16 bit Real x 16 bit Real

Overview

Vector MAC intrinsics with 16 bit real by 16 bit real

Functions
v16acc48	mac16 (v16acc48 acc, v64int16 xbuff, int xstart, unsigned int xoffsets, unsigned int xoffsets_hi, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, unsigned int zoffsets_hi, int zstep)
	Multiply-accumulate intrinsic function .
v16acc48	mac16 (v16acc48 acc, v32int16 xbuff, int xstart, unsigned int xoffsets, unsigned int xoffsets_hi, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, unsigned int zoffsets_hi, int zstep)
	Multiply-accumulate intrinsic function using small X input buffer.
v8acc48	mac8 (v8acc48 acc, v64int16 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Multiply-accumulate intrinsic function .
v8acc48	mac8 (v8acc48 acc, v32int16 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Multiply-accumulate intrinsic function using small X input buffer.
v16acc48	msc16 (v16acc48 acc, v64int16 xbuff, int xstart, unsigned int xoffsets, unsigned int xoffsets_hi, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, unsigned int zoffsets_hi, int zstep)
	Multiply-subtract intrinsic function .
v16acc48	msc16 (v16acc48 acc, v32int16 xbuff, int xstart, unsigned int xoffsets, unsigned int xoffsets_hi, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, unsigned int zoffsets_hi, int zstep)
	Multiply-subtract intrinsic function using small X input buffer.
v8acc48	msc8 (v8acc48 acc, v64int16 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Multiply-subtract intrinsic function .
v8acc48	msc8 (v8acc48 acc, v32int16 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Multiply-subtract intrinsic function using small X input buffer.
v16acc48	mul16 (v64int16 xbuff, int xstart, unsigned int xoffsets, unsigned int xoffsets_hi, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, unsigned int zoffsets_hi, int zstep)
	Multiply intrinsic function .
v16acc48	mul16 (v32int16 xbuff, int xstart, unsigned int xoffsets, unsigned int xoffsets_hi, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, unsigned int zoffsets_hi, int zstep)
	Multiply intrinsic function using small X input buffer.
v8acc48	mul8 (v64int16 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Multiply intrinsic function .
v8acc48	mul8 (v32int16 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Multiply intrinsic function using small X input buffer.
v16acc48	negmul16 (v64int16 xbuff, int xstart, unsigned int xoffsets, unsigned int xoffsets_hi, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, unsigned int zoffsets_hi, int zstep)
	Multiply-negate intrinsic function .
v16acc48	negmul16 (v32int16 xbuff, int xstart, unsigned int xoffsets, unsigned int xoffsets_hi, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, unsigned int zoffsets_hi, int zstep)
	Multiply-negate intrinsic function using small X input buffer.
v8acc48	negmul8 (v64int16 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Multiply-negate intrinsic function .
v8acc48	negmul8 (v32int16 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v16int16 zbuff, int zstart, unsigned int zoffsets, int zstep)
	Multiply-negate intrinsic function using small X input buffer.

Function Documentation

v16acc48 mac16	(	v16acc48	acc,
		v64int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		unsigned int	xoffsets_hi,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		unsigned int	zoffsets_hi,
		int	zstep
	)

Multiply-accumulate intrinsic function .

acc0 += z00*x00 + z01*x01
acc1 += z10*x10 + z11*x11
acc2 += z20*x20 + z21*x21
acc3 += z30*x30 + z31*x31
acc4 += z40*x40 + z41*x41
acc5 += z50*x50 + z51*x51
acc6 += z60*x60 + z61*x61
acc7 += z70*x70 + z71*x71
acc8 += z80*x80 + z81*x81
acc9 += z90*x90 + z91*x91
acc10 += z100*x100 + z101*x101
acc11 += z110*x110 + z111*x111
acc12 += z120*x120 + z121*x121
acc13 += z130*x130 + z131*x131
acc14 += z140*x140 + z141*x141
acc15 += z150*x150 + z151*x151

Returns

Returned accumulation vector (16 x int48 lanes)

Parameters

acc	Incoming accumulation vector (16 x int48 lanes)
xbuff	Input buffer of 64 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane
xoffsets_hi	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to 8th lane
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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
zoffsets_hi	4b offset for each lane, applied to input from Z buffer. LSB apply to 8th lane
zstep	Step between each column for selection in the zbuffer.

Note

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v16acc48 mac16	(	v16acc48	acc,
		v32int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		unsigned int	xoffsets_hi,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		unsigned int	zoffsets_hi,
		int	zstep
	)

Multiply-accumulate intrinsic function using small X input buffer.

acc0 += z00*x00 + z01*x01
acc1 += z10*x10 + z11*x11
acc2 += z20*x20 + z21*x21
acc3 += z30*x30 + z31*x31
acc4 += z40*x40 + z41*x41
acc5 += z50*x50 + z51*x51
acc6 += z60*x60 + z61*x61
acc7 += z70*x70 + z71*x71
acc8 += z80*x80 + z81*x81
acc9 += z90*x90 + z91*x91
acc10 += z100*x100 + z101*x101
acc11 += z110*x110 + z111*x111
acc12 += z120*x120 + z121*x121
acc13 += z130*x130 + z131*x131
acc14 += z140*x140 + z141*x141
acc15 += z150*x150 + z151*x151

Returns

Returned accumulation vector (16 x int48 lanes)

Parameters

acc	Incoming accumulation vector (16 x int48 lanes)
xbuff	Input buffer of 32 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane
xoffsets_hi	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to 8th lane
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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
zoffsets_hi	4b offset for each lane, applied to input from Z buffer. LSB apply to 8th lane
zstep	Step between each column for selection in the zbuffer.

Note

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8acc48 mac8	(	v8acc48	acc,
		v64int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	xstep,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Multiply-accumulate intrinsic function .

acc0 += z00*x00 + z01*x01 + z02*x02 + z03*x03
acc1 += z10*x10 + z11*x11 + z12*x12 + z13*x13
acc2 += z20*x20 + z21*x21 + z22*x22 + z23*x23
acc3 += z30*x30 + z31*x31 + z32*x32 + z33*x33
acc4 += z40*x40 + z41*x41 + z42*x42 + z43*x43
acc5 += z50*x50 + z51*x51 + z52*x52 + z53*x53
acc6 += z60*x60 + z61*x61 + z62*x62 + z63*x63
acc7 += z70*x70 + z71*x71 + z72*x72 + z73*x73

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

acc	Incoming accumulation vector (8 x int48 lanes)
xbuff	Input buffer of 64 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, while each second lane is an offset to the lane before + 1, applied to the x buffer. LSB apply to first lane
xstep	Step between each column for selection in the xbuffer. xstep is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8acc48 mac8	(	v8acc48	acc,
		v32int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	xstep,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Multiply-accumulate intrinsic function using small X input buffer.

acc0 += z00*x00 + z01*x01 + z02*x02 + z03*x03
acc1 += z10*x10 + z11*x11 + z12*x12 + z13*x13
acc2 += z20*x20 + z21*x21 + z22*x22 + z23*x23
acc3 += z30*x30 + z31*x31 + z32*x32 + z33*x33
acc4 += z40*x40 + z41*x41 + z42*x42 + z43*x43
acc5 += z50*x50 + z51*x51 + z52*x52 + z53*x53
acc6 += z60*x60 + z61*x61 + z62*x62 + z63*x63
acc7 += z70*x70 + z71*x71 + z72*x72 + z73*x73

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 int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, while each second lane is an offset to the lane before + 1, applied to the x buffer. LSB apply to first lane
xstep	Step between each column for selection in the xbuffer. xstep is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v16acc48 msc16	(	v16acc48	acc,
		v64int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		unsigned int	xoffsets_hi,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		unsigned int	zoffsets_hi,
		int	zstep
	)

Multiply-subtract intrinsic function .

acc0 -= z00*x00 + z01*x01
acc1 -= z10*x10 + z11*x11
acc2 -= z20*x20 + z21*x21
acc3 -= z30*x30 + z31*x31
acc4 -= z40*x40 + z41*x41
acc5 -= z50*x50 + z51*x51
acc6 -= z60*x60 + z61*x61
acc7 -= z70*x70 + z71*x71
acc8 -= z80*x80 + z81*x81
acc9 -= z90*x90 + z91*x91
acc10 -= z100*x100 + z101*x101
acc11 -= z110*x110 + z111*x111
acc12 -= z120*x120 + z121*x121
acc13 -= z130*x130 + z131*x131
acc14 -= z140*x140 + z141*x141
acc15 -= z150*x150 + z151*x151

Returns

Returned accumulation vector (16 x int48 lanes)

Parameters

acc	Incoming accumulation vector (16 x int48 lanes)
xbuff	Input buffer of 64 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane
xoffsets_hi	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to 8th lane
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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
zoffsets_hi	4b offset for each lane, applied to input from Z buffer. LSB apply to 8th lane
zstep	Step between each column for selection in the zbuffer.

Note

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v16acc48 msc16	(	v16acc48	acc,
		v32int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		unsigned int	xoffsets_hi,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		unsigned int	zoffsets_hi,
		int	zstep
	)

Multiply-subtract intrinsic function using small X input buffer.

acc0 -= z00*x00 + z01*x01
acc1 -= z10*x10 + z11*x11
acc2 -= z20*x20 + z21*x21
acc3 -= z30*x30 + z31*x31
acc4 -= z40*x40 + z41*x41
acc5 -= z50*x50 + z51*x51
acc6 -= z60*x60 + z61*x61
acc7 -= z70*x70 + z71*x71
acc8 -= z80*x80 + z81*x81
acc9 -= z90*x90 + z91*x91
acc10 -= z100*x100 + z101*x101
acc11 -= z110*x110 + z111*x111
acc12 -= z120*x120 + z121*x121
acc13 -= z130*x130 + z131*x131
acc14 -= z140*x140 + z141*x141
acc15 -= z150*x150 + z151*x151

Returns

Returned accumulation vector (16 x int48 lanes)

Parameters

acc	Incoming accumulation vector (16 x int48 lanes)
xbuff	Input buffer of 32 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane
xoffsets_hi	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to 8th lane
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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
zoffsets_hi	4b offset for each lane, applied to input from Z buffer. LSB apply to 8th lane
zstep	Step between each column for selection in the zbuffer.

Note

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8acc48 msc8	(	v8acc48	acc,
		v64int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	xstep,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Multiply-subtract intrinsic function .

acc0 -= z00*x00 + z01*x01 + z02*x02 + z03*x03
acc1 -= z10*x10 + z11*x11 + z12*x12 + z13*x13
acc2 -= z20*x20 + z21*x21 + z22*x22 + z23*x23
acc3 -= z30*x30 + z31*x31 + z32*x32 + z33*x33
acc4 -= z40*x40 + z41*x41 + z42*x42 + z43*x43
acc5 -= z50*x50 + z51*x51 + z52*x52 + z53*x53
acc6 -= z60*x60 + z61*x61 + z62*x62 + z63*x63
acc7 -= z70*x70 + z71*x71 + z72*x72 + z73*x73

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

acc	Incoming accumulation vector (8 x int48 lanes)
xbuff	Input buffer of 64 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, while each second lane is an offset to the lane before + 1, applied to the x buffer. LSB apply to first lane
xstep	Step between each column for selection in the xbuffer. xstep is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8acc48 msc8	(	v8acc48	acc,
		v32int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	xstep,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Multiply-subtract intrinsic function using small X input buffer.

acc0 -= z00*x00 + z01*x01 + z02*x02 + z03*x03
acc1 -= z10*x10 + z11*x11 + z12*x12 + z13*x13
acc2 -= z20*x20 + z21*x21 + z22*x22 + z23*x23
acc3 -= z30*x30 + z31*x31 + z32*x32 + z33*x33
acc4 -= z40*x40 + z41*x41 + z42*x42 + z43*x43
acc5 -= z50*x50 + z51*x51 + z52*x52 + z53*x53
acc6 -= z60*x60 + z61*x61 + z62*x62 + z63*x63
acc7 -= z70*x70 + z71*x71 + z72*x72 + z73*x73

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 int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, while each second lane is an offset to the lane before + 1, applied to the x buffer. LSB apply to first lane
xstep	Step between each column for selection in the xbuffer. xstep is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v16acc48 mul16	(	v64int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		unsigned int	xoffsets_hi,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		unsigned int	zoffsets_hi,
		int	zstep
	)

Multiply intrinsic function .

acc0 = z00*x00 + z01*x01
acc1 = z10*x10 + z11*x11
acc2 = z20*x20 + z21*x21
acc3 = z30*x30 + z31*x31
acc4 = z40*x40 + z41*x41
acc5 = z50*x50 + z51*x51
acc6 = z60*x60 + z61*x61
acc7 = z70*x70 + z71*x71
acc8 = z80*x80 + z81*x81
acc9 = z90*x90 + z91*x91
acc10 = z100*x100 + z101*x101
acc11 = z110*x110 + z111*x111
acc12 = z120*x120 + z121*x121
acc13 = z130*x130 + z131*x131
acc14 = z140*x140 + z141*x141
acc15 = z150*x150 + z151*x151

Returns

Returned accumulation vector (16 x int48 lanes)

Parameters

xbuff	Input buffer of 64 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane
xoffsets_hi	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to 8th lane
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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
zoffsets_hi	4b offset for each lane, applied to input from Z buffer. LSB apply to 8th lane
zstep	Step between each column for selection in the zbuffer.

Note

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v16acc48 mul16	(	v32int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		unsigned int	xoffsets_hi,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		unsigned int	zoffsets_hi,
		int	zstep
	)

Multiply intrinsic function using small X input buffer.

acc0 = z00*x00 + z01*x01
acc1 = z10*x10 + z11*x11
acc2 = z20*x20 + z21*x21
acc3 = z30*x30 + z31*x31
acc4 = z40*x40 + z41*x41
acc5 = z50*x50 + z51*x51
acc6 = z60*x60 + z61*x61
acc7 = z70*x70 + z71*x71
acc8 = z80*x80 + z81*x81
acc9 = z90*x90 + z91*x91
acc10 = z100*x100 + z101*x101
acc11 = z110*x110 + z111*x111
acc12 = z120*x120 + z121*x121
acc13 = z130*x130 + z131*x131
acc14 = z140*x140 + z141*x141
acc15 = z150*x150 + z151*x151

Returns

Returned accumulation vector (16 x int48 lanes)

Parameters

xbuff	Input buffer of 32 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane
xoffsets_hi	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to 8th lane
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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
zoffsets_hi	4b offset for each lane, applied to input from Z buffer. LSB apply to 8th lane
zstep	Step between each column for selection in the zbuffer.

Note

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8acc48 mul8	(	v64int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	xstep,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Multiply intrinsic function .

acc0 = z00*x00 + z01*x01 + z02*x02 + z03*x03
acc1 = z10*x10 + z11*x11 + z12*x12 + z13*x13
acc2 = z20*x20 + z21*x21 + z22*x22 + z23*x23
acc3 = z30*x30 + z31*x31 + z32*x32 + z33*x33
acc4 = z40*x40 + z41*x41 + z42*x42 + z43*x43
acc5 = z50*x50 + z51*x51 + z52*x52 + z53*x53
acc6 = z60*x60 + z61*x61 + z62*x62 + z63*x63
acc7 = z70*x70 + z71*x71 + z72*x72 + z73*x73

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

xbuff	Input buffer of 64 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, while each second lane is an offset to the lane before + 1, applied to the x buffer. LSB apply to first lane
xstep	Step between each column for selection in the xbuffer. xstep is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8acc48 mul8	(	v32int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	xstep,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Multiply intrinsic function using small X input buffer.

acc0 = z00*x00 + z01*x01 + z02*x02 + z03*x03
acc1 = z10*x10 + z11*x11 + z12*x12 + z13*x13
acc2 = z20*x20 + z21*x21 + z22*x22 + z23*x23
acc3 = z30*x30 + z31*x31 + z32*x32 + z33*x33
acc4 = z40*x40 + z41*x41 + z42*x42 + z43*x43
acc5 = z50*x50 + z51*x51 + z52*x52 + z53*x53
acc6 = z60*x60 + z61*x61 + z62*x62 + z63*x63
acc7 = z70*x70 + z71*x71 + z72*x72 + z73*x73

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

xbuff	Input buffer of 32 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, while each second lane is an offset to the lane before + 1, applied to the x buffer. LSB apply to first lane
xstep	Step between each column for selection in the xbuffer. xstep is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v16acc48 negmul16	(	v64int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		unsigned int	xoffsets_hi,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		unsigned int	zoffsets_hi,
		int	zstep
	)

Multiply-negate intrinsic function .

acc0 = -( z00*x00 + z01*x01 )
acc1 = -( z10*x10 + z11*x11 )
acc2 = -( z20*x20 + z21*x21 )
acc3 = -( z30*x30 + z31*x31 )
acc4 = -( z40*x40 + z41*x41 )
acc5 = -( z50*x50 + z51*x51 )
acc6 = -( z60*x60 + z61*x61 )
acc7 = -( z70*x70 + z71*x71 )
acc8 = -( z80*x80 + z81*x81 )
acc9 = -( z90*x90 + z91*x91 )
acc10 = -( z100*x100 + z101*x101 )
acc11 = -( z110*x110 + z111*x111 )
acc12 = -( z120*x120 + z121*x121 )
acc13 = -( z130*x130 + z131*x131 )
acc14 = -( z140*x140 + z141*x141 )
acc15 = -( z150*x150 + z151*x151 )

Returns

Returned accumulation vector (16 x int48 lanes)

Parameters

xbuff	Input buffer of 64 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane
xoffsets_hi	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to 8th lane
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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
zoffsets_hi	4b offset for each lane, applied to input from Z buffer. LSB apply to 8th lane
zstep	Step between each column for selection in the zbuffer.

Note

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v16acc48 negmul16	(	v32int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		unsigned int	xoffsets_hi,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		unsigned int	zoffsets_hi,
		int	zstep
	)

Multiply-negate intrinsic function using small X input buffer.

acc0 = -( z00*x00 + z01*x01 )
acc1 = -( z10*x10 + z11*x11 )
acc2 = -( z20*x20 + z21*x21 )
acc3 = -( z30*x30 + z31*x31 )
acc4 = -( z40*x40 + z41*x41 )
acc5 = -( z50*x50 + z51*x51 )
acc6 = -( z60*x60 + z61*x61 )
acc7 = -( z70*x70 + z71*x71 )
acc8 = -( z80*x80 + z81*x81 )
acc9 = -( z90*x90 + z91*x91 )
acc10 = -( z100*x100 + z101*x101 )
acc11 = -( z110*x110 + z111*x111 )
acc12 = -( z120*x120 + z121*x121 )
acc13 = -( z130*x130 + z131*x131 )
acc14 = -( z140*x140 + z141*x141 )
acc15 = -( z150*x150 + z151*x151 )

Returns

Returned accumulation vector (16 x int48 lanes)

Parameters

xbuff	Input buffer of 32 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane
xoffsets_hi	4b offset for each lane, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to 8th lane
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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
zoffsets_hi	4b offset for each lane, applied to input from Z buffer. LSB apply to 8th lane
zstep	Step between each column for selection in the zbuffer.

Note

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8acc48 negmul8	(	v64int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	xstep,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Multiply-negate intrinsic function .

acc0 = -( z00*x00 + z01*x01 + z02*x02 + z03*x03 )
acc1 = -( z10*x10 + z11*x11 + z12*x12 + z13*x13 )
acc2 = -( z20*x20 + z21*x21 + z22*x22 + z23*x23 )
acc3 = -( z30*x30 + z31*x31 + z32*x32 + z33*x33 )
acc4 = -( z40*x40 + z41*x41 + z42*x42 + z43*x43 )
acc5 = -( z50*x50 + z51*x51 + z52*x52 + z53*x53 )
acc6 = -( z60*x60 + z61*x61 + z62*x62 + z63*x63 )
acc7 = -( z70*x70 + z71*x71 + z72*x72 + z73*x73 )

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

xbuff	Input buffer of 64 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, while each second lane is an offset to the lane before + 1, applied to the x buffer. LSB apply to first lane
xstep	Step between each column for selection in the xbuffer. xstep is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.

v8acc48 negmul8	(	v32int16	xbuff,
		int	xstart,
		unsigned int	xoffsets,
		int	xstep,
		unsigned int	xsquare,
		v16int16	zbuff,
		int	zstart,
		unsigned int	zoffsets,
		int	zstep
	)

Multiply-negate intrinsic function using small X input buffer.

acc0 = -( z00*x00 + z01*x01 + z02*x02 + z03*x03 )
acc1 = -( z10*x10 + z11*x11 + z12*x12 + z13*x13 )
acc2 = -( z20*x20 + z21*x21 + z22*x22 + z23*x23 )
acc3 = -( z30*x30 + z31*x31 + z32*x32 + z33*x33 )
acc4 = -( z40*x40 + z41*x41 + z42*x42 + z43*x43 )
acc5 = -( z50*x50 + z51*x51 + z52*x52 + z53*x53 )
acc6 = -( z60*x60 + z61*x61 + z62*x62 + z63*x63 )
acc7 = -( z70*x70 + z71*x71 + z72*x72 + z73*x73 )

Returns

Returned accumulation vector (8 x int48 lanes)

Parameters

xbuff	Input buffer of 32 elements of type int16
xstart	Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xoffsets	4b offset for each lane, while each second lane is an offset to the lane before + 1, applied to the x buffer. LSB apply to first lane
xstep	Step between each column for selection in the xbuffer. xstep is restricted to multiples of 2 as granularity for xbuff is 32-bit.
xsquare	Select order of the mini-permute square (default=0x3210). LSB apply to first element
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

• This intrinsic uses the 'square' parameter, to have more information on how to use this please go here
• For more information on how data selection works from the buffers go here. For this intrinsic, the data buffer uses the 16bx16b scheme and the coefficient buffer uses the general scheme.
• Parameter 'zstart' must be a compile time constant.