Searched refs:v7 (Results 1 – 25 of 298) sorted by relevance
12345678910>>...12
1 … -fast-isel-abort=1 -mtriple=armv7-apple-ios -verify-machineinstrs | FileCheck %s --check-prefix=v72 …st-isel-abort=1 -mtriple=armv7-linux-gnueabi -verify-machineinstrs | FileCheck %s --check-prefix=v73 …fast-isel-abort=1 -mtriple=thumbv7-apple-ios -verify-machineinstrs | FileCheck %s --check-prefix=v716 ; v7-LABEL: zext_1_8:17 ; v7: and r0, r0, #123 ; v7-LABEL: zext_1_16:24 ; v7: and r0, r0, #130 ; v7-LABEL: zext_1_32:31 ; v7: and r0, r0, #137 ; v7-LABEL: zext_8_16:[all …]
12 cmeq v5.8h, v6.8h, v7.8h14 cmeq v9.4s, v7.4s, v8.4s34 cmhs v5.8h, v6.8h, v7.8h36 cmhs v9.4s, v7.4s, v8.4s42 cmls v5.8h, v7.8h, v6.8h44 cmls v9.4s, v8.4s, v7.4s71 cmge v5.8h, v6.8h, v7.8h73 cmge v9.4s, v7.4s, v8.4s79 cmle v5.8h, v7.8h, v6.8h81 cmle v9.4s, v8.4s, v7.4s[all …]
12 ld1.8b {v7, v8, v9, v10}, [x4]90 st1.2d {v7, v8}, [x10]103 ; CHECK: ld1.8b { v7, v8, v9, v10 }, [x4] ; encoding: [0x87,0x20,0x40,0x0c]182 ; CHECK: st1.2d { v7, v8 }, [x10] ; encoding: [0x47,0xad,0x00,0x4c]237 ld3.2d {v7, v8, v9}, [x9]252 st3.4s {v7, v8, v9}, [x29]270 ; CHECK: ld3.2d { v7, v8, v9 }, [x9] ; encoding: [0x27,0x4d,0x40,0x4c]285 ; CHECK: st3.4s { v7, v8, v9 }, [x29] ; encoding: [0xa7,0x4b,0x00,0x4c]289 ld4.8b {v4, v5, v6, v7}, [x19]290 ld4.16b {v4, v5, v6, v7}, [x19][all …]
10 ins v7.h[7], w1415 mov v7.h[7], w1470 ins v6.h[7], v7.h[5]75 mov v6.h[7], v7.h[5]93 dup v11.4h, v7.h[7]96 dup v11.8h, v7.h[7]
12 @ CHECK-ARMV8: warning: deprecated since v7, use 'isb'13 @ CHECK-THUMBV8: warning: deprecated since v7, use 'isb'14 @ CHECK-ARMV7: warning: deprecated since v7, use 'isb'15 @ CHECK-THUMBV7: warning: deprecated since v7, use 'isb'16 @ CHECK-ARMV6-NOT: warning: deprecated since v7, use 'isb'18 @ CHECK-ARMV8: warning: deprecated since v7, use 'dsb'19 @ CHECK-THUMBV8: warning: deprecated since v7, use 'dsb'20 @ CHECK-ARMV7: warning: deprecated since v7, use 'dsb'21 @ CHECK-THUMBV7: warning: deprecated since v7, use 'dsb'22 @ CHECK-ARMV6-NOT: warning: deprecated since v7, use 'dsb'[all …]
192 dup v7.2s, v6.s[1]193 ld1 {v7.s}[1],[x6],x2 //loads pu1_src_tmp194 umull v4.8h, v7.8b, v1.8b //vmull_u8(vreinterpret_u8_u32(src_tmp2), coeffabs_1)195 dup v7.2s, v7.s[1]196 ld1 {v7.s}[1],[x6],x2198 umlal v4.8h, v7.8b, v2.8b199 dup v7.2s, v7.s[1]200 ld1 {v7.s}[1],[x6]202 umlsl v4.8h, v7.8b, v3.8b239 ld1 {v7.8b},[x6],x2 //load and increment[all …]
191 dup v7.2s, v6.s[1]192 ld1 {v7.s}[1],[x6],x2 //loads pu1_src_tmp193 umull v4.8h, v7.8b, v1.8b //vmull_u8(vreinterpret_u8_u32(src_tmp2), coeffabs_1)194 dup v7.2s, v7.s[1]195 ld1 {v7.s}[1],[x6],x2197 umlal v4.8h, v7.8b, v2.8b198 dup v7.2s, v7.s[1]199 ld1 {v7.s}[1],[x6]201 umlsl v4.8h, v7.8b, v3.8b238 ld1 {v7.8b},[x6],x2 //load and increment[all …]
187 … smull v7.4s, v3.4h, v0.h[0] //vmull_n_s16(pi2_src_val1, (int16_t) wgt0) iii iteration190 add v7.4s, v7.4s , v30.4s //vaddq_s32(i4_tmp1_t, tmp_lvl_shift_t) iii iteration199 sshl v7.4s,v7.4s,v28.4s206 sqxtun v7.4h, v7.4s //vqmovun_s32(sto_res_tmp1) iii iteration216 uqxtn v7.8b, v7.8h //vqmovn_u16(sto_res_tmp3) iii iteration220 st1 {v7.s}[0],[x6],x3 //store pu1_dst i iteration iii iteration
148 …mov v7.8b, v5.8b //mov #1 to d7 to used for inc for row+1 and dec for nt-1-…196 add v5.8b, v5.8b , v7.8b //(1)198 sub v6.8b, v6.8b , v7.8b //(1)211 add v5.8b, v5.8b , v7.8b //(2)212 sub v6.8b, v6.8b , v7.8b //(2)228 add v5.8b, v5.8b , v7.8b //(3)229 sub v6.8b, v6.8b , v7.8b //(3)245 add v5.8b, v5.8b , v7.8b //(4)246 sub v6.8b, v6.8b , v7.8b //(4)261 add v5.8b, v5.8b , v7.8b //(5)[all …]
130 ld1 {v7.8b},[x3],x2 //src_tmp4 = vld1_u8(pu1_src_tmp)//135 umlsl v19.8h, v7.8b, v29.8b //mul_res1 = vmlsl_u8(mul_res1, src_tmp4, coeffabs_7)//159 umlal v20.8h, v7.8b, v28.8b //mul_res2 = vmlal_u8(mul_res2, src_tmp4, coeffabs_6)//173 umlsl v21.8h, v7.8b, v27.8b188 umlal v30.8h, v7.8b, v26.8b195 ld1 {v7.8b},[x3],x2 //src_tmp4 = vld1_u8(pu1_src_tmp)//230 umlsl v19.8h, v7.8b, v29.8b //mul_res1 = vmlsl_u8(mul_res1, src_tmp4, coeffabs_7)//253 umlal v20.8h, v7.8b, v28.8b //mul_res2 = vmlal_u8(mul_res2, src_tmp4, coeffabs_6)//277 umlsl v21.8h, v7.8b, v27.8b299 umlal v30.8h, v7.8b, v26.8b[all …]
149 saddl v7.4s, v0.4h, v2.4h //pi2_src[0] + pi2_src[2]151 shl v7.4s, v7.4s,#6 //e[0] = 64*(pi2_src[0] + pi2_src[2])154 add v19.4s, v7.4s , v6.4s //((e[0] + o[0] )157 sub v20.4s, v7.4s , v6.4s //((e[0] - o[0])183 saddl v7.4s, v0.4h, v2.4h //pi2_src[0] + pi2_src[2]185 shl v7.4s, v7.4s,#6 //e[0] = 64*(pi2_src[0] + pi2_src[2])189 add v19.4s, v7.4s , v6.4s //((e[0] + o[0] )192 sub v20.4s, v7.4s , v6.4s //((e[0] - o[0])
173 ld1 {v7.8b},[x3],x2 //src_tmp4 = vld1_u8(pu1_src_tmp)//178 umlsl v19.8h, v7.8b, v29.8b //mul_res1 = vmlsl_u8(mul_res1, src_tmp4, coeffabs_7)//202 umlal v20.8h, v7.8b, v28.8b //mul_res2 = vmlal_u8(mul_res2, src_tmp4, coeffabs_6)//219 umlsl v21.8h, v7.8b, v27.8b234 umlal v30.8h, v7.8b, v26.8b241 ld1 {v7.8b},[x3],x2 //src_tmp4 = vld1_u8(pu1_src_tmp)//274 umlsl v19.8h, v7.8b, v29.8b //mul_res1 = vmlsl_u8(mul_res1, src_tmp4, coeffabs_7)//298 umlal v20.8h, v7.8b, v28.8b //mul_res2 = vmlal_u8(mul_res2, src_tmp4, coeffabs_6)//322 umlsl v21.8h, v7.8b, v27.8b347 umlal v30.8h, v7.8b, v26.8b[all …]
185 mov v7.h[0],w6 //moved for scalar multiplication189 mov v7.h[1],w8 //moved for scalar multiplication222 smull v4.4s, v0.4h, v7.h[0] //vmull_n_s16(pi2_src1_val1, (int16_t) wgt0)224 smull v5.4s, v1.4h, v7.h[1] //vmull_n_s16(pi2_src2_val1, (int16_t) wgt1)229 … smull v6.4s, v2.4h, v7.h[0] //vmull_n_s16(pi2_src1_val2, (int16_t) wgt0) ii iteration233 … smull v19.4s, v0.4h, v7.h[0] //vmull_n_s16(pi2_src1_val1, (int16_t) wgt0) iii iteration236 … smull v17.4s, v3.4h, v7.h[1] //vmull_n_s16(pi2_src2_val2, (int16_t) wgt1) ii iteration243 … smull v16.4s, v1.4h, v7.h[1] //vmull_n_s16(pi2_src2_val1, (int16_t) wgt1) iii iteration251 … smull v18.4s, v2.4h, v7.h[0] //vmull_n_s16(pi2_src1_val2, (int16_t) wgt0) iv iteration257 … smull v20.4s, v3.4h, v7.h[1] //vmull_n_s16(pi2_src2_val2, (int16_t) wgt1) iv iteration
183 sub v7.8b, v28.8b , v6.8b //32-fract190 umull v24.8h, v12.8b, v7.8b //mul (row 0)200 umull v22.8h, v16.8b, v7.8b //mul (row 1)211 umull v20.8h, v14.8b, v7.8b //mul (row 2)222 umull v18.8h, v23.8b, v7.8b //mul (row 3)233 umull v24.8h, v12.8b, v7.8b //mul (row 4)244 umull v22.8h, v16.8b, v7.8b //mul (row 5)255 umull v20.8h, v14.8b, v7.8b //mul (row 6)259 umull v18.8h, v23.8b, v7.8b //mul (row 7)317 umull v20.8h, v14.8b, v7.8b //mul (row 6)[all …]
303 sub v7.8b, v28.8b , v6.8b //32-fract310 umull v24.8h, v12.8b, v7.8b //mul (row 0)320 umull v22.8h, v16.8b, v7.8b //mul (row 1)331 umull v20.8h, v14.8b, v7.8b //mul (row 2)342 umull v18.8h, v23.8b, v7.8b //mul (row 3)353 umull v24.8h, v12.8b, v7.8b //mul (row 4)364 umull v22.8h, v16.8b, v7.8b //mul (row 5)375 umull v20.8h, v14.8b, v7.8b //mul (row 6)379 umull v18.8h, v23.8b, v7.8b //mul (row 7)438 umull v20.8h, v14.8b, v7.8b //mul (row 6)[all …]
56 #CHECK: 1c67d816 { v22.ub = vsub(v24.ub,{{ *}}v7.ub):sat }57 v22.ub=vsub(v24.ub,v7.ub):sat68 #CHECK: 1c4dc78b { v11.w = vadd(v7.w,{{ *}}v13.w):sat }69 v11.w=vadd(v7.w,v13.w):sat71 #CHECK: 1c48c7a4 { v4.b = vsub(v7.b,{{ *}}v8.b) }72 v4.b=vsub(v7.b,v8.b)80 #CHECK: 1c7ad4a6 { v7:6.h = vadd(v21:20.h,{{ *}}v27:26.h) }81 v7:6.h=vadd(v21:20.h,v27:26.h)83 #CHECK: 1cc7c564 { v4.uw = vabsdiff(v5.w,{{ *}}v7.w) }84 v4.uw=vabsdiff(v5.w,v7.w)[all …]
26 #CHECK: 191de846 { v7:6.w += vtmpy(v9:8.h,r29.b) }27 v7:6.w += vtmpy(v9:8.h,r29.b)38 #CHECK: 1969ff47 { v7.w += vasl(v31.w,r9) }39 v7.w += vasl(v31.w,r9)65 #CHECK: 1948e306 { v7:6.w += vmpy(v3.h,r8.h):sat }66 v7:6.w += vmpy(v3.h,r8.h):sat77 #CHECK: 1946e247 { v7.w += vmpyi(v2.w,r6.b) }78 v7.w += vmpyi(v2.w,r6.b)98 #CHECK: 1c31f026 { v7:6.w += vmpy(v16.h,v17.uh) }99 v7:6.w += vmpy(v16.h,v17.uh)[all …]
175 add v7.4h, v1.4h, v9.4h // x3 = d1 + (d3 >> 1)//179 add v10.4h, v4.4h , v7.4h // x0+x3182 sub v13.4h, v4.4h , v7.4h192 trn2 v7.4h, v12.4h, v13.4h195 trn1 v11.2s, v5.2s, v7.2s // 8197 trn2 v13.2s, v5.2s, v7.2s363 add v7.4h, v1.4h, v9.4h // x3 = d1 + (d3 >> 1)//366 add v10.4h, v4.4h , v7.4h // x0+x3369 sub v13.4h, v4.4h , v7.4h381 trn2 v7.4h, v12.4h, v13.4h[all …]
93 ld2 {v6.8b, v7.8b}, [x0], x1 //D6 = p1u , D7 = p1v101 uaddl v10.8h, v7.8b, v1.8b //Q4,Q5 = q0 + p1107 mov v6.d[1], v7.d[0]119 umlal v28.8h, v7.8b, v31.8b //Q14,Q7 = (X2(p1U) + p0U + q1U)201 ld4 {v4.h, v5.h, v6.h, v7.h}[0], [x0], x1202 ld4 {v4.h, v5.h, v6.h, v7.h}[1], [x0], x1203 ld4 {v4.h, v5.h, v6.h, v7.h}[2], [x0], x1204 ld4 {v4.h, v5.h, v6.h, v7.h}[3], [x0], x1222 mov v6.d[1], v7.d[0]228 uaddl v16.8h, v3.8b, v7.8b //(p0 + q1)[all …]
112 ssubl v7.4s, v0.4h, v3.4h //x3 = x4 - x7115 add v1.4s, v7.4s, v6.4s //pi4_tmp_ptr[1] = x3 + x2117 sub v3.4s, v7.4s, v6.4s //pi4_tmp_ptr[3] = x3 - x2126 trn2 v7.4s, v2.4s, v3.4s130 trn1 v1.2d, v5.2d, v7.2d131 trn2 v3.2d, v5.2d, v7.2d137 sub v7.4s, v0.4s, v3.4s //x3 = x4-x7140 add v1.4s, v7.4s, v6.4s //pi4_tmp_ptr[1] = x3 + x2142 sub v3.4s, v7.4s, v6.4s //pi4_tmp_ptr[3] = x3 - x2
128 ld1 {v6.2s, v7.2s}, [x0], x2 // Vector load from src[3_0]140 uaddl v18.8h, v5.8b, v7.8b // temp3 = src[2_8] + src[3_8]151 uaddl v12.8h, v7.8b, v9.8b168 uaddl v26.8h, v7.8b, v1.8b181 uaddl v18.8h, v7.8b, v5.8b186 ld1 {v6.2s, v7.2s}, [x0], x2192 uaddl v20.8h, v9.8b, v7.8b // temp4 = src[0_8] + src[5_8]210 uaddl v26.8h, v1.8b, v7.8b211 uaddl v18.8h, v5.8b, v7.8b224 uaddl v18.8h, v7.8b, v9.8b[all …]
104 ld1 {v6.8b, v7.8b}, [x0], x1 //p0 values are loaded into q3109 mov v6.d[1], v7.d[0]135 usubl v30.8h, v1.8b, v7.8b //254 ld1 {v6.8b, v7.8b}, [x12] //load p0 to Q3258 mov v6.d[1] , v7.d[0]277 uaddl v26.8h, v5.8b, v7.8b //p0+q0 H353 uaddl v16.8h, v7.8b, v9.8b //p0+q1 H458 ld1 {v7.8b}, [x0], x1 //row12484 trn1 v21.8b, v5.8b, v7.8b485 trn2 v7.8b, v5.8b, v7.8b //row11 & 12[all …]
39 function select(n, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10) { argument47 v6 = v7;48 v7 = v8;56 function select_while(n, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10) { argument65 v6 = v7;66 v7 = v8;
120 ld1 {v7.16b}, [x1], x3125 uabal v30.8h, v6.8b, v7.8b126 uabal2 v30.8h, v6.16b, v7.16b201 ld1 {v7.16b}, [x1], x3206 uabal v30.8h, v6.8b, v7.8b207 uabal2 v30.8h, v6.16b, v7.16b414 ld1 {v7.16b}, [x1], x3423 uabal v31.8h, v7.8b, v8.8b424 uabal2 v31.8h, v7.16b, v8.16b493 ld1 {v7.16b}, [x3], x5[all …]
249 ld1 {v7.8b}, [x14], x2 //// eighth row hence x8 = D7254 urhadd v5.16b, v5.16b , v7.16b //// estimated row 3 = D4, row 7 = D5261 urhadd v7.8b, v7.8b , v8.8b //// estimated row 8 = D7267 st1 {v7.8b}, [x14], x3 //// eighth row hence x8 = D7327 ld1 {v6.8b, v7.8b}, [x14], x2 //row6336 ext v20.8b, v6.8b , v7.8b , #1356 ext v7.8b, v21.8b , v22.8b , #1371 urhadd v7.16b, v7.16b , v21.16b //operate on row6 and row8387 st1 {v7.8b}, [x12], x3 //store row8444 ld1 {v6.8b, v7.8b}, [x14], x2 //row6[all …]