X and Z buffer conjugation

Overview

16 bit complex by 32 bit complex using X and Z buffer conjugation

Functions
v8cacc48	mac4_sym_uct_cc (v8cacc48 acc, v32cint16 xbuff, int xstart, unsigned int xoffsets, int ystart, int uct_shift, v4cint32 zbuff, int zstart, unsigned int zoffsets)
	Complex symetric multiply-accumulate intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer .
v8cacc48	mac4_sym_uct_cc (v8cacc48 acc, v16cint16 xbuff, int xstart, unsigned int xoffsets, int ystart, int uct_shift, v4cint32 zbuff, int zstart, unsigned int zoffsets)
	Complex symetric multiply-accumulate intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer using small X input buffer.
v8cacc48	mac4_sym_uct_cc (v8cacc48 acc, v16cint16 xbuff, int xstart, unsigned int xoffsets, v16cint16 ybuff, int ystart, int uct_shift, v4cint32 zbuff, int zstart, unsigned int zoffsets)
	Complex symetric multiply-accumulate intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x and y input buffers using small X input buffer.
v8cacc48	msc4_sym_uct_cc (v8cacc48 acc, v32cint16 xbuff, int xstart, unsigned int xoffsets, int ystart, int uct_shift, v4cint32 zbuff, int zstart, unsigned int zoffsets)
	Complex symetric multiply-subtract intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer .
v8cacc48	msc4_sym_uct_cc (v8cacc48 acc, v16cint16 xbuff, int xstart, unsigned int xoffsets, int ystart, int uct_shift, v4cint32 zbuff, int zstart, unsigned int zoffsets)
	Complex symetric multiply-subtract intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer using small X input buffer.
v8cacc48	msc4_sym_uct_cc (v8cacc48 acc, v16cint16 xbuff, int xstart, unsigned int xoffsets, v16cint16 ybuff, int ystart, int uct_shift, v4cint32 zbuff, int zstart, unsigned int zoffsets)
	Complex symetric multiply-subtract intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x and y input buffers using small X input buffer.
v8cacc48	mul4_sym_uct_cc (v32cint16 xbuff, int xstart, unsigned int xoffsets, int ystart, int uct_shift, v4cint32 zbuff, int zstart, unsigned int zoffsets)
	Complex symetric multiply intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer .
v8cacc48	mul4_sym_uct_cc (v16cint16 xbuff, int xstart, unsigned int xoffsets, int ystart, int uct_shift, v4cint32 zbuff, int zstart, unsigned int zoffsets)
	Complex symetric multiply intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer using small X input buffer.
v8cacc48	mul4_sym_uct_cc (v16cint16 xbuff, int xstart, unsigned int xoffsets, v16cint16 ybuff, int ystart, int uct_shift, v4cint32 zbuff, int zstart, unsigned int zoffsets)
	Complex symetric multiply intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x and y input buffers using small X input buffer.
v8cacc48	negmul4_sym_uct_cc (v32cint16 xbuff, int xstart, unsigned int xoffsets, int ystart, int uct_shift, v4cint32 zbuff, int zstart, unsigned int zoffsets)
	Complex symetric multiply-negate intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer .
v8cacc48	negmul4_sym_uct_cc (v16cint16 xbuff, int xstart, unsigned int xoffsets, int ystart, int uct_shift, v4cint32 zbuff, int zstart, unsigned int zoffsets)
	Complex symetric multiply-negate intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer using small X input buffer.
v8cacc48	negmul4_sym_uct_cc (v16cint16 xbuff, int xstart, unsigned int xoffsets, v16cint16 ybuff, int ystart, int uct_shift, v4cint32 zbuff, int zstart, unsigned int zoffsets)
	Complex symetric multiply-negate intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x and y input buffers using small X input buffer.

Function Documentation

v8cacc48 mac4_sym_uct_cc	(	v8cacc48	acc,
		v32cint16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		int	uct_shift,
		v4cint32	zbuff,
		int	zstart,
		unsigned int	zoffsets
	)

Complex symetric multiply-accumulate intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer .

acc0 += cnj(z00)*(cnj(x00) + cnj(y00))
acc1 += cnj(z10)*(cnj(x10) + cnj(y10))
acc2 += cnj(z20)*(cnj(x20) + cnj(y20))
acc3 += cnj(z30)*(cnj(x30) + cnj(y30))
acc4 = y00 << uct_shift
acc5 = y10 << uct_shift
acc6 = y20 << uct_shift
acc7 = y30 << uct_shift

Returns

Returned accumulation vector (8 x cint48 lanes)

Parameters

acc	Incoming accumulation vector (8 x cint48 lanes)
xbuff	Input buffer of 32 elements of type cint16
xstart	Starting position offset applied to all lanes of input from X buffer.
xoffsets	4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart	Starting position offset applied to all lanes for input from Y buffer.
uct_shift	Upshift value to the four upshifted lanes
zbuff	Input buffer of 4 elements of type cint32
zstart	Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets	4b offset for each lane, applied to input from Z buffer. LSB apply to first lane

Note

• For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8cacc48 mac4_sym_uct_cc	(	v8cacc48	acc,
		v16cint16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		int	uct_shift,
		v4cint32	zbuff,
		int	zstart,
		unsigned int	zoffsets
	)

Complex symetric multiply-accumulate intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer using small X input buffer.

acc0 += cnj(z00)*(cnj(x00) + cnj(y00))
acc1 += cnj(z10)*(cnj(x10) + cnj(y10))
acc2 += cnj(z20)*(cnj(x20) + cnj(y20))
acc3 += cnj(z30)*(cnj(x30) + cnj(y30))
acc4 = y00 << uct_shift
acc5 = y10 << uct_shift
acc6 = y20 << uct_shift
acc7 = y30 << uct_shift

Returns

Returned accumulation vector (8 x cint48 lanes)

Parameters

acc	Incoming accumulation vector (8 x cint48 lanes)
xbuff	Input buffer of 16 elements of type cint16
xstart	Starting position offset applied to all lanes of input from X buffer.
xoffsets	4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart	Starting position offset applied to all lanes for input from Y buffer.
uct_shift	Upshift value to the four upshifted lanes
zbuff	Input buffer of 4 elements of type cint32
zstart	Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets	4b offset for each lane, applied to input from Z buffer. LSB apply to first lane

Note

• For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8cacc48 mac4_sym_uct_cc	(	v8cacc48	acc,
		v16cint16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		v16cint16	ybuff,
		int	ystart,
		int	uct_shift,
		v4cint32	zbuff,
		int	zstart,
		unsigned int	zoffsets
	)

Complex symetric multiply-accumulate intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x and y input buffers using small X input buffer.

acc0 += cnj(z00)*(cnj(x00) + cnj(y00))
acc1 += cnj(z10)*(cnj(x10) + cnj(y10))
acc2 += cnj(z20)*(cnj(x20) + cnj(y20))
acc3 += cnj(z30)*(cnj(x30) + cnj(y30))
acc4 = y00 << uct_shift
acc5 = y10 << uct_shift
acc6 = y20 << uct_shift
acc7 = y30 << uct_shift

Returns

Returned accumulation vector (8 x cint48 lanes)

Parameters

acc	Incoming accumulation vector (8 x cint48 lanes)
xbuff	Input buffer of 16 elements of type cint16
xstart	Starting position offset applied to all lanes of input from X buffer.
xoffsets	4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff	Right input buffer of 16 elements of type cint16
ystart	Starting position offset applied to all lanes for input from Y buffer.
uct_shift	Upshift value to the four upshifted lanes
zbuff	Input buffer of 4 elements of type cint32
zstart	Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets	4b offset for each lane, applied to input from Z buffer. LSB apply to first lane

Note

• For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8cacc48 msc4_sym_uct_cc	(	v8cacc48	acc,
		v32cint16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		int	uct_shift,
		v4cint32	zbuff,
		int	zstart,
		unsigned int	zoffsets
	)

Complex symetric multiply-subtract intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer .

acc0 -= cnj(z00)*(cnj(x00) + cnj(y00))
acc1 -= cnj(z10)*(cnj(x10) + cnj(y10))
acc2 -= cnj(z20)*(cnj(x20) + cnj(y20))
acc3 -= cnj(z30)*(cnj(x30) + cnj(y30))
acc4 = y00 << uct_shift
acc5 = y10 << uct_shift
acc6 = y20 << uct_shift
acc7 = y30 << uct_shift

Returns

Returned accumulation vector (8 x cint48 lanes)

Parameters

acc	Incoming accumulation vector (8 x cint48 lanes)
xbuff	Input buffer of 32 elements of type cint16
xstart	Starting position offset applied to all lanes of input from X buffer.
xoffsets	4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart	Starting position offset applied to all lanes for input from Y buffer.
uct_shift	Upshift value to the four upshifted lanes
zbuff	Input buffer of 4 elements of type cint32
zstart	Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets	4b offset for each lane, applied to input from Z buffer. LSB apply to first lane

Note

• For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8cacc48 msc4_sym_uct_cc	(	v8cacc48	acc,
		v16cint16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		int	uct_shift,
		v4cint32	zbuff,
		int	zstart,
		unsigned int	zoffsets
	)

Complex symetric multiply-subtract intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer using small X input buffer.

acc0 -= cnj(z00)*(cnj(x00) + cnj(y00))
acc1 -= cnj(z10)*(cnj(x10) + cnj(y10))
acc2 -= cnj(z20)*(cnj(x20) + cnj(y20))
acc3 -= cnj(z30)*(cnj(x30) + cnj(y30))
acc4 = y00 << uct_shift
acc5 = y10 << uct_shift
acc6 = y20 << uct_shift
acc7 = y30 << uct_shift

Returns

Returned accumulation vector (8 x cint48 lanes)

Parameters

acc	Incoming accumulation vector (8 x cint48 lanes)
xbuff	Input buffer of 16 elements of type cint16
xstart	Starting position offset applied to all lanes of input from X buffer.
xoffsets	4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart	Starting position offset applied to all lanes for input from Y buffer.
uct_shift	Upshift value to the four upshifted lanes
zbuff	Input buffer of 4 elements of type cint32
zstart	Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets	4b offset for each lane, applied to input from Z buffer. LSB apply to first lane

Note

• For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8cacc48 msc4_sym_uct_cc	(	v8cacc48	acc,
		v16cint16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		v16cint16	ybuff,
		int	ystart,
		int	uct_shift,
		v4cint32	zbuff,
		int	zstart,
		unsigned int	zoffsets
	)

Complex symetric multiply-subtract intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x and y input buffers using small X input buffer.

acc0 -= cnj(z00)*(cnj(x00) + cnj(y00))
acc1 -= cnj(z10)*(cnj(x10) + cnj(y10))
acc2 -= cnj(z20)*(cnj(x20) + cnj(y20))
acc3 -= cnj(z30)*(cnj(x30) + cnj(y30))
acc4 = y00 << uct_shift
acc5 = y10 << uct_shift
acc6 = y20 << uct_shift
acc7 = y30 << uct_shift

Returns

Returned accumulation vector (8 x cint48 lanes)

Parameters

acc	Incoming accumulation vector (8 x cint48 lanes)
xbuff	Input buffer of 16 elements of type cint16
xstart	Starting position offset applied to all lanes of input from X buffer.
xoffsets	4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff	Right input buffer of 16 elements of type cint16
ystart	Starting position offset applied to all lanes for input from Y buffer.
uct_shift	Upshift value to the four upshifted lanes
zbuff	Input buffer of 4 elements of type cint32
zstart	Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets	4b offset for each lane, applied to input from Z buffer. LSB apply to first lane

Note

• For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8cacc48 mul4_sym_uct_cc	(	v32cint16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		int	uct_shift,
		v4cint32	zbuff,
		int	zstart,
		unsigned int	zoffsets
	)

Complex symetric multiply intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer .

acc0 = cnj(z00)*(cnj(x00) + cnj(y00))
acc1 = cnj(z10)*(cnj(x10) + cnj(y10))
acc2 = cnj(z20)*(cnj(x20) + cnj(y20))
acc3 = cnj(z30)*(cnj(x30) + cnj(y30))
acc4 = y00 << uct_shift
acc5 = y10 << uct_shift
acc6 = y20 << uct_shift
acc7 = y30 << uct_shift

Returns

Returned accumulation vector (8 x cint48 lanes)

Parameters

xbuff	Input buffer of 32 elements of type cint16
xstart	Starting position offset applied to all lanes of input from X buffer.
xoffsets	4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart	Starting position offset applied to all lanes for input from Y buffer.
uct_shift	Upshift value to the four upshifted lanes
zbuff	Input buffer of 4 elements of type cint32
zstart	Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets	4b offset for each lane, applied to input from Z buffer. LSB apply to first lane

Note

• For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8cacc48 mul4_sym_uct_cc	(	v16cint16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		int	uct_shift,
		v4cint32	zbuff,
		int	zstart,
		unsigned int	zoffsets
	)

Complex symetric multiply intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer using small X input buffer.

acc0 = cnj(z00)*(cnj(x00) + cnj(y00))
acc1 = cnj(z10)*(cnj(x10) + cnj(y10))
acc2 = cnj(z20)*(cnj(x20) + cnj(y20))
acc3 = cnj(z30)*(cnj(x30) + cnj(y30))
acc4 = y00 << uct_shift
acc5 = y10 << uct_shift
acc6 = y20 << uct_shift
acc7 = y30 << uct_shift

Returns

Returned accumulation vector (8 x cint48 lanes)

Parameters

xbuff	Input buffer of 16 elements of type cint16
xstart	Starting position offset applied to all lanes of input from X buffer.
xoffsets	4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart	Starting position offset applied to all lanes for input from Y buffer.
uct_shift	Upshift value to the four upshifted lanes
zbuff	Input buffer of 4 elements of type cint32
zstart	Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets	4b offset for each lane, applied to input from Z buffer. LSB apply to first lane

Note

• For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8cacc48 mul4_sym_uct_cc	(	v16cint16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		v16cint16	ybuff,
		int	ystart,
		int	uct_shift,
		v4cint32	zbuff,
		int	zstart,
		unsigned int	zoffsets
	)

Complex symetric multiply intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x and y input buffers using small X input buffer.

acc0 = cnj(z00)*(cnj(x00) + cnj(y00))
acc1 = cnj(z10)*(cnj(x10) + cnj(y10))
acc2 = cnj(z20)*(cnj(x20) + cnj(y20))
acc3 = cnj(z30)*(cnj(x30) + cnj(y30))
acc4 = y00 << uct_shift
acc5 = y10 << uct_shift
acc6 = y20 << uct_shift
acc7 = y30 << uct_shift

Returns

Returned accumulation vector (8 x cint48 lanes)

Parameters

xbuff	Input buffer of 16 elements of type cint16
xstart	Starting position offset applied to all lanes of input from X buffer.
xoffsets	4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff	Right input buffer of 16 elements of type cint16
ystart	Starting position offset applied to all lanes for input from Y buffer.
uct_shift	Upshift value to the four upshifted lanes
zbuff	Input buffer of 4 elements of type cint32
zstart	Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets	4b offset for each lane, applied to input from Z buffer. LSB apply to first lane

Note

• For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8cacc48 negmul4_sym_uct_cc	(	v32cint16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		int	uct_shift,
		v4cint32	zbuff,
		int	zstart,
		unsigned int	zoffsets
	)

Complex symetric multiply-negate intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer .

acc0 = -( cnj(z00)*(cnj(x00) + cnj(y00)) )
acc1 = -( cnj(z10)*(cnj(x10) + cnj(y10)) )
acc2 = -( cnj(z20)*(cnj(x20) + cnj(y20)) )
acc3 = -( cnj(z30)*(cnj(x30) + cnj(y30)) )
acc4 = y00 << uct_shift )
acc5 = y10 << uct_shift )
acc6 = y20 << uct_shift )
acc7 = y30 << uct_shift )

Returns

Returned accumulation vector (8 x cint48 lanes)

Parameters

xbuff	Input buffer of 32 elements of type cint16
xstart	Starting position offset applied to all lanes of input from X buffer.
xoffsets	4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart	Starting position offset applied to all lanes for input from Y buffer.
uct_shift	Upshift value to the four upshifted lanes
zbuff	Input buffer of 4 elements of type cint32
zstart	Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets	4b offset for each lane, applied to input from Z buffer. LSB apply to first lane

Note

• For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8cacc48 negmul4_sym_uct_cc	(	v16cint16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		int	uct_shift,
		v4cint32	zbuff,
		int	zstart,
		unsigned int	zoffsets
	)

Complex symetric multiply-negate intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x input buffer using small X input buffer.

acc0 = -( cnj(z00)*(cnj(x00) + cnj(y00)) )
acc1 = -( cnj(z10)*(cnj(x10) + cnj(y10)) )
acc2 = -( cnj(z20)*(cnj(x20) + cnj(y20)) )
acc3 = -( cnj(z30)*(cnj(x30) + cnj(y30)) )
acc4 = y00 << uct_shift )
acc5 = y10 << uct_shift )
acc6 = y20 << uct_shift )
acc7 = y30 << uct_shift )

Returns

Returned accumulation vector (8 x cint48 lanes)

Parameters

xbuff	Input buffer of 16 elements of type cint16
xstart	Starting position offset applied to all lanes of input from X buffer.
xoffsets	4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart	Starting position offset applied to all lanes for input from Y buffer.
uct_shift	Upshift value to the four upshifted lanes
zbuff	Input buffer of 4 elements of type cint32
zstart	Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets	4b offset for each lane, applied to input from Z buffer. LSB apply to first lane

Note

• For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8cacc48 negmul4_sym_uct_cc	(	v16cint16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		v16cint16	ybuff,
		int	ystart,
		int	uct_shift,
		v4cint32	zbuff,
		int	zstart,
		unsigned int	zoffsets
	)

Complex symetric multiply-negate intrinsic function with unit center-tap optimization with x and z buffer conjugation with pre-add from x and y input buffers using small X input buffer.

acc0 = -( cnj(z00)*(cnj(x00) + cnj(y00)) )
acc1 = -( cnj(z10)*(cnj(x10) + cnj(y10)) )
acc2 = -( cnj(z20)*(cnj(x20) + cnj(y20)) )
acc3 = -( cnj(z30)*(cnj(x30) + cnj(y30)) )
acc4 = y00 << uct_shift )
acc5 = y10 << uct_shift )
acc6 = y20 << uct_shift )
acc7 = y30 << uct_shift )

Returns

Returned accumulation vector (8 x cint48 lanes)

Parameters

xbuff	Input buffer of 16 elements of type cint16
xstart	Starting position offset applied to all lanes of input from X buffer.
xoffsets	4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff	Right input buffer of 16 elements of type cint16
ystart	Starting position offset applied to all lanes for input from Y buffer.
uct_shift	Upshift value to the four upshifted lanes
zbuff	Input buffer of 4 elements of type cint32
zstart	Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets	4b offset for each lane, applied to input from Z buffer. LSB apply to first lane

Note

• For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
• Parameter 'zstart' must be a compile time constant.