| /external/llvm/test/MC/AArch64/ |
| D | neon-simd-misc.s | 11 rev64 v6.4s, v8.4s
117 suqadd v6.4s, v8.4s
118 suqadd v6.2d, v8.2d
137 usqadd v6.4s, v8.4s
138 usqadd v6.2d, v8.2d
157 sqabs v6.4s, v8.4s
158 sqabs v6.2d, v8.2d
177 sqneg v6.4s, v8.4s
178 sqneg v6.2d, v8.2d
197 abs v6.4s, v8.4s
[all …]
|
| D | arm64-simd-ldst.s | 11 ld1.8b {v4, v5, v6}, [x3]
102 ; CHECK: ld1.8b { v4, v5, v6 }, [x3] ; encoding: [0x64,0x60,0x40,0x0c]
223 ld3.8b {v4, v5, v6}, [x19]
224 ld3.16b {v4, v5, v6}, [x19]
225 ld3.4h {v4, v5, v6}, [x19]
226 ld3.8h {v4, v5, v6}, [x19]
227 ld3.2s {v4, v5, v6}, [x19]
228 ld3.4s {v4, v5, v6}, [x19]
229 ld3.2d {v4, v5, v6}, [x19]
236 ld3.4s {v4, v5, v6}, [x29]
[all …]
|
| /external/llvm/test/MC/R600/ |
| D | vop3.s | 12 v_cmp_lt_f32 s[2:3] -v4, v6
15 v_cmp_lt_f32 s[2:3] v4, -v6
18 v_cmp_lt_f32 s[2:3] -v4, -v6
21 v_cmp_lt_f32 s[2:3] |v4|, v6
24 v_cmp_lt_f32 s[2:3] v4, |v6|
27 v_cmp_lt_f32 s[2:3] |v4|, |v6|
30 v_cmp_lt_f32 s[2:3] -|v4|, v6
33 v_cmp_lt_f32 s[2:3] v4, -|v6|
36 v_cmp_lt_f32 s[2:3] -|v4|, -|v6|
43 v_cmp_f_f32 s[2:3], v4, v6
[all …]
|
| D | ds.s | 15 ds_write2_b32 v2, v4, v6 offset0:4
18 ds_write2_b32 v2, v4, v6 offset0:4 offset1:8
21 ds_write2_b32 v2, v4, v6 offset1:8
72 ds_mskor_b32 v2, v4, v6
78 ds_write2_b32 v2, v4, v6
81 ds_write2st64_b32 v2, v4, v6
84 ds_cmpst_b32 v2, v4, v6
87 ds_cmpst_f32 v2, v4, v6
90 ds_min_f32 v2, v4, v6
93 ds_max_f32 v2, v4, v6
[all …]
|
| /external/libavc/common/armv8/ |
| D | ih264_weighted_pred_av8.s | 140 ld1 {v6.s}[0], [x0], x2 //load row 3 in source
141 ld1 {v6.s}[1], [x0], x2 //load row 4 in source
144 uxtl v6.8h, v6.8b //converting rows 3,4 to 16-bit
147 mul v6.8h, v6.8h , v2.h[0] //weight mult. for rows 3,4
151 srshl v6.8h, v6.8h , v0.8h //rounds off the weighted samples from rows 3,4
154 saddw v6.8h, v6.8h , v3.8b //adding offset for rows 3,4
157 sqxtun v6.8b, v6.8h //saturating rows 3,4 to unsigned 8-bit
161 st1 {v6.s}[0], [x1], x3 //store row 3 in destination
162 st1 {v6.s}[1], [x1], x3 //store row 4 in destination
171 ld1 {v6.8b}, [x0], x2 //load row 2 in source
[all …]
|
| D | ih264_inter_pred_filters_luma_horz_av8.s | 129 ld1 {v5.8b, v6.8b, v7.8b}, [x0], x2 //// Load row1
132 ext v28.8b, v5.8b , v6.8b, #5 ////extract a[5] (column1,row1)
134 ext v27.8b, v6.8b , v7.8b, #5 ////extract a[5] (column2,row1)
137 uaddl v16.8h, v27.8b, v6.8b //// a0 + a5 (column2,row1)
140 ext v28.8b, v5.8b , v6.8b, #2 ////extract a[2] (column1,row1)
142 ext v27.8b, v6.8b , v7.8b, #2 ////extract a[2] (column2,row1)
148 ext v28.8b, v5.8b , v6.8b, #3 ////extract a[3] (column1,row1)
150 ext v27.8b, v6.8b , v7.8b, #3 ////extract a[3] (column2,row1)
156 ext v28.8b, v5.8b , v6.8b, #1 ////extract a[1] (column1,row1)
158 ext v27.8b, v6.8b , v7.8b, #1 ////extract a[1] (column2,row1)
[all …]
|
| D | ih264_inter_pred_luma_horz_qpel_av8.s | 137 ld1 {v5.8b, v6.8b, v7.8b}, [x0], x2 //// Load row1
140 ext v28.8b, v5.8b , v6.8b , #5
142 ext v27.8b, v6.8b , v7.8b , #5
145 uaddl v16.8h, v27.8b, v6.8b //// a0 + a5 (column2,row1)
148 ext v28.8b, v5.8b , v6.8b , #2
150 ext v27.8b, v6.8b , v7.8b , #2
156 ext v28.8b, v5.8b , v6.8b , #3
158 ext v27.8b, v6.8b , v7.8b , #3
164 ext v28.8b, v5.8b , v6.8b , #1
166 ext v27.8b, v6.8b , v7.8b , #1
[all …]
|
| D | ih264_padding_neon_av8.s | 200 dup v6.16b, w11
205 st1 {v6.16b}, [x4], x1 // 16 bytes store
220 dup v6.16b, w11
223 st1 {v6.16b}, [x4], x1 // 16 bytes store
248 dup v6.16b, w11
253 st1 {v6.16b}, [x4], #16 // 16 bytes store
258 st1 {v6.16b}, [x4], x6 // 16 bytes store
272 dup v6.16b, w11
275 st1 {v6.16b}, [x4], #16 // 16 bytes store
276 st1 {v6.16b}, [x4], x6 // 16 bytes store
[all …]
|
| D | ih264_deblk_chroma_av8.s | 93 ld2 {v6.8b, v7.8b}, [x0], x1 //D6 = p1u , D7 = p1v
100 uaddl v8.8h, v6.8b, v0.8b //
107 mov v6.d[1], v7.d[0]
108 uabd v26.16b, v6.16b , v4.16b //Q13 = ABS(p1 - p0)
118 umlal v14.8h, v6.8b, v31.8b //
201 ld4 {v4.h, v5.h, v6.h, v7.h}[0], [x0], x1
202 ld4 {v4.h, v5.h, v6.h, v7.h}[1], [x0], x1
203 ld4 {v4.h, v5.h, v6.h, v7.h}[2], [x0], x1
204 ld4 {v4.h, v5.h, v6.h, v7.h}[3], [x0], x1
210 mov v10.16b, v6.16b
[all …]
|
| D | ih264_intra_pred_luma_16x16_av8.s | 436 ld1 {v6.2s, v7.2s}, [x7]
438 uxtl v16.8h, v6.8b
514 dup v6.8h, w12
516 shl v26.8h, v6.8h, #3
519 add v28.8h, v30.8h , v6.8h
524 add v26.8h, v26.8h , v6.8h
525 add v28.8h, v28.8h , v6.8h
529 add v26.8h, v26.8h , v6.8h
530 add v28.8h, v28.8h , v6.8h
534 add v26.8h, v26.8h , v6.8h
[all …]
|
| D | ih264_iquant_itrans_recon_av8.s | 148 smull v6.4s, v19.4h, v23.4h // q3 = p[i] = (x[i] * trns_coeff[i]) where i = 12..15
153 sshl v6.4s, v6.4s, v30.4s // q3 = q[i] = (p[i] << (qp/6)) where i = 12..15
158 sqrshrn v3.4h, v6.4s, #0x4 // d3 = c[i] = ((q[i] + 32) >> 4) where i = 12..15
174 sub v6.4h, v8.4h, v3.4h // x2 = (d1 >> 1) - d3//
180 add v11.4h, v5.4h , v6.4h // x1+x2
181 sub v12.4h, v5.4h , v6.4h // x1-x2
191 trn1 v6.4h, v12.4h, v13.4h
194 trn1 v10.2s, v4.2s, v6.2s // 0
196 trn2 v12.2s, v4.2s, v6.2s // 4
339 smull v6.4s, v19.4h, v23.4h // q3 = p[i] = (x[i] * trns_coeff[i]) where i = 12..15
[all …]
|
| D | ih264_inter_pred_chroma_av8.s | 152 ld1 {v5.8b, v6.8b, v7.8b}, [x0], x2 //// Load row1;
154 ext v8.8b, v5.8b , v6.8b , #2
156 ext v9.8b, v6.8b , v7.8b , #2
165 umlal v22.8h, v6.8b, v30.8b
177 umull v16.8h, v6.8b, v28.8b
190 ld1 {v5.8b, v6.8b, v7.8b}, [x0], x2 //// Load row4;
192 ext v8.8b, v5.8b , v6.8b , #2
194 ext v9.8b, v6.8b , v7.8b , #2
204 umlal v22.8h, v6.8b, v30.8b
222 umull v16.8h, v6.8b, v28.8b
[all …]
|
| D | ih264_inter_pred_luma_horz_hpel_vert_qpel_av8.s | 160 uaddl v6.8h, v0.8b, v5.8b
166 mla v6.8h, v8.8h , v22.8h
170 mls v6.8h, v8.8h , v24.8h
177 st1 {v6.4s}, [x9], x6 // store temp buffer 0
253 saddl v18.4s, v6.4h, v16.4h
257 saddl2 v6.4s, v6.8h, v16.8h
263 smlal2 v6.4s, v30.8h, v22.8h
264 smlsl2 v6.4s, v28.8h, v24.8h
270 sqrshrun v19.4h, v6.4s, #10
291 saddl2 v6.4s, v8.8h, v20.8h
[all …]
|
| D | ih264_deblk_luma_av8.s | 104 ld1 {v6.8b, v7.8b}, [x0], x1 //p0 values are loaded into q3
109 mov v6.d[1], v7.d[0]
110 uabd v26.16b, v8.16b, v6.16b
114 uabd v22.16b, v6.16b, v0.16b
124 uabd v28.16b, v10.16b, v6.16b
136 usubl v24.8h, v0.8b, v6.8b //Q15,Q12 = (q0 - p0)
146 urhadd v16.16b, v6.16b , v0.16b //Q8 = ((p0+q0+1) >> 1)
179 uqadd v16.16b, v6.16b , v18.16b //Q8 = p0 + delta
180 uqsub v6.16b, v6.16b , v18.16b //Q3 = p0 - delta
186 bif v16.16b, v6.16b , v24.16b //Q8 = (i_macro >= 0 ) ? (p0+delta) : (p0-delta)
[all …]
|
| /external/libhevc/common/arm64/ |
| D | ihevc_padding.s | 113 dup v6.16b,w11
141 st1 {v6.16b},[x7],#16 //128/8 = 16 bytes store
142 st1 {v6.16b},[x7],#16 //128/8 = 16 bytes store
143 st1 {v6.16b},[x7],#16 //128/8 = 16 bytes store
144 st1 {v6.16b},[x7],#16 //128/8 = 16 bytes store
145 st1 {v6.16b},[x7] //128/8 = 16 bytes store
231 dup v6.8h,w11
259 st1 {v6.16b},[x7],#16 //128/8 = 16 bytes store
260 st1 {v6.16b},[x7],#16 //128/8 = 16 bytes store
261 st1 {v6.16b},[x7],#16 //128/8 = 16 bytes store
[all …]
|
| D | ihevc_intra_pred_luma_planar.s | 147 dup v6.8b,w9 //nt - 1 - row
190 umlal v27.8h, v6.8b, v3.8b //(1)(nt-1-row) * src[2nt+1+col]
198 sub v6.8b, v6.8b , v7.8b //(1)
206 umlal v30.8h, v6.8b, v3.8b //(2)
212 sub v6.8b, v6.8b , v7.8b //(2)
221 umlal v28.8h, v6.8b, v3.8b //(3)
229 sub v6.8b, v6.8b , v7.8b //(3)
238 umlal v25.8h, v6.8b, v3.8b //(4)
246 sub v6.8b, v6.8b , v7.8b //(4)
255 umlal v16.8h, v6.8b, v3.8b //(5)
[all …]
|
| D | ihevc_itrans_recon_8x8.s | 190 ld1 {v6.4h},[x0],#8
192 smull v24.4s, v6.4h, v0.h[1] //// y1 * cos1(part of b0)
194 smull v26.4s, v6.4h, v0.h[3] //// y1 * cos3(part of b1)
196 smull v28.4s, v6.4h, v1.h[1] //// y1 * sin3(part of b2)
198 smull v30.4s, v6.4h, v1.h[3] //// y1 * sin1(part of b3)
210 smull v6.4s, v3.4h, v0.h[2] //// y2 * cos2(part of d0)
240 smlal v6.4s, v11.4h, v1.h[2] //// d0 = y2 * cos2 + y6 * sin2(part of a0 and a1)
250 add v14.4s, v10.4s , v6.4s //// a0 = c0 + d0(part of x0,x7)
251 sub v10.4s, v10.4s , v6.4s //// a3 = c0 - d0(part of x3,x4)
256 sub v6.4s, v14.4s , v24.4s //// a0 - b0(part of x7)
[all …]
|
| D | ihevc_intra_pred_chroma_horz.s | 125 dup v6.8h, v0.h[5] //duplicate the iii value.
134 st1 { v6.8h},[x2],x3
135 st1 { v6.8h},[x9],x3
141 dup v6.8h, v0.h[1]
150 st1 { v6.8h},[x2],x3
151 st1 { v6.8h},[x9],x3
157 dup v6.8h, v18.h[5]
166 st1 { v6.8h},[x2],x3
167 st1 { v6.8h},[x9],x3
173 dup v6.8h, v18.h[1]
[all …]
|
| D | ihevc_weighted_pred_uni.s | 180 smull v6.4s, v2.4h, v0.h[0] //vmull_n_s16(pi2_src_val2, (int16_t) wgt0) ii iteration
185 add v6.4s, v6.4s , v30.4s //vaddq_s32(i4_tmp2_t, tmp_lvl_shift_t) ii iteration
193 sshl v6.4s,v6.4s,v28.4s
201 sqxtun v6.4h, v6.4s //vqmovun_s32(sto_res_tmp1) ii iteration
213 uqxtn v6.8b, v6.8h //vqmovn_u16(sto_res_tmp3) ii iteration
214 st1 {v6.s}[0],[x6],x3 //store pu1_dst ii iteration
|
| D | ihevc_inter_pred_chroma_vert.s | 148 umull v6.8h, v17.8b, v1.8b //vmull_u8(vreinterpret_u8_u32(src_tmp2), coeffabs_1)
150 …umlsl v6.8h, v5.8b, v0.8b //vmlsl_u8(mul_res1, vreinterpret_u8_u32(src_tmp1), coeffa…
152 …umlal v6.8h, v4.8b, v2.8b //vmlal_u8(mul_res1, vreinterpret_u8_u32(src_tmp3), coeffa…
154 umlsl v6.8h, v16.8b, v3.8b
158 sqrshrun v6.8b, v6.8h,#6 //shifts right
162 st1 {v6.8b},[x1],#8 //stores the loaded value
187 ld1 {v6.s}[0],[x0] //vld1_lane_u32((uint32_t *)pu1_src_tmp, src_tmp1, 0
190 ld1 {v6.s}[1],[x6],x2 //loads pu1_src_tmp
191 dup v7.2s, v6.s[1]
196 umlsl v4.8h, v6.8b, v0.8b
[all …]
|
| D | ihevc_intra_pred_luma_horz.s | 125 dup v6.16b, v0.b[13] //duplicate the iii value.
134 st1 { v6.16b},[x2],x3
135 st1 { v6.16b},[x9],x3
141 dup v6.16b, v0.b[9]
150 st1 { v6.16b},[x2],x3
151 st1 { v6.16b},[x9],x3
157 dup v6.16b, v0.b[5]
166 st1 { v6.16b},[x2],x3
167 st1 { v6.16b},[x9],x3
173 dup v6.16b, v0.b[1]
[all …]
|
| D | ihevc_inter_pred_chroma_vert_w16out.s | 151 umull v6.8h, v17.8b, v1.8b //vmull_u8(vreinterpret_u8_u32(src_tmp2), coeffabs_1)
153 …umlsl v6.8h, v5.8b, v0.8b //vmlsl_u8(mul_res1, vreinterpret_u8_u32(src_tmp1), coeffa…
155 …umlal v6.8h, v4.8b, v2.8b //vmlal_u8(mul_res1, vreinterpret_u8_u32(src_tmp3), coeffa…
158 umlsl v6.8h, v16.8b, v3.8b
163 st1 { v6.8h},[x1],#16 //stores the loaded value
188 ld1 {v6.s}[0],[x0] //vld1_lane_u32((uint32_t *)pu1_src_tmp, src_tmp1, 0
191 ld1 {v6.s}[1],[x6],x2 //loads pu1_src_tmp
192 dup v7.2s, v6.s[1]
197 umlsl v4.8h, v6.8b, v0.8b
237 ld1 {v6.8b},[x6],x2 //load and increment
[all …]
|
| D | ihevc_intra_pred_luma_mode_3_to_9.s | 157 xtn v6.8b, v22.8h
172 and v6.8b, v6.8b , v29.8b //fract values in d1/ idx values in d0
183 sub v7.8b, v28.8b , v6.8b //32-fract
191 umlal v24.8h, v13.8b, v6.8b //mul (row 0)
201 umlal v22.8h, v17.8b, v6.8b //mul (row 1)
212 umlal v20.8h, v15.8b, v6.8b //mul (row 2)
223 umlal v18.8h, v25.8b, v6.8b //mul (row 3)
234 umlal v24.8h, v13.8b, v6.8b //mul (row 4)
245 umlal v22.8h, v17.8b, v6.8b //mul (row 5)
256 umlal v20.8h, v15.8b, v6.8b //mul (row 6)
[all …]
|
| /external/libmpeg2/common/armv8/ |
| D | impeg2_idct.s | 165 ld1 {v6.8b}, [x2], x5
192 uaddw v20.8h, v30.8h , v6.8b
197 sqxtun v6.8b, v20.8h
201 st1 {v6.8b}, [x3], x6
390 ld1 {v6.4h}, [x0], #8
392 smull v24.4s, v6.4h, v0.4h[1] //// y1 * cos1(part of b0)
394 smull v26.4s, v6.4h, v0.4h[3] //// y1 * cos3(part of b1)
396 smull v28.4s, v6.4h, v1.4h[1] //// y1 * sin3(part of b2)
398 smull v30.4s, v6.4h, v1.4h[3] //// y1 * sin1(part of b3)
410 smull v6.4s, v3.4h, v0.4h[2] //// y2 * cos2(part of d0)
[all …]
|
| /external/boringssl/linux-aarch64/crypto/aes/ |
| D | aesv8-armx.S | 48 tbl v6.16b,{v3.16b},v2.16b
51 aese v6.16b,v0.16b
58 eor v6.16b,v6.16b,v1.16b
61 eor v3.16b,v3.16b,v6.16b
66 tbl v6.16b,{v3.16b},v2.16b
69 aese v6.16b,v0.16b
75 eor v6.16b,v6.16b,v1.16b
78 eor v3.16b,v3.16b,v6.16b
80 tbl v6.16b,{v3.16b},v2.16b
83 aese v6.16b,v0.16b
[all …]
|