2018-09-08 19:24:29 +02:00
|
|
|
#define SimdRegElem
|
|
|
|
|
Add a new JIT compiler for CPU code (#693)
* Start of the ARMeilleure project
* Refactoring around the old IRAdapter, now renamed to PreAllocator
* Optimize the LowestBitSet method
* Add CLZ support and fix CLS implementation
* Add missing Equals and GetHashCode overrides on some structs, misc small tweaks
* Implement the ByteSwap IR instruction, and some refactoring on the assembler
* Implement the DivideUI IR instruction and fix 64-bits IDIV
* Correct constant operand type on CSINC
* Move division instructions implementation to InstEmitDiv
* Fix destination type for the ConditionalSelect IR instruction
* Implement UMULH and SMULH, with new IR instructions
* Fix some issues with shift instructions
* Fix constant types for BFM instructions
* Fix up new tests using the new V128 struct
* Update tests
* Move DIV tests to a separate file
* Add support for calls, and some instructions that depends on them
* Start adding support for SIMD & FP types, along with some of the related ARM instructions
* Fix some typos and the divide instruction with FP operands
* Fix wrong method call on Clz_V
* Implement ARM FP & SIMD move instructions, Saddlv_V, and misc. fixes
* Implement SIMD logical instructions and more misc. fixes
* Fix PSRAD x86 instruction encoding, TRN, UABD and UABDL implementations
* Implement float conversion instruction, merge in LDj3SNuD fixes, and some other misc. fixes
* Implement SIMD shift instruction and fix Dup_V
* Add SCVTF and UCVTF (vector, fixed-point) variants to the opcode table
* Fix check with tolerance on tester
* Implement FP & SIMD comparison instructions, and some fixes
* Update FCVT (Scalar) encoding on the table to support the Half-float variants
* Support passing V128 structs, some cleanup on the register allocator, merge LDj3SNuD fixes
* Use old memory access methods, made a start on SIMD memory insts support, some fixes
* Fix float constant passed to functions, save and restore non-volatile XMM registers, other fixes
* Fix arguments count with struct return values, other fixes
* More instructions
* Misc. fixes and integrate LDj3SNuD fixes
* Update tests
* Add a faster linear scan allocator, unwinding support on windows, and other changes
* Update Ryujinx.HLE
* Update Ryujinx.Graphics
* Fix V128 return pointer passing, RCX is clobbered
* Update Ryujinx.Tests
* Update ITimeZoneService
* Stop using GetFunctionPointer as that can't be called from native code, misc. fixes and tweaks
* Use generic GetFunctionPointerForDelegate method and other tweaks
* Some refactoring on the code generator, assert on invalid operations and use a separate enum for intrinsics
* Remove some unused code on the assembler
* Fix REX.W prefix regression on float conversion instructions, add some sort of profiler
* Add hardware capability detection
* Fix regression on Sha1h and revert Fcm** changes
* Add SSE2-only paths on vector extract and insert, some refactoring on the pre-allocator
* Fix silly mistake introduced on last commit on CpuId
* Generate inline stack probes when the stack allocation is too large
* Initial support for the System-V ABI
* Support multiple destination operands
* Fix SSE2 VectorInsert8 path, and other fixes
* Change placement of XMM callee save and restore code to match other compilers
* Rename Dest to Destination and Inst to Instruction
* Fix a regression related to calls and the V128 type
* Add an extra space on comments to match code style
* Some refactoring
* Fix vector insert FP32 SSE2 path
* Port over the ARM32 instructions
* Avoid memory protection races on JIT Cache
* Another fix on VectorInsert FP32 (thanks to LDj3SNuD
* Float operands don't need to use the same register when VEX is supported
* Add a new register allocator, higher quality code for hot code (tier up), and other tweaks
* Some nits, small improvements on the pre allocator
* CpuThreadState is gone
* Allow changing CPU emulators with a config entry
* Add runtime identifiers on the ARMeilleure project
* Allow switching between CPUs through a config entry (pt. 2)
* Change win10-x64 to win-x64 on projects
* Update the Ryujinx project to use ARMeilleure
* Ensure that the selected register is valid on the hybrid allocator
* Allow exiting on returns to 0 (should fix test regression)
* Remove register assignments for most used variables on the hybrid allocator
* Do not use fixed registers as spill temp
* Add missing namespace and remove unneeded using
* Address PR feedback
* Fix types, etc
* Enable AssumeStrictAbiCompliance by default
* Ensure that Spill and Fill don't load or store any more than necessary
2019-08-08 20:56:22 +02:00
|
|
|
using ARMeilleure.State;
|
2018-09-08 19:24:29 +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
|
|
|
using NUnit.Framework;
|
2018-09-08 19:24:29 +02:00
|
|
|
|
|
|
|
namespace Ryujinx.Tests.Cpu
|
|
|
|
{
|
2018-11-01 05:22:09 +01:00
|
|
|
[Category("SimdRegElem")]
|
2018-09-08 19:24:29 +02:00
|
|
|
public sealed class CpuTestSimdRegElem : CpuTest
|
|
|
|
{
|
|
|
|
#if SimdRegElem
|
|
|
|
|
2018-10-14 04:35:16 +02:00
|
|
|
#region "ValueSource (Types)"
|
2018-09-08 19:24:29 +02:00
|
|
|
private static ulong[] _2S_()
|
|
|
|
{
|
|
|
|
return new ulong[] { 0x0000000000000000ul, 0x7FFFFFFF7FFFFFFFul,
|
|
|
|
0x8000000080000000ul, 0xFFFFFFFFFFFFFFFFul };
|
|
|
|
}
|
|
|
|
|
|
|
|
private static ulong[] _4H_()
|
|
|
|
{
|
|
|
|
return new ulong[] { 0x0000000000000000ul, 0x7FFF7FFF7FFF7FFFul,
|
|
|
|
0x8000800080008000ul, 0xFFFFFFFFFFFFFFFFul };
|
|
|
|
}
|
|
|
|
#endregion
|
|
|
|
|
2018-10-14 04:35:16 +02:00
|
|
|
#region "ValueSource (Opcodes)"
|
2021-03-25 23:33:32 +01:00
|
|
|
private static uint[] _Mla_Mls_Mul_Sqdmulh_Sqrdmulh_Ve_4H_8H_()
|
2018-09-08 19:24:29 +02:00
|
|
|
{
|
2018-10-14 04:35:16 +02:00
|
|
|
return new uint[]
|
|
|
|
{
|
2021-03-25 23:33:32 +01:00
|
|
|
0x2F400000u, // MLA V0.4H, V0.4H, V0.H[0]
|
|
|
|
0x2F404000u, // MLS V0.4H, V0.4H, V0.H[0]
|
|
|
|
0x0F408000u, // MUL V0.4H, V0.4H, V0.H[0]
|
|
|
|
0x0F40C000u, // SQDMULH V0.4H, V0.4H, V0.H[0]
|
|
|
|
0x0F40D000u // SQRDMULH V0.4H, V0.4H, V0.H[0]
|
2018-10-14 04:35:16 +02:00
|
|
|
};
|
2018-09-08 19:24:29 +02:00
|
|
|
}
|
|
|
|
|
2021-03-25 23:33:32 +01:00
|
|
|
private static uint[] _Mla_Mls_Mul_Sqdmulh_Sqrdmulh_Ve_2S_4S_()
|
2018-09-08 19:24:29 +02:00
|
|
|
{
|
2018-10-14 04:35:16 +02:00
|
|
|
return new uint[]
|
|
|
|
{
|
2021-03-25 23:33:32 +01:00
|
|
|
0x2F800000u, // MLA V0.2S, V0.2S, V0.S[0]
|
|
|
|
0x2F804000u, // MLS V0.2S, V0.2S, V0.S[0]
|
|
|
|
0x0F808000u, // MUL V0.2S, V0.2S, V0.S[0]
|
|
|
|
0x0F80C000u, // SQDMULH V0.2S, V0.2S, V0.S[0]
|
|
|
|
0x0F80D000u // SQRDMULH V0.2S, V0.2S, V0.S[0]
|
2018-10-14 04:35:16 +02:00
|
|
|
};
|
2018-09-08 19:24:29 +02:00
|
|
|
}
|
2019-01-29 14:54:39 +01:00
|
|
|
|
|
|
|
private static uint[] _SU_Mlal_Mlsl_Mull_Ve_4H4S_8H4S_()
|
|
|
|
{
|
|
|
|
return new uint[]
|
|
|
|
{
|
|
|
|
0x0F402000u, // SMLAL V0.4S, V0.4H, V0.H[0]
|
|
|
|
0x0F406000u, // SMLSL V0.4S, V0.4H, V0.H[0]
|
|
|
|
0x0F40A000u, // SMULL V0.4S, V0.4H, V0.H[0]
|
|
|
|
0x2F402000u, // UMLAL V0.4S, V0.4H, V0.H[0]
|
|
|
|
0x2F406000u, // UMLSL V0.4S, V0.4H, V0.H[0]
|
|
|
|
0x2F40A000u // UMULL V0.4S, V0.4H, V0.H[0]
|
|
|
|
};
|
|
|
|
}
|
|
|
|
|
|
|
|
private static uint[] _SU_Mlal_Mlsl_Mull_Ve_2S2D_4S2D_()
|
|
|
|
{
|
|
|
|
return new uint[]
|
|
|
|
{
|
|
|
|
0x0F802000u, // SMLAL V0.2D, V0.2S, V0.S[0]
|
|
|
|
0x0F806000u, // SMLSL V0.2D, V0.2S, V0.S[0]
|
|
|
|
0x0F80A000u, // SMULL V0.2D, V0.2S, V0.S[0]
|
|
|
|
0x2F802000u, // UMLAL V0.2D, V0.2S, V0.S[0]
|
|
|
|
0x2F806000u, // UMLSL V0.2D, V0.2S, V0.S[0]
|
|
|
|
0x2F80A000u // UMULL V0.2D, V0.2S, V0.S[0]
|
|
|
|
};
|
|
|
|
}
|
2018-10-14 04:35:16 +02:00
|
|
|
#endregion
|
|
|
|
|
2019-06-30 01:02:48 +02:00
|
|
|
private const int RndCnt = 2;
|
|
|
|
private const int RndCntIndex = 2;
|
2018-09-08 19:24:29 +02:00
|
|
|
|
2018-10-14 04:35:16 +02:00
|
|
|
[Test, Pairwise]
|
2021-03-25 23:33:32 +01:00
|
|
|
public void Mla_Mls_Mul_Sqdmulh_Sqrdmulh_Ve_4H_8H([ValueSource(nameof(_Mla_Mls_Mul_Sqdmulh_Sqrdmulh_Ve_4H_8H_))] uint opcodes,
|
|
|
|
[Values(0u)] uint rd,
|
|
|
|
[Values(1u, 0u)] uint rn,
|
|
|
|
[Values(2u, 0u)] uint rm,
|
|
|
|
[ValueSource(nameof(_4H_))] [Random(RndCnt)] ulong z,
|
|
|
|
[ValueSource(nameof(_4H_))] [Random(RndCnt)] ulong a,
|
|
|
|
[ValueSource(nameof(_4H_))] [Random(RndCnt)] ulong b,
|
|
|
|
[Values(0u, 7u)] [Random(1u, 6u, RndCntIndex)] uint index,
|
|
|
|
[Values(0b0u, 0b1u)] uint q) // <4H, 8H>
|
2018-09-08 19:24:29 +02:00
|
|
|
{
|
2018-11-01 05:22:09 +01:00
|
|
|
uint h = (index >> 2) & 1;
|
|
|
|
uint l = (index >> 1) & 1;
|
|
|
|
uint m = index & 1;
|
2018-09-08 19:24:29 +02:00
|
|
|
|
2018-11-01 05:22:09 +01:00
|
|
|
opcodes |= ((rm & 15) << 16) | ((rn & 31) << 5) | ((rd & 31) << 0);
|
|
|
|
opcodes |= (l << 21) | (m << 20) | (h << 11);
|
|
|
|
opcodes |= ((q & 1) << 30);
|
2018-09-08 19:24:29 +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
|
|
|
V128 v0 = MakeVectorE0E1(z, z);
|
|
|
|
V128 v1 = MakeVectorE0E1(a, a * q);
|
|
|
|
V128 v2 = MakeVectorE0E1(b, b * h);
|
2018-09-08 19:24:29 +02:00
|
|
|
|
2018-11-01 05:22:09 +01:00
|
|
|
SingleOpcode(opcodes, v0: v0, v1: v1, v2: v2);
|
2018-09-08 19:24:29 +02:00
|
|
|
|
2021-03-25 23:33:32 +01:00
|
|
|
CompareAgainstUnicorn(fpsrMask: Fpsr.Qc);
|
2018-09-08 19:24:29 +02:00
|
|
|
}
|
|
|
|
|
2018-10-14 04:35:16 +02:00
|
|
|
[Test, Pairwise]
|
2021-03-25 23:33:32 +01:00
|
|
|
public void Mla_Mls_Mul_Sqdmulh_Sqrdmulh_Ve_2S_4S([ValueSource(nameof(_Mla_Mls_Mul_Sqdmulh_Sqrdmulh_Ve_2S_4S_))] uint opcodes,
|
|
|
|
[Values(0u)] uint rd,
|
|
|
|
[Values(1u, 0u)] uint rn,
|
|
|
|
[Values(2u, 0u)] uint rm,
|
|
|
|
[ValueSource(nameof(_2S_))] [Random(RndCnt)] ulong z,
|
|
|
|
[ValueSource(nameof(_2S_))] [Random(RndCnt)] ulong a,
|
|
|
|
[ValueSource(nameof(_2S_))] [Random(RndCnt)] ulong b,
|
|
|
|
[Values(0u, 1u, 2u, 3u)] uint index,
|
|
|
|
[Values(0b0u, 0b1u)] uint q) // <2S, 4S>
|
2018-09-08 19:24:29 +02:00
|
|
|
{
|
2018-11-01 05:22:09 +01:00
|
|
|
uint h = (index >> 1) & 1;
|
|
|
|
uint l = index & 1;
|
2018-09-08 19:24:29 +02:00
|
|
|
|
2018-11-01 05:22:09 +01:00
|
|
|
opcodes |= ((rm & 15) << 16) | ((rn & 31) << 5) | ((rd & 31) << 0);
|
|
|
|
opcodes |= (l << 21) | (h << 11);
|
|
|
|
opcodes |= ((q & 1) << 30);
|
2018-09-08 19:24:29 +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
|
|
|
V128 v0 = MakeVectorE0E1(z, z);
|
|
|
|
V128 v1 = MakeVectorE0E1(a, a * q);
|
|
|
|
V128 v2 = MakeVectorE0E1(b, b * h);
|
2018-09-08 19:24:29 +02:00
|
|
|
|
2018-11-01 05:22:09 +01:00
|
|
|
SingleOpcode(opcodes, v0: v0, v1: v1, v2: v2);
|
2018-09-08 19:24:29 +02:00
|
|
|
|
2021-03-25 23:33:32 +01:00
|
|
|
CompareAgainstUnicorn(fpsrMask: Fpsr.Qc);
|
2018-09-08 19:24:29 +02:00
|
|
|
}
|
2019-01-29 14:54:39 +01:00
|
|
|
|
|
|
|
[Test, Pairwise]
|
|
|
|
public void SU_Mlal_Mlsl_Mull_Ve_4H4S_8H4S([ValueSource("_SU_Mlal_Mlsl_Mull_Ve_4H4S_8H4S_")] uint opcodes,
|
|
|
|
[Values(0u)] uint rd,
|
|
|
|
[Values(1u, 0u)] uint rn,
|
|
|
|
[Values(2u, 0u)] uint rm,
|
|
|
|
[ValueSource("_4H_")] [Random(RndCnt)] ulong z,
|
|
|
|
[ValueSource("_4H_")] [Random(RndCnt)] ulong a,
|
|
|
|
[ValueSource("_4H_")] [Random(RndCnt)] ulong b,
|
2019-06-30 01:02:48 +02:00
|
|
|
[Values(0u, 7u)] [Random(1u, 6u, RndCntIndex)] uint index,
|
2019-01-29 14:54:39 +01:00
|
|
|
[Values(0b0u, 0b1u)] uint q) // <4H4S, 8H4S>
|
|
|
|
{
|
|
|
|
uint h = (index >> 2) & 1;
|
|
|
|
uint l = (index >> 1) & 1;
|
|
|
|
uint m = index & 1;
|
|
|
|
|
|
|
|
opcodes |= ((rm & 15) << 16) | ((rn & 31) << 5) | ((rd & 31) << 0);
|
|
|
|
opcodes |= (l << 21) | (m << 20) | (h << 11);
|
|
|
|
opcodes |= ((q & 1) << 30);
|
|
|
|
|
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
|
|
|
V128 v0 = MakeVectorE0E1(z, z);
|
|
|
|
V128 v1 = MakeVectorE0E1(q == 0u ? a : 0ul, q == 1u ? a : 0ul);
|
|
|
|
V128 v2 = MakeVectorE0E1(b, b * h);
|
2019-01-29 14:54:39 +01:00
|
|
|
|
|
|
|
SingleOpcode(opcodes, v0: v0, v1: v1, v2: v2);
|
|
|
|
|
|
|
|
CompareAgainstUnicorn();
|
|
|
|
}
|
|
|
|
|
|
|
|
[Test, Pairwise]
|
|
|
|
public void SU_Mlal_Mlsl_Mull_Ve_2S2D_4S2D([ValueSource("_SU_Mlal_Mlsl_Mull_Ve_2S2D_4S2D_")] uint opcodes,
|
|
|
|
[Values(0u)] uint rd,
|
|
|
|
[Values(1u, 0u)] uint rn,
|
|
|
|
[Values(2u, 0u)] uint rm,
|
|
|
|
[ValueSource("_2S_")] [Random(RndCnt)] ulong z,
|
|
|
|
[ValueSource("_2S_")] [Random(RndCnt)] ulong a,
|
|
|
|
[ValueSource("_2S_")] [Random(RndCnt)] ulong b,
|
|
|
|
[Values(0u, 1u, 2u, 3u)] uint index,
|
|
|
|
[Values(0b0u, 0b1u)] uint q) // <2S2D, 4S2D>
|
|
|
|
{
|
|
|
|
uint h = (index >> 1) & 1;
|
|
|
|
uint l = index & 1;
|
|
|
|
|
|
|
|
opcodes |= ((rm & 15) << 16) | ((rn & 31) << 5) | ((rd & 31) << 0);
|
|
|
|
opcodes |= (l << 21) | (h << 11);
|
|
|
|
opcodes |= ((q & 1) << 30);
|
|
|
|
|
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
|
|
|
V128 v0 = MakeVectorE0E1(z, z);
|
|
|
|
V128 v1 = MakeVectorE0E1(q == 0u ? a : 0ul, q == 1u ? a : 0ul);
|
|
|
|
V128 v2 = MakeVectorE0E1(b, b * h);
|
2019-01-29 14:54:39 +01:00
|
|
|
|
|
|
|
SingleOpcode(opcodes, v0: v0, v1: v1, v2: v2);
|
|
|
|
|
|
|
|
CompareAgainstUnicorn();
|
|
|
|
}
|
2018-09-08 19:24:29 +02:00
|
|
|
#endif
|
|
|
|
}
|
|
|
|
}
|