AIE architectures offer multiplication instructions that can perform additional operations on on the input arguments. Instead of adding one variant for each possible combination, AIE API offers types that can wrap an existing vector, accumulator of element reference and be passed into the multiplication function. Then the API will merge the operations into a single instruction or apply the operation on the vector before the multiplication, depending on the hardware support,
The following example performs an element-wise multiplication of the absolute of vector a and the conjugate of vector b.
{
return ret;
}
constexpr auto mul(const Vec1 &v1, const Vec2 &v2) -> accum< AccumTag, Vec1::size()>
Definition: aie.hpp:3033
constexpr unary_op< T, Operation::Abs > op_abs(const T &e)
Definition: aie.hpp:480
constexpr unary_op< T, Operation::Conj > op_conj(const T &e)
Definition: aie.hpp:495
Definition: aie_declaration.hpp:71
Definition: aie_declaration.hpp:68
|
| template<typename T > |
| constexpr unary_op< T, Operation::Abs > | aie::op_abs (const T &e) |
| | More...
|
| |
| template<Accum Acc> |
| constexpr unary_op< Acc, Operation::Acc_Add > | aie::op_add (const Acc &acc) |
| | More...
|
| |
| template<typename T > |
| constexpr unary_op< T, Operation::Conj > | aie::op_conj (const T &e) |
| | More...
|
| |
| template<Vector Vec> |
| constexpr binary_op< Vec, Vec, Operation::Max > | aie::op_max (const Vec &a, const Vec &b) |
| | More...
|
| |
| template<Vector Vec> |
| constexpr binary_op< Vec, Vec, Operation::Min > | aie::op_min (const Vec &a, const Vec &b) |
| | More...
|
| |
| template<typename T > |
| constexpr unary_op< T, Operation::None > | aie::op_none (const T &e) |
| | More...
|
| |
| template<Accum Acc> |
| constexpr unary_op< Acc, Operation::Acc_Sub > | aie::op_sub (const Acc &acc) |
| | More...
|
| |
◆ op_abs()
Returns an absolute operation modifier for the given vector or element. On some architecture versions, this operation can be collapsed with a subsequent multiplication.
- Parameters
-
| e | Vector or element to which the operation is performed. The type must meet Vector or Elem. |
◆ op_add()
Returns an addition operation modifier for the given accumulator.
- Parameters
-
| acc | Accumulator to which the operation is performed. |
◆ op_conj()
Returns a conjugate operation modifier for the given complex vector or element. On some architecture versions, this operation can be collapsed with a subsequent multiplication.
- Parameters
-
◆ op_max()
Returns a maximum operation modifier for the given pair of vectors or elements. On some architecture versions, this operation can be collapsed with a subsequent multiplication.
- Parameters
-
| a | First vector or element passed to the operation. The type must meet Vector or Elem. |
| b | Second vector or element passed to the operation. The type must be the same as for a. |
◆ op_min()
Returns a minimum operation modifier for the given pair of vectors or elements. On some architecture versions, this operation can be collapsed with a subsequent multiplication.
- Parameters
-
| a | First vector or element passed to the operation. The type must meet Vector or Elem. |
| b | Second vector or element passed to the operation. The type must be the same as for a. |
◆ op_none()
Returns an empty operation modifier for the given vector or element.
- Parameters
-
| e | Vector or element to which the operation is performed. The type must meet Vector or Elem. |
◆ op_sub()
Returns a subtraction operation modifier for the given accumulator.
- Parameters
-
| acc | Accumulator to which the operation is performed. |