The OpenD Programming Language

Vector Extensions

CPUs often support specialized vector types and vector operations (a.k.a. media instructions). Vector types are a fixed array of floating or integer types, and vector operations operate simultaneously on them.

Specialized type, Vector types provide access to them.

The type, VectorBaseType must be a Static Array. The VectorElementType is the unqualified element type of the static array. The dimension of the static array is the number of elements in the vector.

Which vector types are supported depends on the target. The implementation is expected to only support the vector types and operations that are implemented in the target's hardware.

Rationale: Emulating unsupported vector types and operations can exhibit such poor performance that the user is likely better off selecting a different algorithm than relying on emulation.

Use the declarations in core.simd instead of the language type, Vector grammar.

core.simd

Vector types and operations are introduced by importing core.simd:

import core.simd;
These types and operations will be the ones defined for the architecture the compiler is targeting. If a particular CPU family has varying support for vector types, an additional runtime check may be necessary. The compiler does not emit runtime checks; those must be done by the programmer.
Depending on the target architecture, compiler flags may be required to activate support for SIMD types.

The types defined will all follow the naming convention:

typeNN

where type is the vector element type and NN is the number of those elements in the vector type. The type names will not be keywords.

Properties

Vector types have the property:

Vector Type Properties
Property Description
.arrayReturns static array representation

Vectors support the following properties based on the vector element type. The value produced is that of a vector of the same type with each element set to the value corresponding to the property value for the element type.

Integral Vector Type Properties
Property Description
.minminimum value
.maxmaximum value
Floating Point Vector Type Properties
Property Description
.epsilonsmallest increment to the value 1
.infinityinfinity value
.maxlargest representable value that is not infinity
.min_normalsmallest representable value that is not 0
.nanNaN value

Conversions

Vector types of the same size (number_of_elements * size_of_element) can be implicitly converted among each other, this is done as a reinterpret cast (a type paint). Vector types can be cast to their type, VectorBaseType.

Integers and floating point values can be implicitly converted to their vector equivalents:

int4 v = 7;
v = 3 + v;   // add 3 to each element in v

Accessing Individual Vector Elements

They cannot be accessed directly, but can be when converted to an array type:

int4 v;
(cast(int*)&v)[3] = 2;   // set 3rd element of the 4 int vector
(cast(int[4])v)[3] = 2;  // set 3rd element of the 4 int vector
v.array[3] = 2;          // set 3rd element of the 4 int vector
v.ptr[3] = 2;            // set 3rd element of the 4 int vector

Conditional Compilation

If vector extensions are implemented, the version identifier D_SIMD is set.

Whether a type exists or not can be tested at compile time with an IsExpression:

static if (is(typeNN))
    ... yes, it is supported ...
else
    ... nope, use workaround ...

Whether a particular operation on a type is supported can be tested at compile time with:

float4 a,b;
static if (__traits(compiles, a+b))
    ... yes, add is supported for float4 ...
else
    ... nope, use workaround ...

For runtime testing to see if certain vector instructions are available, see the functions in core.cpuid.

A typical workaround for unsupported vector operations would be to use array operations instead:

float4 a,b;
static if (__traits(compiles, a/b))
    c = a / b;
else
    c[] = a[] / b[];

X86 And X86_64 Vector Extension Implementation

The following describes the specific implementation of the vector types for the X86 and X86_64 architectures.

The vector extensions are currently implemented for the OS X 32 bit target, and all 64 bit targets.

core.simd defines the following types:

Vector Types
Type Name Description gcc Equivalent
void1616 bytes of untyped datano equivalent
byte1616 bytessigned char __attribute__((vector_size(16)))
ubyte1616 ubytesunsigned char __attribute__((vector_size(16)))
short88 shortsshort __attribute__((vector_size(16)))
ushort88 ushortsushort __attribute__((vector_size(16)))
int44 intsint __attribute__((vector_size(16)))
uint44 uintsunsigned __attribute__((vector_size(16)))
long22 longslong __attribute__((vector_size(16)))
ulong22 ulongsunsigned long __attribute__((vector_size(16)))
float44 floatsfloat __attribute__((vector_size(16)))
double22 doublesdouble __attribute__((vector_size(16)))
void3232 bytes of untyped datano equivalent
byte3232 bytessigned char __attribute__((vector_size(32)))
ubyte3232 ubytesunsigned char __attribute__((vector_size(32)))
short1616 shortsshort __attribute__((vector_size(32)))
ushort1616 ushortsushort __attribute__((vector_size(32)))
int88 intsint __attribute__((vector_size(32)))
uint88 uintsunsigned __attribute__((vector_size(32)))
long44 longslong __attribute__((vector_size(32)))
ulong44 ulongsunsigned long __attribute__((vector_size(32)))
float88 floatsfloat __attribute__((vector_size(32)))
double44 doublesdouble __attribute__((vector_size(32)))

Note: for 32 bit gcc and clang, it's long long instead of long.

Supported 128-bit Vector Operators
Operator void16 byte16 ubyte16 short8 ushort8 int4 uint4 long2 ulong2 float4 double2
=XXXXXXXXXXX
XXXXXXXXXX
XXXXXXXXXX
XXXX
/XX
&XXXXXXXX
|XXXXXXXX
^XXXXXXXX
=XXXXXXXXXX
=XXXXXXXXXX
=XXXX
/=XX
&=XXXXXXXX
|=XXXXXXXX
^=XXXXXXXX
==XXXXXXXXXX
!=XXXXXXXXXX
<XXXXXXXXXX
<=XXXXXXXXXX
>=XXXXXXXXXX
>XXXXXXXXXX
unary~XXXXXXXX
unary+XXXXXXXXXX
unary-XXXXXXXXXX
Supported 256-bit Vector Operators
Operator void32 byte32 ubyte32 short16 ushort16 int8 uint8 long4 ulong4 float8 double4
=XXXXXXXXXXX
XXXXXXXXXX
XXXXXXXXXX
XX
/XX
&XXXXXXXX
|XXXXXXXX
^XXXXXXXX
=XXXXXXXXXX
=XXXXXXXXXX
=XX
/=XX
&=XXXXXXXX
|=XXXXXXXX
^=XXXXXXXX
==XXXXXXXXXX
!=XXXXXXXXXX
<XXXXXXXXXX
<=XXXXXXXXXX
>=XXXXXXXXXX
>XXXXXXXXXX
unary~XXXXXXXX
unary+XXXXXXXXXX
unary-XXXXXXXXXX

Operators not listed are not supported at all.

Vector Operation Intrinsics

See core.simd for the supported intrinsics. abi, Application Binary Interface, betterc, Better C