68      * load initial offsets  in arm_mat_mult_f16_2x2_mve()
73      * load {a00 a00 a10 a10 x x x x }  in arm_mat_mult_f16_2x2_mve()
77      * load {b00 b01 b00 b01 x x x x }  in arm_mat_mult_f16_2x2_mve()
92      * load {a01 a01 a11 a11 x x x x}  in arm_mat_mult_f16_2x2_mve()
100      * load {b10, b11, b10, b11, x x x x }  in arm_mat_mult_f16_2x2_mve()
137      * load initial offsets  in arm_mat_mult_f16_3x3_mve()
143      * load {a00 a00 a00 a10 a10 a10 a20 a20}  in arm_mat_mult_f16_3x3_mve()
147      * load {b00 b01 b02 b00 b01 b02 b00 b01}  in arm_mat_mult_f16_3x3_mve()
162      * load {a01 a01 a01 a11 a11 a11 a21 a21}  in arm_mat_mult_f16_3x3_mve()
170      * load {b10, b11, b12, b10, b11, b12, b10, b11}  in arm_mat_mult_f16_3x3_mve()
184      * load {a02 a02 a02 a12 a12 a12 a22 a22}  in arm_mat_mult_f16_3x3_mve()
192      * load {b20, b21, b22, b20, b21, b22, b20, b21}  in arm_mat_mult_f16_3x3_mve()
235      * load initial offsets  in arm_mat_mult_f16_4x4_mve()
241      * load {a00 a00 a00 a00 a10 a10 a10 a10}  in arm_mat_mult_f16_4x4_mve()
245      * load {b00 b01 b02 b03 b00 b01 b02 b03}  in arm_mat_mult_f16_4x4_mve()
258      * load {a20 a20 a20 a20 a30 a30 a30 a30}  in arm_mat_mult_f16_4x4_mve()
271      * load {a01 a01 a01 a01 a11 a11 a11 a11}  in arm_mat_mult_f16_4x4_mve()
279      * load {b10, b11, b12, b13, b10, b11, b12, b13}  in arm_mat_mult_f16_4x4_mve()
292      * load {a21 a21 a21 a21 a31 a31 a31 a31}  in arm_mat_mult_f16_4x4_mve()
306      * load {a02 a02 a02 a02 a12 a12 a12 a12}  in arm_mat_mult_f16_4x4_mve()
314      * load {b20, b21, b22, b23, b20, b21, b22, b23}  in arm_mat_mult_f16_4x4_mve()
328      * load {a22 a22 a22 a22 a32 a32 a32 a32}  in arm_mat_mult_f16_4x4_mve()
342      * load {a03 a03 a03 a03 a13 a13 a13 a13}  in arm_mat_mult_f16_4x4_mve()
350      * load {b30, b31, b32, b33, b30, b31, b32, b33}  in arm_mat_mult_f16_4x4_mve()
363      * load {a23 a23 a23 a23 a33 a33 a33 a33}  in arm_mat_mult_f16_4x4_mve()
457                  * load {bi,4n+0, bi,4n+1, bi,4n+2, bi,4n+3..., bi,4n+7}  in arm_mat_mult_f16()
504                  * load {bi,4n+0, bi,4n+1, bi,4n+2, ..bi,4n+colBLeft-1, 0, ..}  in arm_mat_mult_f16()
561                      * load {bi,4n+0, bi,4n+1, bi,4n+2, bi,4n+3, ...bi,4n+7}  in arm_mat_mult_f16()
594                      * load {bi,4n+0, bi,4n+1, bi,4n+2, ..., bi,4n+colBLeft, 0, ...}  in arm_mat_mult_f16()