![]() |
AI Engine Intrinsics
(AIE) r2p19
|
Vector MAC intrinsics with 8 bit real by 8 bit real
Functions | |
| v16acc48 | mac16 (v16acc48 acc, v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function . | |
| v16acc48 | mac16 (v16acc48 acc, v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function using small X input buffer. | |
| v16acc48 | mac16 (v16acc48 acc, v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function . | |
| v16acc48 | mac16 (v16acc48 acc, v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function using small X input buffer. | |
| v16acc48 | mac16 (v16acc48 acc, v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function . | |
| v16acc48 | mac16 (v16acc48 acc, v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function using small X input buffer. | |
| v16acc48 | mac16 (v16acc48 acc, v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function . | |
| v16acc48 | mac16 (v16acc48 acc, v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function using small X input buffer. | |
| v8acc48 | mac8 (v8acc48 acc, v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function . | |
| v8acc48 | mac8 (v8acc48 acc, v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function using small X input buffer. | |
| v8acc48 | mac8 (v8acc48 acc, v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function . | |
| v8acc48 | mac8 (v8acc48 acc, v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function using small X input buffer. | |
| v8acc48 | mac8 (v8acc48 acc, v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function . | |
| v8acc48 | mac8 (v8acc48 acc, v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function using small X input buffer. | |
| v8acc48 | mac8 (v8acc48 acc, v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function . | |
| v8acc48 | mac8 (v8acc48 acc, v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-accumulate intrinsic function using small X input buffer. | |
| v16acc48 | msc16 (v16acc48 acc, v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function . | |
| v16acc48 | msc16 (v16acc48 acc, v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function using small X input buffer. | |
| v16acc48 | msc16 (v16acc48 acc, v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function . | |
| v16acc48 | msc16 (v16acc48 acc, v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function using small X input buffer. | |
| v16acc48 | msc16 (v16acc48 acc, v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function . | |
| v16acc48 | msc16 (v16acc48 acc, v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function using small X input buffer. | |
| v16acc48 | msc16 (v16acc48 acc, v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function . | |
| v16acc48 | msc16 (v16acc48 acc, v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function using small X input buffer. | |
| v8acc48 | msc8 (v8acc48 acc, v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function . | |
| v8acc48 | msc8 (v8acc48 acc, v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function using small X input buffer. | |
| v8acc48 | msc8 (v8acc48 acc, v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function . | |
| v8acc48 | msc8 (v8acc48 acc, v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function using small X input buffer. | |
| v8acc48 | msc8 (v8acc48 acc, v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function . | |
| v8acc48 | msc8 (v8acc48 acc, v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function using small X input buffer. | |
| v8acc48 | msc8 (v8acc48 acc, v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function . | |
| v8acc48 | msc8 (v8acc48 acc, v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-subtract intrinsic function using small X input buffer. | |
| v16acc48 | mul16 (v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function . | |
| v16acc48 | mul16 (v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function using small X input buffer. | |
| v16acc48 | mul16 (v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function . | |
| v16acc48 | mul16 (v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function using small X input buffer. | |
| v16acc48 | mul16 (v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function . | |
| v16acc48 | mul16 (v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function using small X input buffer. | |
| v16acc48 | mul16 (v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function . | |
| v16acc48 | mul16 (v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function using small X input buffer. | |
| v8acc48 | mul8 (v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function . | |
| v8acc48 | mul8 (v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function using small X input buffer. | |
| v8acc48 | mul8 (v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function . | |
| v8acc48 | mul8 (v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function using small X input buffer. | |
| v8acc48 | mul8 (v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function . | |
| v8acc48 | mul8 (v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function using small X input buffer. | |
| v8acc48 | mul8 (v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function . | |
| v8acc48 | mul8 (v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply intrinsic function using small X input buffer. | |
| v16acc48 | negmul16 (v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function . | |
| v16acc48 | negmul16 (v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function using small X input buffer. | |
| v16acc48 | negmul16 (v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function . | |
| v16acc48 | negmul16 (v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function using small X input buffer. | |
| v16acc48 | negmul16 (v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function . | |
| v16acc48 | negmul16 (v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function using small X input buffer. | |
| v16acc48 | negmul16 (v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function . | |
| v16acc48 | negmul16 (v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function using small X input buffer. | |
| v8acc48 | negmul8 (v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function . | |
| v8acc48 | negmul8 (v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function using small X input buffer. | |
| v8acc48 | negmul8 (v128int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function . | |
| v8acc48 | negmul8 (v64int8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function using small X input buffer. | |
| v8acc48 | negmul8 (v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function . | |
| v8acc48 | negmul8 (v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32int8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function using small X input buffer. | |
| v8acc48 | negmul8 (v128uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function . | |
| v8acc48 | negmul8 (v64uint8 xbuff, int xstart, unsigned int xoffsets, int xstep, unsigned int xsquare, v32uint8 zbuff, int zstart, unsigned int zoffsets, int zstep, unsigned int zsquare) |
| Multiply-negate intrinsic function using small X input buffer. | |
| v16acc48 mac16 | ( | v16acc48 | acc, |
| v128int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mac16 | ( | v16acc48 | acc, |
| v64int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mac16 | ( | v16acc48 | acc, |
| v128int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mac16 | ( | v16acc48 | acc, |
| v64int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mac16 | ( | v16acc48 | acc, |
| v128uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mac16 | ( | v16acc48 | acc, |
| v64uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mac16 | ( | v16acc48 | acc, |
| v128uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mac16 | ( | v16acc48 | acc, |
| v64uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mac8 | ( | v8acc48 | acc, |
| v128int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mac8 | ( | v8acc48 | acc, |
| v64int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mac8 | ( | v8acc48 | acc, |
| v128int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mac8 | ( | v8acc48 | acc, |
| v64int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mac8 | ( | v8acc48 | acc, |
| v128uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mac8 | ( | v8acc48 | acc, |
| v64uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mac8 | ( | v8acc48 | acc, |
| v128uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mac8 | ( | v8acc48 | acc, |
| v64uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-accumulate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 msc16 | ( | v16acc48 | acc, |
| v128int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 msc16 | ( | v16acc48 | acc, |
| v64int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 msc16 | ( | v16acc48 | acc, |
| v128int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 msc16 | ( | v16acc48 | acc, |
| v64int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 msc16 | ( | v16acc48 | acc, |
| v128uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 msc16 | ( | v16acc48 | acc, |
| v64uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 msc16 | ( | v16acc48 | acc, |
| v128uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 msc16 | ( | v16acc48 | acc, |
| v64uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| acc | v16acc48 | Incoming accumulation vector (16 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 msc8 | ( | v8acc48 | acc, |
| v128int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 msc8 | ( | v8acc48 | acc, |
| v64int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 msc8 | ( | v8acc48 | acc, |
| v128int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 msc8 | ( | v8acc48 | acc, |
| v64int8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 msc8 | ( | v8acc48 | acc, |
| v128uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 msc8 | ( | v8acc48 | acc, |
| v64uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 msc8 | ( | v8acc48 | acc, |
| v128uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 msc8 | ( | v8acc48 | acc, |
| v64uint8 | xbuff, | ||
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-subtract intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| acc | v8acc48 | Incoming accumulation vector (8 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mul16 | ( | v128int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mul16 | ( | v64int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mul16 | ( | v128int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mul16 | ( | v64int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mul16 | ( | v128uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mul16 | ( | v64uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mul16 | ( | v128uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 mul16 | ( | v64uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mul8 | ( | v128int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mul8 | ( | v64int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mul8 | ( | v128int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mul8 | ( | v64int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mul8 | ( | v128uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mul8 | ( | v64uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mul8 | ( | v128uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 mul8 | ( | v64uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 negmul16 | ( | v128int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 negmul16 | ( | v64int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 negmul16 | ( | v128int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 negmul16 | ( | v64int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 negmul16 | ( | v128uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 negmul16 | ( | v64uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 negmul16 | ( | v128uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v16acc48 negmul16 | ( | v64uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v16acc48 | Returned accumulation vector (16 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 |
| xstep | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane, applied to input from Z buffer, corresponds to 2x the lane number and each second lane is an offset to the lane before + 1. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 negmul8 | ( | v128int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 negmul8 | ( | v64int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 negmul8 | ( | v128int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v128int8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 negmul8 | ( | v64int8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v64int8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 negmul8 | ( | v128uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 negmul8 | ( | v64uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32int8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32int8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 negmul8 | ( | v128uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function .
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v128uint8 | Input buffer of 128 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| v8acc48 negmul8 | ( | v64uint8 | xbuff, |
| int | xstart, | ||
| unsigned int | xoffsets, | ||
| int | xstep, | ||
| unsigned int | xsquare, | ||
| v32uint8 | zbuff, | ||
| int | zstart, | ||
| unsigned int | zoffsets, | ||
| int | zstep, | ||
| unsigned int | zsquare | ||
| ) |
Multiply-negate intrinsic function using small X input buffer.
| Input/Output | Type | Comments |
|---|---|---|
| return | v8acc48 | Returned accumulation vector (8 x int48 lanes) |
| xbuff | v64uint8 | Input buffer of 64 elements of type int8 |
| xstart | int | Starting position offset applied to all lanes of input from X buffer. xstart is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xoffsets | unsigned int | 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 | unsigned int | Step between each column for selection in the xbuffer. xstep is restricted to multiples of 4 as granularity for xbuff is 32-bit. |
| xsquare | unsigned int | Select order of the mini-permute square (default=0x3210). LSB apply to first element |
| zbuff | v32uint8 | Input buffer of 32 elements of type int8 |
| zstart | int | 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. zstart is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zoffsets | unsigned int | 4b offset for each lane while each second lane is an offset to the lane before + 1, applied to input from Z buffer. LSB apply to first lane |
| zstep | int | Step between each column for selection in the zbuffer. zstep is restricted to multiples of 2 as granularity for zbuff is 16-bit. |
| zsquare | unsigned int | 4b offset for each lane, Select order of the mini-permute square (default=0x3210). LSB apply to first element |