Home
last modified time | relevance | path

Searched refs:MulAdd (Results 1 – 17 of 17) sorted by relevance

/external/tensorflow/tensorflow/compiler/xla/service/cpu/
Dllvm_ir_runtime.cc209 llvm::Value* n = vsl.Floor(vsl.MulAdd(input, cephes_LOG2EF, half)); in GenerateVF32Exp()
257 llvm::Value* z = vsl.MulAdd(x, cephes_exp_p0, cephes_exp_p1); in GenerateVF32Exp()
258 z = vsl.MulAdd(z, x, cephes_exp_p2); in GenerateVF32Exp()
259 z = vsl.MulAdd(z, x, cephes_exp_p3); in GenerateVF32Exp()
260 z = vsl.MulAdd(z, x, cephes_exp_p4); in GenerateVF32Exp()
261 z = vsl.MulAdd(z, x, cephes_exp_p5); in GenerateVF32Exp()
262 z = vsl.MulAdd(z, vsl.Mul(x, x), x); in GenerateVF32Exp()
356 y = vsl.MulAdd(tmp0, cephes_log_p0, cephes_log_p1); in GenerateVF32Log()
357 y1 = vsl.MulAdd(tmp0, cephes_log_p3, cephes_log_p4); in GenerateVF32Log()
358 y2 = vsl.MulAdd(tmp0, cephes_log_p6, cephes_log_p7); in GenerateVF32Log()
[all …]
Dvector_support_library.h87 llvm::Value* MulAdd(llvm::Value* a, llvm::Value* b, llvm::Value* c) { in MulAdd() function
91 llvm::Value* MulAdd(llvm::Value* a, llvm::Value* b, const llvm::APFloat& c) { in MulAdd() function
95 llvm::Value* MulAdd(llvm::Value* a, const llvm::APFloat& b, in MulAdd() function
Dtiled_dot_emitter.cc335 accumulator = vsl_.MulAdd(lhs_tile[i], rhs_tile[i], accumulator); in EmitInnerLoopTiled()
929 vsl->MulAdd(lhs_tile[r_m_i][r_k_i], rhs_tile[r_k_i], in EmitTiledGemm()
/external/gemmlowp/internal/
Dsimd_wrappers_sse.h99 inline void MulAdd(Int32x4 lhs, Int32x4 rhs, Int32x4* acc) {
103 inline void MulAdd(Int32x4 lhs, std::int32_t rhs, Int32x4* acc) {
Dsimd_wrappers_msa.h134 inline void MulAdd(Int32x4 lhs, Int32x4 rhs, Int32x4* acc) {
140 inline void MulAdd(Int32x4 lhs, std::int32_t rhs, Int32x4* acc) {
Dsimd_wrappers_common_neon_sse.h700 MulAdd(lhs.buf.reg[i], p, &acc->buf.reg[i]);
732 MulAdd(lhs.buf.reg[i], p, &acc->buf.reg[i]);
757 MulAdd(lhs.buf.reg[0], rhs.buf.reg[0], &acc->buf.reg[0]);
820 MulAdd(lhs.buf.reg[0], p, &acc->buf.reg[0]);
844 MulAdd(lhs.buf.reg[0], p, &acc->buf.reg[0]);
Dsimd_wrappers_neon.h129 inline void MulAdd(Int32x4 lhs, Int32x4 rhs, Int32x4* acc) {
133 inline void MulAdd(Int32x4 lhs, std::int32_t rhs, Int32x4* acc) {
Dsimd_wrappers.h40 inline void MulAdd(std::int32_t lhs, std::int32_t rhs, std::int32_t* acc) { in MulAdd() function
419 MulAdd(lhs.buf.reg[lhs_r + lhs_c * LhsRows],
/external/tensorflow/tensorflow/core/kernels/
Dsparse_matmul_op.cc409 ALWAYS_INLINE void MulAdd(const Packet a, const bfloat16** binp, float** out) { in MulAdd() function
523 ALWAYS_INLINE void MulAdd(const Packet a, const float** inp, float** out) { in MulAdd() function
732 MulAdd(l, &r, &out); in GEPP()
733 MulAdd(nl, &nr, &nout); in GEPP()
734 MulAdd(n2l, &n2r, &n2out); in GEPP()
735 MulAdd(n3l, &n3r, &n3out); in GEPP()
756 MulAdd(l, &r, &out); in GEPP()
/external/tensorflow/tensorflow/lite/kernels/internal/optimized/
Dneon_tensor_utils.cc1319 inline int64x2x2_t MulAdd(int32x4_t acc, int32x4_t lhs, int32x4_t rhs) { in MulAdd() function
1431 MulAdd(bias_0, rescaled.val[0], layer_norm_weights_s32_0); in NeonApplyLayerNorm()
1433 MulAdd(bias_1, rescaled.val[1], layer_norm_weights_s32_1); in NeonApplyLayerNorm()
1435 MulAdd(bias_2, rescaled.val[2], layer_norm_weights_s32_2); in NeonApplyLayerNorm()
1437 MulAdd(bias_3, rescaled.val[3], layer_norm_weights_s32_3); in NeonApplyLayerNorm()
/external/clang/test/SemaCXX/
Dconstant-expression-cxx11.cpp501 constexpr int MulAdd(int x, int y, int c) { return x * y + c; } in MulAdd() function
502 constexpr int InnerProduct = ZipFoldR(MulAdd, 5, xs, ys, 0);
/external/swiftshader/src/Reactor/
DReactorUnitTests.cpp1366 TEST(ReactorUnitTests, MulAdd) in TEST() argument
1373 MulAdd(Short4(0x1aa, 0x2dd, 0x3ee, 0xF422), in TEST()
DReactor.hpp889 RValue<Int2> MulAdd(RValue<Short4> x, RValue<Short4> y);
1016 RValue<Int4> MulAdd(RValue<Short8> x, RValue<Short8> y);
DSubzeroReactor.cpp2614 RValue<Int2> MulAdd(RValue<Short4> x, RValue<Short4> y) in MulAdd() function
3027 RValue<Int4> MulAdd(RValue<Short8> x, RValue<Short8> y) in MulAdd() function
DLLVMReactor.cpp1991 RValue<Int2> MulAdd(RValue<Short4> x, RValue<Short4> y) in MulAdd() function
2172 RValue<Int4> MulAdd(RValue<Short8> x, RValue<Short8> y) in MulAdd() function
/external/swiftshader/src/Pipeline/
DSamplerCore.cpp1391 uuuu = As<Short4>(MulAdd(uuuu, *Pointer<Short4>(mipmap + OFFSET(Mipmap, onePitchP)))); in computeIndices()
1392 uuu2 = As<Short4>(MulAdd(uuu2, *Pointer<Short4>(mipmap + OFFSET(Mipmap, onePitchP)))); in computeIndices()
/external/swiftshader/src/Shader/
DSamplerCore.cpp1678 uuuu = As<Short4>(MulAdd(uuuu, *Pointer<Short4>(mipmap + OFFSET(Mipmap,onePitchP)))); in computeIndices()
1679 uuu2 = As<Short4>(MulAdd(uuu2, *Pointer<Short4>(mipmap + OFFSET(Mipmap,onePitchP)))); in computeIndices()