AI Engine-ML Intrinsics User Guide (v2024.1)
Loading...
Searching...
No Matches
Multiply-accumulate of 8b x 4b datatypes

Matrix multiplications in which matrix A has data elements of 8 bit and matrix B has data elements of 4 bit. These operations are emulated on top of int8 x int8. More...

Overview

Matrix multiplications in which matrix A has data elements of 8 bit and matrix B has data elements of 4 bit. These operations are emulated on top of int8 x int8.

For an explanation how these operations works see Multiply Accumulate.

4x16_16x8 <br>

v32acc32 mul_4x16_16x8 (v64int8 a, v128int4 b)
 
v32acc32 negmul_4x16_16x8 (v64int8 a, v128int4 b)
 
v32acc32 mac_4x16_16x8 (v64int8 a, v128int4 b, v32acc32 acc)
 
v32acc32 msc_4x16_16x8 (v64int8 a, v128int4 b, v32acc32 acc)
 
v32acc32 negmac_4x16_16x8 (v64int8 a, v128int4 b, v32acc32 acc)
 
v32acc32 negmsc_4x16_16x8 (v64int8 a, v128int4 b, v32acc32 acc)
 
v32acc32 addmac_4x16_16x8 (v64int8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 addmsc_4x16_16x8 (v64int8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submac_4x16_16x8 (v64int8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submsc_4x16_16x8 (v64int8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 mul_4x16_16x8 (v64int8 a, v128uint4 b)
 
v32acc32 negmul_4x16_16x8 (v64int8 a, v128uint4 b)
 
v32acc32 mac_4x16_16x8 (v64int8 a, v128uint4 b, v32acc32 acc)
 
v32acc32 msc_4x16_16x8 (v64int8 a, v128uint4 b, v32acc32 acc)
 
v32acc32 negmac_4x16_16x8 (v64int8 a, v128uint4 b, v32acc32 acc)
 
v32acc32 negmsc_4x16_16x8 (v64int8 a, v128uint4 b, v32acc32 acc)
 
v32acc32 addmac_4x16_16x8 (v64int8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 addmsc_4x16_16x8 (v64int8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submac_4x16_16x8 (v64int8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submsc_4x16_16x8 (v64int8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 mul_4x16_16x8 (v64uint8 a, v128uint4 b)
 
v32acc32 negmul_4x16_16x8 (v64uint8 a, v128uint4 b)
 
v32acc32 mac_4x16_16x8 (v64uint8 a, v128uint4 b, v32acc32 acc)
 
v32acc32 msc_4x16_16x8 (v64uint8 a, v128uint4 b, v32acc32 acc)
 
v32acc32 negmac_4x16_16x8 (v64uint8 a, v128uint4 b, v32acc32 acc)
 
v32acc32 negmsc_4x16_16x8 (v64uint8 a, v128uint4 b, v32acc32 acc)
 
v32acc32 addmac_4x16_16x8 (v64uint8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 addmsc_4x16_16x8 (v64uint8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submac_4x16_16x8 (v64uint8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submsc_4x16_16x8 (v64uint8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 mul_4x16_16x8 (v64uint8 a, v128int4 b)
 
v32acc32 negmul_4x16_16x8 (v64uint8 a, v128int4 b)
 
v32acc32 mac_4x16_16x8 (v64uint8 a, v128int4 b, v32acc32 acc)
 
v32acc32 msc_4x16_16x8 (v64uint8 a, v128int4 b, v32acc32 acc)
 
v32acc32 negmac_4x16_16x8 (v64uint8 a, v128int4 b, v32acc32 acc)
 
v32acc32 negmsc_4x16_16x8 (v64uint8 a, v128int4 b, v32acc32 acc)
 
v32acc32 addmac_4x16_16x8 (v64uint8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 addmsc_4x16_16x8 (v64uint8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submac_4x16_16x8 (v64uint8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submsc_4x16_16x8 (v64uint8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2)
 

4x16_16x8 with dynamic sign <br>

v32acc32 mul_4x16_16x8 (v64int8 a, int sgn_x, v128int4 b, int sgn_y)
 
v32acc32 negmul_4x16_16x8 (v64int8 a, int sgn_x, v128int4 b, int sgn_y)
 
v32acc32 mac_4x16_16x8 (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc)
 
v32acc32 msc_4x16_16x8 (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc)
 
v32acc32 negmac_4x16_16x8 (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc)
 
v32acc32 negmsc_4x16_16x8 (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc)
 
v32acc32 addmac_4x16_16x8 (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 addmsc_4x16_16x8 (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submac_4x16_16x8 (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submsc_4x16_16x8 (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 mul_4x16_16x8 (v64int8 a, int sgn_x, v128uint4 b, int sgn_y)
 
v32acc32 negmul_4x16_16x8 (v64int8 a, int sgn_x, v128uint4 b, int sgn_y)
 
v32acc32 mac_4x16_16x8 (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc)
 
v32acc32 msc_4x16_16x8 (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc)
 
v32acc32 negmac_4x16_16x8 (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc)
 
v32acc32 negmsc_4x16_16x8 (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc)
 
v32acc32 addmac_4x16_16x8 (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 addmsc_4x16_16x8 (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submac_4x16_16x8 (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submsc_4x16_16x8 (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 mul_4x16_16x8 (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y)
 
v32acc32 negmul_4x16_16x8 (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y)
 
v32acc32 mac_4x16_16x8 (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc)
 
v32acc32 msc_4x16_16x8 (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc)
 
v32acc32 negmac_4x16_16x8 (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc)
 
v32acc32 negmsc_4x16_16x8 (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc)
 
v32acc32 addmac_4x16_16x8 (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 addmsc_4x16_16x8 (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submac_4x16_16x8 (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submsc_4x16_16x8 (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 mul_4x16_16x8 (v64uint8 a, int sgn_x, v128int4 b, int sgn_y)
 
v32acc32 negmul_4x16_16x8 (v64uint8 a, int sgn_x, v128int4 b, int sgn_y)
 
v32acc32 mac_4x16_16x8 (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc)
 
v32acc32 msc_4x16_16x8 (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc)
 
v32acc32 negmac_4x16_16x8 (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc)
 
v32acc32 negmsc_4x16_16x8 (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc)
 
v32acc32 addmac_4x16_16x8 (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 addmsc_4x16_16x8 (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submac_4x16_16x8 (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 
v32acc32 submsc_4x16_16x8 (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2)
 

4x16_16x8 with dynamic negation of multiplication result <br>

v32acc32 mul_4x16_16x8_conf (v64int8 a, v128int4 b, int sub_mul)
 
v32acc32 negmul_4x16_16x8_conf (v64int8 a, v128int4 b, int sub_mul)
 
v32acc32 mul_4x16_16x8_conf (v64int8 a, v128uint4 b, int sub_mul)
 
v32acc32 negmul_4x16_16x8_conf (v64int8 a, v128uint4 b, int sub_mul)
 
v32acc32 mul_4x16_16x8_conf (v64uint8 a, v128uint4 b, int sub_mul)
 
v32acc32 negmul_4x16_16x8_conf (v64uint8 a, v128uint4 b, int sub_mul)
 
v32acc32 mul_4x16_16x8_conf (v64uint8 a, v128int4 b, int sub_mul)
 
v32acc32 negmul_4x16_16x8_conf (v64uint8 a, v128int4 b, int sub_mul)
 

4x16_16x8 with dynamic negation of multiplication result, zeroing of acc1, negation of acc1

v32acc32 mac_4x16_16x8_conf (v64int8 a, v128int4 b, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 msc_4x16_16x8_conf (v64int8 a, v128int4 b, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 negmsc_4x16_16x8_conf (v64int8 a, v128int4 b, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 negmac_4x16_16x8_conf (v64int8 a, v128int4 b, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 addmac_4x16_16x8_conf (v64int8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 addmsc_4x16_16x8_conf (v64int8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submac_4x16_16x8_conf (v64int8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submsc_4x16_16x8_conf (v64int8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 mac_4x16_16x8_conf (v64int8 a, v128uint4 b, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 msc_4x16_16x8_conf (v64int8 a, v128uint4 b, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 negmsc_4x16_16x8_conf (v64int8 a, v128uint4 b, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 negmac_4x16_16x8_conf (v64int8 a, v128uint4 b, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 addmac_4x16_16x8_conf (v64int8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 addmsc_4x16_16x8_conf (v64int8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submac_4x16_16x8_conf (v64int8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submsc_4x16_16x8_conf (v64int8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 mac_4x16_16x8_conf (v64uint8 a, v128uint4 b, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 msc_4x16_16x8_conf (v64uint8 a, v128uint4 b, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 negmsc_4x16_16x8_conf (v64uint8 a, v128uint4 b, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 negmac_4x16_16x8_conf (v64uint8 a, v128uint4 b, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 addmac_4x16_16x8_conf (v64uint8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 addmsc_4x16_16x8_conf (v64uint8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submac_4x16_16x8_conf (v64uint8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submsc_4x16_16x8_conf (v64uint8 a, v128uint4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 mac_4x16_16x8_conf (v64uint8 a, v128int4 b, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 msc_4x16_16x8_conf (v64uint8 a, v128int4 b, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 negmsc_4x16_16x8_conf (v64uint8 a, v128int4 b, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 negmac_4x16_16x8_conf (v64uint8 a, v128int4 b, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 addmac_4x16_16x8_conf (v64uint8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 addmsc_4x16_16x8_conf (v64uint8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submac_4x16_16x8_conf (v64uint8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submsc_4x16_16x8_conf (v64uint8 a, v128int4 b, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 

4x16_16x8 with dynamic sign and dynamic negation of multiplication result

v32acc32 mac_4x16_16x8_conf (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 mul_4x16_16x8_conf (v64int8 a, int sgn_x, v128int4 b, int sgn_y, int sub_mul)
 
v32acc32 negmul_4x16_16x8_conf (v64int8 a, int sgn_x, v128int4 b, int sgn_y, int sub_mul)
 
v32acc32 msc_4x16_16x8_conf (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 negmsc_4x16_16x8_conf (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 negmac_4x16_16x8_conf (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 addmac_4x16_16x8_conf (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 addmsc_4x16_16x8_conf (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submac_4x16_16x8_conf (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submsc_4x16_16x8_conf (v64int8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 mac_4x16_16x8_conf (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 mul_4x16_16x8_conf (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, int sub_mul)
 
v32acc32 negmul_4x16_16x8_conf (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, int sub_mul)
 
v32acc32 msc_4x16_16x8_conf (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 negmsc_4x16_16x8_conf (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 negmac_4x16_16x8_conf (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 addmac_4x16_16x8_conf (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 addmsc_4x16_16x8_conf (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submac_4x16_16x8_conf (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submsc_4x16_16x8_conf (v64int8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 mac_4x16_16x8_conf (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 mul_4x16_16x8_conf (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, int sub_mul)
 
v32acc32 negmul_4x16_16x8_conf (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, int sub_mul)
 
v32acc32 msc_4x16_16x8_conf (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 negmsc_4x16_16x8_conf (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 negmac_4x16_16x8_conf (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 addmac_4x16_16x8_conf (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 addmsc_4x16_16x8_conf (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submac_4x16_16x8_conf (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submsc_4x16_16x8_conf (v64uint8 a, int sgn_x, v128uint4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 mac_4x16_16x8_conf (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 mul_4x16_16x8_conf (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, int sub_mul)
 
v32acc32 negmul_4x16_16x8_conf (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, int sub_mul)
 
v32acc32 msc_4x16_16x8_conf (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc, int zero_acc, int shift16, int sub_mul, int sub_acc1)
 
v32acc32 negmsc_4x16_16x8_conf (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 negmac_4x16_16x8_conf (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc, int zero_acc, int sub_mul, int sub_acc1)
 
v32acc32 addmac_4x16_16x8_conf (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 addmsc_4x16_16x8_conf (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int shift16, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submac_4x16_16x8_conf (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 
v32acc32 submsc_4x16_16x8_conf (v64uint8 a, int sgn_x, v128int4 b, int sgn_y, v32acc32 acc1, v32acc32 acc2, int zero_acc1, int sub_mul, int sub_acc1, int sub_acc2)
 

Function Documentation

◆ addmac_4x16_16x8() [1/8]

v32acc32 addmac_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmac_4x16_16x8() [2/8]

v32acc32 addmac_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmac_4x16_16x8() [3/8]

v32acc32 addmac_4x16_16x8 ( v64int8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmac_4x16_16x8() [4/8]

v32acc32 addmac_4x16_16x8 ( v64int8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmac_4x16_16x8() [5/8]

v32acc32 addmac_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmac_4x16_16x8() [6/8]

v32acc32 addmac_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmac_4x16_16x8() [7/8]

v32acc32 addmac_4x16_16x8 ( v64uint8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmac_4x16_16x8() [8/8]

v32acc32 addmac_4x16_16x8 ( v64uint8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmac_4x16_16x8_conf() [1/8]

v32acc32 addmac_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmac_4x16_16x8_conf() [2/8]

v32acc32 addmac_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmac_4x16_16x8_conf() [3/8]

v32acc32 addmac_4x16_16x8_conf ( v64int8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmac_4x16_16x8_conf() [4/8]

v32acc32 addmac_4x16_16x8_conf ( v64int8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmac_4x16_16x8_conf() [5/8]

v32acc32 addmac_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmac_4x16_16x8_conf() [6/8]

v32acc32 addmac_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmac_4x16_16x8_conf() [7/8]

v32acc32 addmac_4x16_16x8_conf ( v64uint8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmac_4x16_16x8_conf() [8/8]

v32acc32 addmac_4x16_16x8_conf ( v64uint8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmsc_4x16_16x8() [1/8]

v32acc32 addmsc_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmsc_4x16_16x8() [2/8]

v32acc32 addmsc_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmsc_4x16_16x8() [3/8]

v32acc32 addmsc_4x16_16x8 ( v64int8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmsc_4x16_16x8() [4/8]

v32acc32 addmsc_4x16_16x8 ( v64int8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmsc_4x16_16x8() [5/8]

v32acc32 addmsc_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmsc_4x16_16x8() [6/8]

v32acc32 addmsc_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmsc_4x16_16x8() [7/8]

v32acc32 addmsc_4x16_16x8 ( v64uint8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmsc_4x16_16x8() [8/8]

v32acc32 addmsc_4x16_16x8 ( v64uint8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ addmsc_4x16_16x8_conf() [1/8]

v32acc32 addmsc_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmsc_4x16_16x8_conf() [2/8]

v32acc32 addmsc_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmsc_4x16_16x8_conf() [3/8]

v32acc32 addmsc_4x16_16x8_conf ( v64int8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmsc_4x16_16x8_conf() [4/8]

v32acc32 addmsc_4x16_16x8_conf ( v64int8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmsc_4x16_16x8_conf() [5/8]

v32acc32 addmsc_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmsc_4x16_16x8_conf() [6/8]

v32acc32 addmsc_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmsc_4x16_16x8_conf() [7/8]

v32acc32 addmsc_4x16_16x8_conf ( v64uint8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ addmsc_4x16_16x8_conf() [8/8]

v32acc32 addmsc_4x16_16x8_conf ( v64uint8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  shift16,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ mac_4x16_16x8() [1/8]

v32acc32 mac_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ mac_4x16_16x8() [2/8]

v32acc32 mac_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ mac_4x16_16x8() [3/8]

v32acc32 mac_4x16_16x8 ( v64int8  a,
v128int4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ mac_4x16_16x8() [4/8]

v32acc32 mac_4x16_16x8 ( v64int8  a,
v128uint4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ mac_4x16_16x8() [5/8]

v32acc32 mac_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ mac_4x16_16x8() [6/8]

v32acc32 mac_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ mac_4x16_16x8() [7/8]

v32acc32 mac_4x16_16x8 ( v64uint8  a,
v128int4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ mac_4x16_16x8() [8/8]

v32acc32 mac_4x16_16x8 ( v64uint8  a,
v128uint4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ mac_4x16_16x8_conf() [1/8]

v32acc32 mac_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ mac_4x16_16x8_conf() [2/8]

v32acc32 mac_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ mac_4x16_16x8_conf() [3/8]

v32acc32 mac_4x16_16x8_conf ( v64int8  a,
v128int4  b,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ mac_4x16_16x8_conf() [4/8]

v32acc32 mac_4x16_16x8_conf ( v64int8  a,
v128uint4  b,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ mac_4x16_16x8_conf() [5/8]

v32acc32 mac_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ mac_4x16_16x8_conf() [6/8]

v32acc32 mac_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ mac_4x16_16x8_conf() [7/8]

v32acc32 mac_4x16_16x8_conf ( v64uint8  a,
v128int4  b,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ mac_4x16_16x8_conf() [8/8]

v32acc32 mac_4x16_16x8_conf ( v64uint8  a,
v128uint4  b,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ msc_4x16_16x8() [1/8]

v32acc32 msc_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ msc_4x16_16x8() [2/8]

v32acc32 msc_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ msc_4x16_16x8() [3/8]

v32acc32 msc_4x16_16x8 ( v64int8  a,
v128int4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ msc_4x16_16x8() [4/8]

v32acc32 msc_4x16_16x8 ( v64int8  a,
v128uint4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ msc_4x16_16x8() [5/8]

v32acc32 msc_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ msc_4x16_16x8() [6/8]

v32acc32 msc_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ msc_4x16_16x8() [7/8]

v32acc32 msc_4x16_16x8 ( v64uint8  a,
v128int4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ msc_4x16_16x8() [8/8]

v32acc32 msc_4x16_16x8 ( v64uint8  a,
v128uint4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ msc_4x16_16x8_conf() [1/8]

v32acc32 msc_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ msc_4x16_16x8_conf() [2/8]

v32acc32 msc_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ msc_4x16_16x8_conf() [3/8]

v32acc32 msc_4x16_16x8_conf ( v64int8  a,
v128int4  b,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ msc_4x16_16x8_conf() [4/8]

v32acc32 msc_4x16_16x8_conf ( v64int8  a,
v128uint4  b,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ msc_4x16_16x8_conf() [5/8]

v32acc32 msc_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ msc_4x16_16x8_conf() [6/8]

v32acc32 msc_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ msc_4x16_16x8_conf() [7/8]

v32acc32 msc_4x16_16x8_conf ( v64uint8  a,
v128int4  b,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ msc_4x16_16x8_conf() [8/8]

v32acc32 msc_4x16_16x8_conf ( v64uint8  a,
v128uint4  b,
v32acc32  acc,
int  zero_acc,
int  shift16,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
shift16Shift mask of input accumulator acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ mul_4x16_16x8() [1/8]

v32acc32 mul_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
Returns
Result of operation

◆ mul_4x16_16x8() [2/8]

v32acc32 mul_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
Returns
Result of operation

◆ mul_4x16_16x8() [3/8]

v32acc32 mul_4x16_16x8 ( v64int8  a,
v128int4  b 
)
Parameters
aMatrix A
bMatrix B
Returns
Result of operation

◆ mul_4x16_16x8() [4/8]

v32acc32 mul_4x16_16x8 ( v64int8  a,
v128uint4  b 
)
Parameters
aMatrix A
bMatrix B
Returns
Result of operation

◆ mul_4x16_16x8() [5/8]

v32acc32 mul_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
Returns
Result of operation

◆ mul_4x16_16x8() [6/8]

v32acc32 mul_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
Returns
Result of operation

◆ mul_4x16_16x8() [7/8]

v32acc32 mul_4x16_16x8 ( v64uint8  a,
v128int4  b 
)
Parameters
aMatrix A
bMatrix B
Returns
Result of operation

◆ mul_4x16_16x8() [8/8]

v32acc32 mul_4x16_16x8 ( v64uint8  a,
v128uint4  b 
)
Parameters
aMatrix A
bMatrix B
Returns
Result of operation

◆ mul_4x16_16x8_conf() [1/8]

v32acc32 mul_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
int  sub_mul 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
sub_mulNegation mask of multiplication result
Returns
Result of operation

◆ mul_4x16_16x8_conf() [2/8]

v32acc32 mul_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
int  sub_mul 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
sub_mulNegation mask of multiplication result
Returns
Result of operation

◆ mul_4x16_16x8_conf() [3/8]

v32acc32 mul_4x16_16x8_conf ( v64int8  a,
v128int4  b,
int  sub_mul 
)
Parameters
aMatrix A
bMatrix B
sub_mulNegation mask for multiplication result. If a bit of sub_mul is set the corresponding vector lane of the output accumulator will be negated.
Returns
Result of operation

◆ mul_4x16_16x8_conf() [4/8]

v32acc32 mul_4x16_16x8_conf ( v64int8  a,
v128uint4  b,
int  sub_mul 
)
Parameters
aMatrix A
bMatrix B
sub_mulNegation mask for multiplication result. If a bit of sub_mul is set the corresponding vector lane of the output accumulator will be negated.
Returns
Result of operation

◆ mul_4x16_16x8_conf() [5/8]

v32acc32 mul_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
int  sub_mul 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
sub_mulNegation mask of multiplication result
Returns
Result of operation

◆ mul_4x16_16x8_conf() [6/8]

v32acc32 mul_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
int  sub_mul 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
sub_mulNegation mask of multiplication result
Returns
Result of operation

◆ mul_4x16_16x8_conf() [7/8]

v32acc32 mul_4x16_16x8_conf ( v64uint8  a,
v128int4  b,
int  sub_mul 
)
Parameters
aMatrix A
bMatrix B
sub_mulNegation mask for multiplication result. If a bit of sub_mul is set the corresponding vector lane of the output accumulator will be negated.
Returns
Result of operation

◆ mul_4x16_16x8_conf() [8/8]

v32acc32 mul_4x16_16x8_conf ( v64uint8  a,
v128uint4  b,
int  sub_mul 
)
Parameters
aMatrix A
bMatrix B
sub_mulNegation mask for multiplication result. If a bit of sub_mul is set the corresponding vector lane of the output accumulator will be negated.
Returns
Result of operation

◆ negmac_4x16_16x8() [1/8]

v32acc32 negmac_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ negmac_4x16_16x8() [2/8]

v32acc32 negmac_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ negmac_4x16_16x8() [3/8]

v32acc32 negmac_4x16_16x8 ( v64int8  a,
v128int4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ negmac_4x16_16x8() [4/8]

v32acc32 negmac_4x16_16x8 ( v64int8  a,
v128uint4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ negmac_4x16_16x8() [5/8]

v32acc32 negmac_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ negmac_4x16_16x8() [6/8]

v32acc32 negmac_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ negmac_4x16_16x8() [7/8]

v32acc32 negmac_4x16_16x8 ( v64uint8  a,
v128int4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ negmac_4x16_16x8() [8/8]

v32acc32 negmac_4x16_16x8 ( v64uint8  a,
v128uint4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ negmac_4x16_16x8_conf() [1/8]

v32acc32 negmac_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmac_4x16_16x8_conf() [2/8]

v32acc32 negmac_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmac_4x16_16x8_conf() [3/8]

v32acc32 negmac_4x16_16x8_conf ( v64int8  a,
v128int4  b,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmac_4x16_16x8_conf() [4/8]

v32acc32 negmac_4x16_16x8_conf ( v64int8  a,
v128uint4  b,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmac_4x16_16x8_conf() [5/8]

v32acc32 negmac_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmac_4x16_16x8_conf() [6/8]

v32acc32 negmac_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmac_4x16_16x8_conf() [7/8]

v32acc32 negmac_4x16_16x8_conf ( v64uint8  a,
v128int4  b,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmac_4x16_16x8_conf() [8/8]

v32acc32 negmac_4x16_16x8_conf ( v64uint8  a,
v128uint4  b,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmsc_4x16_16x8() [1/8]

v32acc32 negmsc_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ negmsc_4x16_16x8() [2/8]

v32acc32 negmsc_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ negmsc_4x16_16x8() [3/8]

v32acc32 negmsc_4x16_16x8 ( v64int8  a,
v128int4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ negmsc_4x16_16x8() [4/8]

v32acc32 negmsc_4x16_16x8 ( v64int8  a,
v128uint4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ negmsc_4x16_16x8() [5/8]

v32acc32 negmsc_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ negmsc_4x16_16x8() [6/8]

v32acc32 negmsc_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
Returns
Result of operation

◆ negmsc_4x16_16x8() [7/8]

v32acc32 negmsc_4x16_16x8 ( v64uint8  a,
v128int4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ negmsc_4x16_16x8() [8/8]

v32acc32 negmsc_4x16_16x8 ( v64uint8  a,
v128uint4  b,
v32acc32  acc 
)
Parameters
aMatrix A
bMatrix B
accAccumulator
Returns
Result of operation

◆ negmsc_4x16_16x8_conf() [1/8]

v32acc32 negmsc_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmsc_4x16_16x8_conf() [2/8]

v32acc32 negmsc_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmsc_4x16_16x8_conf() [3/8]

v32acc32 negmsc_4x16_16x8_conf ( v64int8  a,
v128int4  b,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmsc_4x16_16x8_conf() [4/8]

v32acc32 negmsc_4x16_16x8_conf ( v64int8  a,
v128uint4  b,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmsc_4x16_16x8_conf() [5/8]

v32acc32 negmsc_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmsc_4x16_16x8_conf() [6/8]

v32acc32 negmsc_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmsc_4x16_16x8_conf() [7/8]

v32acc32 negmsc_4x16_16x8_conf ( v64uint8  a,
v128int4  b,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmsc_4x16_16x8_conf() [8/8]

v32acc32 negmsc_4x16_16x8_conf ( v64uint8  a,
v128uint4  b,
v32acc32  acc,
int  zero_acc,
int  sub_mul,
int  sub_acc1 
)
Parameters
aMatrix A
bMatrix B
accInput accumulator
zero_accZeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ negmul_4x16_16x8() [1/8]

v32acc32 negmul_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
Returns
Result of operation

◆ negmul_4x16_16x8() [2/8]

v32acc32 negmul_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
Returns
Result of operation

◆ negmul_4x16_16x8() [3/8]

v32acc32 negmul_4x16_16x8 ( v64int8  a,
v128int4  b 
)
Parameters
aMatrix A
bMatrix B
Returns
Result of operation

◆ negmul_4x16_16x8() [4/8]

v32acc32 negmul_4x16_16x8 ( v64int8  a,
v128uint4  b 
)
Parameters
aMatrix A
bMatrix B
Returns
Result of operation

◆ negmul_4x16_16x8() [5/8]

v32acc32 negmul_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
Returns
Result of operation

◆ negmul_4x16_16x8() [6/8]

v32acc32 negmul_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
Returns
Result of operation

◆ negmul_4x16_16x8() [7/8]

v32acc32 negmul_4x16_16x8 ( v64uint8  a,
v128int4  b 
)
Parameters
aMatrix A
bMatrix B
Returns
Result of operation

◆ negmul_4x16_16x8() [8/8]

v32acc32 negmul_4x16_16x8 ( v64uint8  a,
v128uint4  b 
)
Parameters
aMatrix A
bMatrix B
Returns
Result of operation

◆ negmul_4x16_16x8_conf() [1/8]

v32acc32 negmul_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
int  sub_mul 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
sub_mulNegation mask of multiplication result
Returns
Result of operation

◆ negmul_4x16_16x8_conf() [2/8]

v32acc32 negmul_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
int  sub_mul 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
sub_mulNegation mask of multiplication result
Returns
Result of operation

◆ negmul_4x16_16x8_conf() [3/8]

v32acc32 negmul_4x16_16x8_conf ( v64int8  a,
v128int4  b,
int  sub_mul 
)
Parameters
aMatrix A
bMatrix B
sub_mulNegation mask for multiplication result. If a bit of sub_mul is set the corresponding vector lane of the output accumulator will be negated.
Returns
Result of operation

◆ negmul_4x16_16x8_conf() [4/8]

v32acc32 negmul_4x16_16x8_conf ( v64int8  a,
v128uint4  b,
int  sub_mul 
)
Parameters
aMatrix A
bMatrix B
sub_mulNegation mask for multiplication result. If a bit of sub_mul is set the corresponding vector lane of the output accumulator will be negated.
Returns
Result of operation

◆ negmul_4x16_16x8_conf() [5/8]

v32acc32 negmul_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
int  sub_mul 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
sub_mulNegation mask of multiplication result
Returns
Result of operation

◆ negmul_4x16_16x8_conf() [6/8]

v32acc32 negmul_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
int  sub_mul 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
sub_mulNegation mask of multiplication result
Returns
Result of operation

◆ negmul_4x16_16x8_conf() [7/8]

v32acc32 negmul_4x16_16x8_conf ( v64uint8  a,
v128int4  b,
int  sub_mul 
)
Parameters
aMatrix A
bMatrix B
sub_mulNegation mask for multiplication result. If a bit of sub_mul is set the corresponding vector lane of the output accumulator will be negated.
Returns
Result of operation

◆ negmul_4x16_16x8_conf() [8/8]

v32acc32 negmul_4x16_16x8_conf ( v64uint8  a,
v128uint4  b,
int  sub_mul 
)
Parameters
aMatrix A
bMatrix B
sub_mulNegation mask for multiplication result. If a bit of sub_mul is set the corresponding vector lane of the output accumulator will be negated.
Returns
Result of operation

◆ submac_4x16_16x8() [1/8]

v32acc32 submac_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submac_4x16_16x8() [2/8]

v32acc32 submac_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submac_4x16_16x8() [3/8]

v32acc32 submac_4x16_16x8 ( v64int8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submac_4x16_16x8() [4/8]

v32acc32 submac_4x16_16x8 ( v64int8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submac_4x16_16x8() [5/8]

v32acc32 submac_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submac_4x16_16x8() [6/8]

v32acc32 submac_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submac_4x16_16x8() [7/8]

v32acc32 submac_4x16_16x8 ( v64uint8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submac_4x16_16x8() [8/8]

v32acc32 submac_4x16_16x8 ( v64uint8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submac_4x16_16x8_conf() [1/8]

v32acc32 submac_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submac_4x16_16x8_conf() [2/8]

v32acc32 submac_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submac_4x16_16x8_conf() [3/8]

v32acc32 submac_4x16_16x8_conf ( v64int8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submac_4x16_16x8_conf() [4/8]

v32acc32 submac_4x16_16x8_conf ( v64int8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submac_4x16_16x8_conf() [5/8]

v32acc32 submac_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submac_4x16_16x8_conf() [6/8]

v32acc32 submac_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submac_4x16_16x8_conf() [7/8]

v32acc32 submac_4x16_16x8_conf ( v64uint8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submac_4x16_16x8_conf() [8/8]

v32acc32 submac_4x16_16x8_conf ( v64uint8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submsc_4x16_16x8() [1/8]

v32acc32 submsc_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submsc_4x16_16x8() [2/8]

v32acc32 submsc_4x16_16x8 ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submsc_4x16_16x8() [3/8]

v32acc32 submsc_4x16_16x8 ( v64int8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submsc_4x16_16x8() [4/8]

v32acc32 submsc_4x16_16x8 ( v64int8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submsc_4x16_16x8() [5/8]

v32acc32 submsc_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submsc_4x16_16x8() [6/8]

v32acc32 submsc_4x16_16x8 ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submsc_4x16_16x8() [7/8]

v32acc32 submsc_4x16_16x8 ( v64uint8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submsc_4x16_16x8() [8/8]

v32acc32 submsc_4x16_16x8 ( v64uint8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
Returns
Result of operation

◆ submsc_4x16_16x8_conf() [1/8]

v32acc32 submsc_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submsc_4x16_16x8_conf() [2/8]

v32acc32 submsc_4x16_16x8_conf ( v64int8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submsc_4x16_16x8_conf() [3/8]

v32acc32 submsc_4x16_16x8_conf ( v64int8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submsc_4x16_16x8_conf() [4/8]

v32acc32 submsc_4x16_16x8_conf ( v64int8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submsc_4x16_16x8_conf() [5/8]

v32acc32 submsc_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128int4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submsc_4x16_16x8_conf() [6/8]

v32acc32 submsc_4x16_16x8_conf ( v64uint8  a,
int  sgn_x,
v128uint4  b,
int  sgn_y,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
sgn_xSign mask for matrix A
bMatrix B
sgn_ySign mask for matrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submsc_4x16_16x8_conf() [7/8]

v32acc32 submsc_4x16_16x8_conf ( v64uint8  a,
v128int4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation

◆ submsc_4x16_16x8_conf() [8/8]

v32acc32 submsc_4x16_16x8_conf ( v64uint8  a,
v128uint4  b,
v32acc32  acc1,
v32acc32  acc2,
int  zero_acc1,
int  sub_mul,
int  sub_acc1,
int  sub_acc2 
)
Parameters
aMatrix A
bMatrix B
acc1Input accumulator 1
acc2Input accumulator 2
zero_acc1Zeroing mask for acc1
sub_mulNegation mask of multiplication result
sub_acc1Negation mask of acc1
Returns
Result of operation