/external/llvm-project/llvm/test/CodeGen/Thumb2/ |
D | mve-shuffleext.ll | 12 …%strided.vec = shufflevector <8 x i16> %src, <8 x i16> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 … 13 %out = sext <4 x i16> %strided.vec to <4 x i32> 23 …%strided.vec = shufflevector <8 x i16> %src, <8 x i16> undef, <4 x i32> <i32 1, i32 3, i32 5, i32 … 24 %out = sext <4 x i16> %strided.vec to <4 x i32> 35 …%strided.vec = shufflevector <16 x i16> %src, <16 x i16> undef, <8 x i32> <i32 0, i32 2, i32 4, i3… 36 %out = sext <8 x i16> %strided.vec to <8 x i32> 47 …%strided.vec = shufflevector <16 x i16> %src, <16 x i16> undef, <8 x i32> <i32 1, i32 3, i32 5, i3… 48 %out = sext <8 x i16> %strided.vec to <8 x i32> 58 …%strided.vec = shufflevector <8 x i16> %src, <8 x i16> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 … 59 %out = zext <4 x i16> %strided.vec to <4 x i32> [all …]
|
D | mve-vcvt16.ll | 77 …%strided.vec = shufflevector <4 x float> %src1, <4 x float> %src2, <8 x i32> <i32 0, i32 4, i32 1,… 78 %out = fptrunc <8 x float> %strided.vec to <8 x half> 90 …%strided.vec = shufflevector <4 x float> %src1, <4 x float> %src2, <8 x i32> <i32 4, i32 0, i32 5,… 91 %out = fptrunc <8 x float> %strided.vec to <8 x half> 104 …%strided.vec = shufflevector <8 x float> %src1, <8 x float> %src2, <16 x i32> <i32 0, i32 8, i32 1… 105 %out = fptrunc <16 x float> %strided.vec to <16 x half> 120 …%strided.vec = shufflevector <8 x float> %src1, <8 x float> %src2, <16 x i32> <i32 8, i32 0, i32 9… 121 %out = fptrunc <16 x float> %strided.vec to <16 x half> 320 …%strided.vec = shufflevector <4 x float> %val1, <4 x float> %val2, <8 x i32> <i32 0, i32 4, i32 1,… 321 %out = fptrunc <8 x float> %strided.vec to <8 x half> [all …]
|
/external/llvm-project/polly/docs/experiments/matmul/ |
D | matmul.polly.interchanged+tiled+vector.ll | 191 …%strided.vec = shufflevector <16 x float> %wide.vec21, <16 x float> undef, <4 x i32> <i32 0, i32 4… 192 …%strided.vec11 = shufflevector <16 x float> %wide.vec21, <16 x float> undef, <4 x i32> <i32 1, i32… 193 …%strided.vec12 = shufflevector <16 x float> %wide.vec21, <16 x float> undef, <4 x i32> <i32 2, i32… 194 …%strided.vec13 = shufflevector <16 x float> %wide.vec21, <16 x float> undef, <4 x i32> <i32 3, i32… 198 …%strided.vec15 = shufflevector <16 x float> %wide.vec14, <16 x float> undef, <4 x i32> <i32 0, i32… 199 …%strided.vec16 = shufflevector <16 x float> %wide.vec14, <16 x float> undef, <4 x i32> <i32 1, i32… 200 …%strided.vec17 = shufflevector <16 x float> %wide.vec14, <16 x float> undef, <4 x i32> <i32 2, i32… 201 …%strided.vec18 = shufflevector <16 x float> %wide.vec14, <16 x float> undef, <4 x i32> <i32 3, i32… 202 %11 = fmul <4 x float> %broadcast.splat20, %strided.vec15 203 %12 = fadd <4 x float> %strided.vec, %11 [all …]
|
D | matmul.polly.interchanged+tiled+vector+openmp.ll | 272 …%strided.vec = shufflevector <16 x float> %wide.vec15, <16 x float> undef, <4 x i32> <i32 0, i32 4… 273 …%strided.vec5 = shufflevector <16 x float> %wide.vec15, <16 x float> undef, <4 x i32> <i32 1, i32 … 274 …%strided.vec6 = shufflevector <16 x float> %wide.vec15, <16 x float> undef, <4 x i32> <i32 2, i32 … 275 …%strided.vec7 = shufflevector <16 x float> %wide.vec15, <16 x float> undef, <4 x i32> <i32 3, i32 … 279 …%strided.vec9 = shufflevector <16 x float> %wide.vec8, <16 x float> undef, <4 x i32> <i32 0, i32 4… 280 …%strided.vec10 = shufflevector <16 x float> %wide.vec8, <16 x float> undef, <4 x i32> <i32 1, i32 … 281 …%strided.vec11 = shufflevector <16 x float> %wide.vec8, <16 x float> undef, <4 x i32> <i32 2, i32 … 282 …%strided.vec12 = shufflevector <16 x float> %wide.vec8, <16 x float> undef, <4 x i32> <i32 3, i32 … 283 %16 = fmul <4 x float> %broadcast.splat14, %strided.vec9 284 %17 = fadd <4 x float> %strided.vec, %16 [all …]
|
/external/llvm/test/CodeGen/PowerPC/ |
D | qpx-unal-cons-lds.ll | 53 …%strided.vec = shufflevector <8 x double> %wide.vec, <8 x double> undef, <4 x i32> <i32 0, i32 2, … 54 …%3 = fadd <4 x double> %strided.vec, <double 1.000000e+00, double 1.000000e+00, double 1.000000e+0… 63 …%strided.vec.1 = shufflevector <8 x double> %wide.vec.1, <8 x double> undef, <4 x i32> <i32 0, i32… 64 …%9 = fadd <4 x double> %strided.vec.1, <double 1.000000e+00, double 1.000000e+00, double 1.000000e… 73 …%strided.vec.2 = shufflevector <8 x double> %wide.vec.2, <8 x double> undef, <4 x i32> <i32 0, i32… 74 …%15 = fadd <4 x double> %strided.vec.2, <double 1.000000e+00, double 1.000000e+00, double 1.000000… 83 …%strided.vec.3 = shufflevector <8 x double> %wide.vec.3, <8 x double> undef, <4 x i32> <i32 0, i32… 84 …%21 = fadd <4 x double> %strided.vec.3, <double 1.000000e+00, double 1.000000e+00, double 1.000000… 93 …%strided.vec.4 = shufflevector <8 x double> %wide.vec.4, <8 x double> undef, <4 x i32> <i32 0, i32… 94 …%27 = fadd <4 x double> %strided.vec.4, <double 1.000000e+00, double 1.000000e+00, double 1.000000… [all …]
|
D | p8altivec-shuffles-pred.ll | 8 %strided.vec = shufflevector <4 x i32> %wide.vec, <4 x i32> undef, <2 x i32> <i32 0, i32 2> 9 ret <2 x i32> %strided.vec 20 …%strided.vec = shufflevector <16 x i8> %wide.vec, <16 x i8> undef, <16 x i32> <i32 undef, i32 unde… 21 ret <16 x i8> %strided.vec
|
/external/llvm/test/CodeGen/AArch64/ |
D | aarch64-interleaved-accesses.ll | 10 …%strided.v0 = shufflevector <16 x i8> %wide.vec, <16 x i8> undef, <8 x i32> <i32 0, i32 2, i32 4, … 11 …%strided.v1 = shufflevector <16 x i8> %wide.vec, <16 x i8> undef, <8 x i32> <i32 1, i32 3, i32 5, … 12 %add = add nsw <8 x i8> %strided.v0, %strided.v1 23 …%strided.v2 = shufflevector <12 x i32> %wide.vec, <12 x i32> undef, <4 x i32> <i32 2, i32 5, i32 8… 24 …%strided.v1 = shufflevector <12 x i32> %wide.vec, <12 x i32> undef, <4 x i32> <i32 1, i32 4, i32 7… 25 %add = add nsw <4 x i32> %strided.v2, %strided.v1 36 …%strided.v0 = shufflevector <16 x i32> %wide.vec, <16 x i32> undef, <4 x i32> <i32 0, i32 4, i32 8… 37 …%strided.v2 = shufflevector <16 x i32> %wide.vec, <16 x i32> undef, <4 x i32> <i32 2, i32 6, i32 1… 38 %add = add nsw <4 x i32> %strided.v0, %strided.v2 88 %strided.v0 = shufflevector <4 x i32*> %wide.vec, <4 x i32*> undef, <2 x i32> <i32 0, i32 2> [all …]
|
/external/llvm-project/llvm/test/CodeGen/AArch64/ |
D | falkor-hwpf-fix.mir | 15 $w2 = LDRWui $x1, 0 :: ("aarch64-strided-access" load 4) 36 $q2 = LD1i64 $q2, 0, $x1 :: ("aarch64-strided-access" load 4) 57 $q2 = LD1i8 $q2, 0, $x1 :: ("aarch64-strided-access" load 4) 78 $d2 = LD1Onev1d $x1 :: ("aarch64-strided-access" load 4) 99 $d2_d3 = LD1Twov1d $x1 :: ("aarch64-strided-access" load 4) 120 $q2, $q3 = LDPQi $x1, 3 :: ("aarch64-strided-access" load 4) 141 $x2, $x3 = LDPXi $x1, 3 :: ("aarch64-strided-access" load 4) 166 $x1, $w2 = LDRWpost $x1, 0 :: ("aarch64-strided-access" load 4) 188 $x1, $q2 = LD1i64_POST $q2, 0, $x1, $x1 :: ("aarch64-strided-access" load 4) 210 $x1, $q2 = LD1i8_POST $q2, 0, $x1, $x1 :: ("aarch64-strided-access" load 4) [all …]
|
D | falkor-hwpf.ll | 4 ; Check that strided access metadata is added to loads in inner loops when compiling for Falkor. 7 ; CHECK: load i32, i32* %gep, align 4, !falkor.strided.access !0 8 ; CHECK: load i32, i32* %gep2, align 4, !falkor.strided.access !0 34 ; Check that outer loop strided load isn't marked. 36 ; CHECK: load i32, i32* %gep, align 4, !falkor.strided.access !0 79 ; Check that non-strided load isn't marked. 81 ; CHECK: load i32, i32* %gep, align 4, !falkor.strided.access !0
|
/external/llvm/test/CodeGen/ARM/ |
D | arm-interleaved-accesses.ll | 10 …%strided.v0 = shufflevector <16 x i8> %wide.vec, <16 x i8> undef, <8 x i32> <i32 0, i32 2, i32 4, … 11 …%strided.v1 = shufflevector <16 x i8> %wide.vec, <16 x i8> undef, <8 x i32> <i32 1, i32 3, i32 5, … 12 %add = add nsw <8 x i8> %strided.v0, %strided.v1 23 %strided.v2 = shufflevector <6 x i32> %wide.vec, <6 x i32> undef, <2 x i32> <i32 2, i32 5> 24 %strided.v1 = shufflevector <6 x i32> %wide.vec, <6 x i32> undef, <2 x i32> <i32 1, i32 4> 25 %add = add nsw <2 x i32> %strided.v2, %strided.v1 37 …%strided.v0 = shufflevector <16 x i32> %wide.vec, <16 x i32> undef, <4 x i32> <i32 0, i32 4, i32 8… 38 …%strided.v2 = shufflevector <16 x i32> %wide.vec, <16 x i32> undef, <4 x i32> <i32 2, i32 6, i32 1… 39 %add = add nsw <4 x i32> %strided.v0, %strided.v2 91 %strided.v0 = shufflevector <4 x i32*> %wide.vec, <4 x i32*> undef, <2 x i32> <i32 0, i32 2> [all …]
|
/external/llvm-project/llvm/test/CodeGen/PowerPC/ |
D | p8altivec-shuffles-pred.ll | 8 %strided.vec = shufflevector <4 x i32> %wide.vec, <4 x i32> undef, <2 x i32> <i32 0, i32 2> 9 ret <2 x i32> %strided.vec 20 …%strided.vec = shufflevector <16 x i8> %wide.vec, <16 x i8> undef, <16 x i32> <i32 undef, i32 unde… 21 ret <16 x i8> %strided.vec
|
/external/llvm-project/llvm/test/CodeGen/MIR/AArch64/ |
D | target-memoperands.mir | 19 ; CHECK: [[LOAD1:%[0-9]+]]:_(s32) = G_LOAD [[COPY]](p0) :: ("aarch64-strided-access" load 4) 21 ; CHECK: G_STORE [[LOAD1]](s32), [[COPY]](p0) :: ("aarch64-strided-access" store 4) 25 %2:_(s32) = G_LOAD %0(p0) :: ("aarch64-strided-access" load 4) 27 G_STORE %2(s32), %0(p0) :: ("aarch64-strided-access" store 4)
|
/external/llvm-project/llvm/test/CodeGen/AArch64/GlobalISel/ |
D | irtranslator-atomic-metadata.ll | 23 …_ADD:%[0-9]+]]:_(s32) = G_ATOMICRMW_ADD [[COPY]](p0), [[C]] :: ("aarch64-strided-access" load stor… 26 %oldval = atomicrmw add i32* %ptr, i32 1 monotonic, !falkor.strided.access !0 36 …9]+]]:_(s32) = G_ATOMICRMW_ADD [[COPY]](p0), [[C]] :: (volatile "aarch64-strided-access" load stor… 39 %oldval = atomicrmw volatile add i32* %ptr, i32 1 monotonic, !falkor.strided.access !0 65 …= G_ATOMIC_CMPXCHG_WITH_SUCCESS [[COPY]](p0), [[C]], [[C1]] :: ("aarch64-strided-access" load stor… 68 %val_success = cmpxchg i32* %addr, i32 0, i32 1 monotonic monotonic, !falkor.strided.access !0 80 …C_CMPXCHG_WITH_SUCCESS [[COPY]](p0), [[C]], [[C1]] :: (volatile "aarch64-strided-access" load stor… 83 …%val_success = cmpxchg volatile i32* %addr, i32 0, i32 1 monotonic monotonic, !falkor.strided.acce…
|
D | irtranslator-store-metadata.ll | 46 ; CHECK: G_STORE [[C]](s32), [[COPY]](p0) :: ("aarch64-strided-access" store 4 into %ir.ptr) 48 store i32 0, i32* %ptr, align 4, !falkor.strided.access !0
|
/external/llvm-project/llvm/test/CodeGen/X86/ |
D | shuffle-strided-with-offset-256.ll | 33 …%strided.vec = shufflevector <32 x i8> %vec, <32 x i8> undef, <16 x i32> <i32 1, i32 3, i32 5, i32… 34 store <16 x i8> %strided.vec, <16 x i8>* %S 90 …%strided.vec = shufflevector <16 x i16> %vec, <16 x i16> undef, <8 x i32> <i32 1, i32 3, i32 5, i3… 91 store <8 x i16> %strided.vec, <8 x i16>* %S 110 …%strided.vec = shufflevector <8 x i32> %vec, <8 x i32> undef, <4 x i32> <i32 1, i32 3, i32 5, i32 … 111 store <4 x i32> %strided.vec, <4 x i32>* %S 138 …%strided.vec = shufflevector <32 x i8> %vec, <32 x i8> undef, <8 x i32> <i32 1, i32 5, i32 9, i32 … 139 store <8 x i8> %strided.vec, <8 x i8>* %S 166 …%strided.vec = shufflevector <32 x i8> %vec, <32 x i8> undef, <8 x i32> <i32 2, i32 6, i32 10, i32… 167 store <8 x i8> %strided.vec, <8 x i8>* %S [all …]
|
D | shuffle-strided-with-offset-512.ll | 56 …%strided.vec = shufflevector <64 x i8> %vec, <64 x i8> undef, <32 x i32> <i32 1, i32 3, i32 5, i32… 57 store <32 x i8> %strided.vec, <32 x i8>* %S 106 …%strided.vec = shufflevector <32 x i16> %vec, <32 x i16> undef, <16 x i32> <i32 1, i32 3, i32 5, i… 107 store <16 x i16> %strided.vec, <16 x i16>* %S 149 …%strided.vec = shufflevector <16 x i32> %vec, <16 x i32> undef, <8 x i32> <i32 1, i32 3, i32 5, i3… 150 store <8 x i32> %strided.vec, <8 x i32>* %S 173 …%strided.vec = shufflevector <64 x i8> %vec, <64 x i8> undef, <16 x i32> <i32 1, i32 5, i32 9, i32… 174 store <16 x i8> %strided.vec, <16 x i8>* %S 197 …%strided.vec = shufflevector <64 x i8> %vec, <64 x i8> undef, <16 x i32> <i32 2, i32 6, i32 10, i3… 198 store <16 x i8> %strided.vec, <16 x i8>* %S [all …]
|
D | shuffle-strided-with-offset-128.ll | 42 …%strided.vec = shufflevector <16 x i8> %vec, <16 x i8> undef, <8 x i32> <i32 1, i32 3, i32 5, i32 … 43 store <8 x i8> %strided.vec, <8 x i8>* %S 78 …%strided.vec = shufflevector <8 x i16> %vec, <8 x i16> undef, <4 x i32> <i32 1, i32 3, i32 5, i32 … 79 store <4 x i16> %strided.vec, <4 x i16>* %S 102 %strided.vec = shufflevector <4 x i32> %vec, <4 x i32> undef, <2 x i32> <i32 1, i32 3> 103 store <2 x i32> %strided.vec, <2 x i32>* %S 145 …%strided.vec = shufflevector <16 x i8> %vec, <16 x i8> undef, <4 x i32> <i32 1, i32 5, i32 9, i32 … 146 store <4 x i8> %strided.vec, <4 x i8>* %S 184 …%strided.vec = shufflevector <16 x i8> %vec, <16 x i8> undef, <4 x i32> <i32 2, i32 6, i32 10, i32… 185 store <4 x i8> %strided.vec, <4 x i8>* %S [all …]
|
D | shuffle-vs-trunc-512.ll | 66 …%strided.vec = shufflevector <64 x i8> %vec, <64 x i8> undef, <32 x i32> <i32 0, i32 2, i32 4, i32… 67 store <32 x i8> %strided.vec, <32 x i8>* %S 119 %strided.vec = trunc <32 x i16> %bc to <32 x i8> 120 store <32 x i8> %strided.vec, <32 x i8>* %S 132 …%strided.vec = shufflevector <32 x i16> %vec, <32 x i16> undef, <16 x i32> <i32 0, i32 2, i32 4, i… 133 store <16 x i16> %strided.vec, <16 x i16>* %S 146 %strided.vec = trunc <16 x i32> %bc to <16 x i16> 147 store <16 x i16> %strided.vec, <16 x i16>* %S 159 …%strided.vec = shufflevector <16 x i32> %vec, <16 x i32> undef, <8 x i32> <i32 0, i32 2, i32 4, i3… 160 store <8 x i32> %strided.vec, <8 x i32>* %S [all …]
|
D | shuffle-vs-trunc-128.ll | 67 …%strided.vec = shufflevector <16 x i8> %vec, <16 x i8> undef, <8 x i32> <i32 0, i32 2, i32 4, i32 … 68 store <8 x i8> %strided.vec, <8 x i8>* %S 123 %strided.vec = trunc <8 x i16> %bc to <8 x i8> 124 store <8 x i8> %strided.vec, <8 x i8>* %S 177 …%strided.vec = shufflevector <8 x i16> %vec, <8 x i16> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 … 178 store <4 x i16> %strided.vec, <4 x i16>* %S 232 %strided.vec = trunc <4 x i32> %bc to <4 x i16> 233 store <4 x i16> %strided.vec, <4 x i16>* %S 256 %strided.vec = shufflevector <4 x i32> %vec, <4 x i32> undef, <2 x i32> <i32 0, i32 2> 257 store <2 x i32> %strided.vec, <2 x i32>* %S [all …]
|
D | shuffle-vs-trunc-256.ll | 65 …%strided.vec = shufflevector <32 x i8> %vec, <32 x i8> undef, <16 x i32> <i32 0, i32 2, i32 4, i32… 66 store <16 x i8> %strided.vec, <16 x i8>* %S 128 %strided.vec = trunc <16 x i16> %bc to <16 x i8> 129 store <16 x i8> %strided.vec, <16 x i8>* %S 180 …%strided.vec = shufflevector <16 x i16> %vec, <16 x i16> undef, <8 x i32> <i32 0, i32 2, i32 4, i3… 181 store <8 x i16> %strided.vec, <8 x i16>* %S 244 %strided.vec = trunc <8 x i32> %bc to <8 x i16> 245 store <8 x i16> %strided.vec, <8 x i16>* %S 292 …%strided.vec = shufflevector <8 x i32> %vec, <8 x i32> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 … 293 store <4 x i32> %strided.vec, <4 x i32>* %S [all …]
|
/external/llvm-project/llvm/test/CodeGen/ARM/ |
D | crash-on-pow2-shufflevector.ll | 20 %strided.vec = shufflevector <16 x i32> %wide.vec, <16 x i32> undef, <2 x i32> <i32 0, i32 8> 21 %bin.rdx20 = add <2 x i32> %strided.vec, %strided.vec
|
/external/llvm-project/mlir/test/Dialect/Affine/ |
D | memref-stride-calculation.mlir | 54 // CHECK: MemRefType memref<5xf32, affine_map<(d0)[s0] -> (s0)>> cannot be converted to strided form 56 // CHECK: MemRefType memref<5xf32, affine_map<(d0)[s0] -> (123)>> cannot be converted to strided fo… 63 …fine_map<(d0, d1, d2)[s0, s1] -> (d0 * s0 + d1 * s1 + d2 + 1)>> cannot be converted to strided form 65 …4x5xf32, affine_map<(d0, d1, d2) -> (d0 floordiv 4 + d1 + d2)>> cannot be converted to strided form 67 …x4x5xf32, affine_map<(d0, d1, d2) -> (d0 ceildiv 4 + d1 + d2)>> cannot be converted to strided form 69 …ef<3x4x5xf32, affine_map<(d0, d1, d2) -> (d0 mod 4 + d1 + d2)>> cannot be converted to strided form
|
/external/llvm-project/llvm/test/Transforms/InterleavedAccess/X86/ |
D | interleaved-accesses-64bits-avx.ll | 31 …%strided.v0 = shufflevector <16 x double> %wide.vec, <16 x double> undef, <4 x i32> <i32 0, i32 4,… 32 …%strided.v1 = shufflevector <16 x double> %wide.vec, <16 x double> undef, <4 x i32> <i32 1, i32 5,… 33 …%strided.v2 = shufflevector <16 x double> %wide.vec, <16 x double> undef, <4 x i32> <i32 2, i32 6,… 34 …%strided.v3 = shufflevector <16 x double> %wide.vec, <16 x double> undef, <4 x i32> <i32 3, i32 7,… 35 %add1 = fadd <4 x double> %strided.v0, %strided.v1 36 %add2 = fadd <4 x double> %add1, %strided.v2 37 %add3 = fadd <4 x double> %add2, %strided.v3 66 …%strided.v0 = shufflevector <16 x i64> %wide.vec, <16 x i64> undef, <4 x i32> <i32 0, i32 4, i32 8… 67 …%strided.v1 = shufflevector <16 x i64> %wide.vec, <16 x i64> undef, <4 x i32> <i32 1, i32 5, i32 9… 68 …%strided.v2 = shufflevector <16 x i64> %wide.vec, <16 x i64> undef, <4 x i32> <i32 2, i32 6, i32 1… [all …]
|
/external/llvm-project/llvm/test/CodeGen/Hexagon/loop-idiom/ |
D | hexagon-memmove2.ll | 5 ; Ensure that we don't form a memcpy for strided loops. Briefly, when we taught 6 ; LoopIdiom about memmove and strided loops, this got miscompiled into a memcpy
|
/external/llvm-project/llvm/test/Transforms/LoopVectorize/ |
D | interleaved-accesses-uniform-load.ll | 3 ; Make sure the vectorizer can handle this loop: The strided load is only used 35 ; CHECK: %strided.vec = shufflevector <16 x i32> %wide.vec, <16 x i32> undef, <4 x i32> <i32 0, i32…
|