Lines Matching refs:i16
65 @ vadd.i16 q0, q8, q10
88 vadd.i16 q12, q0, q2 @ temp1 = src[0] + src[2]
90 vsub.i16 q13, q0, q2 @ temp2 = src[0] - src[2]
96 vmov.i16 q8, #\add @ t1 will accumulate here
97 vmov.i16 q9, #\add @ t2 will accumulate here
99 vmul.i16 q10, q1, d0[1] @ t3 = 22 * (src[1])
100 vmul.i16 q11, q3, d0[1] @ t4 = 22 * (src[3])
102 vmla.i16 q8, q12, d0[0] @ t1 = 17 * (temp1) + 4
103 vmla.i16 q9, q13, d0[0] @ t2 = 17 * (temp2) + 4
105 vmla.i16 q10, q3, d1[0] @ t3 += 10 * src[3]
106 vmls.i16 q11, q1, d1[0] @ t4 -= 10 * src[1]
137 vmov.i16 q14, #\add @ t1|t2 will accumulate here
140 vadd.i16 d20, d0, d1 @ temp1 = src[0] + src[32]
141 vsub.i16 d21, d0, d1 @ temp2 = src[0] - src[32]
144 vshl.i16 q15, q2, #4 @ t3|t4 = 16 * (src[16]|src[48])
146 vmla.i16 q14, q10, d0[0] @ t1|t2 = 12 * (temp1|temp2) + 64
151 vmla.i16 q15, q2, d0[1] @ t3|t4 += 6 * (src[48]|src[16])
161 vshl.i16 q8, q1, #4 @ t1|t2 = 16 * (src[8]|src[40])
163 vshl.i16 q12, q3, #4 @ temp3a|temp4a = 16 * src[24]|src[56]
165 vshl.i16 q13, q1, #2 @ temp3b|temp4b = 4 * (src[40]|src[8])
166 vshl.i16 q2, q3, #2 @ temp1|temp2 = 4 * (src[56]|src[24])
168 vsub.i16 q9, q13, q12 @ t3|t4 = - (temp3a|temp4a) + (temp3b|temp4b)
169 vadd.i16 q8, q8, q2 @ t1|t2 += temp1|temp2
170 vmul.i16 q12, q3, d0[3] @ temp3|temp4 = 9 * src[8]|src[24]
171 vmla.i16 q8, q1, d0[3] @ t1|t2 += 9 * (src[40]|src[56])
175 vsub.i16 q11, q14, q15 @ t8|t7 = old t1|t2 - old t3|t4
176 vadd.i16 q10, q14, q15 @ t5|t6 = old t1|t2 + old t3|t4
178 vmov.i16 q15, #1
181 vadd.i16 d18, d18, d24 @ t3 += temp3
182 vsub.i16 d19, d19, d25 @ t4 -= temp4
187 vmla.i16 q9, q1, d0[2] @ t3|t4 += 15 * src[56]|src[40]
188 vmla.i16 q8, q3, d0[2] @ t1|t2 += 15 * src[24]|src[8]
202 vadd.i16 q3, q15, q10 @ line[7,6] = t5|t6 + 1
203 vadd.i16 q2, q15, q11 @ line[5,4] = t7|t8 + 1
241 vmov.i16 q0, #(6 / 2) @ q0 = #6/2
242 vshl.i16 q1, q10, #3 @ t3 = 16/2 * src[16]
243 vshl.i16 q3, q14, #3 @ temp4 = 16/2 * src[48]
244 vmul.i16 q2, q10, q0 @ t4 = 6/2 * src[16]
245 vmla.i16 q1, q14, q0 @ t3 += 6/2 * src[48]
247 vmov.i16 q0, #(12 / 2) @ q0 = #12/2
248 vadd.i16 q10, q8, q12 @ temp1 = src[0] + src[32]
249 vsub.i16 q14, q8, q12 @ temp2 = src[0] - src[32]
251 vmov.i16 q8, #(\add / 2) @ t1 will accumulate here
252 vmov.i16 q12, #(\add / 2) @ t2 will accumulate here
254 vsub.i16 q2, q2, q3 @ t4 = 6/2 * src[16] - 16/2 * src[48]
257 vmla.i16 q8, q10, q0 @ t1 = 12/2 * temp1 + add
258 vmla.i16 q12, q14, q0 @ t2 = 12/2 * temp2 + add
278 vshl.i16 q3, q9, #4 @ t1 = 16 * src[8]
280 vshl.i16 q10, q9, #2 @ t4 = 4 * src[8]
282 vmul.i16 q14, q9, d0[0] @ t2 = 15 * src[8]
283 vmul.i16 q9, q9, d0[1] @ t3 = 9 * src[8]
285 vmla.i16 q3, q11, d0[0] @ t1 += 15 * src[24]
286 vmls.i16 q10, q11, d0[1] @ t4 -= 9 * src[24]
287 vmls.i16 q14, q11, d1[1] @ t2 -= 4 * src[24]
288 vmls.i16 q9, q11, d1[0] @ t3 -= 16 * src[24]
290 vmla.i16 q3, q13, d0[1] @ t1 += 9 * src[40]
291 vmla.i16 q10, q13, d0[0] @ t4 += 15 * src[40]
292 vmls.i16 q14, q13, d1[0] @ t2 -= 16 * src[40]
293 vmla.i16 q9, q13, d1[1] @ t3 += 4 * src[40]
298 vadd.i16 q11, q8, q1 @ t5 = t1 + t3
299 vsub.i16 q1, q8, q1 @ t8 = t1 - t3
300 vadd.i16 q13, q12, q2 @ t6 = t2 + t4
301 vsub.i16 q2, q12, q2 @ t7 = t2 - t4
305 vmov.i16 q12, #1
309 vmla.i16 q3, q15, d1[1] @ t1 += 4 * src[56]
310 vmls.i16 q14, q15, d0[1] @ t2 -= 9 * src[56]
311 vmla.i16 q9, q15, d0[0] @ t3 += 15 * src[56]
312 vmls.i16 q10, q15, d1[0] @ t4 -= 16 * src[56]
342 vadd.i16 q8, q11, q3 @ q8 = t5half + t1
344 vsub.i16 q3, q3, q12 @ q3 = t1 - 1
347 vadd.i16 q0, q13, q14 @ q0 = t6half + t2
349 vsub.i16 q14, q14, q12 @ q14 = t2 - 1
352 vadd.i16 q15, q2, q9 @ q15 = t7half + t3
354 vsub.i16 q9, q9, q12 @ q9 = t3 - 1
359 vsub.i16 q3, q11, q3 @ q3 = t5half - t1 + 1
362 vsub.i16 q14, q13, q14 @ q14 = t6half - t2 + 1
365 vsub.i16 q9, q2, q9 @ q9 = t7half - t3 + 1
370 vadd.i16 q11, q1, q10 @ q11 = t8half + t4
372 vsub.i16 q10, q10, q12 @ q10 = t4 - 1
381 vsub.i16 q10, q1, q10 @ q10 = t8half - t4 + 1
574 vmov.i16 q13, #17
575 vmov.i16 q14, #22
576 vmov.i16 d30, #10 @ only need double-word, not quad-word
581 vmov.i16 q2, #\add @ t1|t2 will accumulate here
583 vadd.i16 d16, d0, d1 @ temp1 = src[0] + src[2]
584 vsub.i16 d17, d0, d1 @ temp2 = src[0] - src[2]
585 vmul.i16 q3, q14, q1 @ t3|t4 = 22 * (src[1]|src[3])
586 vmla.i16 q2, q13, q8 @ t1|t2 = 17 * (temp1|temp2) + add
587 vmla.i16 d6, d30, d3 @ t3 += 10 * src[3]
588 vmls.i16 d7, d30, d2 @ t4 -= 10 * src[1]
590 vadd.i16 q0, q2, q3 @ dst[0,2] = (t1|t2 + t3|t4)
591 vsub.i16 q1, q2, q3 @ dst[3,1] = (t1|t2 - t3|t4)
691 @ Typesize should be i8 or i16.
857 mspel_constants i16, d28, d29, d30, d31, \filter_h_a, \filter_h_b, \filter_h_c, \filter_h_d, q13, r3
1187 vmls.i16 d20, d22, d0[1] @ 2*P1-5*P2
1189 vmls.i16 d18, d2, d0[1] @ 2*P5-5*P6
1192 vmla.i16 d18, d22, d0[1] @ 2*P5-5*P6+5*P7
1198 vsub.i16 d1, d6, d24 @ P4-P5
1199 vmls.i16 d20, d6, d0[0] @ 2*P1-5*P2+5*P3-2*P4
1200 vmla.i16 d4, d24, d0[1] @ 2*P3-5*P4+5*P5
1201 vmls.i16 d4, d2, d0[0] @ 2*P3-5*P4+5*P5-2*P6
1210 vceq.i16 d7, d2, #0 @ test clip == 0
1216 vsub.i16 d1, d1, d4 @ clip_sign - a0_sign
1220 vmul.i16 d0, d4, d0[1] @ a0 >= a3 ? 5*(a0-a3) : 0
1229 vmls.i16 d6, d0, d1 @ invert d depending on clip_sign & a0_sign, or zero it if they match, and accumulate into P4
1230 vmla.i16 d24, d0, d1 @ invert d depending on clip_sign & a0_sign, or zero it if they match, and accumulate into P5
1259 vmls.i16 q3, q8, d0[1] @ 2*P1-5*P2, 2*P5-5*P6
1262 vmla.i16 q3, q9, d0[1] @ 2*P1-5*P2+5*P3, 2*P5-5*P6+5*P7
1263 vmls.i16 q3, q2, d0[0] @ 2*P1-5*P2+5*P3-2*P4, 2*P5-5*P6+5*P7-2*P8
1265 vmls.i16 d20, d4, d0[1] @ 2*P3-5*P4
1266 vmla.i16 d20, d3, d0[1] @ 2*P3-5*P4+5*P5
1267 vsub.i16 d3, d4, d2 @ P4-P5
1268 vmls.i16 d20, d17, d0[0] @ 2*P3-5*P4+5*P5-2*P6
1276 vceq.i16 d18, d5, #0 @ test clip == 0
1280 vsub.i16 d16, d3, d16 @ clip_sign - a0_sign
1285 vmul.i16 d0, d3, d0[1] @ a0 >= a3 ? 5*(a0-a3) : 0
1294 vmla.i16 d2, d0, d16 @ invert d depending on clip_sign & a0_sign, or zero it if they match, and accumulate into P5
1295 vmls.i16 d4, d0, d16 @ invert d depending on clip_sign & a0_sign, or zero it if they match, and accumulate into P4
1327 vmls.i16 q9, q1, d0[1] @ 2*P1-5*P2
1330 vmls.i16 q8, q11, d0[1] @ 2*P5-5*P6
1333 vmla.i16 q8, q1, d0[1] @ 2*P5-5*P6+5*P7
1335 vmla.i16 q9, q13, d0[1] @ 2*P1-5*P2+5*P3
1336 vsub.i16 q13, q3, q1 @ P4-P5
1337 vmls.i16 q2, q3, d0[1] @ 2*P3-5*P4
1338 vmls.i16 q8, q10, d0[0] @ 2*P5-5*P6+5*P7-2*P8
1341 vmls.i16 q9, q3, d0[0] @ 2*P1-5*P2+5*P3-2*P4
1343 vmla.i16 q2, q1, d0[1] @ 2*P3-5*P4+5*P5
1345 vmls.i16 q2, q11, d0[0] @ 2*P3-5*P4+5*P5-2*P6
1346 vceq.i16 q11, q10, #0 @ test clip == 0
1356 vsub.i16 q2, q13, q2 @ clip_sign - a0_sign
1360 vmul.i16 q0, q9, d0[1] @ a0 >= a3 ? 5*(a0-a3) : 0
1374 vmls.i16 q3, q0, q2 @ invert d depending on clip_sign & a0_sign, or zero it if they match, and accumulate into P4
1375 vmla.i16 q1, q0, q2 @ invert d depending on clip_sign & a0_sign, or zero it if they match, and accumulate into P5
1422 vmls.i16 q10, q12, d0[1] @ 2*P1-5*P2
1425 vmls.i16 q11, q13, d0[1] @ 2*P5-5*P6
1428 vmla.i16 q11, q12, d0[1] @ 2*P5-5*P6+5*P7
1430 vmla.i16 q10, q14, d0[1] @ 2*P1-5*P2+5*P3
1431 vsub.i16 q12, q2, q3 @ P4-P5
1432 vmls.i16 q1, q2, d0[1] @ 2*P3-5*P4
1433 vmls.i16 q11, q8, d0[0] @ 2*P5-5*P6+5*P7-2*P8
1436 vmls.i16 q10, q2, d0[0] @ 2*P1-5*P2+5*P3-2*P4
1438 vmla.i16 q1, q3, d0[1] @ 2*P3-5*P4+5*P5
1440 vmls.i16 q1, q13, d0[0] @ 2*P3-5*P4+5*P5-2*P6
1441 vceq.i16 q13, q8, #0 @ test clip == 0
1451 vsub.i16 q1, q12, q1 @ clip_sign - a0_sign
1455 vmul.i16 q0, q10, d0[1] @ a0 >= a3 ? 5*(a0-a3) : 0
1466 vmla.i16 q3, q0, q1 @ invert d depending on clip_sign & a0_sign, or zero it if they match, and accumulate into P5
1467 vmls.i16 q2, q0, q1 @ invert d depending on clip_sign & a0_sign, or zero it if they match, and accumulate into P4
1508 vmls.i16 q8, q10, d0[1] @ 2*P1[0..7]-5*P2[0..7]
1511 vmls.i16 q7, q12, d0[1] @ 2*P5[0..7]-5*P6[0..7]
1515 vmls.i16 q2, q3, d0[1] @ 2*P1[8..15]-5*P2[8..15]
1517 vmls.i16 q10, q4, d0[1] @ 2*P5[8..15]-5*P6[8..15]
1519 vmla.i16 q8, q14, d0[1] @ 2*P1[0..7]-5*P2[0..7]+5*P3[0..7]
1522 vmla.i16 q7, q15, d0[1] @ 2*P5[0..7]-5*P6[0..7]+5*P7[0..7]
1525 vmls.i16 q3, q14, d0[1] @ 2*P3[0..7]-5*P4[0..7]
1526 vmla.i16 q2, q15, d0[1] @ 2*P1[8..15]-5*P2[8..15]+5*P3[8..15]
1529 vmla.i16 q10, q6, d0[1] @ 2*P5[8..15]-5*P6[8..15]+5*P7[8..15]
1532 vmls.i16 q5, q9, d0[1] @ 2*P3[8..15]-5*P4[8..15]
1533 vmls.i16 q8, q14, d0[0] @ 2*P1[0..7]-5*P2[0..7]+5*P3[0..7]-2*P4[0..7]
1534 vmls.i16 q7, q15, d0[0] @ 2*P5[0..7]-5*P6[0..7]+5*P7[0..7]-2*P8[0..7]
1535 vsub.i16 q15, q14, q6 @ P4[0..7]-P5[0..7]
1536 vmla.i16 q3, q6, d0[1] @ 2*P3[0..7]-5*P4[0..7]+5*P5[0..7]
1538 vmls.i16 q2, q9, d0[0] @ 2*P1[8..15]-5*P2[8..15]+5*P3[8..15]-2*P4[8..15]
1540 vmls.i16 q10, q11, d0[0] @ 2*P5[8..15]-5*P6[8..15]+5*P7[8..15]-2*P8[8..15]
1543 vmla.i16 q5, q1, d0[1] @ 2*P3[8..15]-5*P4[8..15]+5*P5[8..15]
1546 vmls.i16 q3, q12, d0[0] @ 2*P3[0..7]-5*P4[0..7]+5*P5[0..7]-2*P6[0..7]
1549 vsub.i16 q12, q9, q1 @ P4[8..15]-P5[8..15]
1551 vmls.i16 q5, q4, d0[0] @ 2*P3[8..15]-5*P4[8..15]+5*P5[8..15]-2*P6[8..15]
1564 vceq.i16 q10, q11, #0 @ test clip[0..7] == 0
1566 vsub.i16 q3, q15, q3 @ clip_sign[0..7] - a0_sign[0..7]
1573 vmul.i16 q15, q15, d0[1] @ a0[0..7] >= a3[0..7] ? 5*(a0[0..7]-a3[0..7]) : 0
1576 vsub.i16 q5, q8, q5 @ clip_sign[8..15] - a0_sign[8..15]
1577 vceq.i16 q8, q7, #0 @ test clip[8..15] == 0
1586 vmul.i16 q0, q13, d0[1] @ a0[8..15] >= a3[8..15] ? 5*(a0[8..15]-a3[8..15]) : 0
1602 vmls.i16 q14, q2, q3 @ invert d[0..7] depending on clip_sign[0..7] & a0_sign[0..7], or zero it if they match, and accumulate into P4[0..7]
1607 vmla.i16 q6, q2, q3 @ invert d[0..7] depending on clip_sign[0..7] & a0_sign[0..7], or zero it if they match, and accumulate into P5[0..7]
1608 vmls.i16 q9, q0, q5 @ invert d[8..15] depending on clip_sign[8..15] & a0_sign[8..15], or zero it if they match, and accumulate into P4[8..15]
1610 vmla.i16 q1, q0, q5 @ invert d[8..15] depending on clip_sign[8..15] & a0_sign[8..15], or zero it if they match, and accumulate into P5[8..15]
1677 vmls.i16 q10, q1, d0[1] @ 2*P1[0..7]-5*P2[0..7]
1681 vmls.i16 q11, q12, d0[1] @ 2*P5[0..7]-5*P6[0..7]
1685 vmla.i16 q10, q1, d0[1] @ 2*P1[0..7]-5*P2[0..7]+5*P3[0..7]
1687 vmls.i16 q2, q14, d0[1] @ 2*P1[8..15]-5*P2[8..15]
1690 vmls.i16 q13, q3, d0[1] @ 2*P5[8..15]-5*P6[8..15]
1692 vmla.i16 q11, q15, d0[1] @ 2*P5[0..7]-5*P6[0..7]+5*P7[0..7]
1695 vmla.i16 q2, q1, d0[1] @ 2*P1[8..15]-5*P2[8..15]+5*P3[8..15]
1698 vmla.i16 q13, q4, d0[1] @ 2*P5[8..15]-5*P6[8..15]+5*P7[8..15]
1701 vmls.i16 q14, q5, d0[1] @ 2*P3[0..7]-5*P4[0..7]
1702 vmls.i16 q15, q6, d0[1] @ 2*P3[8..15]-5*P4[8..15]
1703 vmls.i16 q10, q5, d0[0] @ 2*P1[0..7]-5*P2[0..7]+5*P3[0..7]-2*P4[0..7]
1704 vmls.i16 q11, q1, d0[0] @ 2*P5[0..7]-5*P6[0..7]+5*P7[0..7]-2*P8[0..7]
1705 vsub.i16 q1, q5, q4 @ P4[0..7]-P5[0..7]
1706 vmls.i16 q2, q6, d0[0] @ 2*P1[8..15]-5*P2[8..15]+5*P3[8..15]-2*P4[8..15]
1708 vmls.i16 q13, q7, d0[0] @ 2*P5[8..15]-5*P6[8..15]+5*P7[8..15]-2*P8[8..15]
1709 vsub.i16 q7, q6, q8 @ P4[8..15]-P5[8..15]
1713 vmla.i16 q15, q8, d0[1] @ 2*P3[8..15]-5*P4[8..15]+5*P5[8..15]
1716 vmls.i16 q15, q3, d0[0] @ 2*P3[8..15]-5*P4[8..15]+5*P5[8..15]-2*P6[8..15]
1719 vmls.i16 q14, q12, d0[0] @ 2*P3[0..7]-5*P4[0..7]+5*P5[0..7]-2*P6[0..7]
1737 vsub.i16 q7, q7, q15 @ clip_sign[8..15] - a0_sign[8..15]
1742 vsub.i16 q1, q1, q14 @ clip_sign[0..7] - a0_sign[0..7]
1745 vmul.i16 q13, q15, d0[1] @ a0[8..15] >= a3[8..15] ? 5*(a0[8..15]-a3[8..15]) : 0
1746 vceq.i16 q15, q11, #0 @ test clip[0..7] == 0
1747 vmul.i16 q0, q14, d0[1] @ a0[0..7] >= a3[0..7] ? 5*(a0[0..7]-a3[0..7]) : 0
1749 vceq.i16 q14, q12, #0 @ test clip[8..15] == 0
1766 vmls.i16 q6, q0, q7 @ invert d[8..15] depending on clip_sign[8..15] & a0_sign[8..15], or zero it if they match, and accumulate into P4
1767 vmls.i16 q5, q2, q1 @ invert d[0..7] depending on clip_sign[0..7] & a0_sign[0..7], or zero it if they match, and accumulate into P4
1769 vmla.i16 q4, q2, q1 @ invert d[0..7] depending on clip_sign[0..7] & a0_sign[0..7], or zero it if they match, and accumulate into P5
1772 vmla.i16 q8, q0, q7 @ invert d[8..15] depending on clip_sign[8..15] & a0_sign[8..15], or zero it if they match, and accumulate into P5