Emulated Multiply-accumulate of Complex Float and Float datatypes

Overview

Elementwise operations based on the already emulated FP32 operations (see intr_gpvectorop_mul_emul_float). These operations might not have optimal performance.

For an explanation how these operations works and the different accuracies provided, please refer to Multiply Accumulate.

! /**

Functions
_INLINE v8caccfloat	mac_elem_8_conf (v8float v1, v8cfloat v2, v8caccfloat acc, int zero_acc, int sub_mask, int sub_mul, int sub_acc1)
_INLINE v8caccfloat	msc_elem_8_conf (v8float v1, v8cfloat v2, v8caccfloat acc, int zero_acc, int sub_mask, int sub_mul, int sub_acc1)
_INLINE v8caccfloat	mul_elem_8_conf (v8float v1, v8cfloat v2, int sub_mask, int sub_mul)
_INLINE v8caccfloat	negmul_elem_8_conf (v8float v1, v8cfloat v2, int sub_mask, int sub_mul)

Emulated element-wise multiplication of complex float and complex float
v8caccfloat	mul_elem_8_conf (v8cfloat v1, v8cfloat v2, int sub_mul)
v8caccfloat	negmul_elem_8_conf (v8cfloat v1, v8cfloat v2, int sub_mul)

Emulated element-wise multiplication of float and complex float
v8caccfloat	mul_elem_8_conf (v8float v1, v8cfloat v2, int sub_mul)
v8caccfloat	mul_elem_8_conf (v8cfloat v1, v8float v2, int sub_mul)

Emulated Multiplication of Channel by channel complex multiplication of float and cfloat with dynamic negation of multiplication result, negation of acc1, and zeroing of acc1
v8caccfloat	mac_elem_8_conf (v8float v1, v8cfloat v2, v8caccfloat acc, int zero_acc, int sub_mul, int sub_acc1)
v8caccfloat	mac_elem_8_conf (v8cfloat v1, v8float v2, v8caccfloat acc, int zero_acc, int sub_mul, int sub_acc1)
v8caccfloat	msc_elem_8_conf (v8float v1, v8cfloat v2, v8caccfloat acc, int zero_acc, int sub_mul, int sub_acc1)
v8caccfloat	msc_elem_8_conf (v8cfloat v1, v8float v2, v8caccfloat acc, int zero_acc, int sub_mul, int sub_acc1)

Emulated Multiplication of Channel by channel complex multiplication of cfloat and cfloat with dynamic negation of multiplication result, negation of acc1, and zeroing of acc1
v8caccfloat	mac_elem_8_conf (v8cfloat v1, v8cfloat v2, v8caccfloat acc, int zero_acc, int sub_mul, int sub_acc1)
v8caccfloat	msc_elem_8_conf (v8cfloat v1, v8cfloat v2, v8caccfloat acc, int zero_acc, int sub_mul, int sub_acc1)
v8caccfloat	negmsc_elem_8_conf (v8cfloat v1, v8cfloat v2, v8caccfloat acc, int zero_acc, int sub_mul, int sub_acc1)

Function Documentation

mac_elem_8_conf() [1/4]

v8caccfloat mac_elem_8_conf	(	v8cfloat	v1,
		v8cfloat	v2,
		v8caccfloat	acc,
		int	zero_acc,
		int	sub_mul,
		int	sub_acc1 )

Parameters

v1	Vector A
v2	Vector B
acc	Accumulator 1 input
zero_acc	Zeroing mask for acc1
sub_mul	Negation mask of multiplication result
sub_acc1	Negation mask of acc1

Returns

Result of operation

mac_elem_8_conf() [2/4]

v8caccfloat mac_elem_8_conf	(	v8cfloat	v1,
		v8float	v2,
		v8caccfloat	acc,
		int	zero_acc,
		int	sub_mul,
		int	sub_acc1 )

Parameters

v1	Vector A
v2	Vector B
acc	Accumulator 1 input
zero_acc	Zeroing mask for acc1
sub_mul	Negation mask of multiplication result
sub_acc1	Negation mask of acc1

Returns

Result of operation

mac_elem_8_conf() [3/4]

_INLINE v8caccfloat mac_elem_8_conf	(	v8float	v1,
		v8cfloat	v2,
		v8caccfloat	acc,
		int	zero_acc,
		int	sub_mask,
		int	sub_mul,
		int	sub_acc1 )

Parameters

v1	Matrix A
v2	Matrix B
acc	Accumulator 1 input

Returns

Result of operation

mac_elem_8_conf() [4/4]

v8caccfloat mac_elem_8_conf	(	v8float	v1,
		v8cfloat	v2,
		v8caccfloat	acc,
		int	zero_acc,
		int	sub_mul,
		int	sub_acc1 )

Parameters

v1	Vector A
v2	Vector B
acc	Accumulator 1 input
zero_acc	Zeroing mask for acc1
sub_mul	Negation mask of multiplication result
sub_acc1	Negation mask of acc1

Returns

Result of operation

msc_elem_8_conf() [1/4]

v8caccfloat msc_elem_8_conf	(	v8cfloat	v1,
		v8cfloat	v2,
		v8caccfloat	acc,
		int	zero_acc,
		int	sub_mul,
		int	sub_acc1 )

Parameters

v1	Vector A
v2	Vector B
acc	Accumulator 1 input
zero_acc	Zeroing mask for acc1
sub_mul	Negation mask of multiplication result
sub_acc1	Negation mask of acc1

Returns

Result of operation

msc_elem_8_conf() [2/4]

v8caccfloat msc_elem_8_conf	(	v8cfloat	v1,
		v8float	v2,
		v8caccfloat	acc,
		int	zero_acc,
		int	sub_mul,
		int	sub_acc1 )

Parameters

v1	Vector A
v2	Vector B
acc	Accumulator 1 input
zero_acc	Zeroing mask for acc1
sub_mul	Negation mask of multiplication result
sub_acc1	Negation mask of acc1

Returns

Result of operation

msc_elem_8_conf() [3/4]

_INLINE v8caccfloat msc_elem_8_conf	(	v8float	v1,
		v8cfloat	v2,
		v8caccfloat	acc,
		int	zero_acc,
		int	sub_mask,
		int	sub_mul,
		int	sub_acc1 )

Parameters

v1	Matrix A
v2	Matrix B
acc	Accumulator 1 input

Returns

Result of operation

msc_elem_8_conf() [4/4]

v8caccfloat msc_elem_8_conf	(	v8float	v1,
		v8cfloat	v2,
		v8caccfloat	acc,
		int	zero_acc,
		int	sub_mul,
		int	sub_acc1 )

Parameters

v1	Vector A
v2	Vector B
acc	Accumulator 1 input
zero_acc	Zeroing mask for acc1
sub_mul	Negation mask of multiplication result
sub_acc1	Negation mask of acc1

Returns

Result of operation

mul_elem_8_conf() [1/4]

v8caccfloat mul_elem_8_conf	(	v8cfloat	v1,
		v8cfloat	v2,
		int	sub_mul )

Parameters

v1	Vector A
v2	Vector B
sub_mul	Negation mask of multiplication result

Returns

Result of operation

mul_elem_8_conf() [2/4]

v8caccfloat mul_elem_8_conf	(	v8cfloat	v1,
		v8float	v2,
		int	sub_mul )

Parameters

v1	Vector A
v2	Vector B
sub_mul	Negation mask of multiplication result

Returns

Result of operation

mul_elem_8_conf() [3/4]

_INLINE v8caccfloat mul_elem_8_conf	(	v8float	v1,
		v8cfloat	v2,
		int	sub_mask,
		int	sub_mul )

Parameters

v1	Matrix A
v2	Matrix B

Returns

Result of operation

mul_elem_8_conf() [4/4]

v8caccfloat mul_elem_8_conf	(	v8float	v1,
		v8cfloat	v2,
		int	sub_mul )

Parameters

v1	Vector A
v2	Vector B
sub_mul	Negation mask of multiplication result

Returns

Result of operation

negmsc_elem_8_conf()

v8caccfloat negmsc_elem_8_conf	(	v8cfloat	v1,
		v8cfloat	v2,
		v8caccfloat	acc,
		int	zero_acc,
		int	sub_mul,
		int	sub_acc1 )

Parameters

v1	Vector A
v2	Vector B
acc	Accumulator 1 input
zero_acc	Zeroing mask for acc1
sub_mul	Negation mask of multiplication result
sub_acc1	Negation mask of acc1

Returns

Result of operation

negmul_elem_8_conf() [1/2]

v8caccfloat negmul_elem_8_conf	(	v8cfloat	v1,
		v8cfloat	v2,
		int	sub_mul )

Parameters

v1	Vector A
v2	Vector B
sub_mul	Negation mask of multiplication result

Returns

Result of operation

negmul_elem_8_conf() [2/2]

_INLINE v8caccfloat negmul_elem_8_conf	(	v8float	v1,
		v8cfloat	v2,
		int	sub_mask,
		int	sub_mul )

Parameters

v1	Matrix A
v2	Matrix B

Returns

Result of operation