| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/armv6/ |
| D | copymem8x8_v6.asm | 21 ;push {r4-r5}
22 stmdb sp!, {r4-r5}
37 ldrb r5, [r0, #1]
43 strb r5, [r2, #1]
46 ldrb r5, [r0, #3]
51 strb r5, [r2, #3]
54 ldrb r5, [r0, #5]
57 strb r5, [r2, #5]
60 ldrb r5, [r0, #7]
65 strb r5, [r2, #7]
[all …]
|
| D | copymem8x4_v6.asm | 21 ;push {r4-r5}
22 stmdb sp!, {r4-r5}
37 ldrb r5, [r0, #1]
43 strb r5, [r2, #1]
46 ldrb r5, [r0, #3]
51 strb r5, [r2, #3]
54 ldrb r5, [r0, #5]
57 strb r5, [r2, #5]
60 ldrb r5, [r0, #7]
65 strb r5, [r2, #7]
[all …]
|
| D | vp8_variance_halfpixvar16x16_hv_armv6.asm | 42 ldr r5, [r9, #0] ; load source pixels c, row N+1
51 uhsub8 r5, r5, r7
52 eor r5, r5, r10
54 mvn r5, r5
55 uhsub8 r4, r4, r5
56 ldr r5, [r2, #0] ; load 4 ref pixels
59 usub8 r6, r4, r5 ; calculate difference
62 usub8 r6, r5, r4 ; calculate difference with reversed operands
68 usad8 r5, r6, lr ; calculate sum of negative differences
72 subs r8, r8, r5 ; subtract negative differences from sum
[all …]
|
| D | intra4x4_predict_v6.asm | 55 ldrb r5, [r1], r2 ; Left[1]
61 add r4, r4, r5
85 ldrb r5, [r1], r2 ; Left[1]
98 add r5, r5, r5, lsl #16 ; l[1|1]
107 sadd16 r4, r5, r10 ; l[1|1] + a[2|0] - [tl|tl]
108 sadd16 r5, r5, r11 ; l[1|1] + a[3|1] - [tl|tl]
114 usat16 r5, #8, r5
119 add r12, r4, r5, lsl #8 ; [3|2|1|0]
126 sadd16 r5, r7, r11 ; l[3|3] + a[3|1] - [tl|tl]
131 usat16 r5, #8, r5
[all …]
|
| D | vp8_variance16x16_armv6.asm | 39 ldr r5, [r2, #0] ; load 4 ref pixels
43 usub8 r6, r4, r5 ; calculate difference
46 usub8 r9, r5, r4 ; calculate difference with reversed operands
52 usad8 r5, r6, lr ; calculate sum of negative differences
56 subs r8, r8, r5 ; subtract negative differences from sum
59 uxtb16 r5, r6 ; byte (two pixels) to halfwords
61 smlad r11, r5, r5, r11 ; dual signed multiply, add and accumulate (1)
65 ldr r5, [r2, #4] ; load 4 ref pixels
68 usub8 r6, r4, r5 ; calculate difference
70 usub8 r9, r5, r4 ; calculate difference with reversed operands
[all …]
|
| D | vp8_variance_halfpixvar16x16_h_armv6.asm | 41 ldr r5, [r2, #0] ; load 4 ref pixels
48 usub8 r6, r4, r5 ; calculate difference
51 usub8 r6, r5, r4 ; calculate difference with reversed operands
57 usad8 r5, r6, lr ; calculate sum of negative differences
61 subs r8, r8, r5 ; subtract negative differences from sum
64 uxtb16 r5, r6 ; byte (two pixels) to halfwords
66 smlad r11, r5, r5, r11 ; dual signed multiply, add and accumulate (1)
71 ldr r5, [r2, #4] ; load 4 ref pixels
80 usub8 r6, r4, r5 ; calculate difference
82 usub8 r6, r5, r4 ; calculate difference with reversed operands
[all …]
|
| D | vp8_variance_halfpixvar16x16_v_armv6.asm | 42 ldr r5, [r2, #0] ; load 4 ref pixels
49 usub8 r6, r4, r5 ; calculate difference
52 usub8 r6, r5, r4 ; calculate difference with reversed operands
58 usad8 r5, r6, lr ; calculate sum of negative differences
62 subs r8, r8, r5 ; subtract negative differences from sum
65 uxtb16 r5, r6 ; byte (two pixels) to halfwords
67 smlad r11, r5, r5, r11 ; dual signed multiply, add and accumulate (1)
72 ldr r5, [r2, #4] ; load 4 ref pixels
81 usub8 r6, r4, r5 ; calculate difference
83 usub8 r6, r5, r4 ; calculate difference with reversed operands
[all …]
|
| D | simpleloopfilter_v6.asm | 64 ldr r5, [src] ; q0
174 pkhbt r10, r5, r6, lsl #16
176 ;transpose r7, r8, r9, r10 to r3, r4, r5, r6
177 TRANSPOSE_MATRIX r7, r8, r9, r10, r3, r4, r5, r6
181 uqsub8 r9, r4, r5 ; p0 - q0
182 uqsub8 r10, r5, r4 ; q0 - p0
201 eor r5, r5, r2 ; q0 offset to convert to a signed value
204 qsub8 r6, r5, r4 ; q0 - p0
230 qsub8 r5, r5, r3 ; u = q0 - Filter1
232 eor r5, r5, r2 ; *oq0 = u^0x80
[all …]
|
| D | dc_only_idct_add_v6.asm | 35 uxtab16 r5, r0, r4 ; a1+2 | a1+0
39 usat16 r5, #8, r5
43 orr r5, r5, r4, lsl #8
46 str r5, [r3], r12
50 uxtab16 r5, r0, r4
54 usat16 r5, #8, r5
58 orr r5, r5, r4, lsl #8
60 str r5, [r3], r12
|
| D | iwalsh_v6.asm | 27 ldr r5, [r0, #12] ; [7 | 6]
44 qadd16 r11, r5, r7 ; b1 [7+11 | 6+10]
45 qsub16 r12, r5, r7 ; c1 [7-11 | 6-10]
49 qadd16 r5, r12, lr ; c1 + d1 [7 | 6]
57 qsubaddx r12, r4, r5 ; [c1|a1] [5-6 | 4+7]
58 qaddsubx lr, r4, r5 ; [b1|d1] [5+6 | 4-7]
64 qaddsubx r5, lr, r12 ; [a2|d2] [b1+a1 | d1-c1]
69 qadd16 r5, r5, r10 ; [a2+3|d2+3]
82 asr r12, r5, #19 ; [4]
87 sxth r5, r5
[all …]
|
| D | idct_v6.asm | 33 mov r5, #0x00004E00 ; cos
34 orr r5, r5, #0x0000007B ; cospi8sqrt2minus1
35 orr r5, r5, #1<<31 ; loop counter on top bit
42 smulbt r9, r5, r6 ; (ip[5] * cospi8sqrt2minus1) >> 16
43 smulbb r7, r5, r6 ; (ip[4] * cospi8sqrt2minus1) >> 16
47 smulbt r11, r5, r12 ; (ip[13] * cospi8sqrt2minus1) >> 16
53 smulbb r9, r5, r12 ; (ip[12] * cospi8sqrt2minus1) >> 16
56 subs r5, r5, #1<<31 ; i--
91 smulbt r9, r5, r6 ; (ip[5] * cospi8sqrt2minus1) >> 16
92 smulbt r7, r5, r0 ; (ip[1] * cospi8sqrt2minus1) >> 16
[all …]
|
| D | copymem16x16_v6.asm | 38 ldrb r5, [r0, #1]
46 strb r5, [r2, #1]
51 ldrb r5, [r0, #5]
58 strb r5, [r2, #5]
63 ldrb r5, [r0, #9]
68 strb r5, [r2, #9]
73 ldrb r5, [r0, #13]
80 strb r5, [r2, #13]
87 ldrneb r5, [r0, #1]
102 ldr r5, [r0, #4]
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vpx_scale/arm/neon/ |
| D | vp8_vpxyv12_extendframeborders_neon.asm | 37 sub r5, r1, #32 ; dest_ptr1 = src_ptr1 - Border
62 vst1.8 {q0, q1}, [r5], lr
64 vst1.8 {q4, q5}, [r5], lr
66 vst1.8 {q8, q9}, [r5], lr
68 vst1.8 {q12, q13}, [r5], lr
83 sub r5, r1, lr, asl #5 ; dest_ptr1 = src_ptr1 - (Border * plane_stride)
101 vst1.8 {q0, q1}, [r5]!
103 vst1.8 {q2, q3}, [r5]!
105 vst1.8 {q4, q5}, [r5]!
107 vst1.8 {q6, q7}, [r5]!
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp9/common/arm/neon/ |
| D | vp9_copy_neon.asm | 19 push {r4-r5, lr}
20 ldrd r4, r5, [sp, #28]
39 subs r5, r5, #1
41 pop {r4-r5, pc}
50 subs r5, r5, #2
52 pop {r4-r5, pc}
61 subs r5, r5, #2
63 pop {r4-r5, pc}
72 subs r5, r5, #2
74 pop {r4-r5, pc}
[all …]
|
| D | vp9_avg_neon.asm | 20 ldrd r4, r5, [sp, #32]
47 subs r5, r5, #1
66 subs r5, r5, #2
83 subs r5, r5, #2
99 subs r5, r5, #2
111 subs r5, r5, #2
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/armv6/ |
| D | vp8_short_fdct4x4_armv6.asm | 26 ldrd r4, r5, [r0] ; [i1 | i0] [i3 | i2]
32 ror r5, r5, #16 ; [i2 | i3]
34 qadd16 r6, r4, r5 ; [i1+i2 | i0+i3] = [b1 | a1] without shift
35 qsub16 r7, r4, r5 ; [i1-i2 | i0-i3] = [c1 | d1] without shift
44 smusd r5, r6, lr ; o2 = (i1+i2)*8 - (i0+i3)*8
52 pkhbt r6, r5, r7, lsl #4 ; [o3 | o2]
74 ldrd r4, r5, [r0] ; [i9 | i8] [i11 | i10]
82 ror r5, r5, #16 ; [i10 | i11]
84 qadd16 r6, r4, r5 ; [i9+i10 | i8+i11]=[b1 | a1] without shift
85 qsub16 r7, r4, r5 ; [i9-i10 | i8-i11]=[c1 | d1] without shift
[all …]
|
| D | vp8_mse16x16_armv6.asm | 39 ldr r5, [r0, #0x0] ; load 4 src pixels
44 usub8 r8, r5, r6 ; calculate difference
47 usub8 r9, r6, r5 ; calculate difference with reversed operands
52 usad8 r5, r7, lr ; calculate sum of positive differences
56 ldr r5, [r0, #0x4] ; load 4 src pixels
67 usub8 r8, r5, r6 ; calculate difference
69 usub8 r9, r6, r5 ; calculate difference with reversed operands
73 usad8 r5, r7, lr ; calculate sum of positive differences
76 ldr r5, [r0, #0x8] ; load 4 src pixels
86 usub8 r8, r5, r6 ; calculate difference
[all …]
|
| D | walsh_v6.asm | 27 ldrd r4, r5, [r0], r2
32 qadd16 r3, r4, r5 ; [d1|a1] [1+3 | 0+2]
33 qsub16 r4, r4, r5 ; [c1|b1] [1-3 | 0-2]
37 qadd16 r5, r6, r7 ; [d1|a1] [5+7 | 4+6]
61 lsls r2, r5, #16
62 smuad r12, r5, lr ; B0 = a1<<2 + d1<<2
105 smusd r5, r5, lr ; B3 = a1<<2 - d1<<2
107 add r7, r5, r9 ; d1_3 = B3 + D3
108 sub r5, r5, r9 ; c1_3 = B3 - D3
113 adds r9, r3, r5 ; b2 = b1_3 + c1_3
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/ppc/ |
| D | sad_altivec.asm | 56 lvsl v3, 0, r5 ;# only needs to be done once per block
60 lvx v1, 0, r5
61 lvx v2, r10, r5
63 add r5, r5, r6
76 lvx v1, 0, r5
77 lvx v2, r10, r5
88 add r5, r5, r6
93 lvx v1, 0, r5
95 lvx v2, r10, r5
100 add r5, r5, r6
[all …]
|
| D | variance_altivec.asm | 76 load_aligned_16 v5, r5, r10
80 add r5, r5, r6
109 load_aligned_16 v5, r5, r10
113 add r5, r5, r6
117 load_aligned_16 v0, r5, r10
121 add r5, r5, r6
152 ;# r5 unsigned char *ref_ptr
174 ;# r5 unsigned char *ref_ptr
195 ;# r5 unsigned char *ref_ptr
208 load_aligned_16 v5, r5, r10
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/neon/ |
| D | buildintrapredictorsmby_neon.asm | 41 ldr r5, [sp, #28] ; Left
51 adds r7, r4, r5
73 cmp r5, #0
81 ldrb r5, [r0], r2
86 add r12, r12, r5
91 ldrb r5, [r0], r2
96 add r12, r12, r5
101 ldrb r5, [r0], r2
106 add r12, r12, r5
111 ldrb r5, [r0], r2
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/neon/ |
| D | subtract_neon.asm | 30 ldr r5, [r0, #vp8_block_src_diff]
52 vst1.16 {d20}, [r5], r2 ;store diff
53 vst1.16 {d22}, [r5], r2
54 vst1.16 {d24}, [r5], r2
55 vst1.16 {d26}, [r5], r2
71 add r5, r0, #16 ; second diff pointer
93 vst1.16 {q9}, [r5], r6
95 vst1.16 {q11}, [r5], r6
97 vst1.16 {q13}, [r5], r6
99 vst1.16 {q15}, [r5], r6
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/ppc/ |
| D | encodemb_altivec.asm | 17 ;# r5 unsigned char *vsrc
74 lvsl v5, 0, r5 ;# permutate value for alignment
75 lvx v1, 0, r5 ;# src
78 add r5, r5, r7
86 lvsl v5, 0, r5 ;# permutate value for alignment
87 lvx v1, 0, r5 ;# src
89 add r5, r5, r7
114 ;# r5 unsigned char *pred
128 lvx v2, 0, r5 ;# pred
131 addi r5, r5, 16
|
| /hardware/samsung_slsi/exynos5/libswconverter/ |
| D | csc_tiled_to_linear_uv_deinterleave_neon.s | 62 @r5 i
81 mov r5, #0
83 mul r8, r11, r5 @ src_offset = tiled_width * i
88 mul r12, r12, r5
127 mul r8, r11, r5 @ src_offset = (tiled_width * i) + (j << 3)
131 mul r12, r12, r5
163 add r5, r5, #8
164 cmp r5, r9
173 mul r8, r11, r5 @ src_offset = (tiled_width * i) + (j << 3)
178 mul r12, r12, r5
[all …]
|
| D | csc_tiled_to_linear_uv_neon.s | 59 @r5 i
78 mov r5, #0
80 mul r8, r11, r5 @ src_offset = tiled_width * i
86 mul r12, r2, r5 @ temp1 = width * i + j;
110 mul r8, r11, r5 @ src_offset = (tiled_width * i) + (j << 3);
114 mul r12, r2, r5 @ temp1 = width * i + j;
141 add r5, r5, #8
142 cmp r5, r9
151 mul r8, r11, r5 @ src_offset = (tiled_width * i) + (j << 3)
155 mul r12, r2, r5 @ temp1 = width * i + j;
[all …]
|