Vector and Accumulator Conversions

Vectors can be reinterpreted as vectors with a different element type, as long as they have the same total size. More...

Overview

Vectors can be reinterpreted as vectors with a different element type, as long as they have the same total size.

aie::vector<int16, 32> v;
aie::vector<int32, 16> v2;
aie::vector<cint16, 16> v3;
 
v2 = v.cast_to<int32>(v);
 
v3 = aie::vector_cast<cint16>(v);

Vectors can be converted into accumulators. Their values can be shifted into a larger magnitude to implement fixed point precision schemes (this does not apply to floating point accumulators).

aie::vector<int16, 16> v;
aie::accum<acc32, 16> acc;
 
acc.from_vector(v, shift);

Conversely, accumulators can be converted into vectors. Their values can be shifted down before rounding and saturation is applied (this does not apply to floating point accumulators).

aie::accum<acc32, 16> acc;
aie::vector<int16, 16> v;
 
v = acc.to_vector<int16>(shift);

Functions
template<AccumElemBaseType AccumTag, VectorOrOp Vec>
accum< AccumTag, Vec::size()>	aie::from_vector (const Vec &v, int shift=0)
	Returns the values of the passed vector in an accumulator of the requested type after applying the requested upshift operation.
template<VectorOrOp Vec, typename T2 = Utils::get_prev_integer_type_t<typename Vec::value_type>>
auto	aie::pack (const Vec &v) -> vector< T2, Vec::size()>
	Returns a conversion of the passed vector to the next smaller integer datatype, with the same number of elements.
template<typename TR , typename T > requires (AccumOrOp<T> || MmulOrOp<T>)
vector< TR, T::size()>	aie::to_vector (const T &acc, int shift=0)
	Returns the values of the passed accumulator in a vector of the requested type.
template<VectorOrOp Vec, typename T2 = Utils::get_next_integer_type_t<typename Vec::value_type>> requires (Utils::is_one_of_v<T2, int8, uint8, int16, uint16>)
auto	aie::unpack (const Vec &v) -> vector< T2, Vec::size()>
	Returns a conversion of the passed vector to the next larger integer datatype, with the same number of elements.
template<ElemBaseType DstT, Accum Acc> requires (arch::is(arch::AIE_ML))
auto	aie::vector_cast (const Acc &acc)
	Reinterpret an accumulator as a vector.
template<ElemBaseType DstT, Vector Vec>
auto	aie::vector_cast (const Vec &v)
	Reinterpret a vector using a different element type.

Function Documentation

from_vector()

template<AccumElemBaseType AccumTag, VectorOrOp Vec>

accum< AccumTag, Vec::size()> aie::from_vector	(	const Vec &	v,
		int	shift = 0
	)

Returns the values of the passed vector in an accumulator of the requested type after applying the requested upshift operation.

Parameters

v	Input vector. The type must meet VectorOrOp.
shift	Optional upshift in bits to be applied to output accumulator. This parameter is ignored for floating-point types.

Template Parameters

AccumTag

Accumulator tag that specifies returned accumulator type. The class must be compatible with the vector type (real/complex).

pack()

template<VectorOrOp Vec, typename T2 = Utils::get_prev_integer_type_t<typename Vec::value_type>>

auto aie::pack

(

const Vec &

v

)

-> vector<T2, Vec::size()>

Returns a conversion of the passed vector to the next smaller integer datatype, with the same number of elements.

Parameters

v	Input vector. The type must meet VectorOrOp

Template Parameters

T2	Element type for the returned vector. Can only select between int8/uint8 on AIE1.

to_vector()

template<typename TR , typename T >
requires (AccumOrOp<T> || MmulOrOp<T>)

vector< TR, T::size()> aie::to_vector	(	const T &	acc,
		int	shift = 0
	)

Returns the values of the passed accumulator in a vector of the requested type.

The values can be shifted down before rounding and saturation are applied (does not apply to floating point accumulators).

Parameters

acc	Input accumulator. The type must meet AccumOrOp or MmulOrOp.
shift	Optional downshift in bits to be applied to output data. This parameter is ignored for floating-point types.

Template Parameters

TR	Element type for the returned vector.

unpack()

template<VectorOrOp Vec, typename T2 = Utils::get_next_integer_type_t<typename Vec::value_type>>
requires (Utils::is_one_of_v<T2, int8, uint8, int16, uint16>)

auto aie::unpack

(

const Vec &

v

)

-> vector<T2, Vec::size()>

Returns a conversion of the passed vector to the next larger integer datatype, with the same number of elements.

Parameters

v	Input vector. The type must meet VectorOrOp

vector_cast() [1/2]

template<ElemBaseType DstT, Accum Acc>
requires (arch::is(arch::AIE_ML))

auto aie::vector_cast

(

const Acc &

acc

)

Reinterpret an accumulator as a vector.

The returned vector has the same size in bits as the input accumulator.

Template Parameters

DstT	Element type for the returned vector.

Parameters

acc	Input accumulator.

vector_cast() [2/2]

template<ElemBaseType DstT, Vector Vec>

auto aie::vector_cast

(

const Vec &

v

)

Reinterpret a vector using a different element type.

The returned vector has the same size in bits as the input vector.

Template Parameters

DstT	Element type for the returned vector.

Parameters

v	Input vector.