/external/gemmlowp/fixedpoint/ |
D | fixedpoint_msa.h | 26 struct FixedPointRawTypeTraits<v4i32> { 38 inline v4i32 BitAnd(v4i32 a, v4i32 b) { 39 return reinterpret_cast<v4i32>(__builtin_msa_and_v(reinterpret_cast<v16u8>(a), 50 inline v4i32 BitOr(v4i32 a, v4i32 b) { 51 return reinterpret_cast<v4i32>(__builtin_msa_or_v(reinterpret_cast<v16u8>(a), 62 inline v4i32 BitXor(v4i32 a, v4i32 b) { 63 return reinterpret_cast<v4i32>(__builtin_msa_xor_v(reinterpret_cast<v16u8>(a), 74 inline v4i32 BitNot(v4i32 a) { 75 return reinterpret_cast<v4i32>(__builtin_msa_nor_v(reinterpret_cast<v16u8>(a), 86 inline v4i32 Add(v4i32 a, v4i32 b) { [all …]
|
/external/llvm/test/CodeGen/PowerPC/ |
D | vaddsplat.ll | 8 %v4i32 = type <4 x i32> 12 define void @test_v4i32_pos_even(%v4i32* %P, %v4i32* %S) { 13 %p = load %v4i32, %v4i32* %P 14 %r = add %v4i32 %p, < i32 18, i32 18, i32 18, i32 18 > 15 store %v4i32 %r, %v4i32* %S 23 define void @test_v4i32_neg_even(%v4i32* %P, %v4i32* %S) { 24 %p = load %v4i32, %v4i32* %P 25 %r = add %v4i32 %p, < i32 -28, i32 -28, i32 -28, i32 -28 > 26 store %v4i32 %r, %v4i32* %S 78 define void @test_v4i32_pos_odd(%v4i32* %P, %v4i32* %S) { [all …]
|
/external/libyuv/files/source/ |
D | rotate_msa.cc | 40 out0 = (v16u8)__msa_ilvr_w((v4i32)in1, (v4i32)in0); \ 41 out1 = (v16u8)__msa_ilvl_w((v4i32)in1, (v4i32)in0); \ 42 out2 = (v16u8)__msa_ilvr_w((v4i32)in3, (v4i32)in2); \ 43 out3 = (v16u8)__msa_ilvl_w((v4i32)in3, (v4i32)in2); \ 132 res8 = (v16u8)__msa_ilvr_w((v4i32)reg4, (v4i32)reg0); in TransposeWx16_MSA() 133 res9 = (v16u8)__msa_ilvl_w((v4i32)reg4, (v4i32)reg0); in TransposeWx16_MSA() 137 res8 = (v16u8)__msa_ilvr_w((v4i32)reg5, (v4i32)reg1); in TransposeWx16_MSA() 138 res9 = (v16u8)__msa_ilvl_w((v4i32)reg5, (v4i32)reg1); in TransposeWx16_MSA() 142 res8 = (v16u8)__msa_ilvr_w((v4i32)reg6, (v4i32)reg2); in TransposeWx16_MSA() 143 res9 = (v16u8)__msa_ilvl_w((v4i32)reg6, (v4i32)reg2); in TransposeWx16_MSA() [all …]
|
D | scale_msa.cc | 35 dst0 = (v16u8)__msa_pckod_w((v4i32)src1, (v4i32)src0); in ScaleARGBRowDown2_MSA() 53 vec0 = (v16u8)__msa_pckev_w((v4i32)src1, (v4i32)src0); in ScaleARGBRowDown2Linear_MSA() 54 vec1 = (v16u8)__msa_pckod_w((v4i32)src1, (v4i32)src0); in ScaleARGBRowDown2Linear_MSA() 335 reg0 = (v4u32)__msa_srari_w((v4i32)reg0, 4); in ScaleRowDown4Box_MSA() 336 reg1 = (v4u32)__msa_srari_w((v4i32)reg1, 4); in ScaleRowDown4Box_MSA() 337 reg2 = (v4u32)__msa_srari_w((v4i32)reg2, 4); in ScaleRowDown4Box_MSA() 338 reg3 = (v4u32)__msa_srari_w((v4i32)reg3, 4); in ScaleRowDown4Box_MSA() 370 dst1 = __msa_copy_u_w((v4i32)vec0, 2); in ScaleRowDown38_MSA() 416 vec0 = (v8u16)__msa_pckod_w((v4i32)vec1, (v4i32)vec0); in ScaleRowDown38_2_Box_MSA() 417 vec1 = (v8u16)__msa_pckod_w((v4i32)vec3, (v4i32)vec2); in ScaleRowDown38_2_Box_MSA() [all …]
|
/external/swiftshader/third_party/LLVM/test/CodeGen/CellSPU/useful-harnesses/ |
D | vecoperations.c | 5 typedef int v4i32 __attribute__((ext_vector_type(4))); typedef 50 void print_v4i32(const char *str, v4i32 v) { in print_v4i32() 76 v4i32 v4i32_shuffle_1(v4i32 a) { in v4i32_shuffle_1() 77 v4i32 c2 = a.yzwx; in v4i32_shuffle_1() 81 v4i32 v4i32_shuffle_2(v4i32 a) { in v4i32_shuffle_2() 82 v4i32 c2 = a.zwxy; in v4i32_shuffle_2() 86 v4i32 v4i32_shuffle_3(v4i32 a) { in v4i32_shuffle_3() 87 v4i32 c2 = a.wxyz; in v4i32_shuffle_3() 91 v4i32 v4i32_shuffle_4(v4i32 a) { in v4i32_shuffle_4() 92 v4i32 c2 = a.xyzw; in v4i32_shuffle_4() [all …]
|
/external/llvm/lib/Target/PowerPC/ |
D | PPCInstrAltivec.td | 407 [(int_ppc_altivec_mtvscr v4i32:$vB)]>; 418 [(set v4i32:$vD, (int_ppc_altivec_lvewx xoaddr:$src))]>; 421 [(set v4i32:$vD, (int_ppc_altivec_lvx xoaddr:$src))]>; 424 [(set v4i32:$vD, (int_ppc_altivec_lvxl xoaddr:$src))]>; 445 [(int_ppc_altivec_stvewx v4i32:$rS, xoaddr:$dst)]>; 448 [(int_ppc_altivec_stvx v4i32:$rS, xoaddr:$dst)]>; 451 [(int_ppc_altivec_stvxl v4i32:$rS, xoaddr:$dst)]>; 475 v4i32, v4i32, v16i8>; 476 def VSEL : VA1a_Int_Ty<42, "vsel", int_ppc_altivec_vsel, v4i32>; 498 [(set v4i32:$vD, (add v4i32:$vA, v4i32:$vB))]>; [all …]
|
/external/swiftshader/third_party/LLVM/lib/Target/CellSPU/ |
D | CellSDKIntrinsics.td | 43 [(set (v4i32 VECREG:$rT), (int_spu_si_mpy (v8i16 VECREG:$rA), 49 [(set (v4i32 VECREG:$rT), (int_spu_si_mpyu (v8i16 VECREG:$rA), 55 [(set (v4i32 VECREG:$rT), (int_spu_si_mpyi (v8i16 VECREG:$rA), 61 [(set (v4i32 VECREG:$rT), (int_spu_si_mpyui (v8i16 VECREG:$rA), 67 [(set (v4i32 VECREG:$rT), (int_spu_si_mpya (v8i16 VECREG:$rA), 74 [(set (v4i32 VECREG:$rT), (int_spu_si_mpyh (v4i32 VECREG:$rA), 80 [(set (v4i32 VECREG:$rT), (int_spu_si_mpys (v8i16 VECREG:$rA), 86 [(set (v4i32 VECREG:$rT), (int_spu_si_mpyhh (v8i16 VECREG:$rA), 92 [(set (v4i32 VECREG:$rT), (int_spu_si_mpyhha (v8i16 VECREG:$rA), 100 [(set (v4i32 VECREG:$rT), (int_spu_si_mpyhhu (v8i16 VECREG:$rA), [all …]
|
D | SPUInstrInfo.td | 62 def v4i32: LoadDFormVec<v4i32>; 94 def v4i32: LoadAFormVec<v4i32>; 126 def v4i32: LoadXFormVec<v4i32>; 174 def v4i32: StoreDFormVec<v4i32>; 204 def v4i32: StoreAFormVec<v4i32>; 236 def v4i32: StoreXFormVec<v4i32>; 283 [(set (v4i32 VECREG:$rT), (SPUshufmask dform2_addr:$src))]>; 287 [(set (v4i32 VECREG:$rT), (SPUshufmask xform_addr:$src))]>; 351 def v4i32: ILVecInst<v4i32, s16imm_i32, v4i32SExt16Imm>; 378 def v4i32: ILHUVecInst<v4i32, u16imm_i32, immILHUvec>; [all …]
|
/external/webp/src/dsp/ |
D | msa_macro.h | 27 #define ADDVI_W(a, b) __msa_addvi_w((v4i32)a, b) 30 #define SRAI_W(a, b) __msa_srai_w((v4i32)a, b) 32 #define SLLI_B(a, b) __msa_slli_b((v4i32)a, b) 58 #define LD_SW(...) LD_W(v4i32, __VA_ARGS__) 70 #define ST_SW(...) ST_W(v4i32, __VA_ARGS__) 275 #define LD_SW2(...) LD_W2(v4i32, __VA_ARGS__) 282 #define LD_SW3(...) LD_W3(v4i32, __VA_ARGS__) 289 #define LD_SW4(...) LD_W4(v4i32, __VA_ARGS__) 329 #define ST_SW2(...) ST_W2(v4i32, __VA_ARGS__) 336 #define ST_SW3(...) ST_W3(v4i32, __VA_ARGS__) [all …]
|
/external/libvpx/libvpx/vp8/common/mips/msa/ |
D | vp8_macros_msa.h | 29 #define LD_SW(...) LD_W(v4i32, __VA_ARGS__) 40 #define ST_SW(...) ST_W(v4i32, __VA_ARGS__) 433 out0_m = __msa_copy_u_w((v4i32)in0, idx0); \ 434 out1_m = __msa_copy_u_w((v4i32)in0, idx1); \ 435 out2_m = __msa_copy_u_w((v4i32)in1, idx2); \ 436 out3_m = __msa_copy_u_w((v4i32)in1, idx3); \ 652 #define DOTP_SH4_SW(...) DOTP_SH4(v4i32, __VA_ARGS__) 666 out0 = (RTYPE)__msa_dotp_s_d((v4i32)mult0, (v4i32)cnst0); \ 667 out1 = (RTYPE)__msa_dotp_s_d((v4i32)mult1, (v4i32)cnst1); \ 708 out0 = (RTYPE)__msa_dpadd_s_w((v4i32)out0, (v8i16)mult0, (v8i16)cnst0); \ [all …]
|
/external/libvpx/libvpx/vpx_dsp/mips/ |
D | macros_msa.h | 24 #define LD_SW(...) LD_V(v4i32, __VA_ARGS__) 30 #define ST_SW(...) ST_V(v4i32, __VA_ARGS__) 243 #define LD_SW2(...) LD_V2(v4i32, __VA_ARGS__) 321 #define ST_SW2(...) ST_V2(v4i32, __VA_ARGS__) 378 out0_m = __msa_copy_u_w((v4i32)in, 0); \ 379 out1_m = __msa_copy_u_w((v4i32)in, 1); \ 401 out0_m = __msa_copy_u_w((v4i32)in0, idx0); \ 402 out1_m = __msa_copy_u_w((v4i32)in0, idx1); \ 403 out2_m = __msa_copy_u_w((v4i32)in1, idx2); \ 404 out3_m = __msa_copy_u_w((v4i32)in1, idx3); \ [all …]
|
/external/llvm/test/CodeGen/AMDGPU/ |
D | llvm.amdgcn.image.atomic.ll | 10 …%orig = call i32 @llvm.amdgcn.image.atomic.swap.v4i32(i32 %2, <4 x i32> %1, <8 x i32> %0, i1 0, i1… 44 …%orig = call i32 @llvm.amdgcn.image.atomic.cmpswap.v4i32(i32 %2, i32 %3, <4 x i32> %1, <8 x i32> %… 55 …%orig = call i32 @llvm.amdgcn.image.atomic.add.v4i32(i32 %2, <4 x i32> %1, <8 x i32> %0, i1 0, i1 … 66 …%orig = call i32 @llvm.amdgcn.image.atomic.sub.v4i32(i32 %2, <4 x i32> %1, <8 x i32> %0, i1 0, i1 … 92 …%t0 = call i32 @llvm.amdgcn.image.atomic.smin.v4i32(i32 %2, <4 x i32> %1, <8 x i32> %0, i1 0, i1 0… 93 …%t1 = call i32 @llvm.amdgcn.image.atomic.umin.v4i32(i32 %t0, <4 x i32> %1, <8 x i32> %0, i1 0, i1 … 94 …%t2 = call i32 @llvm.amdgcn.image.atomic.smax.v4i32(i32 %t1, <4 x i32> %1, <8 x i32> %0, i1 0, i1 … 95 …%t3 = call i32 @llvm.amdgcn.image.atomic.umax.v4i32(i32 %t2, <4 x i32> %1, <8 x i32> %0, i1 0, i1 … 96 …%t4 = call i32 @llvm.amdgcn.image.atomic.and.v4i32(i32 %t3, <4 x i32> %1, <8 x i32> %0, i1 0, i1 0… 97 …%t5 = call i32 @llvm.amdgcn.image.atomic.or.v4i32(i32 %t4, <4 x i32> %1, <8 x i32> %0, i1 0, i1 0,… [all …]
|
D | llvm.SI.image.sample.ll | 9 …%r = call <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> unde… 23 …%r = call <4 x float> @llvm.SI.image.sample.cl.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> u… 37 …%r = call <4 x float> @llvm.SI.image.sample.d.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> un… 51 …%r = call <4 x float> @llvm.SI.image.sample.d.cl.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32>… 65 …%r = call <4 x float> @llvm.SI.image.sample.l.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> un… 79 …%r = call <4 x float> @llvm.SI.image.sample.b.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> un… 93 …%r = call <4 x float> @llvm.SI.image.sample.b.cl.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32>… 107 …%r = call <4 x float> @llvm.SI.image.sample.lz.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> u… 121 …%r = call <4 x float> @llvm.SI.image.sample.cd.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> u… 135 …%r = call <4 x float> @llvm.SI.image.sample.cd.cl.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32… [all …]
|
D | llvm.SI.image.sample.o.ll | 9 …%r = call <4 x float> @llvm.SI.image.sample.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> un… 23 …%r = call <4 x float> @llvm.SI.image.sample.cl.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32>… 37 …%r = call <4 x float> @llvm.SI.image.sample.d.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> … 51 …%r = call <4 x float> @llvm.SI.image.sample.d.cl.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i3… 65 …%r = call <4 x float> @llvm.SI.image.sample.l.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> … 79 …%r = call <4 x float> @llvm.SI.image.sample.b.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> … 93 …%r = call <4 x float> @llvm.SI.image.sample.b.cl.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i3… 107 …%r = call <4 x float> @llvm.SI.image.sample.lz.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32>… 121 …%r = call <4 x float> @llvm.SI.image.sample.cd.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32>… 135 …%r = call <4 x float> @llvm.SI.image.sample.cd.cl.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i… [all …]
|
/external/llvm/test/Transforms/InstCombine/ |
D | 2012-04-23-Neon-Intrinsics.ll | 5 …%a = tail call <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16> %x, <4 x i16> zeroinitializer) noun… 13 …%a = tail call <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16> %x, <4 x i16> <i16 1, i16 1, i16 1,… 22 …%a = tail call <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16> <i16 3, i16 3, i16 3, i16 3>, <4 x … 30 …%b = tail call <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16> <i16 -1, i16 -1, i16 -1, i16 -1>, <… 38 …%b = tail call <4 x i32> @llvm.arm.neon.vmullu.v4i32(<4 x i16> <i16 -1, i16 -1, i16 -1, i16 -1>, <… 46 …%a = tail call <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16> <i16 2, i16 2, i16 2, i16 2>, <4 x … 50 ; CHECK-NEXT: %a = tail call <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16> <i16 2, i16 2, i16 2, … 56 …%a = tail call <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16> <i16 3, i16 3, i16 3, i16 3>, <4 x … 64 declare <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16>, <4 x i16>) nounwind readnone 65 declare <4 x i32> @llvm.arm.neon.vmullu.v4i32(<4 x i16>, <4 x i16>) nounwind readnone [all …]
|
/external/llvm/test/Analysis/CostModel/X86/ |
D | scalarize.ll | 13 declare %i4 @llvm.bswap.v4i32(%i4) 16 declare %i4 @llvm.cttz.v4i32(%i4) 24 ; CHECK32: cost of 1 {{.*}}bswap.v4i32 25 ; CHECK64: cost of 1 {{.*}}bswap.v4i32 26 %r2 = call %i4 @llvm.bswap.v4i32(%i4 undef) 31 ; CHECK32: cost of 12 {{.*}}cttz.v4i32 32 ; CHECK64: cost of 12 {{.*}}cttz.v4i32 33 %r4 = call %i4 @llvm.cttz.v4i32(%i4 undef)
|
/external/gemmlowp/internal/ |
D | output_msa.h | 40 v4i32 tmp = __builtin_msa_sat_s_w(input.reg[0], 8); 42 tmp = reinterpret_cast<v4i32>(__builtin_msa_pckev_h( 51 tmp = reinterpret_cast<v4i32>(__builtin_msa_pckev_b( 73 v4i32 tmp_lo = __builtin_msa_sat_s_w(input.reg[0], 8); 74 v4i32 tmp_hi = __builtin_msa_sat_s_w(input.reg[1], 8); 77 tmp_lo = reinterpret_cast<v4i32>(__builtin_msa_pckev_h( 86 tmp_lo = reinterpret_cast<v4i32>(__builtin_msa_pckev_b( 97 v4i32 tmp0 = __builtin_msa_sat_s_w(in0, 8); \ 98 v4i32 tmp1 = __builtin_msa_sat_s_w(in1, 8); \ 99 v4i32 tmp2 = __builtin_msa_sat_s_w(in2, 8); \ [all …]
|
D | pack_msa.h | 155 v4i32 sums_of_4_cells[kCells][4]; in Pack() 158 v4i32 tmp0 = reinterpret_cast<v4i32>(__builtin_msa_ilvr_h( in Pack() 160 v4i32 tmp1 = reinterpret_cast<v4i32>(__builtin_msa_ilvl_h( in Pack() 167 v4i32 s01 = __builtin_msa_addv_w(sums_of_4_cells[cell][0], in Pack() 169 v4i32 s23 = __builtin_msa_addv_w(sums_of_4_cells[cell][2], in Pack() 171 v4i32 s = __builtin_msa_addv_w(s01, s23); in Pack() 174 v4i32 tmp = __builtin_msa_ld_w(sums_of_each_slice_ptr, 0); in Pack() 330 v4i32 sums_of_16[kCells]; in Pack() 333 sums_of_16[cell] = reinterpret_cast<v4i32>( in Pack() 337 __builtin_msa_addv_w(sums_of_16[cell], reinterpret_cast<v4i32>(tmp)); in Pack() [all …]
|
/external/libvpx/libvpx/vp8/encoder/mips/msa/ |
D | dct_msa.c | 48 v4i32 tmp0_m; \ 49 v4i32 one_m = __msa_ldi_w(1); \ 60 v4i32 tmp0_m; \ 62 v4i32 one_m = __msa_ldi_w(1); \ 74 v4i32 out0, out1, out2, out3; in vp8_short_fdct4x4_msa() 87 out0 = (v4i32)__msa_ilvev_h(zero, in1); in vp8_short_fdct4x4_msa() 89 out1 = __msa_splati_w((v4i32)coeff, 0); in vp8_short_fdct4x4_msa() 106 out1 = __msa_splati_w((v4i32)coeff, 1); in vp8_short_fdct4x4_msa() 110 out1 += (v4i32)temp1; in vp8_short_fdct4x4_msa() 121 v4i32 vec0_w, vec1_w, vec2_w, vec3_w; in vp8_short_fdct8x4_msa() [all …]
|
/external/clang/test/CodeGen/ |
D | mips-vector-arg.c | 9 typedef int v4i32 __attribute__ ((__vector_size__ (16))); typedef 24 extern test_v4i32_2(v4i32, int, v4i32); 25 void test_v4i32(v4i32 a1, int a2, v4i32 a3) { in test_v4i32()
|
/external/llvm/lib/Target/X86/ |
D | X86TargetTransformInfo.cpp | 151 { ISD::SHL, MVT::v4i32, 1 }, in getArithmeticInstrCost() 152 { ISD::SRL, MVT::v4i32, 1 }, in getArithmeticInstrCost() 153 { ISD::SRA, MVT::v4i32, 1 }, in getArithmeticInstrCost() 184 { ISD::SHL, MVT::v4i32, 1 }, in getArithmeticInstrCost() 185 { ISD::SRL, MVT::v4i32, 2 }, in getArithmeticInstrCost() 186 { ISD::SRA, MVT::v4i32, 2 }, in getArithmeticInstrCost() 250 { ISD::SHL, MVT::v4i32, 1 }, // pslld in getArithmeticInstrCost() 259 { ISD::SRL, MVT::v4i32, 1 }, // psrld. in getArithmeticInstrCost() 268 { ISD::SRA, MVT::v4i32, 1 }, // psrad. in getArithmeticInstrCost() 275 { ISD::SDIV, MVT::v4i32, 19 }, // pmuludq sequence in getArithmeticInstrCost() [all …]
|
/external/llvm/test/CodeGen/ARM/ |
D | vcvt-v8.ll | 6 %tmp2 = call <4 x i32> @llvm.arm.neon.vcvtas.v4i32.v4f32(<4 x float> %tmp1) 22 %tmp2 = call <4 x i32> @llvm.arm.neon.vcvtns.v4i32.v4f32(<4 x float> %tmp1) 38 %tmp2 = call <4 x i32> @llvm.arm.neon.vcvtps.v4i32.v4f32(<4 x float> %tmp1) 54 %tmp2 = call <4 x i32> @llvm.arm.neon.vcvtms.v4i32.v4f32(<4 x float> %tmp1) 70 %tmp2 = call <4 x i32> @llvm.arm.neon.vcvtau.v4i32.v4f32(<4 x float> %tmp1) 86 %tmp2 = call <4 x i32> @llvm.arm.neon.vcvtnu.v4i32.v4f32(<4 x float> %tmp1) 102 %tmp2 = call <4 x i32> @llvm.arm.neon.vcvtpu.v4i32.v4f32(<4 x float> %tmp1) 118 %tmp2 = call <4 x i32> @llvm.arm.neon.vcvtmu.v4i32.v4f32(<4 x float> %tmp1) 130 declare <4 x i32> @llvm.arm.neon.vcvtas.v4i32.v4f32(<4 x float>) nounwind readnone 132 declare <4 x i32> @llvm.arm.neon.vcvtns.v4i32.v4f32(<4 x float>) nounwind readnone [all …]
|
/external/llvm/test/Transforms/EarlyCSE/AArch64/ |
D | intrinsics.ll | 8 ; CHECK-NOT: call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8 25 call void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32> %3, <4 x i32> %4, i8* %0) 27 %vld2 = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8(i8* %5) 42 ; CHECK-NOT: call void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32> %3, <4 x i32> %3, i8* %0) 43 ; CHECK: call void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32> %3, <4 x i32> %4, i8* %0) 60 call void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32> %3, <4 x i32> %3, i8* %0) 61 call void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32> %3, <4 x i32> %4, i8* %0) 63 %vld2 = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8(i8* %5) 78 ; CHECK: call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8 79 ; CHECK-NOT: call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8 [all …]
|
/external/llvm/lib/Target/AMDGPU/ |
D | SITypeRewriter.cpp | 39 Type *v4i32; member in __anonac5d0d840111::SITypeRewriter 60 v4i32 = VectorType::get(Type::getInt32Ty(M.getContext()), 4); in doInitialization() 81 PointerType::get(v4i32,PtrTy->getPointerAddressSpace())); in visitLoadInst() 108 Args.push_back(Builder.CreateBitCast(Arg, v4i32)); in visitCallInst() 109 Types.push_back(v4i32); in visitCallInst() 144 if (I.getDestTy() != v4i32) { in visitBitCast() 149 if (Op->getSrcTy() == v4i32) { in visitBitCast()
|
/external/llvm/test/CodeGen/AArch64/ |
D | arm64-neon-copyPhysReg-tuple.ll | 10 …%vld = tail call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2lane.v4i32.p0i32(<4 x i32> <i32 -1… 12 …%vld1 = tail call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2lane.v4i32.p0i32(<4 x i32> %extra… 24 …%vld = tail call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld3lane.v4i32.p0i32(<4 x i… 26 …%vld1 = tail call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld3lane.v4i32.p0i32(<4 x … 39 …%vld = tail call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld4lane.v4i32.p… 41 …%vld1 = tail call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld4lane.v4i32.… 46 declare { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2lane.v4i32.p0i32(<4 x i32>, <4 x i32>, i64, … 47 declare { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld3lane.v4i32.p0i32(<4 x i32>, <4 x … 48 declare { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld4lane.v4i32.p0i32(<4 x …
|