Home
last modified time | relevance | path

Searched refs:d0 (Results 1 – 25 of 48) sorted by relevance

12

/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/neon/
Dvp8_subpixelvariance16x16_neon.asm59 vdup.8 d0, d31[0] ;first_pass filter (d0 d1)
70 vmull.u8 q7, d2, d0 ;(src_ptr[0] * Filter[0])
71 vmull.u8 q8, d3, d0
72 vmull.u8 q9, d5, d0
73 vmull.u8 q10, d6, d0
74 vmull.u8 q11, d8, d0
75 vmull.u8 q12, d9, d0
76 vmull.u8 q13, d11, d0
77 vmull.u8 q14, d12, d0
123 vmull.u8 q9, d2, d0 ;(src_ptr[0] * Filter[0])
[all …]
Dsad8_neon.asm28 vld1.8 {d0}, [r0], r1
34 vabdl.u8 q12, d0, d8
46 vld1.8 {d0}, [r0], r1
54 vabal.u8 q12, d0, d8
69 vadd.u32 d0, d0, d1
71 vmov.32 r0, d0[0]
85 vld1.8 {d0}, [r0], r1
91 vabdl.u8 q12, d0, d8
103 vld1.8 {d0}, [r0], r1
111 vabal.u8 q12, d0, d8
[all …]
Dvariance_neon.asm41 vsubl.u8 q11, d0, d4 ;calculate diff
73 vadd.s64 d0, d0, d1
76 ;vmov.32 r0, d0[0] ;this instruction costs a lot
84 vmull.s32 q5, d0, d0
87 vsub.u32 d0, d1, d10
89 vmov.32 r0, d0[0] ;return
114 vsubl.u8 q11, d0, d4 ;calculate diff
143 vadd.s64 d0, d0, d1
146 vmull.s32 q5, d0, d0
149 vsub.u32 d0, d1, d10
[all …]
Dvp8_subpixelvariance8x8_neon.asm44 vdup.8 d0, d31[0] ;first_pass filter (d0 d1)
49 vmull.u8 q6, d2, d0 ;(src_ptr[0] * Filter[0])
50 vmull.u8 q7, d4, d0
51 vmull.u8 q8, d6, d0
52 vmull.u8 q9, d8, d0
76 vmull.u8 q6, d2, d0 ;(src_ptr[0] * Filter[0])
77 vmull.u8 q7, d4, d0
78 vmull.u8 q8, d6, d0
79 vmull.u8 q9, d8, d0
80 vmull.u8 q10, d10, d0
[all …]
Diwalsh_neon.asm21 ; read in all four lines of values: d0->d3
25 vadd.s16 d4, d0, d3 ;a = [0] + [12]
27 vsub.s16 d5, d0, d3 ;d = [0] - [12]
33 vtrn.32 d0, d2 ;d0: 0 1 8 9
38 vtrn.16 d0, d1 ;d0: 0 4 8 12
45 vadd.s16 d4, d0, d3 ;a = [0] + [3]
47 vsub.s16 d5, d0, d3 ;d = [0] - [3]
64 vst1.i16 d0[0], [r1],r2
69 vst1.i16 d0[1], [r1],r2
74 vst1.i16 d0[2], [r1],r2
[all …]
Dvp8_subpixelvariance16x16s_neon.asm43 vld1.u8 {d0, d1, d2, d3}, [r0], r1 ;load src data
66 vsubl.u8 q4, d0, d22 ;diff
92 vmlal.s16 q9, d0, d0 ;sse
110 vadd.s64 d0, d0, d1
113 vmull.s32 q5, d0, d0
116 vsub.u32 d0, d1, d10
118 vmov.32 r0, d0[0] ;return
159 vsubl.u8 q11, d0, d2 ;diff
185 vmlal.s16 q9, d0, d0 ;sse
206 vadd.s64 d0, d0, d1
[all …]
Dsad16_neon.asm33 vabdl.u8 q12, d0, d8
58 vabal.u8 q12, d0, d8
83 vabal.u8 q12, d0, d8
108 vabal.u8 q12, d0, d8
131 vadd.u32 d0, d0, d1
133 vmov.32 r0, d0[0]
152 vabdl.u8 q12, d0, d8
176 vabal.u8 q12, d0, d8
199 vadd.u32 d0, d0, d1
201 vmov.32 r0, d0[0]
Dsixtappredict8x4_neon.asm62 vdup.8 d0, d24[0] ;first_pass filter (d0-d5)
79 vmull.u8 q7, d6, d0 ;(src_ptr[-2] * vp8_filter[0])
80 vmull.u8 q8, d8, d0
81 vmull.u8 q9, d10, d0
82 vmull.u8 q10, d12, d0
156 vmull.u8 q8, d6, d0 ;(src_ptr[-2] * vp8_filter[0])
157 vmull.u8 q9, d8, d0
158 vmull.u8 q10, d10, d0
159 vmull.u8 q11, d12, d0
160 vmull.u8 q12, d14, d0
[all …]
Didct_dequant_full_2x_neon.asm49 vld1.16 {d0}, [r1]
61 vqdmulh.s16 q6, q4, d0[2] ; sinpi8sqrt2
62 vqdmulh.s16 q7, q5, d0[2]
63 vqdmulh.s16 q8, q4, d0[0] ; cospi8sqrt2minus1
64 vqdmulh.s16 q9, q5, d0[0]
114 vqdmulh.s16 q8, q5, d0[2] ; sinpi8sqrt2
115 vqdmulh.s16 q9, q7, d0[2]
116 vqdmulh.s16 q10, q5, d0[0] ; cospi8sqrt2minus1
117 vqdmulh.s16 q11, q7, d0[0]
173 vqmovun.s16 d0, q4 ; lo
[all …]
Dsixtappredict8x8_neon.asm65 vdup.8 d0, d24[0] ;first_pass filter (d0-d5)
83 vmull.u8 q7, d6, d0 ;(src_ptr[-2] * vp8_filter[0])
84 vmull.u8 q8, d8, d0
85 vmull.u8 q9, d10, d0
86 vmull.u8 q10, d12, d0
169 vmull.u8 q8, d6, d0 ;(src_ptr[-2] * vp8_filter[0])
170 vmull.u8 q9, d8, d0
171 vmull.u8 q10, d10, d0
172 vmull.u8 q11, d12, d0
173 vmull.u8 q12, d14, d0
[all …]
Dsixtappredict16x16_neon.asm72 vdup.8 d0, d24[0] ;first_pass filter (d0-d5)
89 vmull.u8 q8, d6, d0 ;(src_ptr[-2] * vp8_filter[0])
90 vmull.u8 q9, d7, d0
91 vmull.u8 q10, d9, d0
92 vmull.u8 q11, d10, d0
93 vmull.u8 q12, d12, d0
94 vmull.u8 q13, d13, d0
211 vdup.8 d0, d14[0] ;second_pass filter parameters (d0-d5)
232 vmull.u8 q3, d18, d0 ;(src_ptr[-2] * vp8_filter[0])
233 vmull.u8 q4, d19, d0
[all …]
Dsixtappredict4x4_neon.asm61 vdup.8 d0, d24[0] ;first_pass filter (d0-d5)
95 vmlal.u8 q7, d6, d0 ;+(src_ptr[-2] * vp8_filter[0])
96 vmlal.u8 q8, d10, d0
162 vmlal.u8 q7, d6, d0 ;+(src_ptr[-2] * vp8_filter[0])
163 vmlal.u8 q8, d10, d0
164 vmlal.u8 q12, d22, d0 ;(src_ptr[-2] * vp8_filter[0])
224 vdup.8 d0, d14[0] ;second_pass filter parameters (d0-d5)
231 vmull.u8 q3, d27, d0 ;(src_ptr[-2] * vp8_filter[0])
232 vmull.u8 q4, d28, d0
276 vdup.8 d0, d24[0] ;first_pass filter (d0-d5)
[all …]
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/neon/
Dvp8_shortwalsh4x4_neon.asm25 vld1.16 {d0}, [r0@64], r2 ; load input
31 ;transpose d0, d1, d2, d3. Then, d0=ip[0], d1=ip[1], d2=ip[2], d3=ip[3]
32 vtrn.32 d0, d2
37 vtrn.16 d0, d1
40 vadd.s16 d4, d0, d2 ; ip[0] + ip[2]
43 vsub.s16 d7, d0, d2 ; ip[0] - ip[2]
51 vadd.s16 d0, d4, d5 ; a1 + d1
56 vsub.s16 d0, d0, d16 ; op[0] = a1 + d1 + (a1 != 0)
59 ;transpose d0, d1, d2, d3, Then, d0=ip[0], d1=ip[4], d2=ip[8], d3=ip[12]
61 vtrn.32 d0, d2
[all …]
Dshortfdct_neon.asm35 vld1.16 {d0}, [r0@64], r2
43 ; transpose d0=ip[0], d1=ip[1], d2=ip[2], d3=ip[3]
44 vtrn.32 d0, d2
47 vtrn.16 d0, d1
50 vadd.s16 d4, d0, d3 ; a1 = ip[0] + ip[3]
53 vsub.s16 d7, d0, d3 ; d1 = ip[0] - ip[3]
58 vadd.s16 d0, d4, d5 ; op[0] = a1 + b1
72 ; transpose d0=ip[0], d1=ip[4], d2=ip[8], d3=ip[12]
73 vtrn.32 d0, d2
75 vtrn.16 d0, d1
[all …]
Dvp8_mse16x16_neon.asm43 vsubl.u8 q11, d0, d4
69 vadd.u64 d0, d2, d3
71 vst1.32 {d0[0]}, [r12]
72 vmov.32 r0, d0[0]
85 vld1.8 {d0}, [r0], r1 ;Load up source and reference
94 vsubl.u8 q11, d0, d4
109 vadd.u64 d0, d2, d3
111 vmov.32 r0, d0[0]
Dfastquantizeb_neon.asm133 vmax.u16 d0, d0, d1
135 vmovl.u16 q0, d0
138 vmax.u32 d0, d0, d1
140 vpmax.u32 d0, d0, d0
146 vst1.8 {d0[0]}, [r4] ; store eob
Dsubtract_neon.asm36 vld1.8 {d0}, [r3], r6 ;load src
45 vsubl.u8 q10, d0, d1
83 vsubl.u8 q8, d0, d2
123 vld1.8 {d0}, [r1], r3 ;load usrc
140 vsubl.u8 q8, d0, d1
159 vld1.8 {d0}, [r2], r3 ;load vsrc
176 vsubl.u8 q8, d0, d1
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp9/common/arm/neon/
Dvp9_reconintra_neon.asm38 vld1.32 {d0[0]}, [r2]
39 vst1.32 {d0[0]}, [r0], r1
40 vst1.32 {d0[0]}, [r0], r1
41 vst1.32 {d0[0]}, [r0], r1
42 vst1.32 {d0[0]}, [r0], r1
55 vld1.8 {d0}, [r2]
56 vst1.8 {d0}, [r0], r1
57 vst1.8 {d0}, [r0], r1
58 vst1.8 {d0}, [r0], r1
59 vst1.8 {d0}, [r0], r1
[all …]
Dvp9_avg_neon.asm88 vld1.8 {d0}, [r0], r1
97 vst1.8 {d0}, [r2@64], r3
104 vld1.32 {d0[0]}, [r0], r1
105 vld1.32 {d0[1]}, [r0], r1
108 vrhadd.u8 d0, d0, d2
109 vst1.32 {d0[0]}, [r2@32], r3
110 vst1.32 {d0[1]}, [r2@32], r3
Dvp9_idct8x8_add_neon.asm26 vdup.16 d0, r3 ; duplicate cospi_28_64
32 vmull.s16 q2, d18, d0
33 vmull.s16 q3, d19, d0
64 vmlal.s16 q2, d30, d0
65 vmlal.s16 q3, d31, d0
76 vdup.16 d0, r7 ; duplicate cospi_16_64
83 vmull.s16 q2, d16, d0
84 vmull.s16 q3, d17, d0
87 vmull.s16 q13, d16, d0
88 vmull.s16 q15, d17, d0
[all …]
Dvp9_iht8x8_add_neon.asm125 vdup.16 d0, r3 ; duplicate cospi_28_64
131 vmull.s16 q2, d18, d0
132 vmull.s16 q3, d19, d0
163 vmlal.s16 q2, d30, d0
164 vmlal.s16 q3, d31, d0
175 vdup.16 d0, r7 ; duplicate cospi_16_64
182 vmull.s16 q2, d16, d0
183 vmull.s16 q3, d17, d0
186 vmull.s16 q13, d16, d0
187 vmull.s16 q15, d17, d0
[all …]
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/
Dreconinter.c285 BLOCKD *d0 = &x->block[i]; in vp8_build_inter4x4_predictors_mbuv() local
288 if (d0->bmi.mv.as_int == d1->bmi.mv.as_int) in vp8_build_inter4x4_predictors_mbuv()
289 build_inter_predictors2b(x, d0, d0->predictor, 8, base_pre, pre_stride); in vp8_build_inter4x4_predictors_mbuv()
292 vp8_build_inter_predictors_b(d0, 8, base_pre, pre_stride, x->subpixel_predict); in vp8_build_inter4x4_predictors_mbuv()
300 BLOCKD *d0 = &x->block[i]; in vp8_build_inter4x4_predictors_mbuv() local
303 if (d0->bmi.mv.as_int == d1->bmi.mv.as_int) in vp8_build_inter4x4_predictors_mbuv()
304 build_inter_predictors2b(x, d0, d0->predictor, 8, base_pre, pre_stride); in vp8_build_inter4x4_predictors_mbuv()
307 vp8_build_inter_predictors_b(d0, 8, base_pre, pre_stride, x->subpixel_predict); in vp8_build_inter4x4_predictors_mbuv()
471 BLOCKD *d0 = &x->block[i]; in build_inter4x4_predictors_mb() local
483 if (d0->bmi.mv.as_int == d1->bmi.mv.as_int) in build_inter4x4_predictors_mb()
[all …]
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/armv6/
Dvp8_subtract_armv6.asm52 usub16 r4, r4, r5 ; [d2 | d0]
57 pkhbt r0, r4, r5, lsl #16 ; [d1 | d0]
98 usub16 r6, r8, r9 ; [d2 | d0] (A)
104 pkhbt r8, r6, r7, lsl #16 ; [d1 | d0] (A)
115 usub16 r6, r8, r9 ; [d2 | d0] (B)
121 pkhbt r8, r6, r7, lsl #16 ; [d1 | d0] (B)
143 usub16 r6, r8, r9 ; [d2 | d0] (A)
149 pkhbt r8, r6, r7, lsl #16 ; [d1 | d0] (A)
160 usub16 r6, r8, r9 ; [d2 | d0] (B)
166 pkhbt r8, r6, r7, lsl #16 ; [d1 | d0] (B)
[all …]
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vpx_scale/arm/neon/
Dvp8_vpxyv12_extendframeborders_neon.asm42 vld1.8 {d0[], d1[]}, [r1], lr
148 vld1.8 {d0[], d1[]}, [r1], lr
274 vld1.8 {d0}, [r1]!
282 vst1.8 {d0}, [r5], lr
284 vst1.8 {d0}, [r5], lr
286 vst1.8 {d0}, [r5], lr
288 vst1.8 {d0}, [r5], lr
290 vst1.8 {d0}, [r5], lr
292 vst1.8 {d0}, [r5], lr
294 vst1.8 {d0}, [r5], lr
[all …]
Dvp8_vpxyv12_copysrcframe_func_neon.asm77 vld1.8 {d0}, [r2]!
81 vst1.8 {d0}, [r3]!
126 vld1.8 {d0}, [r2]!
129 vst1.8 {d0}, [r3]!
183 vld1.8 {d0}, [r2]!
187 vst1.8 {d0}, [r3]!
228 vld1.8 {d0}, [r2]!
231 vst1.8 {d0}, [r3]!

12