2019-02-04 22:26:05 +01:00
|
|
|
using ChocolArm64.Translation;
|
2018-02-05 00:08:20 +01:00
|
|
|
using System;
|
2018-03-14 01:24:17 +01:00
|
|
|
using System.Diagnostics;
|
2018-09-19 17:16:20 +02:00
|
|
|
using System.Runtime.CompilerServices;
|
2018-05-12 01:10:27 +02:00
|
|
|
using System.Runtime.Intrinsics;
|
2018-02-05 00:08:20 +01:00
|
|
|
|
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 static ChocolArm64.Instructions.VectorHelper;
|
|
|
|
|
2018-02-05 00:08:20 +01:00
|
|
|
namespace ChocolArm64.State
|
|
|
|
{
|
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
|
|
|
public class CpuThreadState : ARMeilleure.State.IExecutionContext
|
2018-02-05 00:08:20 +01:00
|
|
|
{
|
2019-04-26 06:55:12 +02:00
|
|
|
private const int MinCountForCheck = 40000;
|
|
|
|
|
2018-02-06 16:15:08 +01:00
|
|
|
internal const int ErgSizeLog2 = 4;
|
|
|
|
internal const int DczSizeLog2 = 4;
|
|
|
|
|
2018-02-05 00:08:20 +01:00
|
|
|
public ulong X0, X1, X2, X3, X4, X5, X6, X7,
|
|
|
|
X8, X9, X10, X11, X12, X13, X14, X15,
|
|
|
|
X16, X17, X18, X19, X20, X21, X22, X23,
|
|
|
|
X24, X25, X26, X27, X28, X29, X30, X31;
|
|
|
|
|
2018-05-12 01:10:27 +02:00
|
|
|
public Vector128<float> V0, V1, V2, V3, V4, V5, V6, V7,
|
|
|
|
V8, V9, V10, V11, V12, V13, V14, V15,
|
|
|
|
V16, V17, V18, V19, V20, V21, V22, V23,
|
|
|
|
V24, V25, V26, V27, V28, V29, V30, V31;
|
2018-02-05 00:08:20 +01:00
|
|
|
|
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
|
|
|
public bool IsAarch32 { get; set; }
|
Implement some ARM32 memory instructions and CMP (#565)
* Implement ARM32 memory instructions: LDM, LDR, LDRB, LDRD, LDRH, LDRSB, LDRSH, STM, STR, STRB, STRD, STRH (immediate and register + immediate variants), implement CMP (immediate and register shifted by immediate variants)
* Rename some opcode classes and flag masks for consistency
* Fix a few suboptimal ARM32 codegen issues, only loads should be considered on decoder when checking if Rt == PC, and only NZCV flags should be considered for comparison optimizations
* Take into account Rt2 for LDRD instructions aswell when checking if the instruction changes PC
* Re-align arm32 instructions on the opcode table
2019-01-29 17:06:11 +01:00
|
|
|
|
|
|
|
public bool Thumb;
|
|
|
|
public bool BigEndian;
|
|
|
|
|
2018-02-05 00:08:20 +01:00
|
|
|
public bool Overflow;
|
|
|
|
public bool Carry;
|
|
|
|
public bool Zero;
|
|
|
|
public bool Negative;
|
|
|
|
|
2019-01-25 02:59:53 +01:00
|
|
|
public int ElrHyp;
|
|
|
|
|
2018-03-12 05:04:52 +01:00
|
|
|
public bool Running { get; set; }
|
2018-09-19 01:36:43 +02:00
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
private bool _interrupted;
|
2018-03-12 05:04:52 +01:00
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
private int _syncCount;
|
2018-09-19 17:16:20 +02:00
|
|
|
|
2018-02-06 16:15:08 +01:00
|
|
|
public long TpidrEl0 { get; set; }
|
|
|
|
public long Tpidr { get; set; }
|
2018-02-05 00:08:20 +01:00
|
|
|
|
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
|
|
|
public int CFpcr { get; set; }
|
|
|
|
public int CFpsr { get; set; }
|
|
|
|
|
|
|
|
public ARMeilleure.State.FPCR Fpcr
|
|
|
|
{
|
|
|
|
get => (ARMeilleure.State.FPCR)CFpcr;
|
|
|
|
set => CFpcr = (int)value;
|
|
|
|
}
|
|
|
|
|
|
|
|
public ARMeilleure.State.FPSR Fpsr
|
|
|
|
{
|
|
|
|
get => (ARMeilleure.State.FPSR)CFpsr;
|
|
|
|
set => CFpsr = (int)value;
|
|
|
|
}
|
2018-02-05 00:08:20 +01:00
|
|
|
|
2018-06-26 06:09:32 +02:00
|
|
|
public int Psr
|
|
|
|
{
|
|
|
|
get
|
|
|
|
{
|
Implement some ARM32 memory instructions and CMP (#565)
* Implement ARM32 memory instructions: LDM, LDR, LDRB, LDRD, LDRH, LDRSB, LDRSH, STM, STR, STRB, STRD, STRH (immediate and register + immediate variants), implement CMP (immediate and register shifted by immediate variants)
* Rename some opcode classes and flag masks for consistency
* Fix a few suboptimal ARM32 codegen issues, only loads should be considered on decoder when checking if Rt == PC, and only NZCV flags should be considered for comparison optimizations
* Take into account Rt2 for LDRD instructions aswell when checking if the instruction changes PC
* Re-align arm32 instructions on the opcode table
2019-01-29 17:06:11 +01:00
|
|
|
return (Negative ? (int)PState.NMask : 0) |
|
|
|
|
(Zero ? (int)PState.ZMask : 0) |
|
|
|
|
(Carry ? (int)PState.CMask : 0) |
|
|
|
|
(Overflow ? (int)PState.VMask : 0);
|
2018-06-26 06:09:32 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-02-06 16:15:08 +01:00
|
|
|
public uint CtrEl0 => 0x8444c004;
|
|
|
|
public uint DczidEl0 => 0x00000004;
|
2018-02-05 00:08:20 +01:00
|
|
|
|
2018-03-14 01:24:17 +01:00
|
|
|
public ulong CntfrqEl0 { get; set; }
|
|
|
|
public ulong CntpctEl0
|
|
|
|
{
|
|
|
|
get
|
|
|
|
{
|
2018-10-31 02:43:02 +01:00
|
|
|
double ticks = _tickCounter.ElapsedTicks * _hostTickFreq;
|
2018-02-12 04:37:20 +01:00
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
return (ulong)(ticks * CntfrqEl0);
|
2018-03-14 01:24:17 +01:00
|
|
|
}
|
|
|
|
}
|
2018-02-05 00:08:20 +01:00
|
|
|
|
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
|
|
|
public event EventHandler<EventArgs> Interrupt;
|
|
|
|
public event EventHandler<ARMeilleure.State.InstExceptionEventArgs> Break;
|
|
|
|
public event EventHandler<ARMeilleure.State.InstExceptionEventArgs> SupervisorCall;
|
|
|
|
public event EventHandler<ARMeilleure.State.InstUndefinedEventArgs> Undefined;
|
2018-02-10 14:24:16 +01:00
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
private static Stopwatch _tickCounter;
|
2018-03-14 01:24:17 +01:00
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
private static double _hostTickFreq;
|
2018-03-14 01:24:17 +01:00
|
|
|
|
2019-02-04 22:26:05 +01:00
|
|
|
internal Translator CurrentTranslator;
|
|
|
|
|
2019-02-19 00:52:06 +01:00
|
|
|
private ulong _exclusiveAddress;
|
|
|
|
|
|
|
|
internal ulong ExclusiveValueLow { get; set; }
|
|
|
|
internal ulong ExclusiveValueHigh { get; set; }
|
|
|
|
|
|
|
|
public CpuThreadState()
|
|
|
|
{
|
|
|
|
ClearExclusiveAddress();
|
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
|
|
|
|
|
|
|
Running = true;
|
2019-02-19 00:52:06 +01:00
|
|
|
}
|
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
static CpuThreadState()
|
2018-03-14 01:24:17 +01:00
|
|
|
{
|
2018-10-31 02:43:02 +01:00
|
|
|
_hostTickFreq = 1.0 / Stopwatch.Frequency;
|
2018-03-14 01:24:17 +01:00
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
_tickCounter = new Stopwatch();
|
2018-03-14 01:24:17 +01:00
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
_tickCounter.Start();
|
2018-03-14 01:24:17 +01:00
|
|
|
}
|
|
|
|
|
2019-02-19 00:52:06 +01:00
|
|
|
internal void SetExclusiveAddress(ulong address)
|
|
|
|
{
|
|
|
|
_exclusiveAddress = GetMaskedExclusiveAddress(address);
|
|
|
|
}
|
|
|
|
|
|
|
|
internal bool CheckExclusiveAddress(ulong address)
|
|
|
|
{
|
|
|
|
return GetMaskedExclusiveAddress(address) == _exclusiveAddress;
|
|
|
|
}
|
|
|
|
|
|
|
|
internal void ClearExclusiveAddress()
|
|
|
|
{
|
|
|
|
_exclusiveAddress = ulong.MaxValue;
|
|
|
|
}
|
|
|
|
|
|
|
|
private ulong GetMaskedExclusiveAddress(ulong address)
|
|
|
|
{
|
|
|
|
return address & ~((4UL << ErgSizeLog2) - 1);
|
|
|
|
}
|
|
|
|
|
2018-09-19 17:16:20 +02:00
|
|
|
[MethodImpl(MethodImplOptions.AggressiveInlining)]
|
2019-04-26 06:55:12 +02:00
|
|
|
internal bool Synchronize()
|
2018-08-17 01:47:36 +02:00
|
|
|
{
|
2019-07-02 04:39:22 +02:00
|
|
|
// Firing a interrupt frequently is expensive, so we only
|
|
|
|
// do it after a given number of instructions has executed.
|
2019-04-26 06:55:12 +02:00
|
|
|
_syncCount++;
|
2018-09-19 01:36:43 +02:00
|
|
|
|
2019-04-26 06:55:12 +02:00
|
|
|
if (_syncCount >= MinCountForCheck)
|
2018-09-19 17:16:20 +02:00
|
|
|
{
|
|
|
|
CheckInterrupt();
|
2018-09-19 01:36:43 +02:00
|
|
|
}
|
|
|
|
|
2018-08-17 01:47:36 +02:00
|
|
|
return Running;
|
|
|
|
}
|
|
|
|
|
2018-09-19 17:16:20 +02:00
|
|
|
[MethodImpl(MethodImplOptions.NoInlining)]
|
|
|
|
private void CheckInterrupt()
|
2018-09-19 01:36:43 +02:00
|
|
|
{
|
2018-10-31 02:43:02 +01:00
|
|
|
_syncCount = 0;
|
2018-09-19 17:16:20 +02:00
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
if (_interrupted)
|
2018-09-19 17:16:20 +02:00
|
|
|
{
|
2018-10-31 02:43:02 +01:00
|
|
|
_interrupted = false;
|
2018-09-19 17:16:20 +02:00
|
|
|
|
|
|
|
Interrupt?.Invoke(this, EventArgs.Empty);
|
|
|
|
}
|
2018-09-19 01:36:43 +02:00
|
|
|
}
|
|
|
|
|
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
|
|
|
public ulong GetX(int index)
|
|
|
|
{
|
|
|
|
switch (index)
|
|
|
|
{
|
|
|
|
case 0: return X0;
|
|
|
|
case 1: return X1;
|
|
|
|
case 2: return X2;
|
|
|
|
case 3: return X3;
|
|
|
|
case 4: return X4;
|
|
|
|
case 5: return X5;
|
|
|
|
case 6: return X6;
|
|
|
|
case 7: return X7;
|
|
|
|
case 8: return X8;
|
|
|
|
case 9: return X9;
|
|
|
|
case 10: return X10;
|
|
|
|
case 11: return X11;
|
|
|
|
case 12: return X12;
|
|
|
|
case 13: return X13;
|
|
|
|
case 14: return X14;
|
|
|
|
case 15: return X15;
|
|
|
|
case 16: return X16;
|
|
|
|
case 17: return X17;
|
|
|
|
case 18: return X18;
|
|
|
|
case 19: return X19;
|
|
|
|
case 20: return X20;
|
|
|
|
case 21: return X21;
|
|
|
|
case 22: return X22;
|
|
|
|
case 23: return X23;
|
|
|
|
case 24: return X24;
|
|
|
|
case 25: return X25;
|
|
|
|
case 26: return X26;
|
|
|
|
case 27: return X27;
|
|
|
|
case 28: return X28;
|
|
|
|
case 29: return X29;
|
|
|
|
case 30: return X30;
|
|
|
|
case 31: return X31;
|
|
|
|
|
|
|
|
default: throw new ArgumentOutOfRangeException(nameof(index));
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
public void SetX(int index, ulong value)
|
|
|
|
{
|
|
|
|
switch (index)
|
|
|
|
{
|
|
|
|
case 0: X0 = value; break;
|
|
|
|
case 1: X1 = value; break;
|
|
|
|
case 2: X2 = value; break;
|
|
|
|
case 3: X3 = value; break;
|
|
|
|
case 4: X4 = value; break;
|
|
|
|
case 5: X5 = value; break;
|
|
|
|
case 6: X6 = value; break;
|
|
|
|
case 7: X7 = value; break;
|
|
|
|
case 8: X8 = value; break;
|
|
|
|
case 9: X9 = value; break;
|
|
|
|
case 10: X10 = value; break;
|
|
|
|
case 11: X11 = value; break;
|
|
|
|
case 12: X12 = value; break;
|
|
|
|
case 13: X13 = value; break;
|
|
|
|
case 14: X14 = value; break;
|
|
|
|
case 15: X15 = value; break;
|
|
|
|
case 16: X16 = value; break;
|
|
|
|
case 17: X17 = value; break;
|
|
|
|
case 18: X18 = value; break;
|
|
|
|
case 19: X19 = value; break;
|
|
|
|
case 20: X20 = value; break;
|
|
|
|
case 21: X21 = value; break;
|
|
|
|
case 22: X22 = value; break;
|
|
|
|
case 23: X23 = value; break;
|
|
|
|
case 24: X24 = value; break;
|
|
|
|
case 25: X25 = value; break;
|
|
|
|
case 26: X26 = value; break;
|
|
|
|
case 27: X27 = value; break;
|
|
|
|
case 28: X28 = value; break;
|
|
|
|
case 29: X29 = value; break;
|
|
|
|
case 30: X30 = value; break;
|
|
|
|
case 31: X31 = value; break;
|
|
|
|
|
|
|
|
default: throw new ArgumentOutOfRangeException(nameof(index));
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
public ARMeilleure.State.V128 GetV(int index)
|
|
|
|
{
|
|
|
|
switch (index)
|
|
|
|
{
|
|
|
|
case 0: return new ARMeilleure.State.V128(VectorExtractIntZx(V0, 0, 3), VectorExtractIntZx(V0, 1, 3));
|
|
|
|
case 1: return new ARMeilleure.State.V128(VectorExtractIntZx(V1, 0, 3), VectorExtractIntZx(V1, 1, 3));
|
|
|
|
case 2: return new ARMeilleure.State.V128(VectorExtractIntZx(V2, 0, 3), VectorExtractIntZx(V2, 1, 3));
|
|
|
|
case 3: return new ARMeilleure.State.V128(VectorExtractIntZx(V3, 0, 3), VectorExtractIntZx(V3, 1, 3));
|
|
|
|
case 4: return new ARMeilleure.State.V128(VectorExtractIntZx(V4, 0, 3), VectorExtractIntZx(V4, 1, 3));
|
|
|
|
case 5: return new ARMeilleure.State.V128(VectorExtractIntZx(V5, 0, 3), VectorExtractIntZx(V5, 1, 3));
|
|
|
|
case 6: return new ARMeilleure.State.V128(VectorExtractIntZx(V6, 0, 3), VectorExtractIntZx(V6, 1, 3));
|
|
|
|
case 7: return new ARMeilleure.State.V128(VectorExtractIntZx(V7, 0, 3), VectorExtractIntZx(V7, 1, 3));
|
|
|
|
case 8: return new ARMeilleure.State.V128(VectorExtractIntZx(V8, 0, 3), VectorExtractIntZx(V8, 1, 3));
|
|
|
|
case 9: return new ARMeilleure.State.V128(VectorExtractIntZx(V9, 0, 3), VectorExtractIntZx(V9, 1, 3));
|
|
|
|
case 10: return new ARMeilleure.State.V128(VectorExtractIntZx(V10, 0, 3), VectorExtractIntZx(V10, 1, 3));
|
|
|
|
case 11: return new ARMeilleure.State.V128(VectorExtractIntZx(V11, 0, 3), VectorExtractIntZx(V11, 1, 3));
|
|
|
|
case 12: return new ARMeilleure.State.V128(VectorExtractIntZx(V12, 0, 3), VectorExtractIntZx(V12, 1, 3));
|
|
|
|
case 13: return new ARMeilleure.State.V128(VectorExtractIntZx(V13, 0, 3), VectorExtractIntZx(V13, 1, 3));
|
|
|
|
case 14: return new ARMeilleure.State.V128(VectorExtractIntZx(V14, 0, 3), VectorExtractIntZx(V14, 1, 3));
|
|
|
|
case 15: return new ARMeilleure.State.V128(VectorExtractIntZx(V15, 0, 3), VectorExtractIntZx(V15, 1, 3));
|
|
|
|
case 16: return new ARMeilleure.State.V128(VectorExtractIntZx(V16, 0, 3), VectorExtractIntZx(V16, 1, 3));
|
|
|
|
case 17: return new ARMeilleure.State.V128(VectorExtractIntZx(V17, 0, 3), VectorExtractIntZx(V17, 1, 3));
|
|
|
|
case 18: return new ARMeilleure.State.V128(VectorExtractIntZx(V18, 0, 3), VectorExtractIntZx(V18, 1, 3));
|
|
|
|
case 19: return new ARMeilleure.State.V128(VectorExtractIntZx(V19, 0, 3), VectorExtractIntZx(V19, 1, 3));
|
|
|
|
case 20: return new ARMeilleure.State.V128(VectorExtractIntZx(V20, 0, 3), VectorExtractIntZx(V20, 1, 3));
|
|
|
|
case 21: return new ARMeilleure.State.V128(VectorExtractIntZx(V21, 0, 3), VectorExtractIntZx(V21, 1, 3));
|
|
|
|
case 22: return new ARMeilleure.State.V128(VectorExtractIntZx(V22, 0, 3), VectorExtractIntZx(V22, 1, 3));
|
|
|
|
case 23: return new ARMeilleure.State.V128(VectorExtractIntZx(V23, 0, 3), VectorExtractIntZx(V23, 1, 3));
|
|
|
|
case 24: return new ARMeilleure.State.V128(VectorExtractIntZx(V24, 0, 3), VectorExtractIntZx(V24, 1, 3));
|
|
|
|
case 25: return new ARMeilleure.State.V128(VectorExtractIntZx(V25, 0, 3), VectorExtractIntZx(V25, 1, 3));
|
|
|
|
case 26: return new ARMeilleure.State.V128(VectorExtractIntZx(V26, 0, 3), VectorExtractIntZx(V26, 1, 3));
|
|
|
|
case 27: return new ARMeilleure.State.V128(VectorExtractIntZx(V27, 0, 3), VectorExtractIntZx(V27, 1, 3));
|
|
|
|
case 28: return new ARMeilleure.State.V128(VectorExtractIntZx(V28, 0, 3), VectorExtractIntZx(V28, 1, 3));
|
|
|
|
case 29: return new ARMeilleure.State.V128(VectorExtractIntZx(V29, 0, 3), VectorExtractIntZx(V29, 1, 3));
|
|
|
|
case 30: return new ARMeilleure.State.V128(VectorExtractIntZx(V30, 0, 3), VectorExtractIntZx(V30, 1, 3));
|
|
|
|
case 31: return new ARMeilleure.State.V128(VectorExtractIntZx(V31, 0, 3), VectorExtractIntZx(V31, 1, 3));
|
|
|
|
|
|
|
|
default: throw new ArgumentOutOfRangeException(nameof(index));
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
public bool GetPstateFlag(ARMeilleure.State.PState flag)
|
|
|
|
{
|
|
|
|
switch (flag)
|
|
|
|
{
|
|
|
|
case ARMeilleure.State.PState.NFlag: return Negative;
|
|
|
|
case ARMeilleure.State.PState.ZFlag: return Zero;
|
|
|
|
case ARMeilleure.State.PState.CFlag: return Carry;
|
|
|
|
case ARMeilleure.State.PState.VFlag: return Overflow;
|
|
|
|
|
|
|
|
default: throw new ArgumentOutOfRangeException(nameof(flag));
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
public void RequestInterrupt()
|
2019-04-26 06:55:12 +02:00
|
|
|
{
|
|
|
|
_interrupted = true;
|
|
|
|
}
|
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
internal void OnBreak(long position, int imm)
|
2018-02-10 14:24:16 +01:00
|
|
|
{
|
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
|
|
|
Break?.Invoke(this, new ARMeilleure.State.InstExceptionEventArgs((ulong)position, imm));
|
2018-02-10 14:24:16 +01:00
|
|
|
}
|
2018-02-05 00:08:20 +01:00
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
internal void OnSvcCall(long position, int imm)
|
2018-02-05 00:08:20 +01:00
|
|
|
{
|
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
|
|
|
SupervisorCall?.Invoke(this, new ARMeilleure.State.InstExceptionEventArgs((ulong)position, imm));
|
2018-02-05 00:08:20 +01:00
|
|
|
}
|
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
internal void OnUndefined(long position, int rawOpCode)
|
2018-02-05 00:08:20 +01:00
|
|
|
{
|
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
|
|
|
Undefined?.Invoke(this, new ARMeilleure.State.InstUndefinedEventArgs((ulong)position, rawOpCode));
|
2018-02-05 00:08:20 +01:00
|
|
|
}
|
2018-10-23 16:12:45 +02:00
|
|
|
|
2019-01-25 02:59:53 +01:00
|
|
|
internal ExecutionMode GetExecutionMode()
|
|
|
|
{
|
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
|
|
|
if (!IsAarch32)
|
2019-01-25 02:59:53 +01:00
|
|
|
{
|
|
|
|
return ExecutionMode.Aarch64;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
return Thumb ? ExecutionMode.Aarch32Thumb : ExecutionMode.Aarch32Arm;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
internal bool GetFpcrFlag(Fpcr flag)
|
2018-10-23 16:12:45 +02:00
|
|
|
{
|
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
|
|
|
return (CFpcr & (1 << (int)flag)) != 0;
|
2018-10-23 16:12:45 +02:00
|
|
|
}
|
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
internal void SetFpsrFlag(Fpsr flag)
|
2018-10-23 16:12:45 +02:00
|
|
|
{
|
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
|
|
|
CFpsr |= 1 << (int)flag;
|
2018-10-23 16:12:45 +02:00
|
|
|
}
|
|
|
|
|
2018-10-31 02:43:02 +01:00
|
|
|
internal RoundMode FPRoundingMode()
|
2018-10-23 16:12:45 +02:00
|
|
|
{
|
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
|
|
|
return (RoundMode)((CFpcr >> (int)State.Fpcr.RMode) & 3);
|
2018-10-23 16:12:45 +02:00
|
|
|
}
|
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
|
|
|
|
|
|
|
public void Dispose() { }
|
2018-02-05 00:08:20 +01:00
|
|
|
}
|
2018-10-23 16:12:45 +02:00
|
|
|
}
|