| /external/libhevc/common/arm/ |
| ihevc_itrans_recon_32x32.s | 211 vmull.s16 q12,d8,d0[1] @// y1 * cos1(part of b0)
212 vmull.s16 q13,d8,d0[3] @// y1 * cos3(part of b1)
213 vmull.s16 q14,d8,d1[1] @// y1 * sin3(part of b2)
214 vmull.s16 q15,d8,d1[3] @// y1 * sin1(part of b3)
216 vmlal.s16 q12,d9,d0[3] @// y1 * cos1 + y3 * cos3(part of b0)
217 vmlal.s16 q13,d9,d2[1] @// y1 * cos3 - y3 * sin1(part of b1)
218 vmlal.s16 q14,d9,d3[3] @// y1 * sin3 - y3 * cos1(part of b2)
219 vmlal.s16 q15,d9,d5[1] @// y1 * sin1 - y3 * sin3(part of b3)
225 vmull.s16 q10,d10,d0[0]
226 vmlal.s16 q10,d11,d0[2
[all...] |
| ihevc_itrans_recon_16x16.s | 242 vmull.s16 q12,d6,d0[1] @// y1 * cos1(part of b0)
243 vmull.s16 q13,d6,d0[3] @// y1 * cos3(part of b1)
244 vmull.s16 q14,d6,d1[1] @// y1 * sin3(part of b2)
245 vmull.s16 q15,d6,d1[3] @// y1 * sin1(part of b3)
247 vmlal.s16 q12,d7,d0[3] @// y1 * cos1 + y3 * cos3(part of b0)
248 vmlal.s16 q13,d7,d2[1] @// y1 * cos3 - y3 * sin1(part of b1)
249 vmlal.s16 q14,d7,d3[3] @// y1 * sin3 - y3 * cos1(part of b2)
250 vmlsl.s16 q15,d7,d2[3] @// y1 * sin1 - y3 * sin3(part of b3)
257 vmull.s16 q6,d10,d0[0]
258 vmlal.s16 q6,d11,d0[2
[all...] |
| ihevc_inter_pred_filters_luma_vert_w16inp.s | 148 vmull.s16 q4,d1,d23 @mul_res1 = vmull_u8(src_tmp2, coeffabs_1)@
150 vmlal.s16 q4,d0,d22 @mul_res1 = vmlal_u8(mul_res1, src_tmp1, coeffabs_0)@
152 vmlal.s16 q4,d2,d24 @mul_res1 = vmlal_u8(mul_res1, src_tmp3, coeffabs_2)@
154 vmlal.s16 q4,d3,d25 @mul_res1 = vmlal_u8(mul_res1, src_tmp4, coeffabs_3)@
156 vmlal.s16 q4,d4,d26 @mul_res1 = vmlal_u8(mul_res1, src_tmp1, coeffabs_4)@
158 vmlal.s16 q4,d5,d27 @mul_res1 = vmlal_u8(mul_res1, src_tmp2, coeffabs_5)@
159 vmlal.s16 q4,d6,d28 @mul_res1 = vmlal_u8(mul_res1, src_tmp3, coeffabs_6)@
160 vmlal.s16 q4,d7,d29 @mul_res1 = vmlal_u8(mul_res1, src_tmp4, coeffabs_7)@
164 vmull.s16 q5,d2,d23 @mul_res2 = vmull_u8(src_tmp3, coeffabs_1)@
166 vmlal.s16 q5,d1,d22 @mul_res2 = vmlal_u8(mul_res2, src_tmp2, coeffabs_0)
[all...] |
| ihevc_inter_pred_luma_vert_w16inp_w16out.s | 158 vmull.s16 q4,d1,d23 @mul_res1 = vmull_u8(src_tmp2, coeffabs_1)@
160 vmlal.s16 q4,d0,d22 @mul_res1 = vmlal_u8(mul_res1, src_tmp1, coeffabs_0)@
162 vmlal.s16 q4,d2,d24 @mul_res1 = vmlal_u8(mul_res1, src_tmp3, coeffabs_2)@
164 vmlal.s16 q4,d3,d25 @mul_res1 = vmlal_u8(mul_res1, src_tmp4, coeffabs_3)@
166 vmlal.s16 q4,d4,d26 @mul_res1 = vmlal_u8(mul_res1, src_tmp1, coeffabs_4)@
168 vmlal.s16 q4,d5,d27 @mul_res1 = vmlal_u8(mul_res1, src_tmp2, coeffabs_5)@
169 vmlal.s16 q4,d6,d28 @mul_res1 = vmlal_u8(mul_res1, src_tmp3, coeffabs_6)@
170 vmlal.s16 q4,d7,d29 @mul_res1 = vmlal_u8(mul_res1, src_tmp4, coeffabs_7)@
174 vmull.s16 q5,d2,d23 @mul_res2 = vmull_u8(src_tmp3, coeffabs_1)@
176 vmlal.s16 q5,d1,d22 @mul_res2 = vmlal_u8(mul_res2, src_tmp2, coeffabs_0)
[all...] |
| /frameworks/av/media/libstagefright/codecs/amrwbenc/src/asm/ARMV7/ |
| Filt_6k_7k_neon.s | 43 VLD1.S16 {D0, D1, D2, D3}, [r0]!
44 VLD1.S16 {D4, D5, D6, D7}, [r0]!
46 VST1.S16 {D0, D1, D2, D3}, [r1]!
47 VST1.S16 {D4, D5, D6}, [r1]!
48 VST1.S16 D7[0], [r1]!
49 VST1.S16 D7[1], [r1]!
63 VLD1.S16 {Q0, Q1}, [r7]! @ signal[0] ~ signal[15]
64 VLD1.S16 {Q2, Q3}, [r7]! @ signal[16] ~ signal[31]
65 VLD1.S16 {Q4, Q5}, [r7]! @ signal[32] ~ signal[47]
66 VLD1.S16 {Q6, Q7}, [r7]! @ signal[48] ~ signal[63
[all...] |
| Dot_p_neon.s | 39 VLD1.S16 {Q0, Q1}, [r0]! @load 16 Word16 x[]
40 VLD1.S16 {Q2, Q3}, [r0]! @load 16 Word16 x[]
41 VLD1.S16 {Q4, Q5}, [r0]! @load 16 Word16 x[]
42 VLD1.S16 {Q6, Q7}, [r0]! @load 16 Word16 x[]
43 VLD1.S16 {Q8, Q9}, [r1]! @load 16 Word16 y[]
44 VLD1.S16 {Q10, Q11}, [r1]! @load 16 Word16 y[]
45 VLD1.S16 {Q12, Q13}, [r1]! @load 16 Word16 y[]
47 VMULL.S16 Q15, D16, D0
48 VMLAL.S16 Q15, D17, D1
49 VMLAL.S16 Q15, D18, D
[all...] |
| /external/libvpx/libvpx/vp8/common/arm/neon/ |
| shortidct4x4llm_neon.asm | 47 vqdmulh.s16 q3, q2, d0[2]
48 vqdmulh.s16 q4, q2, d0[0]
50 vqadd.s16 d12, d2, d3 ;a1
51 vqsub.s16 d13, d2, d3 ;b1
53 vshr.s16 q3, q3, #1
54 vshr.s16 q4, q4, #1
56 vqadd.s16 q3, q3, q2 ;modify since sinpi8sqrt2 > 65536/2 (negtive number)
57 vqadd.s16 q4, q4, q2
64 vqsub.s16 d10, d6, d9 ;c1
65 vqadd.s16 d11, d7, d8 ;d
[all...] |
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/neon/ |
| shortidct4x4llm_neon.asm | 47 vqdmulh.s16 q3, q2, d0[2]
48 vqdmulh.s16 q4, q2, d0[0]
50 vqadd.s16 d12, d2, d3 ;a1
51 vqsub.s16 d13, d2, d3 ;b1
53 vshr.s16 q3, q3, #1
54 vshr.s16 q4, q4, #1
56 vqadd.s16 q3, q3, q2 ;modify since sinpi8sqrt2 > 65536/2 (negtive number)
57 vqadd.s16 q4, q4, q2
64 vqsub.s16 d10, d6, d9 ;c1
65 vqadd.s16 d11, d7, d8 ;d
[all...] |
| /external/chromium_org/third_party/libvpx/source/libvpx/vp8/encoder/arm/neon/ |
| fastquantizeb_neon.asm | 37 vabs.s16 q4, q0 ; calculate x = abs(z)
38 vabs.s16 q5, q1
41 vshr.s16 q2, q0, #15 ; sz
42 vshr.s16 q3, q1, #15
44 vld1.s16 {q6, q7}, [r6@128] ; load round_ptr [0-15]
45 vld1.s16 {q8, q9}, [r5@128] ; load quant_ptr [0-15]
49 vadd.s16 q4, q6 ; x + Round
50 vadd.s16 q5, q7
54 vqdmulh.s16 q4, q8 ; y = ((Round+abs(z)) * Quant) >> 16
55 vqdmulh.s16 q5, q
[all...] |
| shortfdct_neon.asm | 50 vadd.s16 d4, d0, d3 ; a1 = ip[0] + ip[3]
51 vadd.s16 d5, d1, d2 ; b1 = ip[1] + ip[2]
52 vsub.s16 d6, d1, d2 ; c1 = ip[1] - ip[2]
53 vsub.s16 d7, d0, d3 ; d1 = ip[0] - ip[3]
55 vshl.s16 q2, q2, #3 ; (a1, b1) << 3
56 vshl.s16 q3, q3, #3 ; (c1, d1) << 3
58 vadd.s16 d0, d4, d5 ; op[0] = a1 + b1
59 vsub.s16 d2, d4, d5 ; op[2] = a1 - b1
61 vmlal.s16 q9, d7, d16 ; d1*5352 + 14500
62 vmlal.s16 q10, d7, d17 ; d1*2217 + 750
[all...] |
| /external/libvpx/libvpx/vp8/encoder/arm/neon/ |
| fastquantizeb_neon.asm | 37 vabs.s16 q4, q0 ; calculate x = abs(z)
38 vabs.s16 q5, q1
41 vshr.s16 q2, q0, #15 ; sz
42 vshr.s16 q3, q1, #15
44 vld1.s16 {q6, q7}, [r6@128] ; load round_ptr [0-15]
45 vld1.s16 {q8, q9}, [r5@128] ; load quant_ptr [0-15]
49 vadd.s16 q4, q6 ; x + Round
50 vadd.s16 q5, q7
54 vqdmulh.s16 q4, q8 ; y = ((Round+abs(z)) * Quant) >> 16
55 vqdmulh.s16 q5, q
[all...] |
| shortfdct_neon.asm | 50 vadd.s16 d4, d0, d3 ; a1 = ip[0] + ip[3]
51 vadd.s16 d5, d1, d2 ; b1 = ip[1] + ip[2]
52 vsub.s16 d6, d1, d2 ; c1 = ip[1] - ip[2]
53 vsub.s16 d7, d0, d3 ; d1 = ip[0] - ip[3]
55 vshl.s16 q2, q2, #3 ; (a1, b1) << 3
56 vshl.s16 q3, q3, #3 ; (c1, d1) << 3
58 vadd.s16 d0, d4, d5 ; op[0] = a1 + b1
59 vsub.s16 d2, d4, d5 ; op[2] = a1 - b1
61 vmlal.s16 q9, d7, d16 ; d1*5352 + 14500
62 vmlal.s16 q10, d7, d17 ; d1*2217 + 750
[all...] |
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/neon/ |
| fastquantizeb_neon.asm | 37 vabs.s16 q4, q0 ; calculate x = abs(z)
38 vabs.s16 q5, q1
41 vshr.s16 q2, q0, #15 ; sz
42 vshr.s16 q3, q1, #15
44 vld1.s16 {q6, q7}, [r6@128] ; load round_ptr [0-15]
45 vld1.s16 {q8, q9}, [r5@128] ; load quant_ptr [0-15]
49 vadd.s16 q4, q6 ; x + Round
50 vadd.s16 q5, q7
54 vqdmulh.s16 q4, q8 ; y = ((Round+abs(z)) * Quant) >> 16
55 vqdmulh.s16 q5, q
[all...] |
| shortfdct_neon.asm | 50 vadd.s16 d4, d0, d3 ; a1 = ip[0] + ip[3]
51 vadd.s16 d5, d1, d2 ; b1 = ip[1] + ip[2]
52 vsub.s16 d6, d1, d2 ; c1 = ip[1] - ip[2]
53 vsub.s16 d7, d0, d3 ; d1 = ip[0] - ip[3]
55 vshl.s16 q2, q2, #3 ; (a1, b1) << 3
56 vshl.s16 q3, q3, #3 ; (c1, d1) << 3
58 vadd.s16 d0, d4, d5 ; op[0] = a1 + b1
59 vsub.s16 d2, d4, d5 ; op[2] = a1 - b1
61 vmlal.s16 q9, d7, d16 ; d1*5352 + 14500
62 vmlal.s16 q10, d7, d17 ; d1*2217 + 750
[all...] |
| /external/chromium_org/third_party/libvpx/source/libvpx/vp9/common/arm/neon/ |
| vp9_idct8x8_add_neon.asm | 32 vmull.s16 q2, d18, d0
33 vmull.s16 q3, d19, d0
36 vmull.s16 q5, d26, d2
37 vmull.s16 q6, d27, d2
40 vmlsl.s16 q2, d30, d1
41 vmlsl.s16 q3, d31, d1
44 vmlsl.s16 q5, d22, d3
45 vmlsl.s16 q6, d23, d3
56 vmull.s16 q2, d18, d1
57 vmull.s16 q3, d19, d
[all...] |
| vp9_idct16x16_add_neon.asm | 53 vld2.s16 {q8,q9}, [r0]!
54 vld2.s16 {q9,q10}, [r0]!
55 vld2.s16 {q10,q11}, [r0]!
56 vld2.s16 {q11,q12}, [r0]!
57 vld2.s16 {q12,q13}, [r0]!
58 vld2.s16 {q13,q14}, [r0]!
59 vld2.s16 {q14,q15}, [r0]!
60 vld2.s16 {q1,q2}, [r0]!
61 vmov.s16 q15, q1
88 vmull.s16 q2, d18, d
[all...] |
| vp9_iht8x8_add_neon.asm | 131 vmull.s16 q2, d18, d0
132 vmull.s16 q3, d19, d0
135 vmull.s16 q5, d26, d2
136 vmull.s16 q6, d27, d2
139 vmlsl.s16 q2, d30, d1
140 vmlsl.s16 q3, d31, d1
143 vmlsl.s16 q5, d22, d3
144 vmlsl.s16 q6, d23, d3
155 vmull.s16 q2, d18, d1
156 vmull.s16 q3, d19, d
[all...] |
| /external/libvpx/libvpx/vp9/common/arm/neon/ |
| vp9_idct8x8_add_neon.asm | 32 vmull.s16 q2, d18, d0
33 vmull.s16 q3, d19, d0
36 vmull.s16 q5, d26, d2
37 vmull.s16 q6, d27, d2
40 vmlsl.s16 q2, d30, d1
41 vmlsl.s16 q3, d31, d1
44 vmlsl.s16 q5, d22, d3
45 vmlsl.s16 q6, d23, d3
56 vmull.s16 q2, d18, d1
57 vmull.s16 q3, d19, d
[all...] |
| vp9_idct16x16_add_neon.asm | 53 vld2.s16 {q8,q9}, [r0]!
54 vld2.s16 {q9,q10}, [r0]!
55 vld2.s16 {q10,q11}, [r0]!
56 vld2.s16 {q11,q12}, [r0]!
57 vld2.s16 {q12,q13}, [r0]!
58 vld2.s16 {q13,q14}, [r0]!
59 vld2.s16 {q14,q15}, [r0]!
60 vld2.s16 {q1,q2}, [r0]!
61 vmov.s16 q15, q1
88 vmull.s16 q2, d18, d
[all...] |
| vp9_iht8x8_add_neon.asm | 131 vmull.s16 q2, d18, d0
132 vmull.s16 q3, d19, d0
135 vmull.s16 q5, d26, d2
136 vmull.s16 q6, d27, d2
139 vmlsl.s16 q2, d30, d1
140 vmlsl.s16 q3, d31, d1
143 vmlsl.s16 q5, d22, d3
144 vmlsl.s16 q6, d23, d3
155 vmull.s16 q2, d18, d1
156 vmull.s16 q3, d19, d
[all...] |
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp9/common/arm/neon/ |
| vp9_idct8x8_add_neon.asm | 32 vmull.s16 q2, d18, d0
33 vmull.s16 q3, d19, d0
36 vmull.s16 q5, d26, d2
37 vmull.s16 q6, d27, d2
40 vmlsl.s16 q2, d30, d1
41 vmlsl.s16 q3, d31, d1
44 vmlsl.s16 q5, d22, d3
45 vmlsl.s16 q6, d23, d3
56 vmull.s16 q2, d18, d1
57 vmull.s16 q3, d19, d
[all...] |
| vp9_idct16x16_add_neon.asm | 53 vld2.s16 {q8,q9}, [r0]!
54 vld2.s16 {q9,q10}, [r0]!
55 vld2.s16 {q10,q11}, [r0]!
56 vld2.s16 {q11,q12}, [r0]!
57 vld2.s16 {q12,q13}, [r0]!
58 vld2.s16 {q13,q14}, [r0]!
59 vld2.s16 {q14,q15}, [r0]!
60 vld2.s16 {q1,q2}, [r0]!
61 vmov.s16 q15, q1
88 vmull.s16 q2, d18, d
[all...] |
| /frameworks/rs/cpu_ref/ |
| rsCpuIntrinsics_neon_Convolve.S | 69 vmull.s16 q8, d4, d0[0]
70 vmlal.s16 q8, d5, d0[1]
71 vmlal.s16 q8, d6, d0[2]
72 vmlal.s16 q8, d8, d0[3]
73 vmlal.s16 q8, d9, d1[0]
74 vmlal.s16 q8, d10, d1[1]
75 vmlal.s16 q8, d12, d1[2]
76 vmlal.s16 q8, d13, d1[3]
77 vmlal.s16 q8, d14, d2[0]
79 vmull.s16 q9, d5, d0[0
[all...] |
| /frameworks/av/media/libstagefright/codecs/on2/h264dec/omxdl/arm_neon/vc/m4p10/src/ |
| armVCM4P10_TransformResidual4x4_s.s | 59 dIn0 DN D0.S16
60 dIn1 DN D1.S16
61 dIn2 DN D2.S16
62 dIn3 DN D3.S16
65 dZero DN D4.S16
66 de0 DN D5.S16
67 de1 DN D6.S16
68 de2 DN D7.S16
69 de3 DN D8.S16
70 dIn1RS DN D7.S16
[all...] |
| /external/chromium_org/third_party/openmax_dl/dl/sp/src/arm/neon/ |
| armSP_FFT_CToC_SC16_Radix8_fs_unsafe_s.S | 82 #define dXr0 D14.S16
83 #define dXi0 D15.S16
84 #define dXr1 D2.S16
85 #define dXi1 D3.S16
86 #define dXr2 D4.S16
87 #define dXi2 D5.S16
88 #define dXr3 D6.S16
89 #define dXi3 D7.S16
90 #define dXr4 D8.S16
91 #define dXi4 D9.S16
[all...] |
