| /hardware/samsung_slsi/exynos5/libswconverter/ |
| D | csc_tiled_to_linear_uv_deinterleave_neon.s | 69 @r12 temp1
72 stmfd sp!, {r4-r12,r14} @ backup registers
87 mov r12, r3, asr #1 @ temp1 = (width >> 1) * i + (j >> 1)
88 mul r12, r12, r5
92 add r12, r12, r6, asr #1
94 add r7, r0, r12 @ dst_offset = u_dst + temp1
109 add r7, r1, r12 @ dst_offset = v_dst + temp1
130 mov r12, r3, asr #1 @ temp1 = (width >> 1) * i + (j >> 1)
131 mul r12, r12, r5
132 add r12, r12, r6, asr #1
[all …]
|
| D | csc_interleave_memcpy_neon.s | 62 @r12 src2_addr
65 stmfd sp!, {r8-r12,r14} @ backup registers
69 mov r12, r2
84 vld1.8 {q1}, [r12]!
85 vld1.8 {q3}, [r12]!
86 vld1.8 {q5}, [r12]!
87 vld1.8 {q7}, [r12]!
88 vld1.8 {q9}, [r12]!
89 vld1.8 {q11}, [r12]!
90 vld1.8 {q13}, [r12]!
[all …]
|
| D | csc_tiled_to_linear_uv_neon.s | 66 @r12 temp1
69 stmfd sp!, {r4-r12,r14} @ backup registers
86 mul r12, r2, r5 @ temp1 = width * i + j;
88 add r12, r12, r6
91 add r7, r0, r12 @ dst_offset = y_dst + temp1
114 mul r12, r2, r5 @ temp1 = width * i + j;
115 add r12, r12, r6
116 add r7, r0, r12 @ r7 = y_dst + temp1
155 mul r12, r2, r5 @ temp1 = width * i + j;
156 add r12, r12, r6
[all …]
|
| D | csc_tiled_to_linear_y_neon.s | 67 @r12 temp1
70 stmfd sp!, {r4-r12,r14} @ backup registers
91 mul r12, r2, r5 @ temp1 = width * i + j;
94 add r12, r12, r6
96 add r7, r0, r12 @ dst_offset = y_dst + temp1
129 mul r12, r2, r5 @ temp1 = width * i + j;
130 add r12, r12, r6
131 add r7, r0, r12 @ r7 = y_dst + temp1
170 mul r12, r2, r5 @ temp1 = width * i + j;
171 add r12, r12, r6
[all …]
|
| D | csc_ARGB8888_to_YUV420SP_NEON.s | 21 @r12 temp1, nHeight
27 stmfd sp!, {r4-r12,r14} @ backup registers
28 ldr r12, [sp, #40] @ load nHeight
60 stmfd sp!, {r12} @ backup registers
195 ldr r12, [r2], #4 @loadRGB int
198 mov r14, r12 @copy to r10
210 mov r14, r12,lsl #8 @copy to r10
222 mov r14, r12,lsl #16 @copy to r10
247 mov r12,r7
258 sub r12, r12, r11
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vpx_scale/arm/neon/ |
| D | vp8_vpxyv12_copysrcframe_func_neon.asm | 50 mov r12, r5
61 sub r12, r12, #128
62 cmp r12, #128
73 cmp r12, #0
79 sub r12, r12, #8
80 cmp r12, #8
85 cmp r12, #0
90 subs r12, r12, #1
107 mov r12, r5
114 sub r12, r12, #128
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/neon/ |
| D | buildintrapredictorsmby_neon.asm | 44 mov r12, #128
45 vdup.u8 q0, r12
48 mov r12, #0
67 add r12, r4, r6
84 add r12, r12, r3
85 add r12, r12, r4
86 add r12, r12, r5
87 add r12, r12, r6
94 add r12, r12, r3
95 add r12, r12, r4
[all …]
|
| D | mbloopfilter_neon.asm | 32 ldr r12, [sp, #4] ; load thresh
34 vdup.u8 q2, r12 ; thresh
35 add r12, r0, r1, lsr #1 ; move src pointer up by 1 line
38 vld1.u8 {q4}, [r12@128], r1 ; p2
40 vld1.u8 {q6}, [r12@128], r1 ; p0
42 vld1.u8 {q8}, [r12@128], r1 ; q1
44 vld1.u8 {q10}, [r12@128], r1 ; q3
48 sub r12, r12, r1, lsl #2
49 add r0, r12, r1, lsr #1
51 vst1.u8 {q4}, [r12@128],r1 ; store op2
[all …]
|
| D | loopfiltersimpleverticaledge_neon.asm | 26 add r12, r1, r1
29 vld4.8 {d6[0], d7[0], d8[0], d9[0]}, [r0], r12
30 vld4.8 {d6[1], d7[1], d8[1], d9[1]}, [r3], r12
31 vld4.8 {d6[2], d7[2], d8[2], d9[2]}, [r0], r12
32 vld4.8 {d6[3], d7[3], d8[3], d9[3]}, [r3], r12
33 vld4.8 {d6[4], d7[4], d8[4], d9[4]}, [r0], r12
34 vld4.8 {d6[5], d7[5], d8[5], d9[5]}, [r3], r12
35 vld4.8 {d6[6], d7[6], d8[6], d9[6]}, [r0], r12
36 vld4.8 {d6[7], d7[7], d8[7], d9[7]}, [r3], r12
38 vld4.8 {d10[0], d11[0], d12[0], d13[0]}, [r0], r12
[all …]
|
| D | loopfilter_neon.asm | 31 add r12, r2, r1
37 vld1.u8 {q4}, [r12@128], r1 ; p2
39 vld1.u8 {q6}, [r12@128], r1 ; p0
41 vld1.u8 {q8}, [r12@128], r1 ; q1
43 vld1.u8 {q10}, [r12@128] ; q3
46 sub r12, r12, r1, lsl #1
51 vst1.u8 {q6}, [r12@128], r1 ; store op0
53 vst1.u8 {q8}, [r12@128], r1 ; store oq1
69 ldr r12, [sp, #4] ; load thresh
71 vdup.u8 q2, r12 ; duplicate thresh
[all …]
|
| D | idct_dequant_0_2x_neon.asm | 26 add r12, r2, #4
28 vld1.32 {d8[0]}, [r12], r3
30 vld1.32 {d8[1]}, [r12], r3
32 vld1.32 {d10[0]}, [r12], r3
34 vld1.32 {d10[1]}, [r12], r3
36 ldrh r12, [r0] ; lo q
42 sxth r12, r12 ; lo
43 mul r0, r12, r1
|
| D | variance_neon.asm | 33 mov r12, #8
53 subs r12, r12, #1
70 ldr r12, [sp] ;load *sse from stack
79 ;str r1, [r12]
85 vst1.32 {d1[0]}, [r12] ;store sse
106 mov r12, #4
123 subs r12, r12, #1
140 ldr r12, [sp] ;load *sse from stack
147 vst1.32 {d1[0]}, [r12] ;store sse
169 mov r12, #8
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/armv6/ |
| D | loopfilter_v6.asm | 88 ldr r12, [src], pstep ; p0
99 uqsub8 r6, r11, r12 ; p1 - p0
101 uqsub8 r7, r12, r11 ; p0 - p1
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
122 uqsub8 r12, r12, r4 ; compare to flimit
125 orr lr, lr, r12
[all …]
|
| D | dequant_idct_v6.asm | 29 mov r12, #4
44 subs r12, r12, #1
63 ldr r12, [r0, #24]
70 smulwt r11, r3, r12
73 smulwt r7, r4, r12
74 smulwb r9, r3, r12
75 smulwb r10, r4, r12
80 uadd16 r7, r12, r9
104 smulwt r12, r4, r6
109 pkhbt r12, r10, r12, lsl #16
[all …]
|
| D | intra4x4_predict_v6.asm | 34 push {r4-r12, lr}
38 pop {r4-r12, pc} ; default
57 usad8 r12, r8, r9
64 add r4, r4, r12
67 mov r12, r4, asr #3 ; (expected_dc + 4) >> 3
69 add r12, r12, r12, lsl #8
71 add r12, r12, r12, lsl #16
74 str r12, [r3], r0
75 str r12, [r3], r0
76 str r12, [r3], r0
[all …]
|
| D | iwalsh_v6.asm | 22 stmdb sp!, {r4 - r12, lr}
35 qsub16 r12, r4, r6 ; c1 [5-9 | 4-8]
39 qadd16 r4, r12, lr ; c1 + d1 [5 | 4]
41 qsub16 r8, lr, r12 ; d1 - c1 [13 | 12]
45 qsub16 r12, r5, r7 ; c1 [7-11 | 6-10]
49 qadd16 r5, r12, lr ; c1 + d1 [7 | 6]
51 qsub16 r9, lr, r12 ; d1 - c1 [15 | 14]
57 qsubaddx r12, r4, r5 ; [c1|a1] [5-6 | 4+7]
63 qaddsubx r4, r12, lr ; [b2|c2] [c1+d1 | a1-b1]
64 qaddsubx r5, lr, r12 ; [a2|d2] [b1+a1 | d1-c1]
[all …]
|
| D | copymem8x4_v6.asm | 39 mov r12, #4
48 subs r12, r12, #1
83 mov r12, #4
86 subs r12, r12, #1
108 mov r12, #4
115 subs r12, r12, #1
|
| D | copymem8x8_v6.asm | 39 mov r12, #8
48 subs r12, r12, #1
83 mov r12, #8
86 subs r12, r12, #1
108 mov r12, #8
115 subs r12, r12, #1
|
| D | copymem16x16_v6.asm | 42 mov r12, #16
55 subs r12, r12, #1
106 mov r12, #16
109 subs r12, r12, #1
137 mov r12, #16
147 subs r12, r12, #1
165 mov r12, #16
172 subs r12, r12, #1
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp9/common/arm/neon/ |
| D | vp9_idct16x16_1_add_neon.asm | 29 mov r12, #0x2d00
30 add r12, #0x41
33 mul r0, r0, r12 ; input[0] * cospi_16_64
38 mul r0, r0, r12 ; out * cospi_16_64
39 mov r12, r1 ; save dest
69 vst1.64 {d2}, [r12], r0
70 vst1.64 {d3}, [r12], r2
71 vst1.64 {d30}, [r12], r0
72 vst1.64 {d31}, [r12], r2
82 vst1.64 {d2}, [r12], r0
[all …]
|
| D | vp9_idct8x8_1_add_neon.asm | 29 mov r12, #0x2d00
30 add r12, #0x41
33 mul r0, r0, r12 ; input[0] * cospi_16_64
38 mul r0, r0, r12 ; out * cospi_16_64
39 mov r12, r1 ; save dest
67 vst1.64 {d2}, [r12], r2
68 vst1.64 {d3}, [r12], r2
69 vst1.64 {d30}, [r12], r2
70 vst1.64 {d31}, [r12], r2
80 vst1.64 {d2}, [r12], r2
[all …]
|
| D | vp9_idct4x4_1_add_neon.asm | 29 mov r12, #0x2d00
30 add r12, #0x41
33 mul r0, r0, r12 ; input[0] * cospi_16_64
38 mul r0, r0, r12 ; out * cospi_16_64
39 mov r12, r1 ; save dest
60 vst1.32 {d6[0]}, [r12], r2
61 vst1.32 {d6[1]}, [r12], r2
62 vst1.32 {d7[0]}, [r12], r2
63 vst1.32 {d7[1]}, [r12]
|
| D | vp9_dc_only_idct_add_neon.asm | 30 mov r12, #0x2d00
31 add r12, #0x41
34 mul r0, r0, r12 ; input_dc * cospi_16_64
39 mul r0, r0, r12 ; out * cospi_16_64
48 ldr r12, [sp] ; load stride
61 vst1.32 {d2[0]}, [r2], r12
62 vst1.32 {d2[1]}, [r2], r12
63 vst1.32 {d4[0]}, [r2], r12
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/neon/ |
| D | subtract_neon.asm | 68 mov r12, #4
101 subs r12, r12, #1
119 mov r12, #32 ; "diff" stride x2
149 vst1.16 {q8}, [r0], r12 ;store diff
150 vst1.16 {q9}, [r7], r12
151 vst1.16 {q10}, [r0], r12
152 vst1.16 {q11}, [r7], r12
153 vst1.16 {q12}, [r0], r12
154 vst1.16 {q13}, [r7], r12
155 vst1.16 {q14}, [r0], r12
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/armv6/ |
| D | vp8_short_fdct4x4_armv6.asm | 21 stmfd sp!, {r4 - r12, lr}
30 ldr r12, c0x22a453a0 ; [2217*4 | 5352*4]
46 smlad r6, r7, r12, r11 ; o1 = (c1 * 2217 + d1 * 5352 + 14500)
47 smlsdx r7, r7, r12, r10 ; o3 = (d1 * 2217 - c1 * 5352 + 7500)
71 smlad r6, r7, r12, r11 ; o5 = (c1 * 2217 + d1 * 5352 + 14500)
72 smlsdx r7, r7, r12, r10 ; o7 = (d1 * 2217 - c1 * 5352 + 7500)
96 smlad r6, r7, r12, r11 ; o9 = (c1 * 2217 + d1 * 5352 + 14500)
97 smlsdx r7, r7, r12, r10 ; o11 = (d1 * 2217 - c1 * 5352 + 7500)
119 smlad r6, r7, r12, r11 ; o13 = (c1 * 2217 + d1 * 5352 + 14500)
120 smlsdx r7, r7, r12, r10 ; o15 = (d1 * 2217 - c1 * 5352 + 7500)
[all …]
|