| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp9/common/ |
| D | vp9_loopfilter_filters.c | 23 uint8_t p1, uint8_t p0, in filter_mask() argument
29 mask |= (abs(p1 - p0) > limit) * -1; in filter_mask()
33 mask |= (abs(p0 - q0) * 2 + abs(p1 - q1) / 2 > blimit) * -1; in filter_mask()
39 uint8_t p1, uint8_t p0, in flat_mask4() argument
43 mask |= (abs(p1 - p0) > thresh) * -1; in flat_mask4()
45 mask |= (abs(p2 - p0) > thresh) * -1; in flat_mask4()
47 mask |= (abs(p3 - p0) > thresh) * -1; in flat_mask4()
55 uint8_t p0, uint8_t q0, in flat_mask5() argument
58 int8_t mask = ~flat_mask4(thresh, p3, p2, p1, p0, q0, q1, q2, q3); in flat_mask5()
59 mask |= (abs(p4 - p0) > thresh) * -1; in flat_mask5()
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/mips/dspr2/ |
| D | loopfilter_filters_dspr2.c | 48 uint32_t p0, in vp8_filter_mask_vec_mips() argument
120 [p1] "r" (p1), [p0] "r" (p0), [q1] "r" (q1), [q0] "r" (q0), in vp8_filter_mask_vec_mips()
150 : [p0] "r" (p0), [q0] "r" (q0), [p1] "r" (p1), [r3] "r" (r3), in vp8_filter_mask_vec_mips()
355 uint32_t pm1, p0, p1, p2, p3, p4, p5, p6; in vp8_loop_filter_horizontal_edge_mips() local
396 p0 = *((uint32_t *)(s0)); in vp8_loop_filter_horizontal_edge_mips()
400 vp8_filter_mask_vec_mips(limit, flimit, p1, p2, pm1, p0, p3, p4, p5, p6, in vp8_loop_filter_horizontal_edge_mips()
441 p0 = *((uint32_t *)(s0)); in vp8_loop_filter_horizontal_edge_mips()
445 vp8_filter_mask_vec_mips(limit, flimit, p1, p2, pm1, p0, p3, p4, p5, p6, in vp8_loop_filter_horizontal_edge_mips()
486 p0 = *((uint32_t *)(s0)); in vp8_loop_filter_horizontal_edge_mips()
490 vp8_filter_mask_vec_mips(limit, flimit, p1, p2, pm1, p0, p3, p4, p5, p6, in vp8_loop_filter_horizontal_edge_mips()
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp9/common/mips/dspr2/ |
| D | vp9_loopfilter_masks_dspr2.h | 28 uint32_t p1, uint32_t p0, in vp9_filter_hev_mask_dspr2() argument
94 [p1] "r" (p1), [p0] "r" (p0), [q1] "r" (q1), [q0] "r" (q0), in vp9_filter_hev_mask_dspr2()
124 : [p0] "r" (p0), [q0] "r" (q0), [p1] "r" (p1), [r3] "r" (r3), in vp9_filter_hev_mask_dspr2()
135 uint32_t p1, uint32_t p0, in vp9_filter_hev_mask_flatmask4_dspr2() argument
242 [p1] "r" (p1), [p0] "r" (p0), [q1] "r" (q1), [q0] "r" (q0), in vp9_filter_hev_mask_flatmask4_dspr2()
273 : [p0] "r" (p0), [q0] "r" (q0), [p1] "r" (p1), [r3] "r" (r3), in vp9_filter_hev_mask_flatmask4_dspr2()
284 uint32_t p0, uint32_t q0, in vp9_flatmask5() argument
361 [p1] "r" (p1), [p0] "r" (p0), [q0] "r" (q0), [q1] "r" (q1), in vp9_flatmask5()
|
| D | vp9_mbloop_loopfilter_dspr2.c | 36 uint32_t p3, p2, p1, p0, q0, q1, q2, q3; in vp9_lpf_horizontal_8_dspr2() local
78 : [p3] "=&r" (p3), [p2] "=&r" (p2), [p1] "=&r" (p1), [p0] "=&r" (p0), in vp9_lpf_horizontal_8_dspr2()
85 p1, p0, p3, p2, q0, q1, q2, q3, in vp9_lpf_horizontal_8_dspr2()
89 vp9_filter1_dspr2(mask, hev, p1, p0, q0, q1, in vp9_lpf_horizontal_8_dspr2()
126 : [p2] "r" (p2), [p1] "r" (p1), [p0] "r" (p0), in vp9_lpf_horizontal_8_dspr2()
133 vp9_filter1_dspr2(mask, hev, p1, p0, q0, q1, in vp9_lpf_horizontal_8_dspr2()
231 : [p2] "+r" (p2), [p1] "+r" (p1), [p0] "+r" (p0), in vp9_lpf_horizontal_8_dspr2()
334 uint32_t p3, p2, p1, p0, q3, q2, q1, q0; in vp9_lpf_vertical_8_dspr2() local
373 : [p3] "=&r" (p3), [p2] "=&r" (p2), [p1] "=&r" (p1), [p0] "=&r" (p0), in vp9_lpf_vertical_8_dspr2()
411 [p0] "+r" (p0), [p1] "+r" (p1), [p2] "+r" (p2), [p3] "+r" (p3), in vp9_lpf_vertical_8_dspr2()
[all …]
|
| D | vp9_loopfilter_filters_dspr2.c | 32 uint32_t pm1, p0, p1, p2, p3, p4, p5, p6; in vp9_lpf_horizontal_4_dspr2() local
86 : [pm1] "=&r" (pm1), [p0] "=&r" (p0), [p5] "=&r" (p5), in vp9_lpf_horizontal_4_dspr2()
92 pm1, p0, p3, p4, p5, p6, in vp9_lpf_horizontal_4_dspr2()
125 uint32_t pm1, p0, p1, p2, p3, p4, p5, p6; in vp9_lpf_vertical_4_dspr2() local
163 p0 = *((uint32_t *)(s3 - 4)); in vp9_lpf_vertical_4_dspr2()
187 [p2] "+r" (p2), [p1] "+r" (p1), [p0] "+r" (p0), [pm1] "+r" (pm1), in vp9_lpf_vertical_4_dspr2()
221 p0, p3, p4, p5, p6, thresh_vec, in vp9_lpf_vertical_4_dspr2()
|
| D | vp9_loopfilter_filters_dspr2.h | 376 const uint32_t p3 = *op3, p2 = *op2, p1 = *op1, p0 = *op0; in vp9_mbfilter_dspr2() local
435 : [p0] "r" (p0), [q0] "r" (q0), [p1] "r" (p1), [q1] "r" (q1), in vp9_mbfilter_dspr2()
449 uint32_t p1, uint32_t p0, in vp9_mbfilter1_dspr2() argument
513 : [p0] "r" (p0), [q0] "r" (q0), [p1] "r" (p1), [q1] "r" (q1), in vp9_mbfilter1_dspr2()
535 const uint32_t p3 = *op3, p2 = *op2, p1 = *op1, p0 = *op0; in vp9_wide_mbfilter_dspr2() local
564 [p3] "r" (p3), [p2] "r" (p2), [p1] "r" (p1), [p0] "r" (p0), in vp9_wide_mbfilter_dspr2()
650 [p3] "r" (p3), [p2] "r" (p2), [p1] "r" (p1), [p0] "r" (p0), in vp9_wide_mbfilter_dspr2()
|
| D | vp9_loopfilter_macros_dspr2.h | 362 : [p3] "r" (p3), [p2] "r" (p2), [p1] "r" (p1), [p0] "r" (p0), \
402 : [p3] "r" (p3), [p2] "r" (p2), [p1] "r" (p1), [p0] "r" (p0), \
436 : [p2] "=&r" (p2), [p1] "=&r" (p1), [p0] "=&r" (p0), \
|
| D | vp9_mblpf_vert_loopfilter_dspr2.c | 34 uint32_t p7, p6, p5, p4, p3, p2, p1, p0, q0, q1, q2, q3, q4, q5, q6, q7; in vp9_lpf_vertical_16_dspr2() local
78 [p0] "=&r" (p0), [p7] "=&r" (p7), [p6] "=&r" (p6), in vp9_lpf_vertical_16_dspr2()
132 [p0] "+r" (p0), [p1] "+r" (p1), [p2] "+r" (p2), [p3] "+r" (p3), in vp9_lpf_vertical_16_dspr2()
252 p1, p0, p3, p2, q0, q1, q2, q3, in vp9_lpf_vertical_16_dspr2()
255 vp9_flatmask5(p7, p6, p5, p4, p0, q0, q4, q5, q6, q7, &flat2); in vp9_lpf_vertical_16_dspr2()
260 vp9_filter1_dspr2(mask, hev, p1, p0, q0, q1, in vp9_lpf_vertical_16_dspr2()
294 vp9_filter1_dspr2(mask, hev, p1, p0, q0, q1, in vp9_lpf_vertical_16_dspr2()
469 vp9_filter1_dspr2(mask, hev, p1, p0, q0, q1, in vp9_lpf_vertical_16_dspr2()
|
| D | vp9_mblpf_horiz_loopfilter_dspr2.c | 36 uint32_t p7, p6, p5, p4, p3, p2, p1, p0, q0, q1, q2, q3, q4, q5, q6, q7; in vp9_lpf_horizontal_16_dspr2() local
91 : [p3] "=&r" (p3), [p2] "=&r" (p2), [p1] "=&r" (p1), [p0] "=&r" (p0), in vp9_lpf_horizontal_16_dspr2()
114 p1, p0, p3, p2, q0, q1, q2, q3, in vp9_lpf_horizontal_16_dspr2()
117 vp9_flatmask5(p7, p6, p5, p4, p0, q0, q4, q5, q6, q7, &flat2); in vp9_lpf_horizontal_16_dspr2()
122 vp9_filter1_dspr2(mask, hev, p1, p0, q0, q1, in vp9_lpf_horizontal_16_dspr2()
168 [p2] "r" (p2), [p1] "r" (p1), [p0] "r" (p0), in vp9_lpf_horizontal_16_dspr2()
211 : [p2] "r" (p2), [p1] "r" (p1), [p0] "r" (p0), in vp9_lpf_horizontal_16_dspr2()
218 vp9_filter1_dspr2(mask, hev, p1, p0, q0, q1, in vp9_lpf_horizontal_16_dspr2()
402 vp9_filter1_dspr2(mask, hev, p1, p0, q0, q1, in vp9_lpf_horizontal_16_dspr2()
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/x86/ |
| D | loopfilter_mmx.asm | 100 movq mm4, [rsi+rax] ; p0
101 movq mm5, mm4 ; p0
102 psubusb mm4, mm3 ; p0-=p1
103 psubusb mm3, mm5 ; p1-=p0
104 por mm4, mm3 ; abs(p1 - p0)
117 movq mm6, mm5 ; p0
119 psubusb mm5, mm3 ; p0-=q0
120 psubusb mm3, mm6 ; q0-=p0
121 por mm5, mm3 ; abs(p0 - q0)
122 paddusb mm5, mm5 ; abs(p0-q0)*2
[all …]
|
| D | loopfilter_sse2.asm | 111 movdqa xmm4, [rsi+rax] ; p0
114 movlps xmm4, [rsi + rcx*2] ; p0
119 movdqa xmm5, xmm4 ; p0
120 psubusb xmm4, xmm6 ; p0-=p1
122 psubusb xmm6, xmm5 ; p1-=p0
124 por xmm6, xmm4 ; abs(p1 - p0)
144 movdqa xmm6, xmm5 ; p0
147 psubusb xmm5, xmm3 ; p0-=q0
148 psubusb xmm3, xmm6 ; q0-=p0
149 por xmm5, xmm3 ; abs(p0 - q0)
[all …]
|
| D | subpixel_mmx.asm | 67 psrlq mm5, 16 ; mm5 = p0..p5;
68 punpcklbw mm5, mm0 ; mm5 = p0..p3
157 movq mm3, [rsi+rdx] ; mm3 = p0..p8 = row -1
161 movq mm4, [rsi + 4*rdx] ; mm4 = p0..p3 = row 2
165 movq mm4, [rsi + 2*rdx] ; mm4 = p0..p3 = row 0
169 movq mm4, [rsi] ; mm4 = p0..p3 = row -2
175 movq mm4, [rsi + 2*rdx] ; mm4 = p0..p3 = row 1
179 movq mm4, [rsi + 4*rdx] ; mm4 = p0..p3 = row 3
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp9/common/x86/ |
| D | vp9_loopfilter_mmx.asm | 100 movq mm4, [rsi+rax] ; p0
101 movq mm5, mm4 ; p0
102 psubusb mm4, mm3 ; p0-=p1
103 psubusb mm3, mm5 ; p1-=p0
104 por mm4, mm3 ; abs(p1 - p0)
117 movq mm6, mm5 ; p0
119 psubusb mm5, mm3 ; p0-=q0
120 psubusb mm3, mm6 ; q0-=p0
121 por mm5, mm3 ; abs(p0 - q0)
122 paddusb mm5, mm5 ; abs(p0-q0)*2
[all …]
|
| D | vp9_loopfilter_intrin_sse2.c | 388 __m128i p4, p3, p2, p1, p0, q0, q1, q2, q3, q4; in mb_lpf_horizontal_edge_w_sse2_16() local
396 p0 = _mm_loadu_si128((__m128i *)(s - 1 * p)); in mb_lpf_horizontal_edge_w_sse2_16()
407 _mm_store_si128((__m128i *)&ap[0 * 16], p0); in mb_lpf_horizontal_edge_w_sse2_16()
416 const __m128i abs_p1p0 = _mm_or_si128(_mm_subs_epu8(p1, p0), in mb_lpf_horizontal_edge_w_sse2_16()
417 _mm_subs_epu8(p0, p1)); in mb_lpf_horizontal_edge_w_sse2_16()
422 __m128i abs_p0q0 = _mm_or_si128(_mm_subs_epu8(p0, q0), in mb_lpf_horizontal_edge_w_sse2_16()
423 _mm_subs_epu8(q0, p0)); in mb_lpf_horizontal_edge_w_sse2_16()
464 __m128i ps0 = _mm_xor_si128(p0, t80); in mb_lpf_horizontal_edge_w_sse2_16()
512 work = _mm_max_epu8(_mm_or_si128(_mm_subs_epu8(p2, p0), in mb_lpf_horizontal_edge_w_sse2_16()
513 _mm_subs_epu8(p0, p2)), in mb_lpf_horizontal_edge_w_sse2_16()
[all …]
|
| D | vp9_loopfilter_intrin_avx2.c | 402 __m128i p4, p3, p2, p1, p0, q0, q1, q2, q3, q4; in mb_lpf_horizontal_edge_w_avx2_16() local
416 p0 = _mm_loadu_si128((__m128i *) (s - 1 * p)); in mb_lpf_horizontal_edge_w_avx2_16()
424 const __m128i abs_p1p0 = _mm_or_si128(_mm_subs_epu8(p1, p0), in mb_lpf_horizontal_edge_w_avx2_16()
425 _mm_subs_epu8(p0, p1)); in mb_lpf_horizontal_edge_w_avx2_16()
430 __m128i abs_p0q0 = _mm_or_si128(_mm_subs_epu8(p0, q0), in mb_lpf_horizontal_edge_w_avx2_16()
431 _mm_subs_epu8(q0, p0)); in mb_lpf_horizontal_edge_w_avx2_16()
470 __m128i ps0 = _mm_xor_si128(p0, t80); in mb_lpf_horizontal_edge_w_avx2_16()
523 _mm_or_si128(_mm_subs_epu8(p2, p0), _mm_subs_epu8(p0, p2)), in mb_lpf_horizontal_edge_w_avx2_16()
527 _mm_or_si128(_mm_subs_epu8(p3, p0), _mm_subs_epu8(p0, p3)), in mb_lpf_horizontal_edge_w_avx2_16()
531 _mm_or_si128(_mm_subs_epu8(p4, p0), _mm_subs_epu8(p0, p4)), in mb_lpf_horizontal_edge_w_avx2_16()
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/armv6/ |
| D | simpleloopfilter_v6.asm | 43 pkhtb $b3, $a2, $a0, asr #16 ; 33 23 13 03 -- p0
63 ldr r4, [src, -pstep] ; p0
181 uqsub8 r9, r4, r5 ; p0 - q0
182 uqsub8 r10, r5, r4 ; q0 - p0
184 orr r9, r9, r10 ; abs(p0 - q0)
186 uqadd8 r9, r9, r9 ; abs(p0 - q0) * 2
188 uqadd8 r7, r7, r9 ; abs(p0 - q0)*2 + abs(p1 - q1)/2
200 eor r4, r4, r2 ; p0 offset to convert to a signed value
204 qsub8 r6, r5, r4 ; q0 - p0
206 qadd8 r3, r3, r6 ; vp8_filter += q0 - p0
[all …]
|
| D | loopfilter_v6.asm | 45 pkhtb $b3, $a2, $a0, asr #16 ; 33 23 13 03 -- p0
88 ldr r12, [src], pstep ; p0
99 uqsub8 r6, r11, r12 ; p1 - p0
101 uqsub8 r7, r12, r11 ; p0 - p1
104 orr r6, r6, r7 ; abs (p1-p0)
111 uqsub8 r11, r12, r9 ; p0 - q0
112 uqsub8 r12, r9, r12 ; q0 - p0
115 orr r12, r11, r12 ; abs (p0-q0)
117 uqadd8 r12, r12, r12 ; abs (p0-q0) * 2
120 uqadd8 r12, r12, r6 ; abs (p0-q0)*2 + abs (p1-q1)/2
[all …]
|
| D | intra4x4_predict_v6.asm | 300 add r4, r4, r5, lsl #1 ; [p2+2*p3 | p0+2*p1]
301 add r4, r4, r6 ; [p2+2*p3+p4 | p0+2*p1+p2]
302 uxtab16 r4, r4, r12 ; [p2+2*p3+p4+2 | p0+2*p1+p2+2]
427 uadd16 r4, r6, r7 ; [p2+p3 | p0+p1]
428 uhadd16 r4, r4, r2 ; [(p2+p3+1)>>1 | (p0+p1+1)>>1]
431 add r5, r6, r7, lsl #1 ; [p2+2*p3 | p0+2*p1]
432 add r5, r5, r8 ; [p2+2*p3+p4 | p0+2*p1+p2]
433 uxtab16 r5, r5, r12 ; [p2+2*p3+p4+2 | p0+2*p1+p2+2]
501 uadd16 r11, r4, r5 ; [p1+p2 | p0+p1]
502 uhadd16 r11, r11, r2 ; [(p1+p2+1)>>1 | (p0+p1+1)>>1]
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp9/common/arm/neon/ |
| D | vp9_mb_lpf_neon.asm | 47 vld1.u8 {d7}, [r8@64], r1 ; p0
299 ; d7 p0
314 vabd.u8 d21, d6, d7 ; abs(p1 - p0)
321 vmax.u8 d20, d21, d22 ; max(abs(p1 - p0), abs(q1 - q0))
325 vabd.u8 d24, d7, d8 ; abs(p0 - q0)
330 vqadd.u8 d24, d24, d24 ; b = abs(p0 - q0) * 2
336 vabd.u8 d25, d7, d5 ; abs(p0 - p2)
338 vabd.u8 d27, d4, d7 ; abs(p3 - p0)
342 vmax.u8 d25, d25, d26 ; max(abs(p0 - p2), abs(q0 - q2))
343 vmax.u8 d26, d27, d28 ; max(abs(p3 - p0), abs(q3 - q0))
[all …]
|
| D | vp9_loopfilter_16_neon.asm | 60 vld1.u8 {q6}, [r3@64], r1 ; p0
93 ; q6 p0
109 vabd.u8 q13, q5, q6 ; m3 = abs(p1 - p0)
118 vabd.u8 q9, q6, q7 ; abs(p0 - q0)
126 vcgt.u8 q13, q13, q2 ; (abs(p1 - p0) > thresh)*-1
131 vqadd.u8 q9, q9, q9 ; b = abs(p0 - q0) * 2
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/ |
| D | loopfilter_filters.c | 28 uc p3, uc p2, uc p1, uc p0, in vp8_filter_mask() argument
34 mask |= (abs(p1 - p0) > limit); in vp8_filter_mask()
38 mask |= (abs(p0 - q0) * 2 + abs(p1 - q1) / 2 > blimit); in vp8_filter_mask()
43 static signed char vp8_hevmask(uc thresh, uc p1, uc p0, uc q0, uc q1) in vp8_hevmask() argument
46 hev |= (abs(p1 - p0) > thresh) * -1; in vp8_hevmask()
282 static signed char vp8_simple_filter_mask(uc blimit, uc p1, uc p0, uc q0, uc q1) in vp8_simple_filter_mask() argument
288 signed char mask = (abs(p0 - q0) * 2 + abs(p1 - q1) / 2 <= blimit) * -1; in vp8_simple_filter_mask()
296 signed char p0 = (signed char) * op0 ^ 0x80; in vp8_simple_filter() local
302 filter_value = vp8_signed_char_clamp(filter_value + 3 * (q0 - p0)); in vp8_simple_filter()
313 u = vp8_signed_char_clamp(p0 + Filter2); in vp8_simple_filter()
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/neon/ |
| D | loopfiltersimplehorizontaledge_neon.asm | 29 vld1.u8 {q5}, [r3@128], r1 ; p0
33 vabd.u8 q15, q6, q7 ; abs(p0 - q0)
36 vqadd.u8 q15, q15, q15 ; abs(p0 - q0) * 2
40 vqadd.u8 q15, q15, q14 ; abs(p0 - q0) * 2 + abs(p1 - q1) / 2
43 veor q6, q6, q0 ; ps0: p0 offset to convert to a signed value
47 vcge.u8 q15, q1, q15 ; (abs(p0 - q0)*2 + abs(p1-q1)/2 > limit)*-1
|
| D | loopfiltersimpleverticaledge_neon.asm | 53 vabd.u8 q15, q5, q4 ; abs(p0 - q0)
56 vqadd.u8 q15, q15, q15 ; abs(p0 - q0) * 2
60 vqadd.u8 q15, q15, q14 ; abs(p0 - q0) * 2 + abs(p1 - q1) / 2
63 veor q5, q5, q0 ; ps0: p0 offset to convert to a signed value
67 vcge.u8 q15, q1, q15 ; abs(p0 - q0)*2 + abs(p1-q1)/2 > flimit*2 + limit)*-1
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/ppc/ |
| D | recon_altivec.asm | 17 lvx v1, 0, \Pred ;# v1 = pred = p0..p15
19 vmrghb v2, v0, v1 ;# v2 = 16-bit p0..p7
57 lvx v1, 0, \Pred ;# v1 = pred = p0..p15
58 vmrghb v2, v0, v1 ;# v2 = 16-bit p0..p7
144 lvx v1, 0, r10; ;# v1 = pred = p0..p15
150 vmrghb v2, v0, v1; ;# v2 = 16-bit p0..p7
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/armv6/ |
| D | vp8_subtract_armv6.asm | 48 uxtb16 r5, r1 ; [p2 | p0]
94 uxtb16 r9, r7 ; [p2 | p0] (A)
111 uxtb16 r9, r11 ; [p2 | p0] (B)
139 uxtb16 r9, r7 ; [p2 | p0] (A)
156 uxtb16 r9, r11 ; [p2 | p0] (B)
196 uxtb16 r9, r7 ; [p2 | p0] (A)
213 uxtb16 r9, r11 ; [p2 | p0] (B)
230 uxtb16 r9, r11 ; [p2 | p0] (C)
247 uxtb16 r9, r11 ; [p2 | p0] (D)
|