Home
last modified time | relevance | path

Searched refs:s16 (Results 1 – 25 of 670) sorted by relevance

12345678910>>...27

/external/libhevc/common/arm/
Dihevc_itrans_recon_32x32.s217 vmull.s16 q12,d8,d0[1] @// y1 * cos1(part of b0)
218 vmull.s16 q13,d8,d0[3] @// y1 * cos3(part of b1)
219 vmull.s16 q14,d8,d1[1] @// y1 * sin3(part of b2)
220 vmull.s16 q15,d8,d1[3] @// y1 * sin1(part of b3)
222 vmlal.s16 q12,d9,d0[3] @// y1 * cos1 + y3 * cos3(part of b0)
223 vmlal.s16 q13,d9,d2[1] @// y1 * cos3 - y3 * sin1(part of b1)
224 vmlal.s16 q14,d9,d3[3] @// y1 * sin3 - y3 * cos1(part of b2)
225 vmlal.s16 q15,d9,d5[1] @// y1 * sin1 - y3 * sin3(part of b3)
231 vmull.s16 q10,d10,d0[0]
232 vmlal.s16 q10,d11,d0[2]
[all …]
Dihevc_itrans_recon_16x16.s243 vmull.s16 q12,d6,d0[1] @// y1 * cos1(part of b0)
244 vmull.s16 q13,d6,d0[3] @// y1 * cos3(part of b1)
245 vmull.s16 q14,d6,d1[1] @// y1 * sin3(part of b2)
246 vmull.s16 q15,d6,d1[3] @// y1 * sin1(part of b3)
248 vmlal.s16 q12,d7,d0[3] @// y1 * cos1 + y3 * cos3(part of b0)
249 vmlal.s16 q13,d7,d2[1] @// y1 * cos3 - y3 * sin1(part of b1)
250 vmlal.s16 q14,d7,d3[3] @// y1 * sin3 - y3 * cos1(part of b2)
251 vmlsl.s16 q15,d7,d2[3] @// y1 * sin1 - y3 * sin3(part of b3)
258 vmull.s16 q6,d10,d0[0]
259 vmlal.s16 q6,d11,d0[2]
[all …]
Dihevc_inter_pred_filters_luma_vert_w16inp.s153 vmull.s16 q4,d1,d23 @mul_res1 = vmull_u8(src_tmp2, coeffabs_1)@
155 vmlal.s16 q4,d0,d22 @mul_res1 = vmlal_u8(mul_res1, src_tmp1, coeffabs_0)@
157 vmlal.s16 q4,d2,d24 @mul_res1 = vmlal_u8(mul_res1, src_tmp3, coeffabs_2)@
159 vmlal.s16 q4,d3,d25 @mul_res1 = vmlal_u8(mul_res1, src_tmp4, coeffabs_3)@
161 vmlal.s16 q4,d4,d26 @mul_res1 = vmlal_u8(mul_res1, src_tmp1, coeffabs_4)@
163 vmlal.s16 q4,d5,d27 @mul_res1 = vmlal_u8(mul_res1, src_tmp2, coeffabs_5)@
164 vmlal.s16 q4,d6,d28 @mul_res1 = vmlal_u8(mul_res1, src_tmp3, coeffabs_6)@
165 vmlal.s16 q4,d7,d29 @mul_res1 = vmlal_u8(mul_res1, src_tmp4, coeffabs_7)@
169 vmull.s16 q5,d2,d23 @mul_res2 = vmull_u8(src_tmp3, coeffabs_1)@
171 vmlal.s16 q5,d1,d22 @mul_res2 = vmlal_u8(mul_res2, src_tmp2, coeffabs_0)@
[all …]
Dihevc_inter_pred_luma_vert_w16inp_w16out.s163 vmull.s16 q4,d1,d23 @mul_res1 = vmull_u8(src_tmp2, coeffabs_1)@
165 vmlal.s16 q4,d0,d22 @mul_res1 = vmlal_u8(mul_res1, src_tmp1, coeffabs_0)@
167 vmlal.s16 q4,d2,d24 @mul_res1 = vmlal_u8(mul_res1, src_tmp3, coeffabs_2)@
169 vmlal.s16 q4,d3,d25 @mul_res1 = vmlal_u8(mul_res1, src_tmp4, coeffabs_3)@
171 vmlal.s16 q4,d4,d26 @mul_res1 = vmlal_u8(mul_res1, src_tmp1, coeffabs_4)@
173 vmlal.s16 q4,d5,d27 @mul_res1 = vmlal_u8(mul_res1, src_tmp2, coeffabs_5)@
174 vmlal.s16 q4,d6,d28 @mul_res1 = vmlal_u8(mul_res1, src_tmp3, coeffabs_6)@
175 vmlal.s16 q4,d7,d29 @mul_res1 = vmlal_u8(mul_res1, src_tmp4, coeffabs_7)@
179 vmull.s16 q5,d2,d23 @mul_res2 = vmull_u8(src_tmp3, coeffabs_1)@
181 vmlal.s16 q5,d1,d22 @mul_res2 = vmlal_u8(mul_res2, src_tmp2, coeffabs_0)@
[all …]
Dihevc_itrans_recon_8x8.s193 vmull.s16 q10,d2,d0[0] @// y0 * cos4(part of c0 and c1)
195 vmull.s16 q9,d3,d1[2] @// y2 * sin2 (q3 is freed by this time)(part of d1)
198 vmull.s16 q12,d6,d0[1] @// y1 * cos1(part of b0)
200 vmull.s16 q13,d6,d0[3] @// y1 * cos3(part of b1)
202 vmull.s16 q14,d6,d1[1] @// y1 * sin3(part of b2)
204 vmull.s16 q15,d6,d1[3] @// y1 * sin1(part of b3)
206 vmlal.s16 q12,d7,d0[3] @// y1 * cos1 + y3 * cos3(part of b0)
208 vmlsl.s16 q13,d7,d1[3] @// y1 * cos3 - y3 * sin1(part of b1)
210 vmlsl.s16 q14,d7,d0[1] @// y1 * sin3 - y3 * cos1(part of b2)
212 vmlsl.s16 q15,d7,d1[1] @// y1 * sin1 - y3 * sin3(part of b3)
[all …]
Dihevc_inter_pred_chroma_vert_w16inp.s145 vmull.s16 q0,d0,d12 @vmull_s16(src_tmp1, coeff_0)
148 vmull.s16 q4,d2,d12 @vmull_s16(src_tmp2, coeff_0)
150 vmlal.s16 q0,d2,d13
152 vmlal.s16 q4,d3,d13
155 vmlal.s16 q0,d3,d14
156 vmlal.s16 q4,d6,d14
157 vmlal.s16 q0,d6,d15
158 vmlal.s16 q4,d2,d15
161 vqrshrun.s16 d0,q0,#6 @rounding shift
162 vqrshrun.s16 d30,q15,#6 @rounding shift
[all …]
Dihevc_weighted_pred_bi_default.s145 vadd.s16 q0,q0,q2
190 vld1.s16 {d6},[r0]! @load and increment the pi2_src1
192 vld1.s16 {d7},[r1]! @load and increment the pi2_src2
193 vld1.s16 {d8},[r11],r3 @load and increment the pi2_src1 ii iteration
194 vqadd.s16 d18,d6,d7
195 vqadd.s16 d18,d18,d0 @vaddq_s32(i4_tmp1_t1, tmp_lvl_shift_t)
196 vld1.s16 {d9},[r12],r4 @load and increment the pi2_src2 ii iteration
197 vqadd.s16 d20,d8,d9 @vaddq_s32(i4_tmp2_t1, i4_tmp2_t2)
198 vqadd.s16 d19,d20,d0 @vaddq_s32(i4_tmp2_t1, tmp_lvl_shift_t)
199 vqshrun.s16 d20,q9,#7
[all …]
Dihevc_inter_pred_chroma_vert_w16inp_w16out.s145 vmull.s16 q0,d0,d12 @vmull_s16(src_tmp1, coeff_0)
148 vmull.s16 q4,d2,d12 @vmull_s16(src_tmp2, coeff_0)
150 vmlal.s16 q0,d2,d13
152 vmlal.s16 q4,d3,d13
155 vmlal.s16 q0,d3,d14
156 vmlal.s16 q4,d6,d14
157 vmlal.s16 q0,d6,d15
158 vmlal.s16 q4,d2,d15
192 vmull.s16 q15,d0,d12 @vmull_s16(src_tmp1, coeff_0)
194 vmlal.s16 q15,d1,d13
[all …]
Dihevc_itrans_recon_4x4_ttype1.s150 vmull.s16 q3,d1,d4[2] @74 * pi2_src[1]
151 vmlal.s16 q3,d0,d4[0] @74 * pi2_src[1] + 29 * pi2_src[0]
152 vmlal.s16 q3,d3,d4[1] @74 * pi2_src[1] + 29 * pi2_src[0] + 55 * pi2_src[3]
153 …vmlal.s16 q3,d2,d4[3] @pi2_out[0] = 29* pi2_src[0] + 74 * pi2_src[1] + 84* pi2_s…
155 vmull.s16 q4,d1,d4[2] @74 * pi2_src[1]
156 vmlal.s16 q4,d0,d4[1] @74 * pi2_src[1] + 55 * pi2_src[0]
157 vmlsl.s16 q4,d2,d4[0] @74 * pi2_src[1] + 55 * pi2_src[0] - 29 * pi2_src[2]
158 …vmlsl.s16 q4,d3,d4[3] @pi2_out[1] = 74 * pi2_src[1] + 55 * pi2_src[0] - 29 * pi…
160 vmull.s16 q5,d0,d4[2] @ 74 * pi2_src[0]
161 vmlsl.s16 q5,d2,d4[2] @ 74 * pi2_src[0] - 74 * pi2_src[2]
[all …]
/external/libavc/encoder/arm/
Dih264e_half_pel.s170 …vqrshrun.s16 d20, q4, #5 @// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column1,r…
171 …vqrshrun.s16 d21, q5, #5 @// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column2,r…
172 …vqrshrun.s16 d22, q6, #5 @// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column3,r…
173 …vqrshrun.s16 d23, q7, #5 @// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column1,r…
174 …vqrshrun.s16 d24, q8, #5 @// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column2,r…
175 …vqrshrun.s16 d25, q9, #5 @// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column3,r…
316 …vqrshrun.s16 d2, q10, #5 @// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column1,r…
318 …vqrshrun.s16 d3, q11, #5 @// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column2,r…
321 vaddl.s16 q13, d31, d20 @// a0 + a5 (set1)
323 vmlal.s16 q13, d30, d0[1] @// a0 + a5 + 20a2 (set1)
[all …]
/external/swiftshader/third_party/llvm-7.0/llvm/test/CodeGen/AArch64/GlobalISel/
Dmachine-cse-mid-pipeline.mir91 ; CHECK-NEXT: %[[S16:[0-9]+]]:_(s16) = G_SEXT %[[ONE]](s8)
93 ; CHECK-NEXT: %[[S16_Z64:[0-9]+]]:_(s64) = G_ZEXT %[[S16]](s16)
100 %1:_(s16) = G_SEXT %0(s8)
102 %3:_(s64) = G_ZEXT %1(s16)
189 ; CHECK: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $d0
190 …: [[UV0:%[0-9]+]]:_(s16), [[UV1:%[0-9]+]]:_(s16), [[UV2:%[0-9]+]]:_(s16), [[UV3:%[0-9]+]]:_(s16) =…
191 ; CHECK-NEXT: [[ANYEXT0:%[0-9]+]]:_(s32) = G_ANYEXT [[UV0]](s16)
192 ; CHECK-NEXT: [[ANYEXT1:%[0-9]+]]:_(s32) = G_ANYEXT [[UV1]](s16)
193 ; CHECK-NEXT: [[ANYEXT2:%[0-9]+]]:_(s32) = G_ANYEXT [[UV2]](s16)
194 ; CHECK-NEXT: [[ANYEXT3:%[0-9]+]]:_(s32) = G_ANYEXT [[UV3]](s16)
[all …]
/external/libavc/common/arm/
Dih264_iquant_itrans_recon_a9.s137 vld4.s16 {d20, d21, d22, d23}, [r5] @Load pu2_iscal_mat[i], i =0..15
138 vld4.s16 {d26, d27, d28, d29}, [r6] @pu2_weigh_mat[i], i =0..15
139 vmul.s16 q10, q10, q13 @x[i]=(scale[i] * dequant[i]) where i = 0..7
140 vld4.s16 {d16, d17, d18, d19}, [r0] @pi2_src_tmp[i], i =0..15
142 vmul.s16 q11, q11, q14 @x[i]=(scale[i] * dequant[i]) where i = 8..15
147 vmull.s16 q0, d16, d20 @ Q0 = p[i] = (x[i] * trns_coeff[i]) where i = 0..3
148 vmull.s16 q1, d17, d21 @ Q1 = p[i] = (x[i] * trns_coeff[i]) where i = 4..7
149 vmull.s16 q2, d18, d22 @ Q2 = p[i] = (x[i] * trns_coeff[i]) where i = 8..11
150 vmull.s16 q3, d19, d23 @ Q3 = p[i] = (x[i] * trns_coeff[i]) where i = 12..15
168 vadd.s16 d4, d0, d2 @x0 = q0 + q1;
[all …]
Dih264_inter_pred_luma_horz_hpel_vert_hpel_a9q.s124 vmov.s16 d0, #20 @ Filter coeff 20
125 vmov.s16 d1, #5 @ Filter coeff 5
149 vmls.s16 q12, q11, d1[0] @ temp -= temp2 * 5
154 vmls.s16 q13, q10, d1[0] @ temp -= temp2 * 5
158 vmls.s16 q14, q10, d1[0] @ temp -= temp2 * 5
164 vaddl.s16 q1, d20, d24 @// a0 + a5 (column1)
165 vaddl.s16 q15, d21, d25 @// a0 + a5 (column1)
166 vmlal.s16 q1, d22, d0[0] @// a0 + a5 + 20a2 (column1)
167 vmlal.s16 q15, d23, d0[0] @// a0 + a5 + 20a2 (column1)
170 vmlsl.s16 q1, d22, d1[0] @// a0 + a5 + 20a2 + 20a3 - 5a1 (column1)
[all …]
Dih264_intra_pred_chroma_a9q.s132 vqrshrun.s16 d12, q6, #3
133 vqrshrun.s16 d14, q7, #2
151 vqrshrun.s16 d0, q0, #2
164 vqrshrun.s16 d0, q0, #2
406 vmul.s16 q7, q5, q4
407 vmul.s16 q8, q6, q4
410 vpadd.s16 d14, d14
411 vpadd.s16 d15, d15
412 vpadd.s16 d16, d16
413 vpadd.s16 d17, d17
[all …]
Dih264_inter_pred_luma_bilinear_a9q.s150 vqrshrun.s16 d28, q10, #1
151 vqrshrun.s16 d29, q11, #1
153 vqrshrun.s16 d30, q12, #1
154 vqrshrun.s16 d31, q13, #1
158 vqrshrun.s16 d28, q8, #1
160 vqrshrun.s16 d29, q9, #1
162 vqrshrun.s16 d30, q10, #1
164 vqrshrun.s16 d31, q11, #1
174 vqrshrun.s16 d28, q10, #1
176 vqrshrun.s16 d29, q11, #1
[all …]
Dih264_resi_trans_quant_a9.s143 vadd.s16 d8 , d0, d6 @x0 = x4+x7
144 vadd.s16 d9 , d2, d4 @x1 = x5+x6
145 vsub.s16 d10, d2, d4 @x2 = x5-x6
146 vsub.s16 d11, d0, d6 @x3 = x4-x7
148 vshl.s16 d12, d10, #1 @U_SHIFT(x2,1,shft)
149 vshl.s16 d13, d11, #1 @U_SHIFT(x3,1,shft)
151 vadd.s16 d14, d8, d9 @x4 = x0 + x1;
152 vsub.s16 d16, d8, d9 @x6 = x0 - x1;
153 vadd.s16 d15, d13, d10 @x5 = U_SHIFT(x3,1,shft) + x2;
154 vsub.s16 d17, d11, d12 @x7 = x3 - U_SHIFT(x2,1,shft);
[all …]
Dih264_inter_pred_luma_horz_hpel_vert_qpel_a9q.s233 vadd.s16 q14, q4, q7
235 vadd.s16 q15, q5, q6
245 vaddl.s16 q9, d6, d16
249 vaddl.s16 q3, d7, d17
251 vqrshrun.s16 d26, q13, #5
253 vmlal.s16 q9, d30, d22
254 vmlsl.s16 q9, d28, d24
255 vmlal.s16 q3, d31, d22
256 vmlsl.s16 q3, d29, d24
263 vadd.s16 q14, q5, q8
[all …]
Dih264_weighted_bi_pred_a9q.s184 vmul.s16 q2, q2, d2[0] @weight 1 mult. for rows 1,2
185 vmla.s16 q2, q3, d2[2] @weight 2 mult. for rows 1,2
186 vmul.s16 q4, q4, d2[0] @weight 1 mult. for rows 3,4
187 vmla.s16 q4, q5, d2[2] @weight 2 mult. for rows 3,4
190 vrshl.s16 q2, q2, q0 @rounds off the weighted samples from rows 1,2
191 vrshl.s16 q4, q4, q0 @rounds off the weighted samples from rows 3,4
196 vqmovun.s16 d4, q2 @saturating rows 1,2 to unsigned 8-bit
197 vqmovun.s16 d8, q4 @saturating rows 3,4 to unsigned 8-bit
224 vmul.s16 q2, q2, d2[0] @weight 1 mult. for row 1
225 vmla.s16 q2, q3, d2[2] @weight 2 mult. for row 1
[all …]
Dih264_inter_pred_luma_horz_qpel_vert_hpel_a9q.s188 vadd.s16 q0, q12, q13
191 vadd.s16 q12, q12, q13
193 vaddl.s16 q13, d18, d22
194 vmlal.s16 q13, d0, d28
195 vmlsl.s16 q13, d24, d30
197 vaddl.s16 q11, d19, d23
198 vmlal.s16 q11, d1, d28
199 vmlsl.s16 q11, d25, d30
210 vadd.s16 q9, q12, q13
213 vadd.s16 q12, q12, q13
[all …]
Dih264_weighted_pred_a9q.s144 vmul.s16 q2, q2, d2[0] @weight mult. for rows 1,2
145 vmul.s16 q3, q3, d2[0] @weight mult. for rows 3,4
148 vrshl.s16 q2, q2, q0 @rounds off the weighted samples from rows 1,2
149 vrshl.s16 q3, q3, q0 @rounds off the weighted samples from rows 3,4
154 vqmovun.s16 d4, q2 @saturating rows 1,2 to unsigned 8-bit
155 vqmovun.s16 d6, q3 @saturating rows 3,4 to unsigned 8-bit
176 vmul.s16 q2, q2, d2[0] @weight mult. for row 1
178 vmul.s16 q3, q3, d2[0] @weight mult. for row 2
179 vmul.s16 q4, q4, d2[0] @weight mult. for row 3
180 vmul.s16 q5, q5, d2[0] @weight mult. for row 4
[all …]
/external/libmpeg2/common/arm/
Dimpeg2_idct.s152 vdup.s16 q15, r4
160 vqmovun.s16 d0, q4
163 vqmovun.s16 d1, q5
167 vqmovun.s16 d2, q6
171 vqmovun.s16 d3, q7
175 vqmovun.s16 d4, q8
179 vqmovun.s16 d5, q9
183 vqmovun.s16 d6, q10
186 vqmovun.s16 d7, q11
231 vmovl.s16 q4, d2
[all …]
/external/libvpx/config/arm-neon/vpx_dsp/arm/
Dintrapred_neon_asm.asm.S340 vadd.s16 q1, q1, q3
341 vadd.s16 q2, q2, q3
342 vqmovun.s16 d0, q1
343 vqmovun.s16 d1, q2
352 vadd.s16 q1, q1, q3
353 vadd.s16 q2, q2, q3
354 vqmovun.s16 d0, q1
355 vqmovun.s16 d1, q2
390 vadd.s16 q0, q3, q0
391 vadd.s16 q1, q3, q1
[all …]
/external/libvpx/libvpx/vpx_dsp/arm/
Dintrapred_neon_asm.asm315 vadd.s16 q1, q1, q3
316 vadd.s16 q2, q2, q3
317 vqmovun.s16 d0, q1
318 vqmovun.s16 d1, q2
327 vadd.s16 q1, q1, q3
328 vadd.s16 q2, q2, q3
329 vqmovun.s16 d0, q1
330 vqmovun.s16 d1, q2
364 vadd.s16 q0, q3, q0
365 vadd.s16 q1, q3, q1
[all …]
/external/libjpeg-turbo/simd/arm/
Djsimd_neon.S249 vmul.s16 q8, q8, q0
251 vmul.s16 q9, q9, q1
253 vmul.s16 q10, q10, q2
255 vmul.s16 q11, q11, q3
257 vmul.s16 q12, q12, q0
259 vmul.s16 q14, q14, q2
260 vmul.s16 q13, q13, q1
263 vmul.s16 q15, q15, q3
266 vadd.s16 d4, ROW7L, ROW3L
267 vadd.s16 d5, ROW5L, ROW1L
[all …]
/external/libaom/libaom/aom_dsp/simd/
Dv64_intrinsics_c.h29 int16_t s16[4]; member
213 t.s16[c] = (int32_t)a.s16[c] + (int32_t)b.s16[c] > 32767 in c_v64_sadd_s16()
215 : (int32_t)a.s16[c] + (int32_t)b.s16[c] < -32768 in c_v64_sadd_s16()
217 : (int32_t)a.s16[c] + (int32_t)b.s16[c]; in c_v64_sadd_s16()
263 t.s16[c] = (int32_t)a.s16[c] - (int32_t)b.s16[c] < -32768 in c_v64_ssub_s16()
265 : (int32_t)a.s16[c] - (int32_t)b.s16[c] > 32767 in c_v64_ssub_s16()
267 : (int32_t)a.s16[c] - (int32_t)b.s16[c]; in c_v64_ssub_s16()
440 t.s16[3] = (int16_t)a.u8[3 + endian]; in c_v64_unpacklo_u8_s16()
441 t.s16[2] = (int16_t)a.u8[2 + endian]; in c_v64_unpacklo_u8_s16()
442 t.s16[1] = (int16_t)a.u8[1 + endian]; in c_v64_unpacklo_u8_s16()
[all …]

12345678910>>...27