Searched refs:shuffling (Results 1 – 25 of 56) sorted by relevance
123
/external/eigen/bench/tensors/ |
D | tensor_benchmarks_cpu.cc | 45 BM_FuncCPU(shuffling, 4); 46 BM_FuncCPU(shuffling, 8); 47 BM_FuncCPU(shuffling, 12);
|
D | tensor_benchmarks_fp16_gpu.cu | 27 BM_FuncGPU(shuffling);
|
D | tensor_benchmarks_gpu.cu | 27 BM_FuncGPU(shuffling);
|
D | tensor_benchmarks.h | 147 void shuffling(int num_iters) { in shuffling() function
|
/external/tensorflow/tensorflow/core/kernels/mkl/ |
D | mkl_requantization_range_per_channel_op.cc | 65 Eigen::array<int, 2> shuffling({1, 0}); in Compute() local 74 auto transposed_input = input_matrix.shuffle(shuffling); in Compute()
|
/external/curl/tests/data/ |
D | test1608 | 19 verify DNS shuffling
|
/external/tensorflow/tensorflow/core/api_def/base_api/ |
D | api_def_RecordInput.pbtxt | 31 The randomization shuffling buffer.
|
/external/llvm-project/llvm/test/Transforms/GVNSink/ |
D | dither.ll | 4 ; can't sink the zext as we'd need a shuffling PHI in between.
|
/external/llvm-project/llvm/test/Transforms/SLPVectorizer/X86/ |
D | no_alternate_divrem.ll | 53 ;; As such, doing alternate shuffling would be incorrect: 118 ;; As such, doing alternate shuffling would be incorrect:
|
/external/llvm-project/llvm/test/CodeGen/AMDGPU/ |
D | extract_subvector_vec4_vec3.ll | 4 ; We want to see a BUFFER_LOAD, some register shuffling, and a BUFFER_STORE.
|
D | ds_read2_superreg.ll | 113 ; FIXME: Extra moves shuffling superregister 129 ; FIXME: Extra moves shuffling superregister
|
/external/llvm-project/llvm/test/CodeGen/X86/ |
D | avx-cast.ll | 38 ; The next three tests don't need any shuffling. There may or may not be a
|
D | gather-addresses.ll | 10 ; minimize shuffling of each individual element out of the index vector.
|
/external/llvm-project/llvm/test/Transforms/InstCombine/ |
D | shuffle-select-narrow.ll | 86 ; Negative test - mismatched types would require extra shuffling. 101 ; Negative test - mismatched types would require extra shuffling.
|
/external/llvm/test/CodeGen/X86/ |
D | avx-cast.ll | 51 ; The next three tests don't need any shuffling. There may or may not be a
|
D | gather-addresses.ll | 7 ; use an efficient mov/shift sequence rather than shuffling each individual
|
/external/llvm/test/CodeGen/AMDGPU/ |
D | ds_read2_superreg.ll | 113 ; FIXME: Extra moves shuffling superregister 129 ; FIXME: Extra moves shuffling superregister
|
/external/swiftshader/third_party/subzero/docs/ |
D | LOWERING.rst | 43 quality in terms of frame size and register shuffling/spilling, but perhaps not 93 hint would be used to try to reduce the amount of register shuffling.
|
/external/llvm/test/CodeGen/SystemZ/ |
D | fp-copysign-01.ll | 79 ; need any register shuffling here; %a should be tied to %c, with CPSDR
|
/external/llvm-project/llvm/test/CodeGen/SystemZ/ |
D | fp-copysign-01.ll | 79 ; need any register shuffling here; %a should be tied to %c, with CPSDR
|
/external/libvpx/libvpx/third_party/googletest/src/ |
D | CHANGES | 85 * New feature: test shuffling
|
/external/libaom/libaom/third_party/googletest/src/googletest/ |
D | CHANGES | 85 * New feature: test shuffling
|
/external/llvm-project/llvm/lib/Target/AArch64/ |
D | AArch64SchedExynosM3.td | 52 def M3UnitNSHF0 : ProcResource<1>; // Vector shuffling 65 def M3UnitNSHF1 : ProcResource<1>; // Vector shuffling 76 def M3UnitNSHF2 : ProcResource<1>; // Vector shuffling
|
/external/swiftshader/third_party/llvm-10.0/llvm/lib/Target/AArch64/ |
D | AArch64SchedExynosM3.td | 51 def M3UnitNSHF0 : ProcResource<1>; // Vector shuffling 64 def M3UnitNSHF1 : ProcResource<1>; // Vector shuffling 75 def M3UnitNSHF2 : ProcResource<1>; // Vector shuffling
|
/external/skqp/src/compute/hs/ |
D | README.md | 290 intra-warp/inter-lane shuffling ― reenable this capability.
|
123