Lines Matching refs:nt
63 @* @param[in] nt
81 @ word32 nt,
91 @ nt
118 ldr r4,[sp,#40] @loads nt
127 vdup.8 d2, r4 @nt
128 vdup.s16 q8, r4 @nt
130 sub r6, r4, #1 @nt-1
133 vdup.s8 d0, r7 @src[nt-1]
135 add r6, r4, r4,lsl #1 @3nt
136 add r6, r6, #1 @3nt + 1
139 vdup.s8 d1, r7 @src[3nt+1]
141 add r6, r4, r4 @2nt
142 add r14, r6, #1 @2nt+1
143 sub r6, r6, #1 @2nt-1
144 add r6, r6, r0 @&src[2nt-1]
145 add r14, r14, r0 @&src[2nt+1]
148 sub r9, r4, r8 @nt-1-row (row is first 0)
151 vdup.s8 d6, r9 @nt - 1 - row
152 vmov d7, d5 @mov #1 to d7 to used for inc for row+1 and dec for nt-1-row
155 mov r1, r4 @nt (row counter) (dec after every row)
158 mov r0, r14 @&src[2nt+1] (to be reloaded after every row)
168 mov r9, r4, lsr #3 @divide nt by 8
177 mov r10, r4 @nt
178 sub r10, r10, #8 @nt - 8
184 vld1.s8 d4, [r6] @(1-8)src[2nt-1-row]
185 vsub.s8 d9, d2, d8 @(1-8)[nt-1-col]
188 vmlal.u8 q6, d5, d0 @(1)(row+1) * src[nt-1]
190 vld1.s8 d3, [r14] @(1-8)load 8 src[2nt+1+col]
191 vmlal.u8 q6, d8, d1 @(1)(col+1) * src[3nt+1]
194 vmlal.u8 q6, d6, d3 @(1)(nt-1-row) * src[2nt+1+col]
197 vmlal.u8 q6, d9, d20 @(1)(nt-1-col) * src[2nt-1-row]
322 movle r1, r4 @nt reloaded (refresh the value)
329 vld1.s8 d3, [r14] @(1n)(1-8)load 8 src[2nt+1+col]
336 vld1.s8 d4, [r6] @(1n)(1-8)src[2nt-1-row]
337 vsub.s8 d9, d2, d8 @(1n)(1-8)[nt-1-col]
350 vmlal.u8 q6, d5, d0 @(1)(row+1) * src[nt-1]
353 vmlal.u8 q6, d8, d1 @(1)(col+1) * src[3nt+1]
356 vmlal.u8 q6, d6, d3 @(1)(nt-1-row) * src[2nt+1+col]
359 vmlal.u8 q6, d9, d20 @(1)(nt-1-col) * src[2nt-1-row]
370 suble r2, r2, r10 @else go to next set of rows, dst - (nt-8) (cond loop)
382 movle r1, r4 @nt reloaded (refresh the value) (cond loop)
477 vld1.s8 d4, [r6] @(1n)(1-8)src[2nt-1-row]
497 nt-1-row) value
501 vld1.s8 d3, [r14] @(1n)(1-8)load 8 src[2nt+1+col]
505 vsub.s8 d9, d2, d8 @(1n)(1-8)[nt-1-col]
523 vld1.s8 d10, [r14] @load src[2nt+1+col]
527 ldr r7, [r6], #-1 @src[2nt-1-row] (dec to take into account row)
528 vdup.s8 d4, r7 @src[2nt-1-row]
530 vsub.s8 d9, d2, d8 @[nt-1-col]
532 vmull.u8 q6, d5, d0 @(row+1) * src[nt-1]
533 vmlal.u8 q6, d6, d10 @(nt-1-row) * src[2nt+1+col]
534 vmlal.u8 q6, d8, d1 @(col+1) * src[3nt+1]
535 vmlal.u8 q6, d9, d4 @(nt-1-col) * src[2nt-1-row]
536 @ vadd.i16 q6, q6, q8 @add (nt)
543 vsub.s8 d6, d6, d7 @[nt-1-row]--
