core.simd

Builtin SIMD intrinsics

Source

core/simd.d

License:

Boost License 1.0.

Authors:

Walter Bright,

template Vector(T)

Create a vector type.

Parameters

T = one of double[2], float[4], void[16], byte[16], ubyte[16], short[8], ushort[8], int[4], uint[4], long[2], ulong[2]. For 256 bit vectors, one of double[4], float[8], void[32], byte[32], ubyte[32], short[16], ushort[16], int[8], uint[8], long[4], ulong[4]

alias void16 = __vector(void[16]);

alias double2 = __vector(double[2]);

alias float4 = __vector(float[4]);

alias byte16 = __vector(byte[16]);

alias ubyte16 = __vector(ubyte[16]);

alias short8 = __vector(short[8]);

alias ushort8 = __vector(ushort[8]);

alias int4 = __vector(int[4]);

alias uint4 = __vector(uint[4]);

alias long2 = __vector(long[2]);

alias ulong2 = __vector(ulong[2]);

alias void32 = __vector(void[32]);

alias double4 = __vector(double[4]);

alias float8 = __vector(float[8]);

alias byte32 = __vector(byte[32]);

alias ubyte32 = __vector(ubyte[32]);

alias short16 = __vector(short[16]);

alias ushort16 = __vector(ushort[16]);

alias int8 = __vector(int[8]);

alias uint8 = __vector(uint[8]);

alias long4 = __vector(long[4]);

alias ulong4 = __vector(ulong[4]);

enum XMM: int;

XMM opcodes that conform to the following:

opcode xmm1,xmm2/mem

and do not have side effects (i.e. do not write to memory).

pure @safe V1 simd(XMM opcode, V1, V2)(V1 op1, V2 op2)
Constraints: if (is(V1 == __vector) && is(V2 == __vector));

Generate two operand instruction with XMM 128 bit operands.

This is a compiler magic function - it doesn't behave like regular D functions.

Parameters

opcode any of the XMM opcodes; it must be a compile time constant op1 first operand op2 second operand

Returns:

result of opcode

pure @safe V1 simd(XMM opcode, V1)(V1 op1)
Constraints: if (is(V1 == __vector));

Unary SIMD instructions.

pure @safe V1 simd(XMM opcode, V1)(double d)
Constraints: if (is(V1 == __vector));

pure @safe V1 simd(XMM opcode, V1)(float f)
Constraints: if (is(V1 == __vector));

pure @safe V1 simd(XMM opcode, ubyte imm8, V1, V2)(V1 op1, V2 op2)
Constraints: if (is(V1 == __vector) && is(V2 == __vector));

For instructions: CMPPD, CMPSS, CMPSD, CMPPS, PSHUFD, PSHUFHW, PSHUFLW, BLENDPD, BLENDPS, DPPD, DPPS, MPSADBW, PBLENDW, ROUNDPD, ROUNDPS, ROUNDSD, ROUNDSS

Parameters

opcode any of the above XMM opcodes; it must be a compile time constant op1 first operand op2 second operand imm8 third operand; must be a compile time constant

Returns:

result of opcode

pure @safe V1 simd(XMM opcode, ubyte imm8, V1)(V1 op1)
Constraints: if (is(V1 == __vector));

For instructions with the imm8 version: PSLLD, PSLLQ, PSLLW, PSRAD, PSRAW, PSRLD, PSRLQ, PSRLW, PSRLDQ, PSLLDQ

Parameters

opcode any of the XMM opcodes; it must be a compile time constant op1 first operand imm8 second operand; must be a compile time constant

Returns:

result of opcode

@safe V1 simd_sto(XMM opcode, V1, V2)(V1 op1, V2 op2)
Constraints: if (is(V1 == __vector) && is(V2 == __vector));

For "store" operations of the form: op1 op= op2

Returns:

op2 These cannot be marked as pure, as semantic() doesn't check them.

@safe V1 simd_stod(XMM opcode, V1, V2)(double op1, V1 op2)
Constraints: if (is(V1 == __vector));

@safe V1 simd_stof(XMM opcode, V1)(float op1, V1 op2)
Constraints: if (is(V1 == __vector));

void prefetch(bool writeFetch, ubyte locality)(const(void)* address);

Emit prefetch instruction.

Parameters:

const(void)* address	address to be prefetched
writeFetch	true for write fetch, false for read fetch
locality	0..3 (0 meaning least local, 3 meaning most local)

Note

The Intel mappings are:

writeFetch	locality	Instruction
false	0	prefetchnta
false	1	prefetch2
false	2	prefetch1
false	3	prefetch0
false	0	prefetchw
false	1	prefetchw
false	2	prefetchw
false	3	prefetchw

V loadUnaligned(V)(const V* p)
Constraints: if (is(V == void16) || is(V == byte16) || is(V == ubyte16) || is(V == short8) || is(V == ushort8) || is(V == int4) || is(V == uint4) || is(V == long2) || is(V == ulong2));

Load unaligned vector from address. This is a compiler intrinsic.

Parameters:

V* p	pointer to vector

Returns:

vector

V storeUnaligned(V)(V* p, V value)
Constraints: if (is(V == void16) || is(V == byte16) || is(V == ubyte16) || is(V == short8) || is(V == ushort8) || is(V == int4) || is(V == uint4) || is(V == long2) || is(V == ulong2));

Store vector to unaligned address. This is a compiler intrinsic.

Parameters:

V* p	pointer to vector
V value	value to store

Returns:

value