Ryujinx/Ryujinx.HLE/HOS/Kernel/SupervisorCall/SyscallTable.cs

495 lines
23 KiB
C#
Raw Normal View History

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 20:56:22 +02:00
using ARMeilleure.State;
using Ryujinx.Common.Logging;
using Ryujinx.HLE.HOS.Kernel.Common;
using System;
using System.Collections.Generic;
using System.Reflection;
using System.Reflection.Emit;
namespace Ryujinx.HLE.HOS.Kernel.SupervisorCall
{
static class SyscallTable
{
private const int SvcFuncMaxArguments64 = 8;
private const int SvcFuncMaxArguments32 = 4;
private const int SvcMax = 0x80;
public static Action<Syscall32, ExecutionContext>[] SvcTable32 { get; }
public static Action<Syscall64, ExecutionContext>[] SvcTable64 { get; }
static SyscallTable()
{
SvcTable32 = new Action<Syscall32, ExecutionContext>[SvcMax];
SvcTable64 = new Action<Syscall64, ExecutionContext>[SvcMax];
Dictionary<int, string> svcFuncs64 = new Dictionary<int, string>
{
{ 0x01, nameof(Syscall64.SetHeapSize64) },
{ 0x02, nameof(Syscall64.SetMemoryPermission64) },
{ 0x03, nameof(Syscall64.SetMemoryAttribute64) },
{ 0x04, nameof(Syscall64.MapMemory64) },
{ 0x05, nameof(Syscall64.UnmapMemory64) },
{ 0x06, nameof(Syscall64.QueryMemory64) },
{ 0x07, nameof(Syscall64.ExitProcess64) },
{ 0x08, nameof(Syscall64.CreateThread64) },
{ 0x09, nameof(Syscall64.StartThread64) },
{ 0x0a, nameof(Syscall64.ExitThread64) },
{ 0x0b, nameof(Syscall64.SleepThread64) },
{ 0x0c, nameof(Syscall64.GetThreadPriority64) },
{ 0x0d, nameof(Syscall64.SetThreadPriority64) },
{ 0x0e, nameof(Syscall64.GetThreadCoreMask64) },
{ 0x0f, nameof(Syscall64.SetThreadCoreMask64) },
{ 0x10, nameof(Syscall64.GetCurrentProcessorNumber64) },
{ 0x11, nameof(Syscall64.SignalEvent64) },
{ 0x12, nameof(Syscall64.ClearEvent64) },
{ 0x13, nameof(Syscall64.MapSharedMemory64) },
{ 0x14, nameof(Syscall64.UnmapSharedMemory64) },
{ 0x15, nameof(Syscall64.CreateTransferMemory64) },
{ 0x16, nameof(Syscall64.CloseHandle64) },
{ 0x17, nameof(Syscall64.ResetSignal64) },
{ 0x18, nameof(Syscall64.WaitSynchronization64) },
{ 0x19, nameof(Syscall64.CancelSynchronization64) },
{ 0x1a, nameof(Syscall64.ArbitrateLock64) },
{ 0x1b, nameof(Syscall64.ArbitrateUnlock64) },
{ 0x1c, nameof(Syscall64.WaitProcessWideKeyAtomic64) },
{ 0x1d, nameof(Syscall64.SignalProcessWideKey64) },
{ 0x1e, nameof(Syscall64.GetSystemTick64) },
{ 0x1f, nameof(Syscall64.ConnectToNamedPort64) },
{ 0x21, nameof(Syscall64.SendSyncRequest64) },
{ 0x22, nameof(Syscall64.SendSyncRequestWithUserBuffer64) },
{ 0x23, nameof(Syscall64.SendAsyncRequestWithUserBuffer64) },
{ 0x24, nameof(Syscall64.GetProcessId64) },
{ 0x25, nameof(Syscall64.GetThreadId64) },
{ 0x26, nameof(Syscall64.Break64) },
{ 0x27, nameof(Syscall64.OutputDebugString64) },
{ 0x29, nameof(Syscall64.GetInfo64) },
{ 0x2c, nameof(Syscall64.MapPhysicalMemory64) },
{ 0x2d, nameof(Syscall64.UnmapPhysicalMemory64) },
{ 0x30, nameof(Syscall64.GetResourceLimitLimitValue64) },
{ 0x31, nameof(Syscall64.GetResourceLimitCurrentValue64) },
{ 0x32, nameof(Syscall64.SetThreadActivity64) },
{ 0x33, nameof(Syscall64.GetThreadContext364) },
{ 0x34, nameof(Syscall64.WaitForAddress64) },
{ 0x35, nameof(Syscall64.SignalToAddress64) },
{ 0x36, nameof(Syscall64.SynchronizePreemptionState64) },
{ 0x37, nameof(Syscall64.GetResourceLimitPeakValue64) },
{ 0x40, nameof(Syscall64.CreateSession64) },
{ 0x41, nameof(Syscall64.AcceptSession64) },
{ 0x43, nameof(Syscall64.ReplyAndReceive64) },
{ 0x44, nameof(Syscall64.ReplyAndReceiveWithUserBuffer64) },
{ 0x45, nameof(Syscall64.CreateEvent64) },
{ 0x4b, nameof(Syscall64.CreateCodeMemory64) },
{ 0x4c, nameof(Syscall64.ControlCodeMemory64) },
{ 0x51, nameof(Syscall64.MapTransferMemory64) },
{ 0x52, nameof(Syscall64.UnmapTransferMemory64) },
{ 0x65, nameof(Syscall64.GetProcessList64) },
{ 0x6f, nameof(Syscall64.GetSystemInfo64) },
{ 0x70, nameof(Syscall64.CreatePort64) },
{ 0x71, nameof(Syscall64.ManageNamedPort64) },
{ 0x72, nameof(Syscall64.ConnectToPort64) },
{ 0x73, nameof(Syscall64.SetProcessMemoryPermission64) },
{ 0x74, nameof(Syscall64.MapProcessMemory64) },
{ 0x75, nameof(Syscall64.UnmapProcessMemory64) },
{ 0x77, nameof(Syscall64.MapProcessCodeMemory64) },
{ 0x78, nameof(Syscall64.UnmapProcessCodeMemory64) },
{ 0x7B, nameof(Syscall64.TerminateProcess64) },
{ 0x7D, nameof(Syscall64.CreateResourceLimit64) },
{ 0x7E, nameof(Syscall64.SetResourceLimitLimitValue64) }
};
foreach (KeyValuePair<int, string> value in svcFuncs64)
{
SvcTable64[value.Key] = GenerateMethod<Syscall64>(value.Value, SvcFuncMaxArguments64);
}
Dictionary<int, string> svcFuncs32 = new Dictionary<int, string>
{
{ 0x01, nameof(Syscall32.SetHeapSize32) },
{ 0x02, nameof(Syscall32.SetMemoryPermission32) },
{ 0x03, nameof(Syscall32.SetMemoryAttribute32) },
{ 0x04, nameof(Syscall32.MapMemory32) },
{ 0x05, nameof(Syscall32.UnmapMemory32) },
{ 0x06, nameof(Syscall32.QueryMemory32) },
{ 0x07, nameof(Syscall32.ExitProcess32) },
{ 0x08, nameof(Syscall32.CreateThread32) },
{ 0x09, nameof(Syscall32.StartThread32) },
{ 0x0a, nameof(Syscall32.ExitThread32) },
{ 0x0b, nameof(Syscall32.SleepThread32) },
{ 0x0c, nameof(Syscall32.GetThreadPriority32) },
{ 0x0d, nameof(Syscall32.SetThreadPriority32) },
{ 0x0e, nameof(Syscall32.GetThreadCoreMask32) },
{ 0x0f, nameof(Syscall32.SetThreadCoreMask32) },
{ 0x10, nameof(Syscall32.GetCurrentProcessorNumber32) },
{ 0x11, nameof(Syscall32.SignalEvent32) },
{ 0x12, nameof(Syscall32.ClearEvent32) },
{ 0x13, nameof(Syscall32.MapSharedMemory32) },
{ 0x14, nameof(Syscall32.UnmapSharedMemory32) },
{ 0x15, nameof(Syscall32.CreateTransferMemory32) },
{ 0x16, nameof(Syscall32.CloseHandle32) },
{ 0x17, nameof(Syscall32.ResetSignal32) },
{ 0x18, nameof(Syscall32.WaitSynchronization32) },
{ 0x19, nameof(Syscall32.CancelSynchronization32) },
{ 0x1a, nameof(Syscall32.ArbitrateLock32) },
{ 0x1b, nameof(Syscall32.ArbitrateUnlock32) },
{ 0x1c, nameof(Syscall32.WaitProcessWideKeyAtomic32) },
{ 0x1d, nameof(Syscall32.SignalProcessWideKey32) },
{ 0x1e, nameof(Syscall32.GetSystemTick32) },
{ 0x1f, nameof(Syscall32.ConnectToNamedPort32) },
{ 0x21, nameof(Syscall32.SendSyncRequest32) },
{ 0x22, nameof(Syscall32.SendSyncRequestWithUserBuffer32) },
{ 0x24, nameof(Syscall32.GetProcessId32) },
{ 0x25, nameof(Syscall32.GetThreadId32) },
{ 0x26, nameof(Syscall32.Break32) },
{ 0x27, nameof(Syscall32.OutputDebugString32) },
{ 0x29, nameof(Syscall32.GetInfo32) },
{ 0x2c, nameof(Syscall32.MapPhysicalMemory32) },
{ 0x2d, nameof(Syscall32.UnmapPhysicalMemory32) },
{ 0x30, nameof(Syscall32.GetResourceLimitLimitValue32) },
{ 0x31, nameof(Syscall32.GetResourceLimitCurrentValue32) },
{ 0x32, nameof(Syscall32.SetThreadActivity32) },
{ 0x33, nameof(Syscall32.GetThreadContext332) },
{ 0x34, nameof(Syscall32.WaitForAddress32) },
{ 0x35, nameof(Syscall32.SignalToAddress32) },
{ 0x36, nameof(Syscall32.SynchronizePreemptionState32) },
{ 0x37, nameof(Syscall32.GetResourceLimitPeakValue32) },
{ 0x40, nameof(Syscall32.CreateSession32) },
{ 0x41, nameof(Syscall32.AcceptSession32) },
{ 0x43, nameof(Syscall32.ReplyAndReceive32) },
{ 0x45, nameof(Syscall32.CreateEvent32) },
{ 0x4b, nameof(Syscall32.CreateCodeMemory32) },
{ 0x4c, nameof(Syscall32.ControlCodeMemory32) },
{ 0x51, nameof(Syscall32.MapTransferMemory32) },
{ 0x52, nameof(Syscall32.UnmapTransferMemory32) },
{ 0x5F, nameof(Syscall32.FlushProcessDataCache32) },
{ 0x65, nameof(Syscall32.GetProcessList32) },
{ 0x6f, nameof(Syscall32.GetSystemInfo32) },
{ 0x70, nameof(Syscall32.CreatePort32) },
{ 0x71, nameof(Syscall32.ManageNamedPort32) },
{ 0x72, nameof(Syscall32.ConnectToPort32) },
{ 0x73, nameof(Syscall32.SetProcessMemoryPermission32) },
{ 0x74, nameof(Syscall32.MapProcessMemory32) },
{ 0x75, nameof(Syscall32.UnmapProcessMemory32) },
{ 0x77, nameof(Syscall32.MapProcessCodeMemory32) },
{ 0x78, nameof(Syscall32.UnmapProcessCodeMemory32) },
{ 0x7B, nameof(Syscall32.TerminateProcess32) },
{ 0x7D, nameof(Syscall32.CreateResourceLimit32) },
{ 0x7E, nameof(Syscall32.SetResourceLimitLimitValue32) }
};
foreach (KeyValuePair<int, string> value in svcFuncs32)
{
SvcTable32[value.Key] = GenerateMethod<Syscall32>(value.Value, SvcFuncMaxArguments32);
}
}
private static Action<T, ExecutionContext> GenerateMethod<T>(string svcName, int registerCleanCount)
{
Type[] argTypes = new Type[] { typeof(T), typeof(ExecutionContext) };
DynamicMethod method = new DynamicMethod(svcName, null, argTypes);
MethodInfo methodInfo = typeof(T).GetMethod(svcName);
ParameterInfo[] methodArgs = methodInfo.GetParameters();
ILGenerator generator = method.GetILGenerator();
void ConvertToArgType(Type sourceType)
{
CheckIfTypeIsSupported(sourceType, svcName);
switch (Type.GetTypeCode(sourceType))
{
case TypeCode.UInt32: generator.Emit(OpCodes.Conv_U4); break;
case TypeCode.Int32: generator.Emit(OpCodes.Conv_I4); break;
case TypeCode.UInt16: generator.Emit(OpCodes.Conv_U2); break;
case TypeCode.Int16: generator.Emit(OpCodes.Conv_I2); break;
case TypeCode.Byte: generator.Emit(OpCodes.Conv_U1); break;
case TypeCode.SByte: generator.Emit(OpCodes.Conv_I1); break;
case TypeCode.Boolean:
generator.Emit(OpCodes.Conv_I4);
generator.Emit(OpCodes.Ldc_I4_1);
generator.Emit(OpCodes.And);
break;
}
}
void ConvertToFieldType(Type sourceType)
{
CheckIfTypeIsSupported(sourceType, svcName);
switch (Type.GetTypeCode(sourceType))
{
case TypeCode.UInt32:
case TypeCode.Int32:
case TypeCode.UInt16:
case TypeCode.Int16:
case TypeCode.Byte:
case TypeCode.SByte:
case TypeCode.Boolean:
generator.Emit(OpCodes.Conv_U8);
break;
}
}
RAttribute GetRegisterAttribute(ParameterInfo parameterInfo)
{
RAttribute argumentAttribute = (RAttribute)parameterInfo.GetCustomAttribute(typeof(RAttribute));
if (argumentAttribute == null)
{
throw new InvalidOperationException($"Method \"{svcName}\" is missing a {typeof(RAttribute).Name} attribute on parameter \"{parameterInfo.Name}\"");
}
return argumentAttribute;
}
// For functions returning output values, the first registers
// are used to hold pointers where the value will be stored,
// so they can't be used to pass argument and we must
// skip them.
int byRefArgsCount = 0;
for (int index = 0; index < methodArgs.Length; index++)
{
if (methodArgs[index].ParameterType.IsByRef)
{
byRefArgsCount++;
}
}
BindingFlags staticNonPublic = BindingFlags.NonPublic | BindingFlags.Static;
// Print all the arguments for debugging purposes.
int inputArgsCount = methodArgs.Length - byRefArgsCount;
if (inputArgsCount != 0)
{
generator.Emit(OpCodes.Ldc_I4, inputArgsCount);
generator.Emit(OpCodes.Newarr, typeof(object));
string argsFormat = svcName;
for (int index = 0; index < methodArgs.Length; index++)
{
Type argType = methodArgs[index].ParameterType;
// Ignore out argument for printing
if (argType.IsByRef)
{
continue;
}
RAttribute registerAttribute = GetRegisterAttribute(methodArgs[index]);
argsFormat += $" {methodArgs[index].Name}: 0x{{{index}:X8}},";
generator.Emit(OpCodes.Dup);
generator.Emit(OpCodes.Ldc_I4, index);
generator.Emit(OpCodes.Ldarg_1);
generator.Emit(OpCodes.Ldc_I4, registerAttribute.Index);
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 20:56:22 +02:00
MethodInfo info = typeof(ExecutionContext).GetMethod(nameof(ExecutionContext.GetX));
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 20:56:22 +02:00
generator.Emit(OpCodes.Call, info);
generator.Emit(OpCodes.Box, typeof(ulong));
generator.Emit(OpCodes.Stelem_Ref);
}
argsFormat = argsFormat.Substring(0, argsFormat.Length - 1);
generator.Emit(OpCodes.Ldstr, argsFormat);
}
else
{
generator.Emit(OpCodes.Ldnull);
generator.Emit(OpCodes.Ldstr, svcName);
}
MethodInfo printArgsMethod = typeof(SyscallTable).GetMethod(nameof(PrintArguments), staticNonPublic);
generator.Emit(OpCodes.Call, printArgsMethod);
// Call the SVC function handler.
generator.Emit(OpCodes.Ldarg_0);
List<(LocalBuilder, RAttribute)> locals = new List<(LocalBuilder, RAttribute)>();
for (int index = 0; index < methodArgs.Length; index++)
{
Type argType = methodArgs[index].ParameterType;
RAttribute registerAttribute = GetRegisterAttribute(methodArgs[index]);
if (argType.IsByRef)
{
argType = argType.GetElementType();
LocalBuilder local = generator.DeclareLocal(argType);
locals.Add((local, registerAttribute));
if (!methodArgs[index].IsOut)
{
generator.Emit(OpCodes.Ldarg_1);
generator.Emit(OpCodes.Ldc_I4, registerAttribute.Index);
MethodInfo info = typeof(ExecutionContext).GetMethod(nameof(ExecutionContext.GetX));
generator.Emit(OpCodes.Call, info);
ConvertToArgType(argType);
generator.Emit(OpCodes.Stloc, local);
}
generator.Emit(OpCodes.Ldloca, local);
}
else
{
generator.Emit(OpCodes.Ldarg_1);
generator.Emit(OpCodes.Ldc_I4, registerAttribute.Index);
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 20:56:22 +02:00
MethodInfo info = typeof(ExecutionContext).GetMethod(nameof(ExecutionContext.GetX));
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 20:56:22 +02:00
generator.Emit(OpCodes.Call, info);
ConvertToArgType(argType);
}
}
generator.Emit(OpCodes.Call, methodInfo);
Type retType = methodInfo.ReturnType;
// Print result code.
if (retType == typeof(KernelResult))
{
MethodInfo printResultMethod = typeof(SyscallTable).GetMethod(nameof(PrintResult), staticNonPublic);
generator.Emit(OpCodes.Dup);
generator.Emit(OpCodes.Ldstr, svcName);
generator.Emit(OpCodes.Call, printResultMethod);
}
uint registerInUse = 0;
// Save return value into register X0 (when the method has a return value).
if (retType != typeof(void))
{
CheckIfTypeIsSupported(retType, svcName);
LocalBuilder tempLocal = generator.DeclareLocal(retType);
generator.Emit(OpCodes.Stloc, tempLocal);
generator.Emit(OpCodes.Ldarg_1);
generator.Emit(OpCodes.Ldc_I4, 0);
generator.Emit(OpCodes.Ldloc, tempLocal);
ConvertToFieldType(retType);
MethodInfo info = typeof(ExecutionContext).GetMethod(nameof(ExecutionContext.SetX));
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 20:56:22 +02:00
generator.Emit(OpCodes.Call, info);
registerInUse |= 1u << 0;
}
for (int index = 0; index < locals.Count; index++)
{
(LocalBuilder local, RAttribute attribute) = locals[index];
if ((registerInUse & (1u << attribute.Index)) != 0)
{
throw new InvalidSvcException($"Method \"{svcName}\" has conflicting output values at register index \"{attribute.Index}\".");
}
generator.Emit(OpCodes.Ldarg_1);
generator.Emit(OpCodes.Ldc_I4, attribute.Index);
generator.Emit(OpCodes.Ldloc, local);
ConvertToFieldType(local.LocalType);
MethodInfo info = typeof(ExecutionContext).GetMethod(nameof(ExecutionContext.SetX));
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 20:56:22 +02:00
generator.Emit(OpCodes.Call, info);
registerInUse |= 1u << attribute.Index;
}
// Zero out the remaining unused registers.
for (int i = 0; i < registerCleanCount; i++)
{
if ((registerInUse & (1u << i)) != 0)
{
continue;
}
generator.Emit(OpCodes.Ldarg_1);
generator.Emit(OpCodes.Ldc_I4, i);
generator.Emit(OpCodes.Ldc_I8, 0L);
MethodInfo info = typeof(ExecutionContext).GetMethod(nameof(ExecutionContext.SetX));
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 20:56:22 +02:00
generator.Emit(OpCodes.Call, info);
}
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 20:56:22 +02:00
generator.Emit(OpCodes.Ret);
return method.CreateDelegate<Action<T, ExecutionContext>>();
}
private static void CheckIfTypeIsSupported(Type type, string svcName)
{
switch (Type.GetTypeCode(type))
{
case TypeCode.UInt64:
case TypeCode.Int64:
case TypeCode.UInt32:
case TypeCode.Int32:
case TypeCode.UInt16:
case TypeCode.Int16:
case TypeCode.Byte:
case TypeCode.SByte:
case TypeCode.Boolean:
return;
}
throw new InvalidSvcException($"Method \"{svcName}\" has a invalid ref type \"{type.Name}\".");
}
private static void PrintArguments(object[] argValues, string formatOrSvcName)
{
if (argValues != null)
{
Logger.Trace?.Print(LogClass.KernelSvc, string.Format(formatOrSvcName, argValues));
}
else
{
Logger.Trace?.Print(LogClass.KernelSvc, formatOrSvcName);
}
}
private static void PrintResult(KernelResult result, string svcName)
{
if (result != KernelResult.Success &&
result != KernelResult.TimedOut &&
result != KernelResult.Cancelled &&
result != KernelResult.PortRemoteClosed &&
result != KernelResult.InvalidState)
{
Logger.Warning?.Print(LogClass.KernelSvc, $"{svcName} returned error {result}.");
}
else
{
Logger.Trace?.Print(LogClass.KernelSvc, $"{svcName} returned result {result}.");
}
}
}
}