MCPcopy Create free account

hub / github.com/alibaba/BladeDISC / functions

Functions6,088 in github.com/alibaba/BladeDISC

⨍ Functions 6,088◇ Types & classes 1,279↳ Endpoints 6
↓ 1 callersFunctionbuild_mkldnn
(root)
pytorch_blade/common_setup.py:341
↓ 1 callersFunctionbuild_nxgraph_of_nodes
(block)
pytorch_blade/torch_blade/utils.py:77
↓ 1 callersFunctionbuild_onnx_engine
( module, group_id, _try_build_onnx_engine, disable_fallback=False, dynamic_settings=None,
pytorch_blade/torch_blade/onnx_backends/backend_conversion.py:128
↓ 1 callersFunctionbuild_tao_bridge
(root, args)
scripts/python/tao_build.py:480
↓ 1 callersFunctionbuild_tao_compiler
(root, args)
scripts/python/tao_build.py:281
↓ 1 callersFunctionbuild_tao_compiler_add_flags_platform_alibaba
(root, args, flag)
pytorch_blade/common_setup.py:594
↓ 1 callersFunctionbuild_tao_compiler_add_flags_platform_alibaba_cached
(root, flag)
pytorch_blade/common_setup.py:758
↓ 1 callersFunctionbuilder_flags_context
(flags: int)
tensorflow_blade/tf_blade/gpu/tf_to_trt.py:36
↓ 1 callersFunctioncalc_quant_min_max
(bit: int, signed: bool)
tools/torch_quant/torch_quant/observer.py:186
↓ 1 callersMethodcalib_gm
( self, name: str, gm: GraphModule, root: nn.Module, ob_types: ObserverTypes )
tools/torch_quant/torch_quant/quantizer.py:106
↓ 1 callersFunctioncalibrate
(model, dataloader, device)
examples/PyTorch/Quantization/bert_ptq.py:84
↓ 1 callersMethodcall
tao_compiler/mlir/xla/ral/ral_context.cc:106
↓ 1 callersFunctioncanBeHoisted
Checks whether the given op can be hoisted by checking that - the op and none of its contained operations depend on values inside of the loop (by mean
tao_compiler/mlir/disc/transforms/disc_outline_cpu_kernel.cc:228
↓ 1 callersFunctioncanFuseOnCPULegacy
pytorch_blade/pytorch_blade/ltc/disc_compiler/passes/graph_fuser.cpp:1281
↓ 1 callersFunctioncanPropagateShapeByRunningIt
pytorch_blade/pytorch_blade/compiler/jit/torch/shape_analysis.cpp:446
↓ 1 callersFunctioncan_merge
(u, v)
pytorch_blade/torch_blade/clustering/support_fusion_algorithm.py:178
↓ 1 callersFunctioncheck
(args)
scripts/python/bazel_snapshot_client.py:94
↓ 1 callersFunctioncheckGlobalOpContent
Returns true if the globalOp has the same value as `value`.
tao_compiler/mlir/disc/transforms/disc_to_llvm.cc:171
↓ 1 callersFunctioncheckIsOnlyNonShapeUser
tao_compiler/mlir/disc/transforms/disc_transpose_simplifier.cc:630
↓ 1 callersFunctioncheck_init_file_miss
(path, ignore_root=False)
tensorflow_blade/build.py:80
↓ 1 callersFunctioncheck_inputs
( # noqa: C901 simple_graph: SimpleGraph, current_node: SimpleNode, pattern_nodes: Dict[str, Simp
tensorflow_blade/tf_blade/util/tf_graph_transform_util.py:736
↓ 1 callersFunctioncheck_loss
(ref_loss, res_loss)
examples/PyTorch/Train/Lazy/Bert/bert_train_ltc.py:175
↓ 1 callersFunctioncheck_loss
(ref_loss, res_loss)
examples/PyTorch/Train/Dynamo/Bert/test_bert.py:192
↓ 1 callersFunctioncheck_min_max_valid
(min_val: torch.Tensor, max_val: torch.Tensor)
tools/torch_quant/torch_quant/observer.py:37
↓ 1 callersFunctioncheck_results
(results0, results1)
examples/PyTorch/Inference/CPU/albert/main.py:170
↓ 1 callersFunctioncleanUp
tao_compiler/mlir/disc/transforms/disc_shape_optimization.cc:1890
↓ 1 callersMethodclean_tmp
(self)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/foldacc.py:59
↓ 1 callersFunctioncleanup_building_env
(root)
scripts/python/common_setup.py:769
↓ 1 callersFunctioncleanup_building_env
(root)
pytorch_blade/common_setup.py:769
↓ 1 callersMethodclear
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/lru_cache.hpp:96
↓ 1 callersMethodclearError
tao_compiler/mlir/xla/ral/ral_context.cc:180
↓ 1 callersMethodclear_attribute
(self)
pytorch_blade/torch_blade/quantization/prepare_data.py:216
↓ 1 callersMethodclear_buffers
tao/tao_bridge/executable.h:136
↓ 1 callersFunctionclear_class_registry
()
pytorch_blade/torch_blade/testing/common_utils.py:59
↓ 1 callersFunctioncloneFusion
tao_compiler/mlir/disc/transforms/disc_specialize_fusion_with_speculation.cc:94
↓ 1 callersFunctioncloneRegionAndRemapLoad
This method is revised from Region::cloneinto() TODO: any easier ways?
tao_compiler/mlir/disc/transforms/revise_kernel_outlining.cc:146
↓ 1 callersMethodcloneSymbolGroup
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:692
↓ 1 callersFunctioncloneWithBroadcastSimplifying
tao_compiler/mlir/disc/transforms/disc_specialize_fusion_with_speculation.cc:148
↓ 1 callersFunctionclz
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/utils.hpp:145
↓ 1 callersFunctioncmdOptionExists
pytorch_blade/tests/torchscript/shape_analysis_tool.cpp:86
↓ 1 callersMethodcollect
(self, inputs: List[torch.Tensor])
pytorch_blade/torch_blade/quantization/prepare_data.py:36
↓ 1 callersFunctioncollect_engines
Collect all engines in a script_module of the group type
pytorch_blade/torch_blade/tensorrt/__init__.py:40
↓ 1 callersFunctioncollect_engines_in_block
(block)
pytorch_blade/torch_blade/utils.py:174
↓ 1 callersFunctioncompareSymbolicDimProduct
Gives a consistent order of a list op SymbolicDimProducts
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:60
↓ 1 callersMethodcompressed_meta
tao_compiler/mlir/xla/ral/context/cuda_impl_sparse.cc:156
↓ 1 callersMethodcompressed_weight
tao_compiler/mlir/xla/ral/context/cuda_impl_sparse.cc:155
↓ 1 callersFunctioncomputeElementOffsetsFromVectorSliceOffsets
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:1601
↓ 1 callersFunctioncomputeStrides
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:1586
↓ 1 callersFunctioncompute_scales
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/utils.hpp:126
↓ 1 callersFunctionconfig_mkldnn
(root, args)
pytorch_blade/common_setup.py:269
↓ 1 callersFunctionconfigure
(root, args)
scripts/python/tao_build.py:274
↓ 1 callersFunctionconfigureGpuToROCDLConversionLegality
tao_compiler/mlir/disc/transforms/disc_lower_gpu_ops_to_rocdl_ops.cc:124
↓ 1 callersFunctionconfigure_bridge
(root, args)
scripts/python/tao_build.py:174
↓ 1 callersFunctionconfigure_bridge_platform_alibaba
(root, args)
scripts/python/common_setup.py:583
↓ 1 callersFunctionconfigure_compiler
(root, args)
scripts/python/tao_build.py:136
↓ 1 callersFunctionconfigure_compiler_platform_alibaba
(root, args)
scripts/python/common_setup.py:589
↓ 1 callersFunctionconst_unroll
pytorch_blade/pytorch_blade/compiler/jit/torch/const_loop_unroll.cpp:130
↓ 1 callersFunctioncontainsTensorType
pytorch_blade/pytorch_blade/compiler/jit/torch/shape_analysis.cpp:245
↓ 1 callersFunctionconvertQuantizedConv2DWithBiasAndRequantizeOp
tao_compiler/mlir/disc/transforms/lower_tf.cc:1155
↓ 1 callersFunctionconvertToLaunchFuncOp
Replace `parallel` operations with an `disc_ral.cpu_launch` operation launching `kernelFunc`. The kernel func contains the body of the `parallelOp` wi
tao_compiler/mlir/disc/transforms/disc_outline_cpu_kernel.cc:211
↓ 1 callersFunctionconvertValuesIfIntegerType
tao_compiler/mlir/disc/transforms/lhlo_elemental_utils.cc:1381
↓ 1 callersFunctionconvertWorkgroupBuffer
tao_compiler/mlir/disc/transforms/revise_kernel_outlining.cc:357
↓ 1 callersFunctionconvert_chunk_layer
(module, inputs, config, device)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/optimization/torchscript_optimize.py:55
↓ 1 callersFunctioncpu_dealloc
tao_compiler/mlir/xla/ral/context/base/cpu/cpu_context_impl.cc:41
↓ 1 callersFunctioncpu_support_vnni
()
pytorch_blade/tests/disc/pdl/test_e2e/test_quantization.py:26
↓ 1 callersFunctioncreateDiscAssignKernelNamePass
tao_compiler/mlir/disc/transforms/disc_assign_kernel_name.cc:98
↓ 1 callersFunctioncreateDiscAssignMemorySpacePass
tao_compiler/mlir/disc/transforms/disc_assign_memory_space.cc:613
↓ 1 callersFunctioncreateDiscBufferDeallocationPass
tao_compiler/mlir/disc/transforms/disc_buffer_deallocation.cc:116
↓ 1 callersFunctioncreateDiscCompIntensFusionToCUDASourcePass
tao_compiler/mlir/disc/transforms/disc_comp_intens_fusion_to_cuda_source.cc:681
↓ 1 callersFunctioncreateDiscCompIntensFusionToFuncPass
tao_compiler/mlir/disc/transforms/disc_comp_intens_fusion_to_func.cc:114
↓ 1 callersFunctioncreateDiscConstToRALPass
tao_compiler/mlir/disc/transforms/disc_convert_const_to_ral.cc:223
↓ 1 callersFunctioncreateDiscConvRewriter
tao_compiler/mlir/disc/transforms/conv_rewriter.cc:479
↓ 1 callersFunctioncreateDiscConvertFakeQuantOpPass
tao_compiler/mlir/disc/transforms/disc_convert_fake_quant_op.cc:323
↓ 1 callersFunctioncreateDiscConvertForeachThreadOpToParallelOpPass
tao_compiler/mlir/disc/tools/disc-transform/transforms/convert_foreach_thread_op_to_parallel_op.cc:82
↓ 1 callersFunctioncreateDiscConvertHloToStandardPass
tao_compiler/mlir/disc/transforms/disc_hlo_to_std.cc:154
↓ 1 callersFunctioncreateDiscConvertTensorToStandardPass
tao_compiler/mlir/disc/transforms/disc_tensor_to_std.cc:125
↓ 1 callersFunctioncreateDiscCpuMapParallelLoopPass
tao_compiler/mlir/disc/transforms/disc_cpu_map_parallel_loop.cc:170
↓ 1 callersFunctioncreateDiscCustomCallRewriterPass
tao_compiler/mlir/disc/transforms/disc_custom_call_rewriter.cc:148
↓ 1 callersFunctioncreateDiscDenseToSparsePass
tao_compiler/mlir/disc/transforms/disc_dense_to_sparse.cc:551
↓ 1 callersFunctioncreateDiscDotMergePass
tao_compiler/mlir/disc/transforms/disc_dot_merge.cc:901
↓ 1 callersFunctioncreateDiscDotRewriterPass
tao_compiler/mlir/disc/transforms/dot_rewriter.cc:855
↓ 1 callersFunctioncreateDiscDuplicateComputationAfterFusionPass
tao_compiler/mlir/disc/transforms/disc_duplicate_computation_after_fusion.cc:109
↓ 1 callersFunctioncreateDiscDuplicateComputationForFusionPass
tao_compiler/mlir/disc/transforms/disc_duplicate_computation_for_fusion.cc:354
↓ 1 callersFunctioncreateDiscDynamicSliceConverterPass
tao_compiler/mlir/disc/transforms/disc_dynamic_slice_converter.cc:156
↓ 1 callersFunctioncreateDiscElementTypeConverterPass
tao_compiler/mlir/disc/transforms/element_type_converter.cc:265
↓ 1 callersFunctioncreateDiscFlattenMemrefAccessPass
tao_compiler/mlir/disc/transforms/disc_flatten_memref_access.cc:101
↓ 1 callersFunctioncreateDiscFuseSplatConstPass
tao_compiler/mlir/disc/transforms/disc_fuse_splat_const.cc:113
↓ 1 callersFunctioncreateDiscFusionPass
tao_compiler/mlir/disc/transforms/lhlo_fusion.cc:916
↓ 1 callersFunctioncreateDiscGPUSourceToLibPass
tao_compiler/mlir/disc/transforms/disc_gpu_source_to_lib.cc:302
↓ 1 callersFunctioncreateDiscGpuConvPaddingLegalization
tao_compiler/mlir/disc/transforms/gpu_conv_padding.cc:227
↓ 1 callersFunctioncreateDiscInputInlineFusionPass
This pass works after LhloLegalizeRootsToParallelLoops pass for the XLA-style fusion codegen.
tao_compiler/mlir/disc/transforms/input_inline_fusion_pass.cc:59
↓ 1 callersFunctioncreateDiscLegalizeLmhloFusionToLinalgPass
tao_compiler/mlir/disc/tools/disc-transform/transforms/legalize_lmhlo_fusion_to_linalg.cc:445
↓ 1 callersFunctioncreateDiscLegalizeToLhloPass
tao_compiler/mlir/disc/transforms/disc_hlo_legalize_to_lhlo.cc:291
↓ 1 callersFunctioncreateDiscLhloLegalizeRootsToParallelLoopsPass
tao_compiler/mlir/disc/transforms/lhlo_legalize_roots_to_loops.cc:5670
↓ 1 callersFunctioncreateDiscLowerGpuOpsToNVVMOpsPass
tao_compiler/mlir/disc/transforms/disc_lower_gpu_ops_to_nvvm_ops.cc:211
↓ 1 callersFunctioncreateDiscLowerGpuOpsToROCDLOpsPass
tao_compiler/mlir/disc/transforms/disc_lower_gpu_ops_to_rocdl_ops.cc:264
↓ 1 callersFunctioncreateDiscLowerGpuQuantizeAndDequantizePass
tao_compiler/mlir/disc/transforms/disc_lower_quantize_and_dequantize.cc:249
↓ 1 callersFunctioncreateDiscLowerQuantizeAndDequantizePass
tao_compiler/mlir/disc/transforms/disc_lower_quantize_and_dequantize.cc:244
↓ 1 callersFunctioncreateDiscLowerToLibraryCallPass
tao_compiler/mlir/disc/transforms/disc_lower_to_library_call.cc:937
↓ 1 callersFunctioncreateDiscMarkShapeCalcOpPass
tao_compiler/mlir/disc/transforms/mhlo_mark_shape_calc.cc:183
← previousnext →1,901–2,000 of 6,088, ranked by callers