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.Memory;
|
|
|
|
using ARMeilleure.State;
|
2018-12-18 06:33:36 +01:00
|
|
|
using Ryujinx.HLE.HOS.Kernel.Common;
|
|
|
|
using Ryujinx.HLE.HOS.Kernel.Process;
|
|
|
|
using Ryujinx.HLE.HOS.Kernel.Threading;
|
|
|
|
|
|
|
|
namespace Ryujinx.HLE.HOS.Kernel.SupervisorCall
|
|
|
|
{
|
|
|
|
partial class SvcHandler
|
|
|
|
{
|
|
|
|
public KernelResult CreateThread64(
|
|
|
|
ulong entrypoint,
|
|
|
|
ulong argsPtr,
|
|
|
|
ulong stackTop,
|
|
|
|
int priority,
|
|
|
|
int cpuCore,
|
|
|
|
out int handle)
|
|
|
|
{
|
|
|
|
return CreateThread(entrypoint, argsPtr, stackTop, priority, cpuCore, out handle);
|
|
|
|
}
|
|
|
|
|
|
|
|
private KernelResult CreateThread(
|
|
|
|
ulong entrypoint,
|
|
|
|
ulong argsPtr,
|
|
|
|
ulong stackTop,
|
|
|
|
int priority,
|
|
|
|
int cpuCore,
|
|
|
|
out int handle)
|
|
|
|
{
|
|
|
|
handle = 0;
|
|
|
|
|
|
|
|
KProcess currentProcess = _system.Scheduler.GetCurrentProcess();
|
|
|
|
|
|
|
|
if (cpuCore == -2)
|
|
|
|
{
|
|
|
|
cpuCore = currentProcess.DefaultCpuCore;
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((uint)cpuCore >= KScheduler.CpuCoresCount || !currentProcess.IsCpuCoreAllowed(cpuCore))
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidCpuCore;
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((uint)priority >= KScheduler.PrioritiesCount || !currentProcess.IsPriorityAllowed(priority))
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidPriority;
|
|
|
|
}
|
|
|
|
|
|
|
|
long timeout = KTimeManager.ConvertMillisecondsToNanoseconds(100);
|
|
|
|
|
|
|
|
if (currentProcess.ResourceLimit != null &&
|
|
|
|
!currentProcess.ResourceLimit.Reserve(LimitableResource.Thread, 1, timeout))
|
|
|
|
{
|
|
|
|
return KernelResult.ResLimitExceeded;
|
|
|
|
}
|
|
|
|
|
|
|
|
KThread thread = new KThread(_system);
|
|
|
|
|
|
|
|
KernelResult result = currentProcess.InitializeThread(
|
|
|
|
thread,
|
|
|
|
entrypoint,
|
|
|
|
argsPtr,
|
|
|
|
stackTop,
|
|
|
|
priority,
|
|
|
|
cpuCore);
|
|
|
|
|
2019-01-18 23:26:39 +01:00
|
|
|
if (result == KernelResult.Success)
|
2018-12-18 06:33:36 +01:00
|
|
|
{
|
2019-01-18 23:26:39 +01:00
|
|
|
result = _process.HandleTable.GenerateHandle(thread, out handle);
|
2018-12-18 06:33:36 +01:00
|
|
|
}
|
2019-01-18 23:26:39 +01:00
|
|
|
else
|
2018-12-18 06:33:36 +01:00
|
|
|
{
|
|
|
|
currentProcess.ResourceLimit?.Release(LimitableResource.Thread, 1);
|
|
|
|
}
|
|
|
|
|
2019-01-18 23:26:39 +01:00
|
|
|
thread.DecrementReferenceCount();
|
|
|
|
|
2018-12-18 06:33:36 +01:00
|
|
|
return result;
|
|
|
|
}
|
|
|
|
|
|
|
|
public KernelResult StartThread64(int handle)
|
|
|
|
{
|
|
|
|
return StartThread(handle);
|
|
|
|
}
|
|
|
|
|
|
|
|
private KernelResult StartThread(int handle)
|
|
|
|
{
|
2019-01-18 23:26:39 +01:00
|
|
|
KThread thread = _process.HandleTable.GetKThread(handle);
|
2018-12-18 06:33:36 +01:00
|
|
|
|
|
|
|
if (thread != null)
|
|
|
|
{
|
2019-01-18 23:26:39 +01:00
|
|
|
thread.IncrementReferenceCount();
|
|
|
|
|
|
|
|
KernelResult result = thread.Start();
|
|
|
|
|
|
|
|
if (result == KernelResult.Success)
|
|
|
|
{
|
|
|
|
thread.IncrementReferenceCount();
|
|
|
|
}
|
|
|
|
|
|
|
|
thread.DecrementReferenceCount();
|
|
|
|
|
|
|
|
return result;
|
2018-12-18 06:33:36 +01:00
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidHandle;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
public void ExitThread64()
|
|
|
|
{
|
|
|
|
ExitThread();
|
|
|
|
}
|
|
|
|
|
|
|
|
private void ExitThread()
|
|
|
|
{
|
|
|
|
KThread currentThread = _system.Scheduler.GetCurrentThread();
|
|
|
|
|
|
|
|
_system.Scheduler.ExitThread(currentThread);
|
|
|
|
|
|
|
|
currentThread.Exit();
|
|
|
|
}
|
|
|
|
|
|
|
|
public void SleepThread64(long timeout)
|
|
|
|
{
|
|
|
|
SleepThread(timeout);
|
|
|
|
}
|
|
|
|
|
|
|
|
private void SleepThread(long timeout)
|
|
|
|
{
|
|
|
|
KThread currentThread = _system.Scheduler.GetCurrentThread();
|
|
|
|
|
|
|
|
if (timeout < 1)
|
|
|
|
{
|
|
|
|
switch (timeout)
|
|
|
|
{
|
|
|
|
case 0: currentThread.Yield(); break;
|
|
|
|
case -1: currentThread.YieldWithLoadBalancing(); break;
|
|
|
|
case -2: currentThread.YieldAndWaitForLoadBalancing(); break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
currentThread.Sleep(timeout);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
public KernelResult GetThreadPriority64(int handle, out int priority)
|
|
|
|
{
|
|
|
|
return GetThreadPriority(handle, out priority);
|
|
|
|
}
|
|
|
|
|
|
|
|
private KernelResult GetThreadPriority(int handle, out int priority)
|
|
|
|
{
|
|
|
|
KThread thread = _process.HandleTable.GetKThread(handle);
|
|
|
|
|
|
|
|
if (thread != null)
|
|
|
|
{
|
|
|
|
priority = thread.DynamicPriority;
|
|
|
|
|
|
|
|
return KernelResult.Success;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
priority = 0;
|
|
|
|
|
|
|
|
return KernelResult.InvalidHandle;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
public KernelResult SetThreadPriority64(int handle, int priority)
|
|
|
|
{
|
|
|
|
return SetThreadPriority(handle, priority);
|
|
|
|
}
|
|
|
|
|
|
|
|
public KernelResult SetThreadPriority(int handle, int priority)
|
|
|
|
{
|
2019-07-02 04:39:22 +02:00
|
|
|
// TODO: NPDM check.
|
2018-12-18 06:33:36 +01:00
|
|
|
|
|
|
|
KThread thread = _process.HandleTable.GetKThread(handle);
|
|
|
|
|
|
|
|
if (thread == null)
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidHandle;
|
|
|
|
}
|
|
|
|
|
|
|
|
thread.SetPriority(priority);
|
|
|
|
|
|
|
|
return KernelResult.Success;
|
|
|
|
}
|
|
|
|
|
|
|
|
public KernelResult GetThreadCoreMask64(int handle, out int preferredCore, out long affinityMask)
|
|
|
|
{
|
|
|
|
return GetThreadCoreMask(handle, out preferredCore, out affinityMask);
|
|
|
|
}
|
|
|
|
|
|
|
|
private KernelResult GetThreadCoreMask(int handle, out int preferredCore, out long affinityMask)
|
|
|
|
{
|
|
|
|
KThread thread = _process.HandleTable.GetKThread(handle);
|
|
|
|
|
|
|
|
if (thread != null)
|
|
|
|
{
|
|
|
|
preferredCore = thread.PreferredCore;
|
|
|
|
affinityMask = thread.AffinityMask;
|
|
|
|
|
|
|
|
return KernelResult.Success;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
preferredCore = 0;
|
|
|
|
affinityMask = 0;
|
|
|
|
|
|
|
|
return KernelResult.InvalidHandle;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
public KernelResult SetThreadCoreMask64(int handle, int preferredCore, long affinityMask)
|
|
|
|
{
|
|
|
|
return SetThreadCoreMask(handle, preferredCore, affinityMask);
|
|
|
|
}
|
|
|
|
|
|
|
|
private KernelResult SetThreadCoreMask(int handle, int preferredCore, long affinityMask)
|
|
|
|
{
|
|
|
|
KProcess currentProcess = _system.Scheduler.GetCurrentProcess();
|
|
|
|
|
|
|
|
if (preferredCore == -2)
|
|
|
|
{
|
|
|
|
preferredCore = currentProcess.DefaultCpuCore;
|
|
|
|
|
|
|
|
affinityMask = 1 << preferredCore;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
if ((currentProcess.Capabilities.AllowedCpuCoresMask | affinityMask) !=
|
|
|
|
currentProcess.Capabilities.AllowedCpuCoresMask)
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidCpuCore;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (affinityMask == 0)
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidCombination;
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((uint)preferredCore > 3)
|
|
|
|
{
|
|
|
|
if ((preferredCore | 2) != -1)
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidCpuCore;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else if ((affinityMask & (1 << preferredCore)) == 0)
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidCombination;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
KThread thread = _process.HandleTable.GetKThread(handle);
|
|
|
|
|
|
|
|
if (thread == null)
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidHandle;
|
|
|
|
}
|
|
|
|
|
|
|
|
return thread.SetCoreAndAffinityMask(preferredCore, affinityMask);
|
|
|
|
}
|
|
|
|
|
|
|
|
public int GetCurrentProcessorNumber64()
|
|
|
|
{
|
|
|
|
return _system.Scheduler.GetCurrentThread().CurrentCore;
|
|
|
|
}
|
|
|
|
|
|
|
|
public KernelResult GetThreadId64(int handle, out long threadUid)
|
|
|
|
{
|
|
|
|
return GetThreadId(handle, out threadUid);
|
|
|
|
}
|
|
|
|
|
|
|
|
private KernelResult GetThreadId(int handle, out long threadUid)
|
|
|
|
{
|
|
|
|
KThread thread = _process.HandleTable.GetKThread(handle);
|
|
|
|
|
|
|
|
if (thread != null)
|
|
|
|
{
|
|
|
|
threadUid = thread.ThreadUid;
|
|
|
|
|
|
|
|
return KernelResult.Success;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
threadUid = 0;
|
|
|
|
|
|
|
|
return KernelResult.InvalidHandle;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
public KernelResult SetThreadActivity64(int handle, bool pause)
|
|
|
|
{
|
|
|
|
return SetThreadActivity(handle, pause);
|
|
|
|
}
|
|
|
|
|
|
|
|
private KernelResult SetThreadActivity(int handle, bool pause)
|
|
|
|
{
|
|
|
|
KThread thread = _process.HandleTable.GetObject<KThread>(handle);
|
|
|
|
|
|
|
|
if (thread == null)
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidHandle;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (thread.Owner != _system.Scheduler.GetCurrentProcess())
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidHandle;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (thread == _system.Scheduler.GetCurrentThread())
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidThread;
|
|
|
|
}
|
|
|
|
|
|
|
|
return thread.SetActivity(pause);
|
|
|
|
}
|
|
|
|
|
|
|
|
public KernelResult GetThreadContext364(ulong address, int handle)
|
|
|
|
{
|
|
|
|
return GetThreadContext3(address, handle);
|
|
|
|
}
|
|
|
|
|
|
|
|
private KernelResult GetThreadContext3(ulong address, int handle)
|
|
|
|
{
|
|
|
|
KProcess currentProcess = _system.Scheduler.GetCurrentProcess();
|
|
|
|
KThread currentThread = _system.Scheduler.GetCurrentThread();
|
|
|
|
|
|
|
|
KThread thread = _process.HandleTable.GetObject<KThread>(handle);
|
|
|
|
|
|
|
|
if (thread == null)
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidHandle;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (thread.Owner != currentProcess)
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidHandle;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (currentThread == thread)
|
|
|
|
{
|
|
|
|
return KernelResult.InvalidThread;
|
|
|
|
}
|
|
|
|
|
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
|
|
|
IMemoryManager memory = currentProcess.CpuMemory;
|
|
|
|
|
|
|
|
memory.WriteUInt64((long)address + 0x0, thread.Context.GetX(0));
|
|
|
|
memory.WriteUInt64((long)address + 0x8, thread.Context.GetX(1));
|
|
|
|
memory.WriteUInt64((long)address + 0x10, thread.Context.GetX(2));
|
|
|
|
memory.WriteUInt64((long)address + 0x18, thread.Context.GetX(3));
|
|
|
|
memory.WriteUInt64((long)address + 0x20, thread.Context.GetX(4));
|
|
|
|
memory.WriteUInt64((long)address + 0x28, thread.Context.GetX(5));
|
|
|
|
memory.WriteUInt64((long)address + 0x30, thread.Context.GetX(6));
|
|
|
|
memory.WriteUInt64((long)address + 0x38, thread.Context.GetX(7));
|
|
|
|
memory.WriteUInt64((long)address + 0x40, thread.Context.GetX(8));
|
|
|
|
memory.WriteUInt64((long)address + 0x48, thread.Context.GetX(9));
|
|
|
|
memory.WriteUInt64((long)address + 0x50, thread.Context.GetX(10));
|
|
|
|
memory.WriteUInt64((long)address + 0x58, thread.Context.GetX(11));
|
|
|
|
memory.WriteUInt64((long)address + 0x60, thread.Context.GetX(12));
|
|
|
|
memory.WriteUInt64((long)address + 0x68, thread.Context.GetX(13));
|
|
|
|
memory.WriteUInt64((long)address + 0x70, thread.Context.GetX(14));
|
|
|
|
memory.WriteUInt64((long)address + 0x78, thread.Context.GetX(15));
|
|
|
|
memory.WriteUInt64((long)address + 0x80, thread.Context.GetX(16));
|
|
|
|
memory.WriteUInt64((long)address + 0x88, thread.Context.GetX(17));
|
|
|
|
memory.WriteUInt64((long)address + 0x90, thread.Context.GetX(18));
|
|
|
|
memory.WriteUInt64((long)address + 0x98, thread.Context.GetX(19));
|
|
|
|
memory.WriteUInt64((long)address + 0xa0, thread.Context.GetX(20));
|
|
|
|
memory.WriteUInt64((long)address + 0xa8, thread.Context.GetX(21));
|
|
|
|
memory.WriteUInt64((long)address + 0xb0, thread.Context.GetX(22));
|
|
|
|
memory.WriteUInt64((long)address + 0xb8, thread.Context.GetX(23));
|
|
|
|
memory.WriteUInt64((long)address + 0xc0, thread.Context.GetX(24));
|
|
|
|
memory.WriteUInt64((long)address + 0xc8, thread.Context.GetX(25));
|
|
|
|
memory.WriteUInt64((long)address + 0xd0, thread.Context.GetX(26));
|
|
|
|
memory.WriteUInt64((long)address + 0xd8, thread.Context.GetX(27));
|
|
|
|
memory.WriteUInt64((long)address + 0xe0, thread.Context.GetX(28));
|
|
|
|
memory.WriteUInt64((long)address + 0xe8, thread.Context.GetX(29));
|
|
|
|
memory.WriteUInt64((long)address + 0xf0, thread.Context.GetX(30));
|
|
|
|
memory.WriteUInt64((long)address + 0xf8, thread.Context.GetX(31));
|
2019-02-24 08:24:35 +01:00
|
|
|
|
|
|
|
memory.WriteInt64((long)address + 0x100, thread.LastPc);
|
|
|
|
|
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
|
|
|
memory.WriteUInt64((long)address + 0x108, (ulong)GetPsr(thread.Context));
|
|
|
|
|
|
|
|
memory.WriteVector128((long)address + 0x110, thread.Context.GetV(0));
|
|
|
|
memory.WriteVector128((long)address + 0x120, thread.Context.GetV(1));
|
|
|
|
memory.WriteVector128((long)address + 0x130, thread.Context.GetV(2));
|
|
|
|
memory.WriteVector128((long)address + 0x140, thread.Context.GetV(3));
|
|
|
|
memory.WriteVector128((long)address + 0x150, thread.Context.GetV(4));
|
|
|
|
memory.WriteVector128((long)address + 0x160, thread.Context.GetV(5));
|
|
|
|
memory.WriteVector128((long)address + 0x170, thread.Context.GetV(6));
|
|
|
|
memory.WriteVector128((long)address + 0x180, thread.Context.GetV(7));
|
|
|
|
memory.WriteVector128((long)address + 0x190, thread.Context.GetV(8));
|
|
|
|
memory.WriteVector128((long)address + 0x1a0, thread.Context.GetV(9));
|
|
|
|
memory.WriteVector128((long)address + 0x1b0, thread.Context.GetV(10));
|
|
|
|
memory.WriteVector128((long)address + 0x1c0, thread.Context.GetV(11));
|
|
|
|
memory.WriteVector128((long)address + 0x1d0, thread.Context.GetV(12));
|
|
|
|
memory.WriteVector128((long)address + 0x1e0, thread.Context.GetV(13));
|
|
|
|
memory.WriteVector128((long)address + 0x1f0, thread.Context.GetV(14));
|
|
|
|
memory.WriteVector128((long)address + 0x200, thread.Context.GetV(15));
|
|
|
|
memory.WriteVector128((long)address + 0x210, thread.Context.GetV(16));
|
|
|
|
memory.WriteVector128((long)address + 0x220, thread.Context.GetV(17));
|
|
|
|
memory.WriteVector128((long)address + 0x230, thread.Context.GetV(18));
|
|
|
|
memory.WriteVector128((long)address + 0x240, thread.Context.GetV(19));
|
|
|
|
memory.WriteVector128((long)address + 0x250, thread.Context.GetV(20));
|
|
|
|
memory.WriteVector128((long)address + 0x260, thread.Context.GetV(21));
|
|
|
|
memory.WriteVector128((long)address + 0x270, thread.Context.GetV(22));
|
|
|
|
memory.WriteVector128((long)address + 0x280, thread.Context.GetV(23));
|
|
|
|
memory.WriteVector128((long)address + 0x290, thread.Context.GetV(24));
|
|
|
|
memory.WriteVector128((long)address + 0x2a0, thread.Context.GetV(25));
|
|
|
|
memory.WriteVector128((long)address + 0x2b0, thread.Context.GetV(26));
|
|
|
|
memory.WriteVector128((long)address + 0x2c0, thread.Context.GetV(27));
|
|
|
|
memory.WriteVector128((long)address + 0x2d0, thread.Context.GetV(28));
|
|
|
|
memory.WriteVector128((long)address + 0x2e0, thread.Context.GetV(29));
|
|
|
|
memory.WriteVector128((long)address + 0x2f0, thread.Context.GetV(30));
|
|
|
|
memory.WriteVector128((long)address + 0x300, thread.Context.GetV(31));
|
|
|
|
|
|
|
|
memory.WriteInt32((long)address + 0x310, (int)thread.Context.Fpcr);
|
|
|
|
memory.WriteInt32((long)address + 0x314, (int)thread.Context.Fpsr);
|
|
|
|
memory.WriteInt64((long)address + 0x318, thread.Context.Tpidr);
|
2018-12-18 06:33:36 +01:00
|
|
|
|
|
|
|
return KernelResult.Success;
|
|
|
|
}
|
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
|
|
|
|
|
|
|
private static int GetPsr(IExecutionContext context)
|
|
|
|
{
|
|
|
|
return (context.GetPstateFlag(PState.NFlag) ? (1 << 31) : 0) |
|
|
|
|
(context.GetPstateFlag(PState.ZFlag) ? (1 << 30) : 0) |
|
|
|
|
(context.GetPstateFlag(PState.CFlag) ? (1 << 29) : 0) |
|
|
|
|
(context.GetPstateFlag(PState.VFlag) ? (1 << 28) : 0);
|
|
|
|
}
|
2018-12-18 06:33:36 +01:00
|
|
|
}
|
|
|
|
}
|