mirror of
https://github.com/Ryujinx/Ryujinx.git
synced 2024-12-26 19:01:22 -08:00
5e724cf24e
* Delete DelegateTypes.cs * Delete DelegateCache.cs * Add files via upload * Update Horizon.cs * Update Program.cs * Update MainWindow.cs * Update Aot.cs * Update RelocEntry.cs * Update Translator.cs * Update MemoryManager.cs * Update InstEmitMemoryHelper.cs * Update Delegates.cs * Nit. * Nit. * Nit. * 10 fewer MSIL bytes for us * Add comment. Nits. * Update Translator.cs * Update Aot.cs * Nits. * Opt.. * Opt.. * Opt.. * Opt.. * Allow to change compression level. * Update MemoryManager.cs * Update Translator.cs * Manage corner cases during the save phase. Nits. * Update Aot.cs * Translator response tweak for Aot disabled. Nit. * Nit. * Nits. * Create DelegateHelpers.cs * Update Delegates.cs * Nit. * Nit. * Nits. * Fix due to #784. * Fixes due to #757 & #841. * Fix due to #846. * Fix due to #847. * Use MethodInfo for managed method calls. Use IR methods instead of managed methods about Max/Min (S/U). Follow-ups & Nits. * Add missing exception messages. Reintroduce slow path for Fmov_Vi. Implement slow path for Fmov_Si. * Switch to the new folder structure. Nits. * Impl. index-based relocation information. Impl. cache file version field. * Nit. * Address gdkchan comments. Mainly: - fixed cache file corruption issue on exit; - exposed a way to disable AOT on the GUI. * Address AcK77 comment. * Address Thealexbarney, jduncanator & emmauss comments. Header magic, CpuId (FI) & Aot -> Ptc. * Adaptation to the new application reloading system. Improvements to the call system of managed methods. Follow-ups. Nits. * Get the same boot times as on master when PTC is disabled. * Profiled Aot. * A32 support (#897). * #975 support (1 of 2). * #975 support (2 of 2). * Rebase fix & nits. * Some fixes and nits (still one bug left). * One fix & nits. * Tests fix (by gdk) & nits. * Support translations not only in high quality and rejit. Nits. * Added possibility to skip translations and continue execution, using `ESC` key. * Update SettingsWindow.cs * Update GLRenderer.cs * Update Ptc.cs * Disabled Profiled PTC by default as requested in the past by gdk. * Fix rejit bug. Increased number of parallel translations. Add stack unwinding stuffs support (1 of 2). Nits. * Add stack unwinding stuffs support (2 of 2). Tuned number of parallel translations. * Restored the ability to assemble jumps with 8-bit offset when Profiled PTC is disabled or during profiling. Modifications due to rebase. Nits. * Limited profiling of the functions to be translated to the addresses belonging to the range of static objects only. * Nits. * Nits. * Update Delegates.cs * Nit. * Update InstEmitSimdArithmetic.cs * Address riperiperi comments. * Fixed the issue of unjustifiably longer boot times at the second boot than at the first boot, measured at the same time or reference point and with the same number of translated functions. * Implemented a simple redundant load/save mechanism. Halved the value of Decoder.MaxInstsPerFunction more appropriate for the current performance of the Translator. Replaced by Logger.PrintError to Logger.PrintDebug in TexturePool.cs about the supposed invalid texture format to avoid the spawn of the log. Nits. * Nit. Improved Logger.PrintError in TexturePool.cs to avoid log spawn. Added missing code for FZ handling (in output) for fp max/min instructions (slow paths). * Add configuration migration for PTC Co-authored-by: Thog <me@thog.eu>
339 lines
11 KiB
C#
339 lines
11 KiB
C#
using ARMeilleure.CodeGen.RegisterAllocators;
|
|
using ARMeilleure.Common;
|
|
using ARMeilleure.IntermediateRepresentation;
|
|
using ARMeilleure.Translation.PTC;
|
|
using System.Collections.Generic;
|
|
using System.Diagnostics;
|
|
using System.IO;
|
|
|
|
namespace ARMeilleure.CodeGen.X86
|
|
{
|
|
class CodeGenContext
|
|
{
|
|
private const int ReservedBytesForJump = 1;
|
|
|
|
private Stream _stream;
|
|
|
|
private PtcInfo _ptcInfo;
|
|
private bool _ptcDisabled;
|
|
|
|
public int StreamOffset => (int)_stream.Length;
|
|
|
|
public AllocationResult AllocResult { get; }
|
|
|
|
public Assembler Assembler { get; }
|
|
|
|
public BasicBlock CurrBlock { get; private set; }
|
|
|
|
public int CallArgsRegionSize { get; }
|
|
public int XmmSaveRegionSize { get; }
|
|
|
|
private long[] _blockOffsets;
|
|
|
|
private struct Jump
|
|
{
|
|
public bool IsConditional { get; }
|
|
|
|
public X86Condition Condition { get; }
|
|
|
|
public BasicBlock Target { get; }
|
|
|
|
public long JumpPosition { get; }
|
|
|
|
public long RelativeOffset { get; set; }
|
|
|
|
public int InstSize { get; set; }
|
|
|
|
public Jump(BasicBlock target, long jumpPosition, int instSize = 0)
|
|
{
|
|
IsConditional = false;
|
|
Condition = 0;
|
|
Target = target;
|
|
JumpPosition = jumpPosition;
|
|
|
|
RelativeOffset = 0;
|
|
|
|
InstSize = instSize;
|
|
}
|
|
|
|
public Jump(X86Condition condition, BasicBlock target, long jumpPosition, int instSize = 0)
|
|
{
|
|
IsConditional = true;
|
|
Condition = condition;
|
|
Target = target;
|
|
JumpPosition = jumpPosition;
|
|
|
|
RelativeOffset = 0;
|
|
|
|
InstSize = instSize;
|
|
}
|
|
}
|
|
|
|
private List<Jump> _jumps;
|
|
|
|
private X86Condition _jNearCondition;
|
|
|
|
private long _jNearPosition;
|
|
private int _jNearLength;
|
|
|
|
public CodeGenContext(Stream stream, AllocationResult allocResult, int maxCallArgs, int blocksCount, PtcInfo ptcInfo = null)
|
|
{
|
|
_stream = stream;
|
|
|
|
AllocResult = allocResult;
|
|
|
|
Assembler = new Assembler(stream, ptcInfo);
|
|
|
|
CallArgsRegionSize = GetCallArgsRegionSize(allocResult, maxCallArgs, out int xmmSaveRegionSize);
|
|
XmmSaveRegionSize = xmmSaveRegionSize;
|
|
|
|
_blockOffsets = new long[blocksCount];
|
|
|
|
_jumps = new List<Jump>();
|
|
|
|
_ptcInfo = ptcInfo;
|
|
_ptcDisabled = ptcInfo == null;
|
|
}
|
|
|
|
private int GetCallArgsRegionSize(AllocationResult allocResult, int maxCallArgs, out int xmmSaveRegionSize)
|
|
{
|
|
// We need to add 8 bytes to the total size, as the call to this
|
|
// function already pushed 8 bytes (the return address).
|
|
int intMask = CallingConvention.GetIntCalleeSavedRegisters() & allocResult.IntUsedRegisters;
|
|
int vecMask = CallingConvention.GetVecCalleeSavedRegisters() & allocResult.VecUsedRegisters;
|
|
|
|
xmmSaveRegionSize = BitUtils.CountBits(vecMask) * 16;
|
|
|
|
int calleeSaveRegionSize = BitUtils.CountBits(intMask) * 8 + xmmSaveRegionSize + 8;
|
|
|
|
int argsCount = maxCallArgs;
|
|
|
|
if (argsCount < 0)
|
|
{
|
|
// When the function has no calls, argsCount is -1.
|
|
// In this case, we don't need to allocate the shadow space.
|
|
argsCount = 0;
|
|
}
|
|
else if (argsCount < 4)
|
|
{
|
|
// The ABI mandates that the space for at least 4 arguments
|
|
// is reserved on the stack (this is called shadow space).
|
|
argsCount = 4;
|
|
}
|
|
|
|
int frameSize = calleeSaveRegionSize + allocResult.SpillRegionSize;
|
|
|
|
// TODO: Instead of always multiplying by 16 (the largest possible size of a variable,
|
|
// since a V128 has 16 bytes), we should calculate the exact size consumed by the
|
|
// arguments passed to the called functions on the stack.
|
|
int callArgsAndFrameSize = frameSize + argsCount * 16;
|
|
|
|
// Ensure that the Stack Pointer will be aligned to 16 bytes.
|
|
callArgsAndFrameSize = (callArgsAndFrameSize + 0xf) & ~0xf;
|
|
|
|
return callArgsAndFrameSize - frameSize;
|
|
}
|
|
|
|
public void EnterBlock(BasicBlock block)
|
|
{
|
|
_blockOffsets[block.Index] = _stream.Position;
|
|
|
|
CurrBlock = block;
|
|
}
|
|
|
|
public void JumpTo(BasicBlock target)
|
|
{
|
|
if (_ptcDisabled)
|
|
{
|
|
_jumps.Add(new Jump(target, _stream.Position));
|
|
|
|
WritePadding(ReservedBytesForJump);
|
|
}
|
|
else
|
|
{
|
|
_jumps.Add(new Jump(target, _stream.Position, 5));
|
|
|
|
WritePadding(5);
|
|
}
|
|
}
|
|
|
|
public void JumpTo(X86Condition condition, BasicBlock target)
|
|
{
|
|
if (_ptcDisabled)
|
|
{
|
|
_jumps.Add(new Jump(condition, target, _stream.Position));
|
|
|
|
WritePadding(ReservedBytesForJump);
|
|
}
|
|
else
|
|
{
|
|
_jumps.Add(new Jump(condition, target, _stream.Position, 6));
|
|
|
|
WritePadding(6);
|
|
}
|
|
}
|
|
|
|
public void JumpToNear(X86Condition condition)
|
|
{
|
|
_jNearCondition = condition;
|
|
_jNearPosition = _stream.Position;
|
|
_jNearLength = Assembler.GetJccLength(0, _ptcDisabled);
|
|
|
|
_stream.Seek(_jNearLength, SeekOrigin.Current);
|
|
}
|
|
|
|
public void JumpHere()
|
|
{
|
|
long currentPosition = _stream.Position;
|
|
|
|
_stream.Seek(_jNearPosition, SeekOrigin.Begin);
|
|
|
|
long offset = currentPosition - (_jNearPosition + _jNearLength);
|
|
|
|
Debug.Assert(_jNearLength == Assembler.GetJccLength(offset, _ptcDisabled), "Relative offset doesn't fit on near jump.");
|
|
|
|
Assembler.Jcc(_jNearCondition, offset);
|
|
|
|
_stream.Seek(currentPosition, SeekOrigin.Begin);
|
|
}
|
|
|
|
private void WritePadding(int size)
|
|
{
|
|
while (size-- > 0)
|
|
{
|
|
_stream.WriteByte(0);
|
|
}
|
|
}
|
|
|
|
public byte[] GetCode()
|
|
{
|
|
// Write jump relative offsets.
|
|
bool modified;
|
|
|
|
do
|
|
{
|
|
modified = false;
|
|
|
|
for (int index = 0; index < _jumps.Count; index++)
|
|
{
|
|
Jump jump = _jumps[index];
|
|
|
|
long jumpTarget = _blockOffsets[jump.Target.Index];
|
|
|
|
long offset = jumpTarget - jump.JumpPosition;
|
|
|
|
if (_ptcDisabled)
|
|
{
|
|
if (offset < 0)
|
|
{
|
|
for (int index2 = index - 1; index2 >= 0; index2--)
|
|
{
|
|
Jump jump2 = _jumps[index2];
|
|
|
|
if (jump2.JumpPosition < jumpTarget)
|
|
{
|
|
break;
|
|
}
|
|
|
|
offset -= jump2.InstSize - ReservedBytesForJump;
|
|
}
|
|
}
|
|
else
|
|
{
|
|
for (int index2 = index + 1; index2 < _jumps.Count; index2++)
|
|
{
|
|
Jump jump2 = _jumps[index2];
|
|
|
|
if (jump2.JumpPosition >= jumpTarget)
|
|
{
|
|
break;
|
|
}
|
|
|
|
offset += jump2.InstSize - ReservedBytesForJump;
|
|
}
|
|
|
|
offset -= ReservedBytesForJump;
|
|
}
|
|
|
|
if (jump.IsConditional)
|
|
{
|
|
jump.InstSize = Assembler.GetJccLength(offset);
|
|
}
|
|
else
|
|
{
|
|
jump.InstSize = Assembler.GetJmpLength(offset);
|
|
}
|
|
|
|
// The jump is relative to the next instruction, not the current one.
|
|
// Since we didn't know the next instruction address when calculating
|
|
// the offset (as the size of the current jump instruction was not known),
|
|
// we now need to compensate the offset with the jump instruction size.
|
|
// It's also worth noting that:
|
|
// - This is only needed for backward jumps.
|
|
// - The GetJmpLength and GetJccLength also compensates the offset
|
|
// internally when computing the jump instruction size.
|
|
if (offset < 0)
|
|
{
|
|
offset -= jump.InstSize;
|
|
}
|
|
}
|
|
else
|
|
{
|
|
offset -= jump.InstSize;
|
|
}
|
|
|
|
if (jump.RelativeOffset != offset)
|
|
{
|
|
modified = true;
|
|
}
|
|
|
|
jump.RelativeOffset = offset;
|
|
|
|
_jumps[index] = jump;
|
|
}
|
|
}
|
|
while (modified);
|
|
|
|
// Write the code, ignoring the dummy bytes after jumps, into a new stream.
|
|
_stream.Seek(0, SeekOrigin.Begin);
|
|
|
|
using (MemoryStream codeStream = new MemoryStream())
|
|
{
|
|
Assembler assembler = new Assembler(codeStream, _ptcInfo);
|
|
|
|
byte[] buffer;
|
|
|
|
for (int index = 0; index < _jumps.Count; index++)
|
|
{
|
|
Jump jump = _jumps[index];
|
|
|
|
buffer = new byte[jump.JumpPosition - _stream.Position];
|
|
|
|
_stream.Read(buffer, 0, buffer.Length);
|
|
_stream.Seek(_ptcDisabled ? ReservedBytesForJump : jump.InstSize, SeekOrigin.Current);
|
|
|
|
codeStream.Write(buffer);
|
|
|
|
if (jump.IsConditional)
|
|
{
|
|
assembler.Jcc(jump.Condition, jump.RelativeOffset);
|
|
}
|
|
else
|
|
{
|
|
assembler.Jmp(jump.RelativeOffset);
|
|
}
|
|
}
|
|
|
|
buffer = new byte[_stream.Length - _stream.Position];
|
|
|
|
_stream.Read(buffer, 0, buffer.Length);
|
|
|
|
codeStream.Write(buffer);
|
|
|
|
_ptcInfo?.WriteCode(codeStream);
|
|
|
|
return codeStream.ToArray();
|
|
}
|
|
}
|
|
}
|
|
} |