35 @@ along with RandomX. If not, see<http://www.gnu.org/licenses/>.
#include "Pcg32.hpp"
#include "common.hpp"
#include "instructions.hpp"
#include <stdexcept>
namespace RandomX {
CompiledVirtualMachine::CompiledVirtualMachine(bool softAes) : VirtualMachine(softAes) {
#if !defined(_M_X64) && !defined(__x86_64__)
throw std::runtime_error("Compiled VM only supports x86-64 CPUs");
#endif
}
void CompiledVirtualMachine::initializeDataset(const void* seed, bool lightClient) {
if (lightClient) {
throw std::runtime_error("Compiled VM does not support light-client mode");
}
VirtualMachine::initializeDataset(seed, lightClient);
}
void CompiledVirtualMachine::initializeProgram(const void* seed) {
Pcg32 gen(seed);
for (unsigned i = 0; i < sizeof(reg) / sizeof(Pcg32::result_type); ++i) {
*(((uint32_t*)&reg) + i) = gen();
}
for (unsigned i = 0; i < ProgramLength; ++i) {
gen(); gen(); gen(); gen();
}
compiler.generateProgram(gen);
mem.ma = (gen() ^ *(((uint32_t*)seed) + 4)) & ~7;
mem.mx = *(((uint32_t*)seed) + 5);
}
void CompiledVirtualMachine::execute() {
executeProgram(reg, mem, readDataset, scratchpad);
compiler.getProgramFunc()(reg, mem, scratchpad);
#ifdef TRACE
for (int32_t i = InstructionCount - 1; i >= 0; --i) {
std::cout << std::hex << tracepad[i].u64 << std::endl;

View File

@ -20,17 +20,21 @@ along with RandomX. If not, see<http://www.gnu.org/licenses/>.
#pragma once
//#define TRACE
#include "VirtualMachine.hpp"
#include "Program.hpp"
#include <sstream>
#include "JitCompilerX86.hpp"
namespace RandomX {
class CompiledVirtualMachine : public VirtualMachine {
public:
CompiledVirtualMachine(bool softAes) : VirtualMachine(softAes) {}
virtual void initializeProgram(const void* seed) override;
CompiledVirtualMachine(bool softAes);
void initializeDataset(const void* seed, bool light = false) override;
void initializeProgram(const void* seed) override;
virtual void execute() override;
void* getProgram() {
return compiler.getCode();
}
private:
JitCompilerX86 compiler;
#ifdef TRACE
convertible_t tracepad[InstructionCount];
#endif

747
src/JitCompilerX86.cpp Normal file
View File

@ -0,0 +1,747 @@
/*
Copyright (c) 2018 tevador
This file is part of RandomX.
RandomX is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
RandomX is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with RandomX. If not, see<http://www.gnu.org/licenses/>.
*/
#include "JitCompilerX86.hpp"
#include "Pcg32.hpp"
#include <cstring>
#include <stdexcept>
#ifdef _WIN32
#include <windows.h>
#elif defined(__linux__)
#include <unistd.h>
#include <malloc.h>
#include <sys/mman.h>
#else
#error "Unsupported operating system"
#endif
namespace RandomX {
/*
REGISTER ALLOCATION:
rax -> temporary
rbx -> MemoryRegisters& memory
rcx -> temporary
rdx -> temporary
rsi -> convertible_t& scratchpad
rdi -> "ic" (instruction counter)
rbp -> beginning of VM stack
rsp -> end of VM stack
r8 -> "r0"
r9 -> "r1"
r10 -> "r2"
r11 -> "r3"
r12 -> "r4"
r13 -> "r5"
r14 -> "r6"
r15 -> "r7"
xmm0 -> temporary
xmm1 -> temporary
xmm2 -> "f2"
xmm3 -> "f3"
xmm4 -> "f4"
xmm5 -> "f5"
xmm6 -> "f6"
xmm7 -> "f7"
xmm8 -> "f0"
xmm9 -> "f1"
STACK STRUCTURE:
|
|
| saved registers
|
v
[rbp] RegisterFile& registerFile
|
|
| VM stack
|
v
[rsp] last element of VM stack
*/
const uint8_t prologue[] = {
0x53, //push rbx
0x55, //push rbp
#ifdef _WIN32
0x57, //push rdi
0x56, //push rsi
#endif
0x41, 0x54, //push r12
0x41, 0x55, //push r13
0x41, 0x56, //push r14
0x41, 0x57, //push r15
#ifdef _WIN32
0x48, 0x83, 0xec, 0x48, //sub rsp,0x48
0xf3, 0x0f, 0x7f, 0x74, 0x24, 0x30, //movdqu XMMWORD PTR[rsp + 0x30],xmm6
0xf3, 0x0f, 0x7f, 0x7c, 0x24, 0x20, //movdqu XMMWORD PTR[rsp + 0x20],xmm7
0xf3, 0x44, 0x0f, 0x7f, 0x44, 0x24, 0x10, //movdqu XMMWORD PTR[rsp + 0x10],xmm8
0xf3, 0x44, 0x0f, 0x7f, 0x0c, 0x24, //movdqu XMMWORD PTR[rsp],xmm9
0x51, //push rcx
0x48, 0x8b, 0xda, //mov rbx,rdx
0x49, 0x8b, 0xf0, //mov rsi,r8
#else
0x57, //push rdi
0x48, 0x8b, 0xde, //mov rbx, rsi
0x48, 0x8b, 0xf2, //mov rsi, rdx
0x48, 0x8b, 0xcf, //mov rcx, rdi
#endif
0x48, 0x8b, 0xec, //mov rbp,rsp
0x48, 0xc7, 0xc7, 0x00, 0x00, 0x10, 0x00, //mov rdi,0x100000
0x4c, 0x8b, 0x01, //mov r8,QWORD PTR[rcx]
0x4c, 0x8b, 0x49, 0x08, //mov r9,QWORD PTR[rcx+0x8]
0x4c, 0x8b, 0x51, 0x10, //mov r10,QWORD PTR[rcx+0x10]
0x4c, 0x8b, 0x59, 0x18, //mov r11,QWORD PTR[rcx+0x18]
0x4c, 0x8b, 0x61, 0x20, //mov r12,QWORD PTR[rcx+0x20]
0x4c, 0x8b, 0x69, 0x28, //mov r13,QWORD PTR[rcx+0x28]
0x4c, 0x8b, 0x71, 0x30, //mov r14,QWORD PTR[rcx+0x30]
0x4c, 0x8b, 0x79, 0x38, //mov r15,QWORD PTR[rcx+0x38]
0xc7, 0x44, 0x24, 0xf8, 0xc0, 0x9f, 0x00, //mov DWORD PTR[rsp-0x8],0x9fc0
0x00,
0x0f, 0xae, 0x54, 0x24, 0xf8, //ldmxcsr DWORD PTR[rsp-0x8]
0xf2, 0x4c, 0x0f, 0x2a, 0x41, 0x40, //cvtsi2sd xmm8,QWORD PTR[rcx+0x40]
0xf2, 0x4c, 0x0f, 0x2a, 0x49, 0x48, //cvtsi2sd xmm9,QWORD PTR[rcx+0x48]
0xf2, 0x48, 0x0f, 0x2a, 0x51, 0x50, //cvtsi2sd xmm2,QWORD PTR[rcx+0x50]
0xf2, 0x48, 0x0f, 0x2a, 0x59, 0x58, //cvtsi2sd xmm3,QWORD PTR[rcx+0x58]
0xf2, 0x48, 0x0f, 0x2a, 0x61, 0x60, //cvtsi2sd xmm4,QWORD PTR[rcx+0x60]
0xf2, 0x48, 0x0f, 0x2a, 0x69, 0x68, //cvtsi2sd xmm5,QWORD PTR[rcx+0x68]
0xf2, 0x48, 0x0f, 0x2a, 0x71, 0x70, //cvtsi2sd xmm6,QWORD PTR[rcx+0x70]
0xf2, 0x48, 0x0f, 0x2a, 0x79, 0x78, //cvtsi2sd xmm7,QWORD PTR[rcx+0x78]
};
const uint8_t epilogue[] = {
0x48, 0x8b, 0xe5, //mov rsp,rbp
0x59, //pop rcx
0x4c, 0x89, 0x01, //mov QWORD PTR [rcx],r8
0x4c, 0x89, 0x49, 0x08, //mov QWORD PTR [rcx+0x8],r9
0x4c, 0x89, 0x51, 0x10, //mov QWORD PTR [rcx+0x10],r10
0x4c, 0x89, 0x59, 0x18, //mov QWORD PTR [rcx+0x18],r11
0x4c, 0x89, 0x61, 0x20, //mov QWORD PTR [rcx+0x20],r12
0x4c, 0x89, 0x69, 0x28, //mov QWORD PTR [rcx+0x28],r13
0x4c, 0x89, 0x71, 0x30, //mov QWORD PTR [rcx+0x30],r14
0x4c, 0x89, 0x79, 0x38, //mov QWORD PTR [rcx+0x38],r15
0x66, 0x4c, 0x0f, 0x7e, 0x41, 0x40, //movq QWORD PTR [rcx+0x40],xmm8
0x66, 0x4c, 0x0f, 0x7e, 0x49, 0x48, //movq QWORD PTR [rcx+0x48],xmm9
0x66, 0x48, 0x0f, 0x7e, 0x51, 0x50, //movq QWORD PTR [rcx+0x50],xmm2
0x66, 0x48, 0x0f, 0x7e, 0x59, 0x58, //movq QWORD PTR [rcx+0x58],xmm3
0x66, 0x48, 0x0f, 0x7e, 0x61, 0x60, //movq QWORD PTR [rcx+0x60],xmm4
0x66, 0x48, 0x0f, 0x7e, 0x69, 0x68, //movq QWORD PTR [rcx+0x68],xmm5
0x66, 0x48, 0x0f, 0x7e, 0x71, 0x70, //movq QWORD PTR [rcx+0x70],xmm6
0x66, 0x48, 0x0f, 0x7e, 0x79, 0x78, //movq QWORD PTR [rcx+0x78],xmm7
#ifdef _WIN32
0xf3, 0x44, 0x0f, 0x6f, 0x0c, 0x24, //movdqu xmm9,XMMWORD PTR [rsp]
0xf3, 0x44, 0x0f, 0x6f, 0x44, 0x24, 0x10, //movdqu xmm8,XMMWORD PTR [rsp+0x10]
0xf3, 0x0f, 0x6f, 0x7c, 0x24, 0x20, //movdqu xmm7,XMMWORD PTR [rsp+0x20]
0xf3, 0x0f, 0x6f, 0x74, 0x24, 0x30, //movdqu xmm6,XMMWORD PTR [rsp+0x30]
0x48, 0x83, 0xc4, 0x48, //add rsp,0x48
#endif
0x41, 0x5f, //pop r15
0x41, 0x5e, //pop r14
0x41, 0x5d, //pop r13
0x41, 0x5c, //pop r12
#ifdef _WIN32
0x5e, //pop rsi
0x5f, //pop rdi
#endif
0x5d, //pop rbp
0x5b, //pop rbx
0xc3, //ret
};
//41 bytes -> 1 cache line
const uint8_t readDatasetSub[] = {
0x8b, 0x13, //mov edx,DWORD PTR [rbx]
0x48, 0x8b, 0x43, 0x08, //mov rax,QWORD PTR [rbx+0x8]
0x48, 0x8b, 0x04, 0x10, //mov rax,QWORD PTR [rax+rdx*1]
0x83, 0x03, 0x08, //add DWORD PTR [rbx],0x8
0x33, 0x4b, 0x04, //xor ecx,DWORD PTR [rbx+0x4]
0x89, 0x4b, 0x04, //mov DWORD PTR [rbx+0x4],ecx
0xf7, 0xc1, 0xf8, 0xff, 0x00, 0x00, //test ecx,0xfff8
0x75, 0x0d, //jne
0x83, 0xe1, 0xf8, //and ecx,0xfffffff8
0x89, 0x0b, //mov DWORD PTR [rbx],ecx
0x48, 0x8b, 0x53, 0x08, //mov rdx,QWORD PTR [rbx+0x8]
0x0f, 0x18, 0x0c, 0x0a, //prefetcht0 BYTE PTR [rdx+rcx*1]
0xc3, //ret
};
constexpr int getNumCacheLines(size_t size) {
return (size + (CacheLineSize - 1)) / CacheLineSize;
}
constexpr int32_t align(int32_t pos, int32_t align) {
return ((pos - 1) / align + 1) * align;
}
constexpr int32_t readDatasetSubOffset = CodeSize - CacheLineSize * getNumCacheLines(sizeof(readDatasetSub));
constexpr int32_t epilogueOffset = readDatasetSubOffset - CacheLineSize * getNumCacheLines(sizeof(epilogue));
constexpr int32_t startOffsetAligned = align(sizeof(prologue), CacheLineSize);
JitCompilerX86::JitCompilerX86() {
#ifdef _WIN32
code = (uint8_t*)VirtualAlloc(nullptr, CodeSize, MEM_COMMIT, PAGE_EXECUTE_READWRITE);
if (code == nullptr)
throw std::runtime_error("VirtualAlloc failed");
#else
auto pagesize = sysconf(_SC_PAGE_SIZE);
if (pagesize == -1)
throw std::runtime_error("sysconf failed");
code = (uint8_t*)memalign(pagesize, CodeSize);
if (code == nullptr)
throw std::runtime_error("memalign failed");
if (mprotect(code, CodeSize, PROT_READ | PROT_WRITE | PROT_EXEC) == -1)
throw std::runtime_error("mprotect failed");
#endif
memcpy(code, prologue, sizeof(prologue));
if (startOffsetAligned - sizeof(prologue) > 4) {
codePos = sizeof(prologue);
emitByte(0xeb);
emitByte(startOffsetAligned - (codePos + 1));
}
memcpy(code + readDatasetSubOffset, readDatasetSub, sizeof(readDatasetSub));
memcpy(code + epilogueOffset, epilogue, sizeof(epilogue));
}
void JitCompilerX86::generateProgram(Pcg32& gen) {
instructionOffsets.clear();
callOffsets.clear();
codePos = startOffsetAligned;
Instruction instr;
for (unsigned i = 0; i < ProgramLength; ++i) {
for (unsigned j = 0; j < sizeof(instr) / sizeof(Pcg32::result_type); ++j) {
*(((uint32_t*)&instr) + j) = gen();
}
generateCode(instr, i);
}
emitByte(0xe9);
emit(instructionOffsets[0] - (codePos + 4));
fixCallOffsets();
}
void JitCompilerX86::generateCode(Instruction& instr, int i) {
instructionOffsets.push_back(codePos);
emit(0x880fcfff); //dec edx; js <epilogue>
emit(epilogueOffset - (codePos + 4)); //jump offset (RIP-relative)
gena(instr);
auto generator = engine[instr.opcode];
(this->*generator)(instr, i);
}
void JitCompilerX86::fixCallOffsets() {
for (CallOffset& co : callOffsets) {
*reinterpret_cast<int32_t*>(code + co.pos) = instructionOffsets[co.index] - (co.pos + 4);
}
}
void JitCompilerX86::gena(Instruction& instr) {
emit(uint16_t(0x8149)); //xor
emitByte(0xf0 + (instr.rega % RegistersCount));
emit(instr.addr0);
int32_t pc;
switch (instr.loca & 7)
{
case 0:
case 1:
case 2:
case 3:
emit(uint16_t(0x8b41)); //mov
emitByte(0xc8 + (instr.rega % RegistersCount)); //ecx, rega
emitByte(0xe8); //call
emit(readDatasetSubOffset - (codePos + 4));
return;
case 4:
emit(uint16_t(0x8b41)); //mov
emitByte(0xc0 + (instr.rega % RegistersCount)); //eax, rega
emitByte(0x25); //and
emit(ScratchpadL2 - 1); //whole scratchpad
emit(0xc6048b48); // mov rax,QWORD PTR [rsi+rax*8]
return;
default:
emit(uint16_t(0x8b41)); //mov
emitByte(0xc0 + (instr.rega % RegistersCount)); //eax, rega
emitByte(0x25); //and
emit(ScratchpadL1 - 1); //first 16 KiB of scratchpad
emit(0xc6048b48); // mov rax,QWORD PTR [rsi+rax*8]
return;
}
}
void JitCompilerX86::genbr0(Instruction& instr, uint16_t opcodeReg, uint16_t opcodeImm) {
if ((instr.locb & 7) <= 5) {
emit(uint16_t(0x8b49)); //mov
emitByte(0xc8 + (instr.regb % RegistersCount)); //rcx, regb
emitByte(0x48); //REX.W
emit(opcodeReg); //xxx rax, cl
}
else {
emitByte(0x48); //REX.W
emit(opcodeImm); //xxx rax, imm8
emitByte((instr.imm0 & 63));
}
}
void JitCompilerX86::genbr1(Instruction& instr, uint16_t opcodeReg, uint16_t opcodeImm) {
if ((instr.locb & 7) <= 5) {
emit(opcodeReg); // xxx rax, r64
emitByte(0xc0 + (instr.regb % RegistersCount));
}
else {
emit(opcodeImm); // xxx rax, imm32
emit(instr.imm1);
}
}
void JitCompilerX86::genbr132(Instruction& instr, uint16_t opcodeReg, uint8_t opcodeImm) {
if ((instr.locb & 7) <= 5) {
emit(opcodeReg); // xxx eax, r32
emitByte(0xc0 + (instr.regb % RegistersCount));
}
else {
emitByte(opcodeImm); // xxx eax, imm32
emit(instr.imm1);
}
}
void JitCompilerX86::genbf(Instruction& instr, uint8_t opcode) {
emit(0x48f2fffff8002548); //and rax,0xfffffffffffff800; cvtsi2sd xmm0,rax
emit(uint16_t(0x2a0f));
emitByte(0xc0);
if ((instr.locb & 7) <= 5) {
int regb = (instr.regb % RegistersCount);
emitByte(0xf2); //xxxsd xmm0,regb
if (regb <= 1) {
emitByte(0x41); //REX
}
emitByte(0x0f);
emitByte(opcode);
emitByte(0xc0 + regb);
}
else {
convertible_t bimm;
bimm.f64 = (double)instr.imm1;
emit(uint16_t(0xb848)); //movabs rax,imm64
emit(bimm.i64);
emitByte(0x66); //movq xmm1,rax
emit(0xc86e0f48);
emit(uint16_t(0x0ff2)); //xxxsd xmm0,xmm1
emitByte(opcode);
emitByte(0xc1);
}
}
void JitCompilerX86::gencr(Instruction& instr) {
switch (instr.locc & 7)
{
case 0:
emit(0x41c88b48); //mov rcx, rax; REX
emitByte(0x8b); // mov
emitByte(0xc0 + (instr.regc % RegistersCount)); //eax, regc
emitByte(0x35); // xor eax
emit(instr.addr1);
emitByte(0x25); //and
emit(ScratchpadL2 - 1); //whole scratchpad
emit(0xc60c8948); // mov QWORD PTR [rsi+rax*8],rcx
break;
case 1:
case 2:
case 3:
emit(0x41c88b48); //mov rcx, rax; REX
emitByte(0x8b); // mov
emitByte(0xc0 + (instr.regc % RegistersCount)); //eax, regc
emitByte(0x35); // xor eax
emit(instr.addr1);
emitByte(0x25); //and
emit(ScratchpadL1 - 1); //first 16 KiB of scratchpad
emit(0xc60c8948); // mov QWORD PTR [rsi+rax*8],rcx
break;
default:
emit(uint16_t(0x8b4c)); //mov
emitByte(0xc0 + 8 * (instr.regc % RegistersCount)); //regc, rax
break;
}
}
void JitCompilerX86::gencf(Instruction& instr) {
int regc = (instr.regc % RegistersCount);
switch (instr.locc & 7)
{
case 0:
emit(uint16_t(0x8b41)); //mov
emitByte(0xc0 + regc); //eax, regc
emitByte(0x35); // xor eax
emit(instr.addr1);
emitByte(0x25); //and
emit(ScratchpadL2 - 1); //whole scratchpad
emit(uint16_t(0x4866)); //prefix
emit(0xc6047e0f); // movq QWORD PTR [rsi+rax*8],xmm0
break;
case 1:
case 2:
case 3:
emit(uint16_t(0x8b41)); //mov
emitByte(0xc0 + regc); //eax, regc
emitByte(0x35); // xor eax
emit(instr.addr1);
emitByte(0x25); //and
emit(ScratchpadL1 - 1); //first 16 KiB of scratchpad
emit(uint16_t(0x4866)); //prefix
emit(0xc6047e0f); // movq QWORD PTR [rsi+rax*8],xmm0
break;
default:
emitByte(0xf2);
if (regc <= 1) {
emitByte(0x44); //REX
}
emit(uint16_t(0x100f)); //movsd
emitByte(0xc0 + 8 * regc); // regc, xmm0
break;
}
}
void JitCompilerX86::h_ADD_64(Instruction& instr, int i) {
genbr1(instr, 0x0349, 0x0548);
gencr(instr);
}
void JitCompilerX86::h_ADD_32(Instruction& instr, int i) {
genbr132(instr, 0x0341, 0x05);
gencr(instr);
}
void JitCompilerX86::h_SUB_64(Instruction& instr, int i) {
genbr1(instr, 0x2b49, 0x2d48);
gencr(instr);
}
void JitCompilerX86::h_SUB_32(Instruction& instr, int i) {
genbr132(instr, 0x2b41, 0x2d);
gencr(instr);
}
void JitCompilerX86::h_MUL_64(Instruction& instr, int i) {
if ((instr.locb & 7) <= 5) {
emitByte(0x49); //REX
emit(uint16_t(0xaf0f)); // imul rax, r64
emitByte(0xc0 + (instr.regb % RegistersCount));
}
else {
emitByte(0x48); //REX
emit(uint16_t(0xc069)); // imul rax, rax, imm32
emit(instr.imm1);
}
gencr(instr);
}
void JitCompilerX86::h_MULH_64(Instruction& instr, int i) {
if ((instr.locb & 7) <= 5) {
emit(uint16_t(0x8b49)); //mov rcx, r64
emitByte(0xc8 + (instr.regb % RegistersCount));
}
else {
emitByte(0x48);
emit(uint16_t(0xc1c7)); // mov rcx, imm32
emit(instr.imm1);
}
emitByte(0x48);
emit(uint16_t(0xe1f7)); // mul rcx
emitByte(0x48);
emit(uint16_t(0xc28b)); // mov rax,rdx
gencr(instr);
}
void JitCompilerX86::h_MUL_32(Instruction& instr, int i) {
emit(uint16_t(0xc88b)); //mov ecx, eax
if ((instr.locb & 7) <= 5) {
emit(uint16_t(0x8b41)); // mov eax, r32
emitByte(0xc0 + (instr.regb % RegistersCount));
}
else {
emitByte(0xb8); // mov eax, imm32
emit(instr.imm1);
}
emit(0xc1af0f48); //imul rax,rcx
gencr(instr);
}
void JitCompilerX86::h_IMUL_32(Instruction& instr, int i) {
emitByte(0x48);
emit(uint16_t(0xc863)); //movsxd rcx,eax
if ((instr.locb & 7) <= 5) {
emit(uint16_t(0x6349)); //movsxd rax,r32
emitByte(0xc0 + (instr.regb % RegistersCount));
}
else {
emitByte(0x48);
emit(uint16_t(0xc0c7)); // mov rax, imm32
emit(instr.imm1);
}
emit(0xc1af0f48); //imul rax,rcx
gencr(instr);
}
void JitCompilerX86::h_IMULH_64(Instruction& instr, int i) {
if ((instr.locb & 7) <= 5) {
emit(uint16_t(0x8b49)); //mov rcx, r64
emitByte(0xc8 + (instr.regb % RegistersCount));
}
else {
emitByte(0x48);
emit(uint16_t(0xc1c7)); // mov rcx, imm32
emit(instr.imm1);
}
emitByte(0x48);
emit(uint16_t(0xe9f7)); // imul rcx
emitByte(0x48);
emit(uint16_t(0xc28b)); // mov rax,rdx
gencr(instr);
}
void JitCompilerX86::h_DIV_64(Instruction& instr, int i) {
if ((instr.locb & 7) <= 5) {
emitByte(0xb9); //mov ecx, 1
emit(1);
emit(uint16_t(0x8b41)); //mov edx, r32
emitByte(0xd0 + (instr.regb % RegistersCount));
emit(0x450fd285); //test edx, edx; cmovne ecx,edx
emitByte(0xca);
}
else {
emitByte(0xb9); //mov ecx, imm32
emit(instr.imm1 != 0 ? instr.imm1 : 1);
}
emit(0xf748d233); //xor edx,edx; div rcx
emitByte(0xf1);
gencr(instr);
}
void JitCompilerX86::h_IDIV_64(Instruction& instr, int i) {
if ((instr.locb & 7) <= 5) {
emit(uint16_t(0x8b41)); //mov edx, r32
emitByte(0xd0 + (instr.regb % RegistersCount));
}
else {
emitByte(0xba); // xxx edx, imm32
emit(instr.imm1);
}
emit(0xc88b480b75fffa83);
emit(0x1274c9ff48c1d148);
emit(0x0fd28500000001b9);
emit(0x489948c96348ca45);
emit(uint16_t(0xf9f7)); //idiv rcx
gencr(instr);
}
void JitCompilerX86::h_AND_64(Instruction& instr, int i) {
genbr1(instr, 0x2349, 0x2548);
gencr(instr);
}
void JitCompilerX86::h_AND_32(Instruction& instr, int i) {
genbr132(instr, 0x2341, 0x25);
gencr(instr);
}
void JitCompilerX86::h_OR_64(Instruction& instr, int i) {
genbr1(instr, 0x0b49, 0x0d48);
gencr(instr);
}
void JitCompilerX86::h_OR_32(Instruction& instr, int i) {
genbr132(instr, 0x0b41, 0x0d);
gencr(instr);
}
void JitCompilerX86::h_XOR_64(Instruction& instr, int i) {
genbr1(instr, 0x3349, 0x3548);
gencr(instr);
}
void JitCompilerX86::h_XOR_32(Instruction& instr, int i) {
genbr132(instr, 0x3341, 0x35);
gencr(instr);
}
void JitCompilerX86::h_SHL_64(Instruction& instr, int i) {
genbr0(instr, 0xe0d3, 0xe0c1);
gencr(instr);
}
void JitCompilerX86::h_SHR_64(Instruction& instr, int i) {
genbr0(instr, 0xe8d3, 0xe8c1);
gencr(instr);
}
void JitCompilerX86::h_SAR_64(Instruction& instr, int i) {
genbr0(instr, 0xf8d3, 0xf8c1);
gencr(instr);
}
void JitCompilerX86::h_ROL_64(Instruction& instr, int i) {
genbr0(instr, 0xc0d3, 0xc0c1);
gencr(instr);
}
void JitCompilerX86::h_ROR_64(Instruction& instr, int i) {
genbr0(instr, 0xc8d3, 0xc8c1);
gencr(instr);
}
void JitCompilerX86::h_FPADD(Instruction& instr, int i) {
genbf(instr, 0x58);
gencf(instr);
}
void JitCompilerX86::h_FPSUB(Instruction& instr, int i) {
genbf(instr, 0x5c);
gencf(instr);
}
void JitCompilerX86::h_FPMUL(Instruction& instr, int i) {
emit(uint16_t(0x0d48)); //or rax,0x800
emit(0x00000800);
genbf(instr, 0x59);
gencf(instr);
}
void JitCompilerX86::h_FPDIV(Instruction& instr, int i) {
emit(uint16_t(0x0d48)); //or rax,0x800
emit(0x00000800);
genbf(instr, 0x5e);
gencf(instr);
}
void JitCompilerX86::h_FPSQRT(Instruction& instr, int i) {
emit(uint16_t(0xb948)); //or movabs rcx, imm64
emit(0x7ffffffffffff800);
emit(0xc02a0f48f2c12348); //and rax,rcx; cvtsi2sd xmm0,rax
emit(0xc0510ff2); //sqrtsd xmm0,xmm0
gencf(instr);
}
void JitCompilerX86::h_FPROUND(Instruction& instr, int i) {
emit(0x81480de0c1c88b48);
emit(0x600025fffff800e1);
emit(0x0dc12a0f48f20000);
emit(0xf824448900009fc0);
emit(0x2454ae0f); //ldmxcsr DWORD PTR [rsp-0x8]
emitByte(0xf8);
gencf(instr);
}
void JitCompilerX86::h_CALL(Instruction& instr, int i) {
if ((instr.locb & 7) <= 5) {
emit(uint16_t(0x8141)); //cmp regb, imm32
emitByte(0xf8 + (instr.regb % RegistersCount));
emit(instr.imm1);
if ((instr.locc & 7) <= 3) {
emit(uint16_t(0x1676)); //jmp
}
else {
emit(uint16_t(0x0576)); //jmp
}
gencr(instr);
emit(uint16_t(0x06eb)); //jmp to next
}
emitByte(0x50); //push rax
emitByte(0xe8); //call
i = wrapInstr(i + (instr.imm0 & 127) + 2);
if (i < instructionOffsets.size()) {
emit(instructionOffsets[i] - (codePos + 4));
}
else {
callOffsets.push_back(CallOffset(codePos, i));
codePos += 4;
}
}
void JitCompilerX86::h_RET(Instruction& instr, int i) {
int crlen = 0;
int blen = 0;
if ((instr.locc & 7) <= 3) {
crlen = 17;
}
if ((instr.locb & 7) <= 5) {
blen = 9;
}
emit(0x74e53b48); //cmp rsp, rbp; je
emitByte(11 + blen + crlen);
if ((instr.locb & 7) <= 5) {
emit(uint16_t(0x8141)); //cmp regb, imm32
emitByte(0xf8 + (instr.regb % RegistersCount));
emit(instr.imm1);
emitByte(0x77); //jmp
emitByte(11 + crlen);
}
emitByte(0x48);
emit(0x08244433); //xor rax,QWORD PTR [rsp+0x8]
gencr(instr);
emitByte(0xc2); //ret 8
emit(uint16_t(0x0008));
gencr(instr);
}
#include "instructionWeights.hpp"
#define INST_HANDLE(x) REPN(&JitCompilerX86::h_##x, WT(x))
InstructionGeneratorX86 JitCompilerX86::engine[256] = {
INST_HANDLE(ADD_64)
INST_HANDLE(ADD_32)
INST_HANDLE(SUB_64)
INST_HANDLE(SUB_32)
INST_HANDLE(MUL_64)
INST_HANDLE(MULH_64)
INST_HANDLE(MUL_32)
INST_HANDLE(IMUL_32)
INST_HANDLE(IMULH_64)
INST_HANDLE(DIV_64)
INST_HANDLE(IDIV_64)
INST_HANDLE(AND_64)
INST_HANDLE(AND_32)
INST_HANDLE(OR_64)
INST_HANDLE(OR_32)
INST_HANDLE(XOR_64)
INST_HANDLE(XOR_32)
INST_HANDLE(SHL_64)
INST_HANDLE(SHR_64)
INST_HANDLE(SAR_64)
INST_HANDLE(ROL_64)
INST_HANDLE(ROR_64)
INST_HANDLE(FPADD)
INST_HANDLE(FPSUB)
INST_HANDLE(FPMUL)
INST_HANDLE(FPDIV)
INST_HANDLE(FPSQRT)
INST_HANDLE(FPROUND)
INST_HANDLE(CALL)
INST_HANDLE(RET)
};
}

114
src/JitCompilerX86.hpp Normal file
View File

@ -0,0 +1,114 @@
/*
Copyright (c) 2018 tevador
This file is part of RandomX.
RandomX is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
RandomX is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with RandomX. If not, see<http://www.gnu.org/licenses/>.
*/
#pragma once
#include "common.hpp"
#include "Instruction.hpp"
#include <cstring>
#include <vector>
class Pcg32;
namespace RandomX {
class JitCompilerX86;
typedef void(JitCompilerX86::*InstructionGeneratorX86)(Instruction&, int);
constexpr uint32_t CodeSize = 64 * 1024;
constexpr uint32_t CacheLineSize = 64;
struct CallOffset {
CallOffset(int32_t p, int32_t i) : pos(p), index(i) {}
int32_t pos;
int32_t index;
};
class JitCompilerX86 {
public:
JitCompilerX86();
void generateProgram(Pcg32&);
ProgramFunc getProgramFunc() {
return (ProgramFunc)code;
}
uint8_t* getCode() {
return code;
}
private:
static InstructionGeneratorX86 engine[256];
uint8_t* code;
int32_t codePos;
std::vector<int32_t> instructionOffsets;
std::vector<CallOffset> callOffsets;
void gena(Instruction&);
void genbr0(Instruction&, uint16_t, uint16_t);
void genbr1(Instruction&, uint16_t, uint16_t);
void genbr132(Instruction&, uint16_t, uint8_t);
void genbf(Instruction&, uint8_t);
void gencr(Instruction&);
void gencf(Instruction&);
void generateCode(Instruction&, int);
void fixCallOffsets();
void emitByte(uint8_t val) {
code[codePos] = val;
codePos++;
}
template<typename T>
void emit(T val) {
*reinterpret_cast<T*>(code + codePos) = val;
codePos += sizeof(T);
}
void h_ADD_64(Instruction&, int);
void h_ADD_32(Instruction&, int);
void h_SUB_64(Instruction&, int);
void h_SUB_32(Instruction&, int);
void h_MUL_64(Instruction&, int);
void h_MULH_64(Instruction&, int);
void h_MUL_32(Instruction&, int);
void h_IMUL_32(Instruction&, int);
void h_IMULH_64(Instruction&, int);
void h_DIV_64(Instruction&, int);
void h_IDIV_64(Instruction&, int);
void h_AND_64(Instruction&, int);
void h_AND_32(Instruction&, int);
void h_OR_64(Instruction&, int);
void h_OR_32(Instruction&, int);
void h_XOR_64(Instruction&, int);
void h_XOR_32(Instruction&, int);
void h_SHL_64(Instruction&, int);
void h_SHR_64(Instruction&, int);
void h_SAR_64(Instruction&, int);
void h_ROL_64(Instruction&, int);
void h_ROR_64(Instruction&, int);
void h_FPADD(Instruction&, int);
void h_FPSUB(Instruction&, int);
void h_FPMUL(Instruction&, int);
void h_FPDIV(Instruction&, int);
void h_FPSQRT(Instruction&, int);
void h_FPROUND(Instruction&, int);
void h_CALL(Instruction&, int);
void h_RET(Instruction&, int);
};
}

View File

@ -72,4 +72,8 @@ namespace RandomX {
memcpy(scratchpad, mem.dataset + ScratchpadSize * index, ScratchpadSize);
}
}
void VirtualMachine::getResult(void* out) {
}
}

View File

@ -27,10 +27,11 @@ namespace RandomX {
public:
VirtualMachine(bool softAes);
virtual ~VirtualMachine() {}
void initializeDataset(const void* seed, bool light = false);
virtual void initializeDataset(const void* seed, bool light = false);
void initializeScratchpad(uint32_t index);
virtual void initializeProgram(const void* seed) = 0;
virtual void execute() = 0;
void getResult(void*);
const RegisterFile& getRegisterFile() const {
return reg;
}

View File

@ -70,6 +70,10 @@ namespace RandomX {
constexpr uint32_t ScratchpadL2 = ScratchpadSize / sizeof(convertible_t);
constexpr int RegistersCount = 8;
inline int wrapInstr(int i) {
return i % RandomX::ProgramLength;
}
struct LightClientMemory {
uint8_t* cache;
uint8_t* block;
@ -107,7 +111,9 @@ namespace RandomX {
typedef convertible_t(*DatasetReadFunc)(addr_t, MemoryRegisters&);
typedef void(*ProgramFunc)(RegisterFile&, MemoryRegisters&, convertible_t*);
extern "C" {
void executeProgram(RegisterFile& registerFile, MemoryRegisters& memory, DatasetReadFunc readFunc, convertible_t* scratchpad);
void executeProgram(RegisterFile& registerFile, MemoryRegisters& memory, convertible_t* scratchpad);
}
}

View File

@ -1,10 +0,0 @@
#include "common.hpp"
#include <stdexcept>
namespace RandomX {
extern "C" {
void executeProgram(RegisterFile& registerFile, MemoryRegisters& memory, DatasetReadFunc readFunc, convertible_t* scratchpad) {
throw std::runtime_error("not implemented");
}
}
}

View File

@ -54,8 +54,7 @@ executeProgram PROC
; | saved registers
; |
; v
; [rbp+8] RegisterFile& registerFile
; [rbp] DatasetReadFunc readFunc
; [rbp] RegisterFile& registerFile
; |
; |
; | VM stack
@ -72,7 +71,7 @@ executeProgram PROC
push r13
push r14
push r15
sub rsp, 64
sub rsp, 72
movdqu xmmword ptr [rsp+48], xmm6
movdqu xmmword ptr [rsp+32], xmm7
movdqu xmmword ptr [rsp+16], xmm8
@ -81,8 +80,7 @@ executeProgram PROC
; function arguments
push rcx ; RegisterFile& registerFile
mov rbx, rdx ; MemoryRegisters& memory
push r8 ; DatasetReadFunc readFunc
mov rsi, r9 ; convertible_t& scratchpad
mov rsi, r8 ; convertible_t& scratchpad
mov rbp, rsp ; beginning of VM stack
mov rdi, 1048576 ; number of VM instructions to execute
@ -96,8 +94,8 @@ executeProgram PROC
mov r13, qword ptr [rcx+40]
mov r14, qword ptr [rcx+48]
mov r15, qword ptr [rcx+56]
mov dword ptr [rsp - 8], 40896
ldmxcsr dword ptr [rsp - 8]
mov dword ptr [rsp-8], 40896
ldmxcsr dword ptr [rsp-8]
cvtsi2sd xmm8, qword ptr [rcx+64]
cvtsi2sd xmm9, qword ptr [rcx+72]
cvtsi2sd xmm2, qword ptr [rcx+80]
@ -114,10 +112,9 @@ executeProgram PROC
rx_finish:
; unroll the stack
mov rsp, rbp
add rsp, 16
; save VM register values
mov rcx, qword ptr [rbp+8]
pop rcx
mov qword ptr [rcx+0], r8
mov qword ptr [rcx+8], r9
mov qword ptr [rcx+16], r10
@ -136,11 +133,11 @@ rx_finish:
movd qword ptr [rcx+120], xmm7
; load callee-saved registers
movdqu xmm9, xmmword ptr [rsp+0]
movdqu xmm9, xmmword ptr [rsp]
movdqu xmm8, xmmword ptr [rsp+16]
movdqu xmm7, xmmword ptr [rsp+32]
movdqu xmm6, xmmword ptr [rsp+48]
add rsp, 64
add rsp, 72
pop r15
pop r14
pop r13

View File

@ -109,13 +109,13 @@ int main(int argc, char** argv) {
RandomX::VirtualMachine* vm;
try {
if (compiled) {
vm = new RandomX::CompiledVirtualMachine(softAes);
}
else {
vm = new RandomX::InterpretedVirtualMachine(softAes);
}
try {
std::cout << "Initializing..." << std::endl;
Stopwatch sw(true);
vm->initializeDataset(seed, lightClient);