Ryujinx/ARMeilleure/Instructions/NativeInterface.cs
riperiperi d904706fc0
Use a Jump Table for direct and indirect calls/jumps, removing transitions to managed (#975)
* Implement Jump Table for Native Calls

NOTE: this slows down rejit considerably! Not recommended to be used
without codegen optimisation or AOT.

- Does not work on Linux
- A32 needs an additional commit.

* A32 Support

(WIP)

* Actually write Direct Call pointers to the table

That would help.

* Direct Calls: Rather than returning to the translator, attempt to keep within the native stack frame.

A return to the translator can still happen, but only by exceptionally
bubbling up to it.

Also:
- Always translate lowCq as a function. Faster interop with the direct
jumps, and this will be useful in future if we want to do speculative
translation.
- Tail Call Detection: after the decoding stage, detect if we do a tail
call, and avoid translating into it. Detected if a jump is made to an
address outwith the contiguous sequence of blocks surrounding the entry
point. The goal is to reduce code touched by jit and rejit.

* A32 Support

* Use smaller max function size for lowCq, fix exceptional returns

When a return has an unexpected value and there is no code block
following this one, we now return the value rather than continuing.

* CompareAndSwap (buggy)

* Ensure CompareAndSwap does not get optimized away.

* Use CompareAndSwap to make the dynamic table thread safe.

* Tail call for linux, throw on too many arguments.

* Combine CompareAndSwap 128 and 32/64.

They emit different IR instructions since their PreAllocator behaviour
is different, but now they just have one function on EmitterContext.

* Fix issues separating from optimisations.

* Use a stub to find and execute missing functions.

This allows us to skip doing many runtime comparisons and branches, and reduces the amount of code we need to emit significantly.

For the indirect call table, this stub also does the work of moving in the highCq address to the table when one is found.

* Make Jump Tables and Jit Cache dynmically resize

Reserve virtual memory, commit as needed.

* Move TailCallRemover to its own class.

* Multithreaded Translation (based on heuristic)

A poor one, at that. Need to get core count for a better one, which
means a lot of OS specific garbage.

* Better priority management for background threads.

* Bound core limit a bit more

Past a certain point the load is not paralellizable and starts stealing from the main thread. Likely due to GC, memory, heap allocation thread contention. Reduce by one core til optimisations come to improve the situation.

* Fix memory management on linux.

* Temporary solution to some sync problems.

This will make sure threads exit correctly, most of the time. There is a potential race where setting the sync counter to 0 does nothing (counter stays at what it was before, thread could take too long to exit), but we need to find a better way to do this anyways. Synchronization frequency has been tightened as we never enter blockwise segments of code. Essentially this means, check every x functions or loop iterations, before lowcq blocks existed and were worth just as much. Ideally it should be done in a better way, since functions can be anywhere from 1 to 5000 instructions. (maybe based on host timer, or an interrupt flag from a scheduler thread)

* Address feedback minus CompareAndSwap change.

* Use default ReservedRegion granularity.

* Merge CompareAndSwap with its V128 variant.

* We already got the source, no need to do it again.

* Make sure all background translation threads exit.

* Fix CompareAndSwap128

Detection criteria was a bit scuffed.

* Address Comments.
2020-03-12 14:20:55 +11:00

433 lines
12 KiB
C#

using ARMeilleure.Memory;
using ARMeilleure.State;
using ARMeilleure.Translation;
using System;
using System.Runtime.InteropServices;
namespace ARMeilleure.Instructions
{
static class NativeInterface
{
private const int ErgSizeLog2 = 4;
private class ThreadContext
{
public ExecutionContext Context { get; }
public MemoryManager Memory { get; }
public Translator Translator { get; }
public ulong ExclusiveAddress { get; set; }
public ulong ExclusiveValueLow { get; set; }
public ulong ExclusiveValueHigh { get; set; }
public ThreadContext(ExecutionContext context, MemoryManager memory, Translator translator)
{
Context = context;
Memory = memory;
Translator = translator;
ExclusiveAddress = ulong.MaxValue;
}
}
[ThreadStatic]
private static ThreadContext _context;
public static void RegisterThread(ExecutionContext context, MemoryManager memory, Translator translator)
{
_context = new ThreadContext(context, memory, translator);
}
public static void UnregisterThread()
{
_context = null;
}
public static void Break(ulong address, int imm)
{
Statistics.PauseTimer();
GetContext().OnBreak(address, imm);
Statistics.ResumeTimer();
}
public static void SupervisorCall(ulong address, int imm)
{
Statistics.PauseTimer();
GetContext().OnSupervisorCall(address, imm);
Statistics.ResumeTimer();
}
public static void Undefined(ulong address, int opCode)
{
Statistics.PauseTimer();
GetContext().OnUndefined(address, opCode);
Statistics.ResumeTimer();
}
#region "System registers"
public static ulong GetCtrEl0()
{
return (ulong)GetContext().CtrEl0;
}
public static ulong GetDczidEl0()
{
return (ulong)GetContext().DczidEl0;
}
public static ulong GetFpcr()
{
return (ulong)GetContext().Fpcr;
}
public static ulong GetFpsr()
{
return (ulong)GetContext().Fpsr;
}
public static uint GetFpscr()
{
ExecutionContext context = GetContext();
uint result = (uint)(context.Fpsr & FPSR.A32Mask) | (uint)(context.Fpcr & FPCR.A32Mask);
result |= context.GetFPstateFlag(FPState.NFlag) ? (1u << 31) : 0;
result |= context.GetFPstateFlag(FPState.ZFlag) ? (1u << 30) : 0;
result |= context.GetFPstateFlag(FPState.CFlag) ? (1u << 29) : 0;
result |= context.GetFPstateFlag(FPState.VFlag) ? (1u << 28) : 0;
return result;
}
public static ulong GetTpidrEl0()
{
return (ulong)GetContext().TpidrEl0;
}
public static uint GetTpidrEl032()
{
return (uint)GetContext().TpidrEl0;
}
public static ulong GetTpidr()
{
return (ulong)GetContext().Tpidr;
}
public static uint GetTpidr32()
{
return (uint)GetContext().Tpidr;
}
public static ulong GetCntfrqEl0()
{
return GetContext().CntfrqEl0;
}
public static ulong GetCntpctEl0()
{
return GetContext().CntpctEl0;
}
public static void SetFpcr(ulong value)
{
GetContext().Fpcr = (FPCR)value;
}
public static void SetFpsr(ulong value)
{
GetContext().Fpsr = (FPSR)value;
}
public static void SetFpscr(uint value)
{
ExecutionContext context = GetContext();
context.SetFPstateFlag(FPState.NFlag, (value & (1u << 31)) != 0);
context.SetFPstateFlag(FPState.ZFlag, (value & (1u << 30)) != 0);
context.SetFPstateFlag(FPState.CFlag, (value & (1u << 29)) != 0);
context.SetFPstateFlag(FPState.VFlag, (value & (1u << 28)) != 0);
context.Fpsr = FPSR.A32Mask & (FPSR)value;
context.Fpcr = FPCR.A32Mask & (FPCR)value;
}
public static void SetTpidrEl0(ulong value)
{
GetContext().TpidrEl0 = (long)value;
}
public static void SetTpidrEl032(uint value)
{
GetContext().TpidrEl0 = (long)value;
}
#endregion
#region "Read"
public static byte ReadByte(ulong address)
{
return GetMemoryManager().ReadByte((long)address);
}
public static ushort ReadUInt16(ulong address)
{
return GetMemoryManager().ReadUInt16((long)address);
}
public static uint ReadUInt32(ulong address)
{
return GetMemoryManager().ReadUInt32((long)address);
}
public static ulong ReadUInt64(ulong address)
{
return GetMemoryManager().ReadUInt64((long)address);
}
public static V128 ReadVector128(ulong address)
{
return GetMemoryManager().ReadVector128((long)address);
}
#endregion
#region "Read exclusive"
public static byte ReadByteExclusive(ulong address)
{
byte value = _context.Memory.ReadByte((long)address);
_context.ExclusiveAddress = GetMaskedExclusiveAddress(address);
_context.ExclusiveValueLow = value;
_context.ExclusiveValueHigh = 0;
return value;
}
public static ushort ReadUInt16Exclusive(ulong address)
{
ushort value = _context.Memory.ReadUInt16((long)address);
_context.ExclusiveAddress = GetMaskedExclusiveAddress(address);
_context.ExclusiveValueLow = value;
_context.ExclusiveValueHigh = 0;
return value;
}
public static uint ReadUInt32Exclusive(ulong address)
{
uint value = _context.Memory.ReadUInt32((long)address);
_context.ExclusiveAddress = GetMaskedExclusiveAddress(address);
_context.ExclusiveValueLow = value;
_context.ExclusiveValueHigh = 0;
return value;
}
public static ulong ReadUInt64Exclusive(ulong address)
{
ulong value = _context.Memory.ReadUInt64((long)address);
_context.ExclusiveAddress = GetMaskedExclusiveAddress(address);
_context.ExclusiveValueLow = value;
_context.ExclusiveValueHigh = 0;
return value;
}
public static V128 ReadVector128Exclusive(ulong address)
{
V128 value = _context.Memory.AtomicLoadInt128((long)address);
_context.ExclusiveAddress = GetMaskedExclusiveAddress(address);
_context.ExclusiveValueLow = value.GetUInt64(0);
_context.ExclusiveValueHigh = value.GetUInt64(1);
return value;
}
#endregion
#region "Write"
public static void WriteByte(ulong address, byte value)
{
GetMemoryManager().WriteByte((long)address, value);
}
public static void WriteUInt16(ulong address, ushort value)
{
GetMemoryManager().WriteUInt16((long)address, value);
}
public static void WriteUInt32(ulong address, uint value)
{
GetMemoryManager().WriteUInt32((long)address, value);
}
public static void WriteUInt64(ulong address, ulong value)
{
GetMemoryManager().WriteUInt64((long)address, value);
}
public static void WriteVector128(ulong address, V128 value)
{
GetMemoryManager().WriteVector128((long)address, value);
}
#endregion
#region "Write exclusive"
public static int WriteByteExclusive(ulong address, byte value)
{
bool success = _context.ExclusiveAddress == GetMaskedExclusiveAddress(address);
if (success)
{
success = _context.Memory.AtomicCompareExchangeByte(
(long)address,
(byte)_context.ExclusiveValueLow,
(byte)value);
if (success)
{
ClearExclusive();
}
}
return success ? 0 : 1;
}
public static int WriteUInt16Exclusive(ulong address, ushort value)
{
bool success = _context.ExclusiveAddress == GetMaskedExclusiveAddress(address);
if (success)
{
success = _context.Memory.AtomicCompareExchangeInt16(
(long)address,
(short)_context.ExclusiveValueLow,
(short)value);
if (success)
{
ClearExclusive();
}
}
return success ? 0 : 1;
}
public static int WriteUInt32Exclusive(ulong address, uint value)
{
bool success = _context.ExclusiveAddress == GetMaskedExclusiveAddress(address);
if (success)
{
success = _context.Memory.AtomicCompareExchangeInt32(
(long)address,
(int)_context.ExclusiveValueLow,
(int)value);
if (success)
{
ClearExclusive();
}
}
return success ? 0 : 1;
}
public static int WriteUInt64Exclusive(ulong address, ulong value)
{
bool success = _context.ExclusiveAddress == GetMaskedExclusiveAddress(address);
if (success)
{
success = _context.Memory.AtomicCompareExchangeInt64(
(long)address,
(long)_context.ExclusiveValueLow,
(long)value);
if (success)
{
ClearExclusive();
}
}
return success ? 0 : 1;
}
public static int WriteVector128Exclusive(ulong address, V128 value)
{
bool success = _context.ExclusiveAddress == GetMaskedExclusiveAddress(address);
if (success)
{
V128 expected = new V128(_context.ExclusiveValueLow, _context.ExclusiveValueHigh);
success = _context.Memory.AtomicCompareExchangeInt128((long)address, expected, value);
if (success)
{
ClearExclusive();
}
}
return success ? 0 : 1;
}
#endregion
private static ulong GetMaskedExclusiveAddress(ulong address)
{
return address & ~((4UL << ErgSizeLog2) - 1);
}
public static ulong GetFunctionAddress(ulong address)
{
TranslatedFunction function = _context.Translator.GetOrTranslate(address, GetContext().ExecutionMode);
return (ulong)function.GetPointer().ToInt64();
}
public static ulong GetIndirectFunctionAddress(ulong address, ulong entryAddress)
{
TranslatedFunction function = _context.Translator.GetOrTranslate(address, GetContext().ExecutionMode);
ulong ptr = (ulong)function.GetPointer().ToInt64();
if (function.HighCq)
{
// Rewrite the host function address in the table to point to the highCq function.
Marshal.WriteInt64((IntPtr)entryAddress, 8, (long)ptr);
}
return ptr;
}
public static void ClearExclusive()
{
_context.ExclusiveAddress = ulong.MaxValue;
}
public static bool CheckSynchronization()
{
Statistics.PauseTimer();
ExecutionContext context = GetContext();
context.CheckInterrupt();
Statistics.ResumeTimer();
return context.Running;
}
public static ExecutionContext GetContext()
{
return _context.Context;
}
public static MemoryManager GetMemoryManager()
{
return _context.Memory;
}
}
}