32 bit Real x 16 bit Real

Overview

Vector MAC intrinsics with partial pre-adding with 32 bit real by 16 bit real

Functions
v8acc80	lmac8_sym_ct (v8acc80 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-accumulate intrinsic function with center-tap optimization with pre-add from x input buffer .
v8acc80	lmac8_sym_ct (v8acc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-accumulate intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.
v8acc80	lmac8_sym_ct (v8acc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-accumulate intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.
v8acc80	lmsc8_sym_ct (v8acc80 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-subtract intrinsic function with center-tap optimization with pre-add from x input buffer .
v8acc80	lmsc8_sym_ct (v8acc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-subtract intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.
v8acc80	lmsc8_sym_ct (v8acc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-subtract intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.
v8acc80	lmul8_sym_ct (v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply intrinsic function with center-tap optimization with pre-add from x input buffer .
v8acc80	lmul8_sym_ct (v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.
v8acc80	lmul8_sym_ct (v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.
v8acc80	lnegmul8_sym_ct (v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-negate intrinsic function with center-tap optimization with pre-add from x input buffer .
v8acc80	lnegmul8_sym_ct (v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-negate intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.
v8acc80	lnegmul8_sym_ct (v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-negate intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.
v8acc48	mac8_sym_ct (v8acc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-accumulate intrinsic function with center-tap optimization with pre-add from x input buffer .
v8acc48	mac8_sym_ct (v8acc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-accumulate intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.
v8acc48	mac8_sym_ct (v8acc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-accumulate intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.
v8acc48	msc8_sym_ct (v8acc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-subtract intrinsic function with center-tap optimization with pre-add from x input buffer .
v8acc48	msc8_sym_ct (v8acc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-subtract intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.
v8acc48	msc8_sym_ct (v8acc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-subtract intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.
v8acc48	mul8_sym_ct (v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply intrinsic function with center-tap optimization with pre-add from x input buffer .
v8acc48	mul8_sym_ct (v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.
v8acc48	mul8_sym_ct (v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.
v8acc48	negmul8_sym_ct (v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-negate intrinsic function with center-tap optimization with pre-add from x input buffer .
v8acc48	negmul8_sym_ct (v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-negate intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.
v8acc48	negmul8_sym_ct (v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, unsigned int mtap, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Symetric multiply-negate intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.

Function Documentation

v8acc80 lmac8_sym_ct	(	v8acc80	acc,
		v32int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-accumulate intrinsic function with center-tap optimization with pre-add from x input buffer .

acc0 += z00*(x00 + y00) + z01*x01
acc1 += z10*(x10 + y10) + z11*x11
acc2 += z20*(x20 + y20) + z21*x21
acc3 += z30*(x30 + y30) + z31*x31
acc4 += z40*(x40 + y40) + z41*x41
acc5 += z50*(x50 + y50) + z51*x51
acc6 += z60*(x60 + y60) + z61*x61
acc7 += z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int80 lanes)

Parameters

acc	Incoming accumulation vector (8 x int80 lanes)
xbuff	Input buffer of 32 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc80 lmac8_sym_ct	(	v8acc80	acc,
		v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-accumulate intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.

acc0 += z00*(x00 + y00) + z01*x01
acc1 += z10*(x10 + y10) + z11*x11
acc2 += z20*(x20 + y20) + z21*x21
acc3 += z30*(x30 + y30) + z31*x31
acc4 += z40*(x40 + y40) + z41*x41
acc5 += z50*(x50 + y50) + z51*x51
acc6 += z60*(x60 + y60) + z61*x61
acc7 += z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int80 lanes)

Parameters

acc	Incoming accumulation vector (8 x int80 lanes)
xbuff	Input buffer of 16 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc80 lmac8_sym_ct	(	v8acc80	acc,
		v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		v16int32	ybuff,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-accumulate intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.

acc0 += z00*(x00 + y00) + z01*x01
acc1 += z10*(x10 + y10) + z11*x11
acc2 += z20*(x20 + y20) + z21*x21
acc3 += z30*(x30 + y30) + z31*x31
acc4 += z40*(x40 + y40) + z41*x41
acc5 += z50*(x50 + y50) + z51*x51
acc6 += z60*(x60 + y60) + z61*x61
acc7 += z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int80 lanes)

Parameters

acc	Incoming accumulation vector (8 x int80 lanes)
xbuff	Input buffer of 16 elements of type int32
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 int32
ystart	Starting position offset applied to all lanes for input from Y buffer.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc80 lmsc8_sym_ct	(	v8acc80	acc,
		v32int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-subtract intrinsic function with center-tap optimization with pre-add from x input buffer .

acc0 -= z00*(x00 + y00) + z01*x01
acc1 -= z10*(x10 + y10) + z11*x11
acc2 -= z20*(x20 + y20) + z21*x21
acc3 -= z30*(x30 + y30) + z31*x31
acc4 -= z40*(x40 + y40) + z41*x41
acc5 -= z50*(x50 + y50) + z51*x51
acc6 -= z60*(x60 + y60) + z61*x61
acc7 -= z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int80 lanes)

Parameters

acc	Incoming accumulation vector (8 x int80 lanes)
xbuff	Input buffer of 32 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc80 lmsc8_sym_ct	(	v8acc80	acc,
		v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-subtract intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.

acc0 -= z00*(x00 + y00) + z01*x01
acc1 -= z10*(x10 + y10) + z11*x11
acc2 -= z20*(x20 + y20) + z21*x21
acc3 -= z30*(x30 + y30) + z31*x31
acc4 -= z40*(x40 + y40) + z41*x41
acc5 -= z50*(x50 + y50) + z51*x51
acc6 -= z60*(x60 + y60) + z61*x61
acc7 -= z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int80 lanes)

Parameters

acc	Incoming accumulation vector (8 x int80 lanes)
xbuff	Input buffer of 16 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc80 lmsc8_sym_ct	(	v8acc80	acc,
		v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		v16int32	ybuff,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-subtract intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.

acc0 -= z00*(x00 + y00) + z01*x01
acc1 -= z10*(x10 + y10) + z11*x11
acc2 -= z20*(x20 + y20) + z21*x21
acc3 -= z30*(x30 + y30) + z31*x31
acc4 -= z40*(x40 + y40) + z41*x41
acc5 -= z50*(x50 + y50) + z51*x51
acc6 -= z60*(x60 + y60) + z61*x61
acc7 -= z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int80 lanes)

Parameters

acc	Incoming accumulation vector (8 x int80 lanes)
xbuff	Input buffer of 16 elements of type int32
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 int32
ystart	Starting position offset applied to all lanes for input from Y buffer.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc80 lmul8_sym_ct	(	v32int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply intrinsic function with center-tap optimization with pre-add from x input buffer .

acc0 = z00*(x00 + y00) + z01*x01
acc1 = z10*(x10 + y10) + z11*x11
acc2 = z20*(x20 + y20) + z21*x21
acc3 = z30*(x30 + y30) + z31*x31
acc4 = z40*(x40 + y40) + z41*x41
acc5 = z50*(x50 + y50) + z51*x51
acc6 = z60*(x60 + y60) + z61*x61
acc7 = z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int80 lanes)

Parameters

xbuff	Input buffer of 32 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc80 lmul8_sym_ct	(	v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.

acc0 = z00*(x00 + y00) + z01*x01
acc1 = z10*(x10 + y10) + z11*x11
acc2 = z20*(x20 + y20) + z21*x21
acc3 = z30*(x30 + y30) + z31*x31
acc4 = z40*(x40 + y40) + z41*x41
acc5 = z50*(x50 + y50) + z51*x51
acc6 = z60*(x60 + y60) + z61*x61
acc7 = z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int80 lanes)

Parameters

xbuff	Input buffer of 16 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc80 lmul8_sym_ct	(	v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		v16int32	ybuff,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.

acc0 = z00*(x00 + y00) + z01*x01
acc1 = z10*(x10 + y10) + z11*x11
acc2 = z20*(x20 + y20) + z21*x21
acc3 = z30*(x30 + y30) + z31*x31
acc4 = z40*(x40 + y40) + z41*x41
acc5 = z50*(x50 + y50) + z51*x51
acc6 = z60*(x60 + y60) + z61*x61
acc7 = z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int80 lanes)

Parameters

xbuff	Input buffer of 16 elements of type int32
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 int32
ystart	Starting position offset applied to all lanes for input from Y buffer.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc80 lnegmul8_sym_ct	(	v32int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-negate intrinsic function with center-tap optimization with pre-add from x input buffer .

acc0 = -( z00*(x00 + y00) + z01*x01 )
acc1 = -( z10*(x10 + y10) + z11*x11 )
acc2 = -( z20*(x20 + y20) + z21*x21 )
acc3 = -( z30*(x30 + y30) + z31*x31 )
acc4 = -( z40*(x40 + y40) + z41*x41 )
acc5 = -( z50*(x50 + y50) + z51*x51 )
acc6 = -( z60*(x60 + y60) + z61*x61 )
acc7 = -( z70*(x70 + y70) + z71*x71 )

Returns

Returned accumulation vector (8 x int80 lanes)

Parameters

xbuff	Input buffer of 32 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc80 lnegmul8_sym_ct	(	v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-negate intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.

acc0 = -( z00*(x00 + y00) + z01*x01 )
acc1 = -( z10*(x10 + y10) + z11*x11 )
acc2 = -( z20*(x20 + y20) + z21*x21 )
acc3 = -( z30*(x30 + y30) + z31*x31 )
acc4 = -( z40*(x40 + y40) + z41*x41 )
acc5 = -( z50*(x50 + y50) + z51*x51 )
acc6 = -( z60*(x60 + y60) + z61*x61 )
acc7 = -( z70*(x70 + y70) + z71*x71 )

Returns

Returned accumulation vector (8 x int80 lanes)

Parameters

xbuff	Input buffer of 16 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc80 lnegmul8_sym_ct	(	v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		v16int32	ybuff,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-negate intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.

acc0 = -( z00*(x00 + y00) + z01*x01 )
acc1 = -( z10*(x10 + y10) + z11*x11 )
acc2 = -( z20*(x20 + y20) + z21*x21 )
acc3 = -( z30*(x30 + y30) + z31*x31 )
acc4 = -( z40*(x40 + y40) + z41*x41 )
acc5 = -( z50*(x50 + y50) + z51*x51 )
acc6 = -( z60*(x60 + y60) + z61*x61 )
acc7 = -( z70*(x70 + y70) + z71*x71 )

Returns

Returned accumulation vector (8 x int80 lanes)

Parameters

xbuff	Input buffer of 16 elements of type int32
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 int32
ystart	Starting position offset applied to all lanes for input from Y buffer.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc48 mac8_sym_ct	(	v8acc48	acc,
		v32int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-accumulate intrinsic function with center-tap optimization with pre-add from x input buffer .

acc0 += z00*(x00 + y00) + z01*x01
acc1 += z10*(x10 + y10) + z11*x11
acc2 += z20*(x20 + y20) + z21*x21
acc3 += z30*(x30 + y30) + z31*x31
acc4 += z40*(x40 + y40) + z41*x41
acc5 += z50*(x50 + y50) + z51*x51
acc6 += z60*(x60 + y60) + z61*x61
acc7 += z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

acc	Incoming accumulation vector (8 x int48 lanes)
xbuff	Input buffer of 32 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc48 mac8_sym_ct	(	v8acc48	acc,
		v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-accumulate intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.

acc0 += z00*(x00 + y00) + z01*x01
acc1 += z10*(x10 + y10) + z11*x11
acc2 += z20*(x20 + y20) + z21*x21
acc3 += z30*(x30 + y30) + z31*x31
acc4 += z40*(x40 + y40) + z41*x41
acc5 += z50*(x50 + y50) + z51*x51
acc6 += z60*(x60 + y60) + z61*x61
acc7 += z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

acc	Incoming accumulation vector (8 x int48 lanes)
xbuff	Input buffer of 16 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc48 mac8_sym_ct	(	v8acc48	acc,
		v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		v16int32	ybuff,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-accumulate intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.

acc0 += z00*(x00 + y00) + z01*x01
acc1 += z10*(x10 + y10) + z11*x11
acc2 += z20*(x20 + y20) + z21*x21
acc3 += z30*(x30 + y30) + z31*x31
acc4 += z40*(x40 + y40) + z41*x41
acc5 += z50*(x50 + y50) + z51*x51
acc6 += z60*(x60 + y60) + z61*x61
acc7 += z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

acc	Incoming accumulation vector (8 x int48 lanes)
xbuff	Input buffer of 16 elements of type int32
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 int32
ystart	Starting position offset applied to all lanes for input from Y buffer.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc48 msc8_sym_ct	(	v8acc48	acc,
		v32int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-subtract intrinsic function with center-tap optimization with pre-add from x input buffer .

acc0 -= z00*(x00 + y00) + z01*x01
acc1 -= z10*(x10 + y10) + z11*x11
acc2 -= z20*(x20 + y20) + z21*x21
acc3 -= z30*(x30 + y30) + z31*x31
acc4 -= z40*(x40 + y40) + z41*x41
acc5 -= z50*(x50 + y50) + z51*x51
acc6 -= z60*(x60 + y60) + z61*x61
acc7 -= z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

acc	Incoming accumulation vector (8 x int48 lanes)
xbuff	Input buffer of 32 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc48 msc8_sym_ct	(	v8acc48	acc,
		v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-subtract intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.

acc0 -= z00*(x00 + y00) + z01*x01
acc1 -= z10*(x10 + y10) + z11*x11
acc2 -= z20*(x20 + y20) + z21*x21
acc3 -= z30*(x30 + y30) + z31*x31
acc4 -= z40*(x40 + y40) + z41*x41
acc5 -= z50*(x50 + y50) + z51*x51
acc6 -= z60*(x60 + y60) + z61*x61
acc7 -= z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

acc	Incoming accumulation vector (8 x int48 lanes)
xbuff	Input buffer of 16 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc48 msc8_sym_ct	(	v8acc48	acc,
		v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		v16int32	ybuff,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-subtract intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.

acc0 -= z00*(x00 + y00) + z01*x01
acc1 -= z10*(x10 + y10) + z11*x11
acc2 -= z20*(x20 + y20) + z21*x21
acc3 -= z30*(x30 + y30) + z31*x31
acc4 -= z40*(x40 + y40) + z41*x41
acc5 -= z50*(x50 + y50) + z51*x51
acc6 -= z60*(x60 + y60) + z61*x61
acc7 -= z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

acc	Incoming accumulation vector (8 x int48 lanes)
xbuff	Input buffer of 16 elements of type int32
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 int32
ystart	Starting position offset applied to all lanes for input from Y buffer.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc48 mul8_sym_ct	(	v32int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply intrinsic function with center-tap optimization with pre-add from x input buffer .

acc0 = z00*(x00 + y00) + z01*x01
acc1 = z10*(x10 + y10) + z11*x11
acc2 = z20*(x20 + y20) + z21*x21
acc3 = z30*(x30 + y30) + z31*x31
acc4 = z40*(x40 + y40) + z41*x41
acc5 = z50*(x50 + y50) + z51*x51
acc6 = z60*(x60 + y60) + z61*x61
acc7 = z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

xbuff	Input buffer of 32 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc48 mul8_sym_ct	(	v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.

acc0 = z00*(x00 + y00) + z01*x01
acc1 = z10*(x10 + y10) + z11*x11
acc2 = z20*(x20 + y20) + z21*x21
acc3 = z30*(x30 + y30) + z31*x31
acc4 = z40*(x40 + y40) + z41*x41
acc5 = z50*(x50 + y50) + z51*x51
acc6 = z60*(x60 + y60) + z61*x61
acc7 = z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

xbuff	Input buffer of 16 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc48 mul8_sym_ct	(	v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		v16int32	ybuff,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.

acc0 = z00*(x00 + y00) + z01*x01
acc1 = z10*(x10 + y10) + z11*x11
acc2 = z20*(x20 + y20) + z21*x21
acc3 = z30*(x30 + y30) + z31*x31
acc4 = z40*(x40 + y40) + z41*x41
acc5 = z50*(x50 + y50) + z51*x51
acc6 = z60*(x60 + y60) + z61*x61
acc7 = z70*(x70 + y70) + z71*x71

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

xbuff	Input buffer of 16 elements of type int32
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 int32
ystart	Starting position offset applied to all lanes for input from Y buffer.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc48 negmul8_sym_ct	(	v32int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-negate intrinsic function with center-tap optimization with pre-add from x input buffer .

acc0 = -( z00*(x00 + y00) + z01*x01 )
acc1 = -( z10*(x10 + y10) + z11*x11 )
acc2 = -( z20*(x20 + y20) + z21*x21 )
acc3 = -( z30*(x30 + y30) + z31*x31 )
acc4 = -( z40*(x40 + y40) + z41*x41 )
acc5 = -( z50*(x50 + y50) + z51*x51 )
acc6 = -( z60*(x60 + y60) + z61*x61 )
acc7 = -( z70*(x70 + y70) + z71*x71 )

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

xbuff	Input buffer of 32 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc48 negmul8_sym_ct	(	v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-negate intrinsic function with center-tap optimization with pre-add from x input buffer using small X input buffer.

acc0 = -( z00*(x00 + y00) + z01*x01 )
acc1 = -( z10*(x10 + y10) + z11*x11 )
acc2 = -( z20*(x20 + y20) + z21*x21 )
acc3 = -( z30*(x30 + y30) + z31*x31 )
acc4 = -( z40*(x40 + y40) + z41*x41 )
acc5 = -( z50*(x50 + y50) + z51*x51 )
acc6 = -( z60*(x60 + y60) + z61*x61 )
acc7 = -( z70*(x70 + y70) + z71*x71 )

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

xbuff	Input buffer of 16 elements of type int32
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.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.

v8acc48 negmul8_sym_ct	(	v16int32	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		v16int32	ybuff,
		int	ystart,
		unsigned int	mtap,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Symetric multiply-negate intrinsic function with center-tap optimization with pre-add from x and y input buffers using small X input buffer.

acc0 = -( z00*(x00 + y00) + z01*x01 )
acc1 = -( z10*(x10 + y10) + z11*x11 )
acc2 = -( z20*(x20 + y20) + z21*x21 )
acc3 = -( z30*(x30 + y30) + z31*x31 )
acc4 = -( z40*(x40 + y40) + z41*x41 )
acc5 = -( z50*(x50 + y50) + z51*x51 )
acc6 = -( z60*(x60 + y60) + z61*x61 )
acc7 = -( z70*(x70 + y70) + z71*x71 )

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

xbuff	Input buffer of 16 elements of type int32
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 int32
ystart	Starting position offset applied to all lanes for input from Y buffer.
mtap	Selector for partial pre-adding relative to xstart
zbuff	Input buffer of 16 elements of type int16
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
zstep	Step between each column for selection in the zbuffer.

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.