Lines Matching refs:dst
26 static INLINE void dc_store_4x4(uint8_t *dst, ptrdiff_t stride, in dc_store_4x4() argument
30 for (i = 0; i < 4; ++i, dst += stride) { in dc_store_4x4()
31 vst1_lane_u32((uint32_t *)dst, vreinterpret_u32_u8(dc_dup), 0); in dc_store_4x4()
35 void vpx_dc_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_predictor_4x4_neon() argument
45 dc_store_4x4(dst, stride, dc); in vpx_dc_predictor_4x4_neon()
48 void vpx_dc_left_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_left_predictor_4x4_neon() argument
53 dc_store_4x4(dst, stride, dc); in vpx_dc_left_predictor_4x4_neon()
56 void vpx_dc_top_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_top_predictor_4x4_neon() argument
61 dc_store_4x4(dst, stride, dc); in vpx_dc_top_predictor_4x4_neon()
64 void vpx_dc_128_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_128_predictor_4x4_neon() argument
69 dc_store_4x4(dst, stride, dc); in vpx_dc_128_predictor_4x4_neon()
82 static INLINE void dc_store_8x8(uint8_t *dst, ptrdiff_t stride, in dc_store_8x8() argument
86 for (i = 0; i < 8; ++i, dst += stride) { in dc_store_8x8()
87 vst1_u8(dst, dc_dup); in dc_store_8x8()
91 void vpx_dc_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_predictor_8x8_neon() argument
102 dc_store_8x8(dst, stride, dc); in vpx_dc_predictor_8x8_neon()
105 void vpx_dc_left_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_left_predictor_8x8_neon() argument
110 dc_store_8x8(dst, stride, dc); in vpx_dc_left_predictor_8x8_neon()
113 void vpx_dc_top_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_top_predictor_8x8_neon() argument
118 dc_store_8x8(dst, stride, dc); in vpx_dc_top_predictor_8x8_neon()
121 void vpx_dc_128_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_128_predictor_8x8_neon() argument
126 dc_store_8x8(dst, stride, dc); in vpx_dc_128_predictor_8x8_neon()
140 static INLINE void dc_store_16x16(uint8_t *dst, ptrdiff_t stride, in dc_store_16x16() argument
144 for (i = 0; i < 16; ++i, dst += stride) { in dc_store_16x16()
145 vst1q_u8(dst, dc_dup); in dc_store_16x16()
149 void vpx_dc_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_predictor_16x16_neon() argument
161 dc_store_16x16(dst, stride, dc); in vpx_dc_predictor_16x16_neon()
164 void vpx_dc_left_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_left_predictor_16x16_neon() argument
170 dc_store_16x16(dst, stride, dc); in vpx_dc_left_predictor_16x16_neon()
173 void vpx_dc_top_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_top_predictor_16x16_neon() argument
179 dc_store_16x16(dst, stride, dc); in vpx_dc_top_predictor_16x16_neon()
182 void vpx_dc_128_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_128_predictor_16x16_neon() argument
188 dc_store_16x16(dst, stride, dc); in vpx_dc_128_predictor_16x16_neon()
204 static INLINE void dc_store_32x32(uint8_t *dst, ptrdiff_t stride, in dc_store_32x32() argument
210 for (i = 0; i < 32; ++i, dst += stride) { in dc_store_32x32()
211 vst2q_u8(dst, dc_dup); in dc_store_32x32()
215 void vpx_dc_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_predictor_32x32_neon() argument
231 dc_store_32x32(dst, stride, dc); in vpx_dc_predictor_32x32_neon()
234 void vpx_dc_left_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_left_predictor_32x32_neon() argument
240 dc_store_32x32(dst, stride, dc); in vpx_dc_left_predictor_32x32_neon()
243 void vpx_dc_top_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_top_predictor_32x32_neon() argument
249 dc_store_32x32(dst, stride, dc); in vpx_dc_top_predictor_32x32_neon()
252 void vpx_dc_128_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, in vpx_dc_128_predictor_32x32_neon() argument
258 dc_store_32x32(dst, stride, dc); in vpx_dc_128_predictor_32x32_neon()
263 void vpx_d45_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, in vpx_d45_predictor_4x4_neon() argument
278 vst1_lane_u32((uint32_t *)(dst + 0 * stride), r0, 0); in vpx_d45_predictor_4x4_neon()
279 vst1_lane_u32((uint32_t *)(dst + 1 * stride), r1, 0); in vpx_d45_predictor_4x4_neon()
280 vst1_lane_u32((uint32_t *)(dst + 2 * stride), r2, 0); in vpx_d45_predictor_4x4_neon()
281 vst1_lane_u32((uint32_t *)(dst + 3 * stride), r3, 0); in vpx_d45_predictor_4x4_neon()
282 vst1_lane_u8(dst + 3 * stride + 3, ABCDEFGH, 7); in vpx_d45_predictor_4x4_neon()
285 static INLINE void d45_store_8(uint8_t **dst, const ptrdiff_t stride, in d45_store_8() argument
288 vst1_u8(*dst, *row); in d45_store_8()
289 *dst += stride; in d45_store_8()
292 void vpx_d45_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, in vpx_d45_predictor_8x8_neon() argument
302 vst1_u8(dst, row); in vpx_d45_predictor_8x8_neon()
303 dst += stride; in vpx_d45_predictor_8x8_neon()
304 d45_store_8(&dst, stride, above_right, &row); in vpx_d45_predictor_8x8_neon()
305 d45_store_8(&dst, stride, above_right, &row); in vpx_d45_predictor_8x8_neon()
306 d45_store_8(&dst, stride, above_right, &row); in vpx_d45_predictor_8x8_neon()
307 d45_store_8(&dst, stride, above_right, &row); in vpx_d45_predictor_8x8_neon()
308 d45_store_8(&dst, stride, above_right, &row); in vpx_d45_predictor_8x8_neon()
309 d45_store_8(&dst, stride, above_right, &row); in vpx_d45_predictor_8x8_neon()
310 vst1_u8(dst, above_right); in vpx_d45_predictor_8x8_neon()
313 static INLINE void d45_store_16(uint8_t **dst, const ptrdiff_t stride, in d45_store_16() argument
316 vst1q_u8(*dst, *row); in d45_store_16()
317 *dst += stride; in d45_store_16()
320 void vpx_d45_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, in vpx_d45_predictor_16x16_neon() argument
330 vst1q_u8(dst, row); in vpx_d45_predictor_16x16_neon()
331 dst += stride; in vpx_d45_predictor_16x16_neon()
332 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
333 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
334 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
335 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
336 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
337 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
338 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
339 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
340 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
341 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
342 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
343 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
344 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
345 d45_store_16(&dst, stride, above_right, &row); in vpx_d45_predictor_16x16_neon()
346 vst1q_u8(dst, above_right); in vpx_d45_predictor_16x16_neon()
349 void vpx_d45_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, in vpx_d45_predictor_32x32_neon() argument
365 vst1q_u8(dst, row_0); in vpx_d45_predictor_32x32_neon()
366 dst += 16; in vpx_d45_predictor_32x32_neon()
367 vst1q_u8(dst, row_1); in vpx_d45_predictor_32x32_neon()
368 dst += stride - 16; in vpx_d45_predictor_32x32_neon()
373 vst1q_u8(dst, row_0); in vpx_d45_predictor_32x32_neon()
374 dst += 16; in vpx_d45_predictor_32x32_neon()
375 vst1q_u8(dst, row_1); in vpx_d45_predictor_32x32_neon()
376 dst += stride - 16; in vpx_d45_predictor_32x32_neon()
379 vst1q_u8(dst, above_right); in vpx_d45_predictor_32x32_neon()
380 dst += 16; in vpx_d45_predictor_32x32_neon()
381 vst1q_u8(dst, row_1); in vpx_d45_predictor_32x32_neon()
386 void vpx_d135_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, in vpx_d135_predictor_4x4_neon() argument
402 vst1_lane_u32((uint32_t *)dst, r0, 0); in vpx_d135_predictor_4x4_neon()
403 dst += stride; in vpx_d135_predictor_4x4_neon()
404 vst1_lane_u32((uint32_t *)dst, r1, 0); in vpx_d135_predictor_4x4_neon()
405 dst += stride; in vpx_d135_predictor_4x4_neon()
406 vst1_lane_u32((uint32_t *)dst, r2, 0); in vpx_d135_predictor_4x4_neon()
407 dst += stride; in vpx_d135_predictor_4x4_neon()
408 vst1_lane_u32((uint32_t *)dst, r3, 0); in vpx_d135_predictor_4x4_neon()
411 void vpx_d135_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, in vpx_d135_predictor_8x8_neon() argument
435 vst1_u8(dst, r0); in vpx_d135_predictor_8x8_neon()
436 dst += stride; in vpx_d135_predictor_8x8_neon()
437 vst1_u8(dst, r1); in vpx_d135_predictor_8x8_neon()
438 dst += stride; in vpx_d135_predictor_8x8_neon()
439 vst1_u8(dst, r2); in vpx_d135_predictor_8x8_neon()
440 dst += stride; in vpx_d135_predictor_8x8_neon()
441 vst1_u8(dst, r3); in vpx_d135_predictor_8x8_neon()
442 dst += stride; in vpx_d135_predictor_8x8_neon()
443 vst1_u8(dst, r4); in vpx_d135_predictor_8x8_neon()
444 dst += stride; in vpx_d135_predictor_8x8_neon()
445 vst1_u8(dst, r5); in vpx_d135_predictor_8x8_neon()
446 dst += stride; in vpx_d135_predictor_8x8_neon()
447 vst1_u8(dst, r6); in vpx_d135_predictor_8x8_neon()
448 dst += stride; in vpx_d135_predictor_8x8_neon()
449 vst1_u8(dst, row_0); in vpx_d135_predictor_8x8_neon()
453 uint8_t **dst, const ptrdiff_t stride, const uint8x16_t row_0, in d135_store_16x8() argument
457 vst1q_u8(*dst, row_0); in d135_store_16x8()
458 *dst += stride; in d135_store_16x8()
459 vst1q_u8(*dst, row_1); in d135_store_16x8()
460 *dst += stride; in d135_store_16x8()
461 vst1q_u8(*dst, row_2); in d135_store_16x8()
462 *dst += stride; in d135_store_16x8()
463 vst1q_u8(*dst, row_3); in d135_store_16x8()
464 *dst += stride; in d135_store_16x8()
465 vst1q_u8(*dst, row_4); in d135_store_16x8()
466 *dst += stride; in d135_store_16x8()
467 vst1q_u8(*dst, row_5); in d135_store_16x8()
468 *dst += stride; in d135_store_16x8()
469 vst1q_u8(*dst, row_6); in d135_store_16x8()
470 *dst += stride; in d135_store_16x8()
471 vst1q_u8(*dst, row_7); in d135_store_16x8()
472 *dst += stride; in d135_store_16x8()
475 void vpx_d135_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, in vpx_d135_predictor_16x16_neon() argument
508 d135_store_16x8(&dst, stride, r_0, r_1, r_2, r_3, r_4, r_5, r_6, r_7); in vpx_d135_predictor_16x16_neon()
509 d135_store_16x8(&dst, stride, r_8, r_9, r_a, r_b, r_c, r_d, r_e, row_0); in vpx_d135_predictor_16x16_neon()
512 static INLINE void d135_store_32x2(uint8_t **dst, const ptrdiff_t stride, in d135_store_32x2() argument
516 uint8_t *dst2 = *dst; in d135_store_32x2()
524 *dst += stride; in d135_store_32x2()
527 void vpx_d135_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, in vpx_d135_predictor_32x32_neon() argument
566 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
573 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
580 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
587 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
594 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
601 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
608 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
615 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
622 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
629 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
636 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
643 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
650 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
657 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
664 d135_store_32x2(&dst, stride, r_0, r_1, r_2); in vpx_d135_predictor_32x32_neon()
667 d135_store_32x2(&dst, stride, row_0, row_1, row_2); in vpx_d135_predictor_32x32_neon()
672 void vpx_v_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, in vpx_v_predictor_4x4_neon() argument
678 for (i = 0; i < 4; i++, dst += stride) { in vpx_v_predictor_4x4_neon()
679 *(uint32_t *)dst = d; in vpx_v_predictor_4x4_neon()
683 void vpx_v_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, in vpx_v_predictor_8x8_neon() argument
689 for (i = 0; i < 8; i++, dst += stride) { in vpx_v_predictor_8x8_neon()
690 vst1_u8(dst, d); in vpx_v_predictor_8x8_neon()
694 void vpx_v_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, in vpx_v_predictor_16x16_neon() argument
700 for (i = 0; i < 16; i++, dst += stride) { in vpx_v_predictor_16x16_neon()
701 vst1q_u8(dst, d); in vpx_v_predictor_16x16_neon()
705 void vpx_v_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, in vpx_v_predictor_32x32_neon() argument
716 vst1q_u8(dst, d0); in vpx_v_predictor_32x32_neon()
717 dst += 16; in vpx_v_predictor_32x32_neon()
718 vst1q_u8(dst, d1); in vpx_v_predictor_32x32_neon()
719 dst += stride - 16; in vpx_v_predictor_32x32_neon()
725 void vpx_h_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, in vpx_h_predictor_4x4_neon() argument
734 vst1_lane_u32((uint32_t *)dst, vreinterpret_u32_u8(d), 0); in vpx_h_predictor_4x4_neon()
735 dst += stride; in vpx_h_predictor_4x4_neon()
737 vst1_lane_u32((uint32_t *)dst, vreinterpret_u32_u8(d), 0); in vpx_h_predictor_4x4_neon()
738 dst += stride; in vpx_h_predictor_4x4_neon()
740 vst1_lane_u32((uint32_t *)dst, vreinterpret_u32_u8(d), 0); in vpx_h_predictor_4x4_neon()
741 dst += stride; in vpx_h_predictor_4x4_neon()
743 vst1_lane_u32((uint32_t *)dst, vreinterpret_u32_u8(d), 0); in vpx_h_predictor_4x4_neon()
746 void vpx_h_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, in vpx_h_predictor_8x8_neon() argument
753 vst1_u8(dst, d); in vpx_h_predictor_8x8_neon()
754 dst += stride; in vpx_h_predictor_8x8_neon()
756 vst1_u8(dst, d); in vpx_h_predictor_8x8_neon()
757 dst += stride; in vpx_h_predictor_8x8_neon()
759 vst1_u8(dst, d); in vpx_h_predictor_8x8_neon()
760 dst += stride; in vpx_h_predictor_8x8_neon()
762 vst1_u8(dst, d); in vpx_h_predictor_8x8_neon()
763 dst += stride; in vpx_h_predictor_8x8_neon()
765 vst1_u8(dst, d); in vpx_h_predictor_8x8_neon()
766 dst += stride; in vpx_h_predictor_8x8_neon()
768 vst1_u8(dst, d); in vpx_h_predictor_8x8_neon()
769 dst += stride; in vpx_h_predictor_8x8_neon()
771 vst1_u8(dst, d); in vpx_h_predictor_8x8_neon()
772 dst += stride; in vpx_h_predictor_8x8_neon()
774 vst1_u8(dst, d); in vpx_h_predictor_8x8_neon()
777 static INLINE void h_store_16x8(uint8_t **dst, const ptrdiff_t stride, in h_store_16x8() argument
788 vst1q_u8(*dst, row_0); in h_store_16x8()
789 *dst += stride; in h_store_16x8()
790 vst1q_u8(*dst, row_1); in h_store_16x8()
791 *dst += stride; in h_store_16x8()
792 vst1q_u8(*dst, row_2); in h_store_16x8()
793 *dst += stride; in h_store_16x8()
794 vst1q_u8(*dst, row_3); in h_store_16x8()
795 *dst += stride; in h_store_16x8()
796 vst1q_u8(*dst, row_4); in h_store_16x8()
797 *dst += stride; in h_store_16x8()
798 vst1q_u8(*dst, row_5); in h_store_16x8()
799 *dst += stride; in h_store_16x8()
800 vst1q_u8(*dst, row_6); in h_store_16x8()
801 *dst += stride; in h_store_16x8()
802 vst1q_u8(*dst, row_7); in h_store_16x8()
803 *dst += stride; in h_store_16x8()
806 void vpx_h_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, in vpx_h_predictor_16x16_neon() argument
811 h_store_16x8(&dst, stride, vget_low_u8(left_u8q)); in vpx_h_predictor_16x16_neon()
812 h_store_16x8(&dst, stride, vget_high_u8(left_u8q)); in vpx_h_predictor_16x16_neon()
815 static INLINE void h_store_32x8(uint8_t **dst, const ptrdiff_t stride, in h_store_32x8() argument
826 vst1q_u8(*dst, row_0); // Note clang-3.8 produced poor code w/vst2q_u8 in h_store_32x8()
827 *dst += 16; in h_store_32x8()
828 vst1q_u8(*dst, row_0); in h_store_32x8()
829 *dst += stride - 16; in h_store_32x8()
830 vst1q_u8(*dst, row_1); in h_store_32x8()
831 *dst += 16; in h_store_32x8()
832 vst1q_u8(*dst, row_1); in h_store_32x8()
833 *dst += stride - 16; in h_store_32x8()
834 vst1q_u8(*dst, row_2); in h_store_32x8()
835 *dst += 16; in h_store_32x8()
836 vst1q_u8(*dst, row_2); in h_store_32x8()
837 *dst += stride - 16; in h_store_32x8()
838 vst1q_u8(*dst, row_3); in h_store_32x8()
839 *dst += 16; in h_store_32x8()
840 vst1q_u8(*dst, row_3); in h_store_32x8()
841 *dst += stride - 16; in h_store_32x8()
842 vst1q_u8(*dst, row_4); in h_store_32x8()
843 *dst += 16; in h_store_32x8()
844 vst1q_u8(*dst, row_4); in h_store_32x8()
845 *dst += stride - 16; in h_store_32x8()
846 vst1q_u8(*dst, row_5); in h_store_32x8()
847 *dst += 16; in h_store_32x8()
848 vst1q_u8(*dst, row_5); in h_store_32x8()
849 *dst += stride - 16; in h_store_32x8()
850 vst1q_u8(*dst, row_6); in h_store_32x8()
851 *dst += 16; in h_store_32x8()
852 vst1q_u8(*dst, row_6); in h_store_32x8()
853 *dst += stride - 16; in h_store_32x8()
854 vst1q_u8(*dst, row_7); in h_store_32x8()
855 *dst += 16; in h_store_32x8()
856 vst1q_u8(*dst, row_7); in h_store_32x8()
857 *dst += stride - 16; in h_store_32x8()
860 void vpx_h_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, in vpx_h_predictor_32x32_neon() argument
867 h_store_32x8(&dst, stride, vget_low_u8(left_u8)); in vpx_h_predictor_32x32_neon()
868 h_store_32x8(&dst, stride, vget_high_u8(left_u8)); in vpx_h_predictor_32x32_neon()
878 void vpx_tm_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, in vpx_tm_predictor_4x4_neon() argument
895 vst1_lane_u32((uint32_t *)dst, d, 0); in vpx_tm_predictor_4x4_neon()
896 dst += stride; in vpx_tm_predictor_4x4_neon()
897 vst1_lane_u32((uint32_t *)dst, d, 1); in vpx_tm_predictor_4x4_neon()
898 dst += stride; in vpx_tm_predictor_4x4_neon()
903 vst1_lane_u32((uint32_t *)dst, d, 0); in vpx_tm_predictor_4x4_neon()
904 dst += stride; in vpx_tm_predictor_4x4_neon()
905 vst1_lane_u32((uint32_t *)dst, d, 1); in vpx_tm_predictor_4x4_neon()
908 static INLINE void tm_8_kernel(uint8_t **dst, const ptrdiff_t stride, in tm_8_kernel() argument
912 vst1_u8(*dst, d); in tm_8_kernel()
913 *dst += stride; in tm_8_kernel()
916 void vpx_tm_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, in vpx_tm_predictor_8x8_neon() argument
930 tm_8_kernel(&dst, stride, left_dup, sub); in vpx_tm_predictor_8x8_neon()
932 tm_8_kernel(&dst, stride, left_dup, sub); in vpx_tm_predictor_8x8_neon()
934 tm_8_kernel(&dst, stride, left_dup, sub); in vpx_tm_predictor_8x8_neon()
936 tm_8_kernel(&dst, stride, left_dup, sub); in vpx_tm_predictor_8x8_neon()
940 static INLINE void tm_16_kernel(uint8_t **dst, const ptrdiff_t stride, in tm_16_kernel() argument
947 vst1_u8(*dst, d0); in tm_16_kernel()
948 *dst += 8; in tm_16_kernel()
949 vst1_u8(*dst, d1); in tm_16_kernel()
950 *dst += stride - 8; in tm_16_kernel()
953 void vpx_tm_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, in vpx_tm_predictor_16x16_neon() argument
971 tm_16_kernel(&dst, stride, left_dup, sub0, sub1); in vpx_tm_predictor_16x16_neon()
973 tm_16_kernel(&dst, stride, left_dup, sub0, sub1); in vpx_tm_predictor_16x16_neon()
975 tm_16_kernel(&dst, stride, left_dup, sub0, sub1); in vpx_tm_predictor_16x16_neon()
977 tm_16_kernel(&dst, stride, left_dup, sub0, sub1); in vpx_tm_predictor_16x16_neon()
980 tm_16_kernel(&dst, stride, left_dup, sub0, sub1); in vpx_tm_predictor_16x16_neon()
982 tm_16_kernel(&dst, stride, left_dup, sub0, sub1); in vpx_tm_predictor_16x16_neon()
984 tm_16_kernel(&dst, stride, left_dup, sub0, sub1); in vpx_tm_predictor_16x16_neon()
986 tm_16_kernel(&dst, stride, left_dup, sub0, sub1); in vpx_tm_predictor_16x16_neon()
990 static INLINE void tm_32_kernel(uint8_t **dst, const ptrdiff_t stride, in tm_32_kernel() argument
1003 vst1q_u8(*dst, vcombine_u8(d0, d1)); in tm_32_kernel()
1004 *dst += 16; in tm_32_kernel()
1005 vst1q_u8(*dst, vcombine_u8(d2, d3)); in tm_32_kernel()
1006 *dst += stride - 16; in tm_32_kernel()
1009 void vpx_tm_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, in vpx_tm_predictor_32x32_neon() argument
1031 tm_32_kernel(&dst, stride, left_dup, sub0, sub1, sub2, sub3); in vpx_tm_predictor_32x32_neon()
1033 tm_32_kernel(&dst, stride, left_dup, sub0, sub1, sub2, sub3); in vpx_tm_predictor_32x32_neon()
1035 tm_32_kernel(&dst, stride, left_dup, sub0, sub1, sub2, sub3); in vpx_tm_predictor_32x32_neon()
1037 tm_32_kernel(&dst, stride, left_dup, sub0, sub1, sub2, sub3); in vpx_tm_predictor_32x32_neon()