Add a new JIT compiler for CPU code (#693)
* Start of the ARMeilleure project
* Refactoring around the old IRAdapter, now renamed to PreAllocator
* Optimize the LowestBitSet method
* Add CLZ support and fix CLS implementation
* Add missing Equals and GetHashCode overrides on some structs, misc small tweaks
* Implement the ByteSwap IR instruction, and some refactoring on the assembler
* Implement the DivideUI IR instruction and fix 64-bits IDIV
* Correct constant operand type on CSINC
* Move division instructions implementation to InstEmitDiv
* Fix destination type for the ConditionalSelect IR instruction
* Implement UMULH and SMULH, with new IR instructions
* Fix some issues with shift instructions
* Fix constant types for BFM instructions
* Fix up new tests using the new V128 struct
* Update tests
* Move DIV tests to a separate file
* Add support for calls, and some instructions that depends on them
* Start adding support for SIMD & FP types, along with some of the related ARM instructions
* Fix some typos and the divide instruction with FP operands
* Fix wrong method call on Clz_V
* Implement ARM FP & SIMD move instructions, Saddlv_V, and misc. fixes
* Implement SIMD logical instructions and more misc. fixes
* Fix PSRAD x86 instruction encoding, TRN, UABD and UABDL implementations
* Implement float conversion instruction, merge in LDj3SNuD fixes, and some other misc. fixes
* Implement SIMD shift instruction and fix Dup_V
* Add SCVTF and UCVTF (vector, fixed-point) variants to the opcode table
* Fix check with tolerance on tester
* Implement FP & SIMD comparison instructions, and some fixes
* Update FCVT (Scalar) encoding on the table to support the Half-float variants
* Support passing V128 structs, some cleanup on the register allocator, merge LDj3SNuD fixes
* Use old memory access methods, made a start on SIMD memory insts support, some fixes
* Fix float constant passed to functions, save and restore non-volatile XMM registers, other fixes
* Fix arguments count with struct return values, other fixes
* More instructions
* Misc. fixes and integrate LDj3SNuD fixes
* Update tests
* Add a faster linear scan allocator, unwinding support on windows, and other changes
* Update Ryujinx.HLE
* Update Ryujinx.Graphics
* Fix V128 return pointer passing, RCX is clobbered
* Update Ryujinx.Tests
* Update ITimeZoneService
* Stop using GetFunctionPointer as that can't be called from native code, misc. fixes and tweaks
* Use generic GetFunctionPointerForDelegate method and other tweaks
* Some refactoring on the code generator, assert on invalid operations and use a separate enum for intrinsics
* Remove some unused code on the assembler
* Fix REX.W prefix regression on float conversion instructions, add some sort of profiler
* Add hardware capability detection
* Fix regression on Sha1h and revert Fcm** changes
* Add SSE2-only paths on vector extract and insert, some refactoring on the pre-allocator
* Fix silly mistake introduced on last commit on CpuId
* Generate inline stack probes when the stack allocation is too large
* Initial support for the System-V ABI
* Support multiple destination operands
* Fix SSE2 VectorInsert8 path, and other fixes
* Change placement of XMM callee save and restore code to match other compilers
* Rename Dest to Destination and Inst to Instruction
* Fix a regression related to calls and the V128 type
* Add an extra space on comments to match code style
* Some refactoring
* Fix vector insert FP32 SSE2 path
* Port over the ARM32 instructions
* Avoid memory protection races on JIT Cache
* Another fix on VectorInsert FP32 (thanks to LDj3SNuD
* Float operands don't need to use the same register when VEX is supported
* Add a new register allocator, higher quality code for hot code (tier up), and other tweaks
* Some nits, small improvements on the pre allocator
* CpuThreadState is gone
* Allow changing CPU emulators with a config entry
* Add runtime identifiers on the ARMeilleure project
* Allow switching between CPUs through a config entry (pt. 2)
* Change win10-x64 to win-x64 on projects
* Update the Ryujinx project to use ARMeilleure
* Ensure that the selected register is valid on the hybrid allocator
* Allow exiting on returns to 0 (should fix test regression)
* Remove register assignments for most used variables on the hybrid allocator
* Do not use fixed registers as spill temp
* Add missing namespace and remove unneeded using
* Address PR feedback
* Fix types, etc
* Enable AssumeStrictAbiCompliance by default
* Ensure that Spill and Fill don't load or store any more than necessary
2019-08-08 18:56:22 +00:00
|
|
|
using ARMeilleure.Memory;
|
2018-08-15 18:59:51 +00:00
|
|
|
using System;
|
2019-07-22 17:15:46 +00:00
|
|
|
using System.Runtime.InteropServices;
|
2018-08-15 18:59:51 +00:00
|
|
|
|
2018-11-28 22:18:09 +00:00
|
|
|
namespace Ryujinx.HLE
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
|
|
|
class DeviceMemory : IDisposable
|
|
|
|
{
|
|
|
|
public const long RamSize = 4L * 1024 * 1024 * 1024;
|
|
|
|
|
2019-02-24 07:24:35 +00:00
|
|
|
public IntPtr RamPointer { get; }
|
2018-08-15 18:59:51 +00:00
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
private unsafe byte* _ramPtr;
|
2018-08-15 18:59:51 +00:00
|
|
|
|
|
|
|
public unsafe DeviceMemory()
|
|
|
|
{
|
2019-02-24 07:24:35 +00:00
|
|
|
RamPointer = MemoryManagement.AllocateWriteTracked(RamSize);
|
2018-08-15 18:59:51 +00:00
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
_ramPtr = (byte*)RamPointer;
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public sbyte ReadSByte(long position)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
return (sbyte)ReadByte(position);
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public short ReadInt16(long position)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
return (short)ReadUInt16(position);
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public int ReadInt32(long position)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
return (int)ReadUInt32(position);
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public long ReadInt64(long position)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
return (long)ReadUInt64(position);
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public unsafe byte ReadByte(long position)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
return *(_ramPtr + position);
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public unsafe ushort ReadUInt16(long position)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
return *((ushort*)(_ramPtr + position));
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public unsafe uint ReadUInt32(long position)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
return *((uint*)(_ramPtr + position));
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public unsafe ulong ReadUInt64(long position)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
return *((ulong*)(_ramPtr + position));
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public void WriteSByte(long position, sbyte value)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
WriteByte(position, (byte)value);
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public void WriteInt16(long position, short value)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
WriteUInt16(position, (ushort)value);
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public void WriteInt32(long position, int value)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
WriteUInt32(position, (uint)value);
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public void WriteInt64(long position, long value)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
WriteUInt64(position, (ulong)value);
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public unsafe void WriteByte(long position, byte value)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
*(_ramPtr + position) = value;
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public unsafe void WriteUInt16(long position, ushort value)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
*((ushort*)(_ramPtr + position)) = value;
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public unsafe void WriteUInt32(long position, uint value)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
*((uint*)(_ramPtr + position)) = value;
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public unsafe void WriteUInt64(long position, ulong value)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
*((ulong*)(_ramPtr + position)) = value;
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2019-07-22 17:15:46 +00:00
|
|
|
public unsafe void WriteStruct<T>(long position, T value)
|
|
|
|
{
|
|
|
|
Marshal.StructureToPtr(value, (IntPtr)(_ramPtr + position), false);
|
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
public void FillWithZeros(long position, int size)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
int size8 = size & ~(8 - 1);
|
2018-08-15 18:59:51 +00:00
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
for (int offs = 0; offs < size8; offs += 8)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
WriteInt64(position + offs, 0);
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
for (int offs = size8; offs < (size - size8); offs++)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2018-12-06 11:16:24 +00:00
|
|
|
WriteByte(position + offs, 0);
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2019-01-18 22:26:39 +00:00
|
|
|
public void Set(ulong address, byte value, ulong size)
|
|
|
|
{
|
|
|
|
if (address + size < address)
|
|
|
|
{
|
|
|
|
throw new ArgumentOutOfRangeException(nameof(size));
|
|
|
|
}
|
|
|
|
|
|
|
|
if (address + size > RamSize)
|
|
|
|
{
|
|
|
|
throw new ArgumentOutOfRangeException(nameof(address));
|
|
|
|
}
|
|
|
|
|
|
|
|
ulong size8 = size & ~7UL;
|
|
|
|
|
|
|
|
ulong valueRep = (ulong)value * 0x0101010101010101;
|
|
|
|
|
|
|
|
for (ulong offs = 0; offs < size8; offs += 8)
|
|
|
|
{
|
|
|
|
WriteUInt64((long)(address + offs), valueRep);
|
|
|
|
}
|
|
|
|
|
|
|
|
for (ulong offs = size8; offs < (size - size8); offs++)
|
|
|
|
{
|
|
|
|
WriteByte((long)(address + offs), value);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
public void Copy(ulong dst, ulong src, ulong size)
|
|
|
|
{
|
|
|
|
if (dst + size < dst || src + size < src)
|
|
|
|
{
|
|
|
|
throw new ArgumentOutOfRangeException(nameof(size));
|
|
|
|
}
|
|
|
|
|
|
|
|
if (dst + size > RamSize)
|
|
|
|
{
|
|
|
|
throw new ArgumentOutOfRangeException(nameof(dst));
|
|
|
|
}
|
|
|
|
|
|
|
|
if (src + size > RamSize)
|
|
|
|
{
|
|
|
|
throw new ArgumentOutOfRangeException(nameof(src));
|
|
|
|
}
|
|
|
|
|
|
|
|
ulong size8 = size & ~7UL;
|
|
|
|
|
|
|
|
for (ulong offs = 0; offs < size8; offs += 8)
|
|
|
|
{
|
|
|
|
WriteUInt64((long)(dst + offs), ReadUInt64((long)(src + offs)));
|
|
|
|
}
|
|
|
|
|
|
|
|
for (ulong offs = size8; offs < (size - size8); offs++)
|
|
|
|
{
|
|
|
|
WriteByte((long)(dst + offs), ReadByte((long)(src + offs)));
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-08-15 18:59:51 +00:00
|
|
|
public void Dispose()
|
|
|
|
{
|
|
|
|
Dispose(true);
|
|
|
|
}
|
|
|
|
|
2018-12-06 11:16:24 +00:00
|
|
|
protected virtual void Dispose(bool disposing)
|
2018-08-15 18:59:51 +00:00
|
|
|
{
|
2019-02-24 07:24:35 +00:00
|
|
|
MemoryManagement.Free(RamPointer);
|
2018-08-15 18:59:51 +00:00
|
|
|
}
|
|
|
|
}
|
2019-07-22 17:15:46 +00:00
|
|
|
}
|