| /system/core/libpixelflinger/ |
| col32cb16blend_neon.S | 63 vaddl.u8 q3, d6, d7 // add top bit into alpha
64 vsub.u16 q3, q15, q3 // invert alpha
83 vmul.u16 q8, q8, q3 // multiply dst red0 by src alpha
85 vmul.u16 q9, q9, q3 // multiply dst green0 by src alpha
87 vmul.u16 q10, q10, q3 // multiply dst blue0 by src alpha
91 vmul.u16 q13, q13, q3 // multiply dst red1 by src alpha
93 vmul.u16 q12, q12, q3 // multiply dst green1 by src alpha
95 vmul.u16 q11, q11, q3 // multiply dst blue1 by src alpha
|
| /external/libvpx/vp8/decoder/arm/neon/ |
| dequant_idct_neon.asm | 28 vld1.16 {q3, q4}, [r0]
40 vmul.i16 q1, q3, q5 ;input for short_idct4x4llm_neon
47 vqdmulh.s16 q3, q2, d0[2]
53 vshr.s16 q3, q3, #1
56 vqadd.s16 q3, q3, q2
76 vqdmulh.s16 q3, q2, d0[2]
84 vshr.s16 q3, q3, #
[all...] |
| idct_dequant_dc_full_2x_neon.asm | 28 vld1.16 {q2, q3}, [r0] ; l q
51 vmul.i16 q3, q3, q1
61 ; q2: l0r0 q3: l8r8
76 vqadd.s16 q10, q2, q3 ; a1 = 0 + 8
77 vqsub.s16 q11, q2, q3 ; b1 = 0 - 8
98 vqadd.s16 q3, q4, q7
104 vqadd.s16 q4, q10, q3
107 vqsub.s16 q7, q10, q3
130 vqsub.s16 q3, q4, q6 ; b1 = 0 -
[all...] |
| idct_dequant_full_2x_neon.asm | 28 vld1.16 {q2, q3}, [r0] ; l q
47 vmul.i16 q3, q3, q1
53 ; q2: l0r0 q3: l8r8
68 vqadd.s16 q10, q2, q3 ; a1 = 0 + 8
69 vqsub.s16 q11, q2, q3 ; b1 = 0 - 8
90 vqadd.s16 q3, q4, q7
96 vqadd.s16 q4, q10, q3
99 vqsub.s16 q7, q10, q3
122 vqsub.s16 q3, q4, q6 ; b1 = 0 -
[all...] |
| /external/libvpx/vp8/common/arm/neon/ |
| shortidct4x4llm_neon.asm | 40 vqdmulh.s16 q3, q2, d0[2]
46 vshr.s16 q3, q3, #1
49 vqadd.s16 q3, q3, q2 ;modify since sinpi8sqrt2 > 65536/2 (negtive number)
72 vqdmulh.s16 q3, q2, d0[2]
78 vshr.s16 q3, q3, #1
81 vqadd.s16 q3, q3, q2 ;modify since sinpi8sqrt2 > 65536/2 (negtive number
[all...] |
| buildintrapredictorsmby_neon.asm | 61 vpaddl.u16 q3, q2
62 vpaddl.u32 q4, q3
181 vdup.u8 q3, r6
185 vst1.u8 {q3}, [r1]!
194 vdup.u8 q3, r6
198 vst1.u8 {q3}, [r1]!
208 vdup.u8 q3, r6
212 vst1.u8 {q3}, [r1]!
221 vdup.u8 q3, r6
225 vst1.u8 {q3}, [r1]
[all...] |
| recon16x16mb_neon.asm | 38 vmovl.u8 q3, d27
53 vadd.s16 q3, q3, q11
62 vqmovun.s16 d3, q3
74 vst1.u8 {q3}, [r2], r3
89 vmovl.u8 q3, d27
100 vadd.s16 q3, q3, q11
109 vqmovun.s16 d3, q3
|
| loopfiltersimplehorizontaledge_neon.asm | 61 vsubl.s8 q3, d15, d13
67 vadd.s16 q12, q3, q3
72 vadd.s16 q3, q3, q12
75 vaddw.s8 q3, q3, d9
79 vqmovn.s16 d9, q3
|
| mbloopfilter_neon.asm | 41 vld1.u8 {q3}, [r0], r1 ; p3
50 vld1.u8 {q10}, [r0], r1 ; q3
103 vld1.u8 {d20}, [r0], r1 ; q3
104 vld1.u8 {d21}, [r3], r1 ; q3
168 vtrn.32 q3, q7
173 vtrn.16 q3, q5
178 vtrn.8 q3, q4
186 vst1.u8 {q3}, [r12]!
197 vld1.u8 {q3}, [sp]!
201 vtrn.32 q3, q
[all...] |
| sixtappredict4x4_neon.asm | 49 vld1.u8 {q3}, [r0], r1 ;load first 4-line src data
77 vmov q4, q3 ;keep original src data in q4 q6
89 vshr.u64 q3, q4, #32 ;construct src_ptr[2]
103 vshr.u64 q3, q4, #24 ;construct src_ptr[1]
113 vld1.u8 {q3}, [r0], r1 ;load rest 5-line src data
143 vmov q4, q3 ;keep original src data in q4 q6
157 vshr.u64 q3, q4, #32 ;construct src_ptr[2]
177 vshr.u64 q3, q4, #24 ;construct src_ptr[1]
220 vmull.u8 q3, d27, d0 ;(src_ptr[-2] * vp8_filter[0])
226 vmlsl.u8 q3, d29, d4 ;-(src_ptr[2] * vp8_filter[4]
[all...] |
| loopfilter_neon.asm | 43 vld1.u8 {q3}, [r2], r1 ; p3
50 vld1.u8 {q10}, [r2] ; q3
89 vld1.u8 {d20}, [r3] ; q3
101 vld1.u8 {d21}, [r12] ; q3
162 vtrn.32 q3, q7
167 vtrn.16 q3, q5
172 vtrn.8 q3, q4
246 vtrn.32 q3, q7
251 vtrn.16 q3, q5
256 vtrn.8 q3, q
[all...] |
| sixtappredict8x8_neon.asm | 59 vld1.u8 {q3}, [r0], r1 ;load src data
122 vmull.u8 q3, d28, d3 ;(src_ptr[1] * vp8_filter[3])
129 vqadd.s16 q7, q3 ;sum of all (src_data*filter_parameters)
134 vld1.u8 {q3}, [r0], r1 ;load src data
152 ;vld1.u8 {q3}, [r0], r1 ;load src data
218 vmull.u8 q3, d27, d3 ;(src_ptr[1] * vp8_filter[3])
224 vqadd.s16 q8, q3 ;sum of all (src_data*filter_parameters)
261 vmull.u8 q3, d18, d0 ;(src_ptr[-2] * vp8_filter[0])
266 vmlsl.u8 q3, d19, d1 ;-(src_ptr[-1] * vp8_filter[1])
271 vmlsl.u8 q3, d22, d4 ;-(src_ptr[2] * vp8_filter[4]
[all...] |
| sixtappredict8x4_neon.asm | 56 vld1.u8 {q3}, [r0], r1 ;load src data
118 vmull.u8 q3, d28, d3 ;(src_ptr[1] * vp8_filter[3])
123 vqadd.s16 q7, q3 ;sum of all (src_data*filter_parameters)
128 vld1.u8 {q3}, [r0], r1 ;load src data
205 vmull.u8 q3, d27, d3 ;(src_ptr[1] * vp8_filter[3])
211 vqadd.s16 q8, q3 ;sum of all (src_data*filter_parameters)
243 vmull.u8 q3, d22, d0 ;(src_ptr[-2] * vp8_filter[0])
248 vmlsl.u8 q3, d23, d1 ;-(src_ptr[-1] * vp8_filter[1])
253 vmlsl.u8 q3, d26, d4 ;-(src_ptr[2] * vp8_filter[4])
258 vmlal.u8 q3, d24, d2 ;(src_ptr[0] * vp8_filter[2]
[all...] |
| bilinearpredict8x4_neon.asm | 42 vld1.u8 {q3}, [r0], r1
87 vmull.u8 q3, d24, d0
92 vmlal.u8 q3, d25, d1
99 vqrshrn.u16 d4, q3, #7
|
| sixtappredict16x16_neon.asm | 167 vmull.u8 q3, d6, d3
173 vqadd.s16 q13, q3
221 vmull.u8 q3, d18, d0 ;(src_ptr[-2] * vp8_filter[0])
226 vmlsl.u8 q3, d19, d1 ;-(src_ptr[-1] * vp8_filter[1])
231 vmlsl.u8 q3, d22, d4 ;-(src_ptr[2] * vp8_filter[4])
236 vmlal.u8 q3, d20, d2 ;(src_ptr[0] * vp8_filter[2])
241 vmlal.u8 q3, d23, d5 ;(src_ptr[3] * vp8_filter[5])
253 vqadd.s16 q7, q3 ;sum of all (src_data*filter_parameters)
371 vst1.u8 {q3}, [r4], r5 ;store result
411 vmull.u8 q3, d18, d0 ;(src_ptr[-2] * vp8_filter[0]
[all...] |
| bilinearpredict16x16_neon.asm | 140 vmull.u8 q3, d14, d0
147 vmlal.u8 q3, d14, d1
163 vqrshrn.u16 d18, q3, #7
189 vmull.u8 q3, d24, d0
200 vmlal.u8 q3, d26, d1
211 vqrshrn.u16 d4, q3, #7
313 vmull.u8 q3, d24, d0
324 vmlal.u8 q3, d26, d1
333 vqrshrn.u16 d4, q3, #7
|
| /external/libvpx/vp8/encoder/arm/neon/ |
| vp8_subpixelvariance16x16s_neon.asm | 56 vext.8 q3, q2, q3, #1
62 vrhadd.u8 q1, q2, q3
64 vrhadd.u8 q3, q6, q7
73 vsubl.u8 q3, d7, d29
100 vpadal.s16 q8, q3
148 vld1.8 {q3}, [r2], r3
166 vsubl.u8 q3, d13, d15
196 vpadal.s16 q8, q3
254 vext.8 q3, q2, q3, #1 ;construct src_ptr[1
[all...] |
| vp8_memcpy_neon.asm | 32 vld1.8 {q2, q3}, [r1]!
35 vst1.8 {q2, q3}, [r0]!
|
| fastfdct8x4_neon.asm | 34 vld1.16 {q3}, [r0], r2
147 vclt.s16 q7, q3, #0
152 vsub.s16 q3, q3, q7
157 vshr.s16 q3, q3, #1
161 vst1.16 {q3, q4}, [r1]
|
| fastquantizeb_neon.asm | 53 vshr.s16 q3, q1, #15
74 veor.s16 q5, q3
79 vsub.s16 q5, q3
|
| fastfdct4x4_neon.asm | 49 vshl.i16 q3, q3, #1 ; a1, b1
99 vclt.s16 q3, q1, #0
102 vsub.s16 q1, q1, q3
|
| vp8_subpixelvariance16x16_neon.asm | 143 vmull.u8 q3, d14, d0
150 vmlal.u8 q3, d14, d1
166 vqrshrn.u16 d18, q3, #7
195 vmull.u8 q3, d24, d0
206 vmlal.u8 q3, d26, d1
217 vqrshrn.u16 d4, q3, #7
322 vmull.u8 q3, d24, d0
333 vmlal.u8 q3, d26, d1
342 vqrshrn.u16 d4, q3, #7
374 vld1.8 {q3}, [r4], r
[all...] |
| /external/libvpx/vpx_scale/arm/neon/ |
| vp8_vpxyv12_copysrcframe_func_neon.asm | 54 vld1.8 {q2, q3}, [r2]!
64 vst1.8 {q2, q3}, [r3]!
110 vld1.8 {q2, q3}, [r2]!
116 vst1.8 {q2, q3}, [r3]!
168 vld1.8 {q2, q3}, [r2]!
174 vst1.8 {q2, q3}, [r3]!
216 vld1.8 {q2, q3}, [r2]!
220 vst1.8 {q2, q3}, [r3]!
|
| vp8_vpxyv12_copyframe_func_neon.asm | 60 vld1.8 {q2, q3}, [r8]!
71 vst1.8 {q2, q3}, [r9]!
118 vld1.8 {q2, q3}, [r8]!
125 vst1.8 {q2, q3}, [r9]!
|
| vp8_vpxyv12_copyframeyonly_neon.asm | 50 vld1.8 {q2, q3}, [r8]!
61 vst1.8 {q2, q3}, [r9]!
120 vmov q3, q2
131 vst1.8 {q2, q3}, [r6], lr
155 vld1.8 {q2, q3}, [r1]!
169 vst1.8 {q2, q3}, [r5]!
285 vld1.8 {q2, q3}, [r1]!
299 vst1.8 {q2, q3}, [r5]!
427 vld1.8 {q2, q3}, [r8]!
438 vst1.8 {q2, q3}, [r9]
[all...] |
