lib/Runtime/Language/SimdUtils.h

//-------------------------------------------------------------------------------------------------------
// Copyright (C) Microsoft Corporation and contributors. All rights reserved.
// Licensed under the MIT license. See LICENSE.txt file in the project root for full license information.
//-------------------------------------------------------------------------------------------------------

#pragma once

#define SIMD128_TYPE_SPEC_FLAG Js::Configuration::Global.flags.Simd128TypeSpec

// The representations below assume little-endian.
#define SIMD_X 0
#define SIMD_Y 1
#define SIMD_Z 2
#define SIMD_W 3
#define FLOAT64_SIZE 8
#define FLOAT32_SIZE 4
#define INT32_SIZE   4
#define INT16_SIZE   2
#define INT8_SIZE    1
#define SIMD_INDEX_VALUE_MAX     5
#define SIMD_STRING_BUFFER_MAX   1024
#define SIMD_DATA     \
    Field(int32)   i32[4];\
    Field(int16)   i16[8];\
    Field(int8)    i8[16];\
    Field(uint32)  u32[4];\
    Field(uint16)  u16[8];\
    Field(uint8)   u8[16];\
    Field(float)   f32[4];\
    Field(double)  f64[2];
#define SIMD_TEMP_SIZE 3
struct _SIMDValue
{
    union{
        SIMD_DATA
    };

    void SetValue(_SIMDValue value)
    {
        i32[SIMD_X] = value.i32[SIMD_X];
        i32[SIMD_Y] = value.i32[SIMD_Y];
        i32[SIMD_Z] = value.i32[SIMD_Z];
        i32[SIMD_W] = value.i32[SIMD_W];
    }
    void Zero()
    {
        f64[SIMD_X] = f64[SIMD_Y] = 0;
    }
    bool operator==(const _SIMDValue& r)
    {
        // don't compare f64/f32 because NaN bit patterns will not be considered equal.
        return (this->i32[SIMD_X] == r.i32[SIMD_X] &&
            this->i32[SIMD_Y] == r.i32[SIMD_Y] &&
            this->i32[SIMD_Z] == r.i32[SIMD_Z] &&
            this->i32[SIMD_W] == r.i32[SIMD_W]);
    }
    bool IsZero()
    {
        return (i32[SIMD_X] == 0 && i32[SIMD_Y] == 0 && i32[SIMD_Z] == 0 && i32[SIMD_W] == 0);
    }

};
typedef _SIMDValue SIMDValue;

// For dictionary use
template <>
struct DefaultComparer<_SIMDValue>
{
    __forceinline static bool Equals(_SIMDValue x, _SIMDValue y)
    {
        return x == y;
    }

    __forceinline static hash_t GetHashCode(_SIMDValue d)
    {
        return (hash_t)(d.i32[SIMD_X] ^ d.i32[SIMD_Y] ^ d.i32[SIMD_Z] ^ d.i32[SIMD_W]);
    }
};

#if _M_IX86 || _M_AMD64
struct _x86_SIMDValue
{
    union{
        SIMD_DATA
        __m128  m128_value;
        __m128d m128d_value;
        __m128i m128i_value;
    };

    static _x86_SIMDValue ToX86SIMDValue(const SIMDValue& val)
    {
        _x86_SIMDValue result;
        result.m128_value = _mm_loadu_ps((float*) &val);
        return result;
    }

    static SIMDValue ToSIMDValue(const _x86_SIMDValue& val)
    {
        SIMDValue result;
        _mm_storeu_ps((float*) &result, val.m128_value);
        return result;
    }
};

#pragma warning(push)
#pragma warning(disable:4838) // conversion from 'unsigned int' to 'int32' requires a narrowing conversion

// These global values are 16-byte aligned.
const _x86_SIMDValue X86_ABS_MASK_F4 = { 0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff };
const _x86_SIMDValue X86_ABS_MASK_I4 = { 0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff };
const _x86_SIMDValue X86_ABS_MASK_D2 = { 0xffffffff, 0x7fffffff, 0xffffffff, 0x7fffffff };

const _x86_SIMDValue X86_NEG_MASK_F4 = { 0x80000000, 0x80000000, 0x80000000, 0x80000000 };
const _x86_SIMDValue X86_NEG_MASK_D2 = { 0x00000000, 0x80000000, 0x00000000, 0x80000000 };

const _x86_SIMDValue X86_ALL_ONES_F4 = { 0x3f800000, 0x3f800000, 0x3f800000, 0x3f800000 }; // {1.0, 1.0, 1.0, 1.0}
const _x86_SIMDValue X86_ALL_ONES_I4 = { 0x00000001, 0x00000001, 0x00000001, 0x00000001 }; // {1, 1, 1, 1}
const _x86_SIMDValue X86_ALL_ONES_D2 = { 0x00000000, 0x3ff00000, 0x00000000, 0x3ff00000 }; // {1.0, 1.0}

const _x86_SIMDValue X86_ALL_NEG_ONES = { 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff };
const _x86_SIMDValue X86_ALL_NEG_ONES_F4 = { 0xBF800000, 0xBF800000, 0xBF800000, 0xBF800000 }; //-1.0, -1.0, -1.0, -1.0

const _x86_SIMDValue X86_ALL_ONES_I8  = { 0x00010001, 0x00010001, 0x00010001, 0x00010001 }; // {1, 1, 1, 1, 1, 1, 1, 1}

const _x86_SIMDValue X86_ALL_ZEROS          = { 0x00000000, 0x00000000, 0x00000000, 0x00000000 };
const _x86_SIMDValue X86_ALL_ONES_I16       = { 0x01010101, 0x01010101, 0x01010101, 0x01010101 }; // {1, 1, 1, 1,1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1}
const _x86_SIMDValue X86_LANE0_ONES_I16     = { 0x000000ff, 0x00000000, 0x00000000, 0x00000000 };
const _x86_SIMDValue X86_LOWBYTES_MASK      = { 0x00ff00ff, 0x00ff00ff, 0x00ff00ff, 0x00ff00ff };
const _x86_SIMDValue X86_HIGHBYTES_MASK     = { 0xff00ff00, 0xff00ff00, 0xff00ff00, 0xff00ff00 };

const _x86_SIMDValue X86_LANE_W_ZEROS = { 0xffffffff, 0xffffffff, 0xffffffff, 0x00000000 };

const _x86_SIMDValue X86_TWO_31_F4          = { 0x4f000000, 0x4f000000, 0x4f000000, 0x4f000000 }; // f32(2^31), ....
const _x86_SIMDValue X86_NEG_TWO_31_F4      = { 0xcf000000, 0xcf000000, 0xcf000000, 0xcf000000 }; // f32(-2^31), ....
const _x86_SIMDValue X86_TWO_32_F4          = { 0x4f800000, 0x4f800000, 0x4f800000, 0x4f800000 }; // f32(2^32), ....
const _x86_SIMDValue X86_TWO_31_I4          = X86_NEG_MASK_F4;                                    // 2^31, ....
const _x86_SIMDValue X86_WORD_SIGNBITS      = { 0x80008000, 0x80008000, 0x80008000, 0x80008000 };
const _x86_SIMDValue X86_DWORD_SIGNBITS     = { 0x80000000, 0x80000000, 0x80000000, 0x80000000 };
const _x86_SIMDValue X86_BYTE_SIGNBITS      = { 0x80808080, 0x80808080, 0x80808080, 0x80808080 };
const _x86_SIMDValue X86_4LANES_MASKS[]     = {{ 0xffffffff, 0x00000000, 0x00000000, 0x00000000 },
                                               { 0x00000000, 0xffffffff, 0x00000000, 0x00000000 },
                                               { 0x00000000, 0x00000000, 0xffffffff, 0x00000000 },
                                               { 0x00000000, 0x00000000, 0x00000000, 0xffffffff }};


#pragma warning(pop)

#if ENABLE_NATIVE_CODEGEN && defined(ENABLE_SIMDJS)
// auxiliary SIMD values in memory to help JIT'ed code. E.g. used for Int8x16 shuffle.
extern _x86_SIMDValue X86_TEMP_SIMD[];
#endif

typedef _x86_SIMDValue X86SIMDValue;
CompileAssert(sizeof(X86SIMDValue) == 16);
#endif

typedef SIMDValue     AsmJsSIMDValue; // alias for asmjs
CompileAssert(sizeof(SIMDValue) == 16);

class ValueType;

namespace Js {
    enum class OpCode : ushort;
    ///////////////////////////////////////////////////////////////
    //Class with static helper methods for manipulating SIMD Data.
    ///////////////////////////////////////////////////////////////
    class SIMDUtils
    {
    private:
        template<typename SIMDType, typename T>
        static SIMDValue SIMD128SetLaneValue(const Var aVar, const uint32 laneValue, const T value)
        {
            Assert(IsSimdType(aVar));
            SIMDType *jsVal = SIMDType::FromVar(aVar);
            SIMDValue simdValue = jsVal->GetValue();

            switch (GetSIMDLaneCount(aVar))
            {
            case 4:
                return (JavascriptSIMDFloat32x4::Is(aVar)) ?
                       SIMD128InnerReplaceLaneF4(simdValue, laneValue, static_cast<float>(value)):
                       SIMD128InnerReplaceLaneI4(simdValue, laneValue, static_cast<int32>(value));
            case 8:
                return SIMD128InnerReplaceLaneI8(simdValue, laneValue, static_cast<int16>(value));
            case 16:
                return SIMD128InnerReplaceLaneI16(simdValue, laneValue, static_cast<int8>(value));
            default:
                Assert(UNREACHED);
            }
            return simdValue; //Unreached
        };

        template<typename SIMDType, uint32 laneCount, typename T>
        static T SIMD128GetLaneValue(const Var aVar, const uint32 laneValue)
        {
            Assert(IsSimdType(aVar));
            SIMDType *jsVal = SIMDType::FromVar(aVar);

            switch (laneCount)
            {
            case 4:
                return (JavascriptSIMDFloat32x4::Is(aVar)) ?
                       static_cast<T>(SIMD128InnerExtractLaneF4(jsVal->GetValue(), laneValue)):
                       static_cast<T>(SIMD128InnerExtractLaneI4(jsVal->GetValue(), laneValue));
            case 8:
                return static_cast<T>(SIMD128InnerExtractLaneI8(jsVal->GetValue(), laneValue));
            case 16:
                return static_cast<T>(SIMD128InnerExtractLaneI16(jsVal->GetValue(), laneValue));
            default:
                Assert(UNREACHED);
            }
            return 0; //unreached
        };

    public:
        static bool IsSimdType(const Var aVar);
        static uint32 GetSIMDLaneCount(const Var aVar);
        static inline uint32 SIMDCheckTypedArrayIndex(ScriptContext* scriptContext, const Var index);
        static uint32 SIMDCheckLaneIndex(ScriptContext* scriptContext, const Var lane, const uint32 range = 4);
        static uint32 inline SIMDGetShiftAmountMask(uint32 eleSizeInBytes) { return (eleSizeInBytes << 3) - 1; }

        ////////////////////////////////////////////
        //SIMD Extract Lane / Replace Lane Helpers
        ////////////////////////////////////////////
        static inline SIMDValue SIMD128InnerReplaceLaneF4(SIMDValue simdVal, const uint32 lane, const float value)
        {
            simdVal.f32[lane] = value;
            return simdVal;
        };
        static inline SIMDValue SIMD128InnerReplaceLaneI4(SIMDValue simdVal, const uint32 lane, const int32 value)
        {
            simdVal.i32[lane] = value;
            return simdVal;
        };
        static inline SIMDValue SIMD128InnerReplaceLaneI8(SIMDValue simdVal, const uint32 lane, const int16 value)
        {
            simdVal.i16[lane] = value;
            return simdVal;
        };
        static inline SIMDValue SIMD128InnerReplaceLaneI16(SIMDValue simdVal, const uint32 lane, const int8 value)
        {
            simdVal.i8[lane] = value;
            return simdVal;
        };

        static inline float SIMD128InnerExtractLaneF4(const SIMDValue src1, const uint32 lane) { return src1.f32[lane]; };
        static inline int32 SIMD128InnerExtractLaneI4(const SIMDValue src1, const uint32 lane) { return src1.i32[lane]; };
        static inline int16 SIMD128InnerExtractLaneI8(const SIMDValue src1, const uint32 lane) { return src1.i16[lane]; };
        static inline int8 SIMD128InnerExtractLaneI16(const SIMDValue src1, const uint32 lane) { return src1.i8[lane];  };

        template<class SIMDType, uint32 laneCount, typename T>
        static inline T SIMD128ExtractLane(const Var src, const Var lane, ScriptContext* scriptContext)
        {
            uint32 laneValue = SIMDUtils::SIMDCheckLaneIndex(scriptContext, lane, laneCount);
            Assert(laneValue >= 0 && laneValue < laneCount);
            return SIMD128GetLaneValue<SIMDType, laneCount, T>(src, laneValue);
        }

        template<class SIMDType, uint32 laneCount, typename T>
        static inline SIMDValue SIMD128ReplaceLane(const Var src, const Var lane, const T value, ScriptContext* scriptContext)
        {
            uint32 laneValue = SIMDUtils::SIMDCheckLaneIndex(scriptContext, lane, laneCount);
            Assert(laneValue >= 0 && laneValue < laneCount);
            return SIMD128SetLaneValue<SIMDType, T>(src, laneValue, value);
        }

        ////////////////////////////////////////////
        // SIMD Shuffle Swizzle helpers
        ////////////////////////////////////////////
        static SIMDValue SIMD128InnerShuffle(const SIMDValue src1, const SIMDValue src2, uint32 laneCount, const uint32* lanes = nullptr);

        template <class SIMDType>
        static Var SIMD128SlowShuffle(Var src1, Var src2, Var *lanes, const uint32 laneCount, const uint32 range, ScriptContext* scriptContext);

        ///////////////////////////////////////////
        // SIMD Type conversion
        ///////////////////////////////////////////
        template<class SIMDType1, class SIMDType2>
        static inline Var SIMDConvertTypeFromBits(SIMDType1 &instance, ScriptContext& requestContext)
        {
            SIMDValue result = FromSimdBits(instance.GetValue());
            return SIMDType2::New(&result, &requestContext);
        }
        static SIMDValue FromSimdBits(const SIMDValue value);

        ///////////////////////////////////////////
        // SIMD Data load/store
        ///////////////////////////////////////////
        static SIMDValue SIMDLdData(const SIMDValue *data, uint8 dataWidth);
        static void SIMDStData(SIMDValue *data, const SIMDValue simdValue, uint8 dataWidth);
        template<bool acceptNegZero = false>
        static uint32 SIMDCheckUint32Number(ScriptContext* scriptContext, const Var value);
        static bool SIMDIsSupportedTypedArray(Var value);
        static SIMDValue*  SIMDCheckTypedArrayAccess(Var arg1, Var arg2, TypedArrayBase **tarray, int32 *index, uint32 dataWidth, ScriptContext *scriptContext);

        template <class SIMDType>
        static Var SIMD128TypedArrayLoad(Var arg1, Var arg2, uint32 dataWidth, ScriptContext *scriptContext)
        {
            Assert(dataWidth >= 4 && dataWidth <= 16);
            TypedArrayBase *tarray = NULL;
            int32 index = -1;
            SIMDValue* data = NULL;

            data = SIMDUtils::SIMDCheckTypedArrayAccess(arg1, arg2, &tarray, &index, dataWidth, scriptContext);
            Assert(tarray != NULL);
            Assert(index >= 0);
            Assert(data != NULL);

            SIMDValue result = SIMDUtils::SIMDLdData(data, (uint8)dataWidth);
            return SIMDType::New(&result, scriptContext);
        }

        template <class SIMDType>
        static void SIMD128TypedArrayStore(Var arg1, Var arg2, Var simdVar, uint32 dataWidth, ScriptContext *scriptContext)
        {
            Assert(dataWidth >= 4 && dataWidth <= 16);
            TypedArrayBase *tarray = NULL;
            int32 index = -1;
            SIMDValue* data = NULL;

            data = SIMDUtils::SIMDCheckTypedArrayAccess(arg1, arg2, &tarray, &index, dataWidth, scriptContext);
            Assert(tarray != NULL);
            Assert(index >= 0);
            Assert(data != NULL);

            SIMDValue simdValue = SIMDType::FromVar(simdVar)->GetValue();
            SIMDUtils::SIMDStData(data, simdValue, (uint8)dataWidth);
        }
        static uint32 SimdOpcodeAsIndex(Js::OpCode op);
    };
}