MCPcopy Create free account

hub / github.com/alibaba/BladeDISC / functions

Functions6,088 in github.com/alibaba/BladeDISC

↓ 1 callersMethodensureInitialized
tao_compiler/mlir/xla/ral/context/common_context_impl.cc:727
↓ 1 callersFunctionensure_divisibility
Ensure that numerator is divisible by the denominator.
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/distributed/comm.py:19
↓ 1 callersFunctionevaluate
(model, dataloader, device)
tools/torch_quant/bert_ptq_demo.py:58
↓ 1 callersFunctionexpandMemRef
tao_compiler/mlir/disc/transforms/revise_kernel_outlining.cc:283
↓ 1 callersFunctionexport_torch_to_onnx
(model, inputs, output_file: str)
examples/PyTorch/Inference/CUDA/BERT/main.py:148
↓ 1 callersFunctionexport_torch_to_onnx
(model, output_file : str, inputs)
examples/PyTorch/Inference/CPU/albert/main.py:29
↓ 1 callersFunctionextra_acl_patch_dir
(root)
scripts/python/common_setup.py:264
↓ 1 callersFunctionextra_acl_patch_dir
(root)
pytorch_blade/common_setup.py:264
↓ 1 callersFunctionextractBeneficiaryOps
For a given operation `op`, computes whether it is beneficial to sink the operation into the kernel. An operation can be sunk if doing so does not int
tao_compiler/mlir/disc/transforms/disc_outline_cpu_kernel.cc:64
↓ 1 callersFunctionextractConvParams
tao_compiler/mlir/disc/transforms/conv_rewriter.cc:107
↓ 1 callersMethodfallback
(self, model: nn.Module, num: int)
tools/torch_quant/torch_quant/quantizer.py:176
↓ 1 callersFunctionfinal_init_
(weights)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/ops.py:72
↓ 1 callersFunctionfindCandidates
tao_compiler/mlir/disc/transforms/disc_buffer_deallocation.cc:38
↓ 1 callersMethodfindOpsOfSkeletonGroup
tao_compiler/mlir/disc/transforms/fusion_utils.cc:798
↓ 1 callersFunctionfindValidReductionOps
tao_compiler/mlir/disc/transforms/fusion_utils_stitch_gpu.cc:21
↓ 1 callersFunctionflatten_final_dims
(t, no_dims)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/utils.py:57
↓ 1 callersMethodfold
tao_compiler/mlir/disc/tools/disc-transform/LinalgExt/LinalgExtOps.cc:298
↓ 1 callersFunctionformat_package_name
()
tao/setup.py:44
↓ 1 callersMethodforward
(self, m: torch.Tensor, z: torch.Tensor, msa_mask: torch.Tensor, pair_mask: to
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/evoformer.py:248
↓ 1 callersMethodforward
(self, m: torch.Tensor, z: torch.Tensor, msa_mask: Optional[torch.Tensor] = None,
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/evoformer.py:350
↓ 1 callersMethodforward
(self, x)
pytorch_blade/torch_blade/testing/common_utils.py:35
↓ 1 callersMethodforward
(self, x)
pytorch_blade/tests/tensorrt/test_holder_serialize.py:182
↓ 1 callersMethodforward
(self, x)
pytorch_blade/tests/tensorrt/test_optimize.py:41
↓ 1 callersMethodforward
(self, y)
pytorch_blade/tests/disc/ops/test_matmul.py:23
↓ 1 callersFunctionforward_hook
(module, *args, **kwargs)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/optimization/torchscript_optimize.py:120
↓ 1 callersMethodforward_loop
(self, z: torch.Tensor, mask: torch.Tensor, chunk_size: Optional[int] = None, )
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/template.py:236
↓ 1 callersMethodforward_single
(self, z, mask, chunk_size=None)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/template.py:298
↓ 1 callersMethodfree
tensorflow_blade/src/tensorrt/bridge/tensorrt_tf_allocator.cc:74
↓ 1 callersFunctionfreeze_rank_analysis
(outer_block)
pytorch_blade/torch_blade/pass_manager.py:192
↓ 1 callersFunctionfromNumberType
pytorch_blade/pytorch_blade/compiler/jit/tool_funcs.cpp:152
↓ 1 callersFunctionfusion_block
(block)
pytorch_blade/torch_blade/mlir/disc_engine_conversion.py:205
↓ 1 callersFunctiongatherTensorAlongSingleAxis
pytorch_blade/pytorch_blade/torch-mlir/lib/Conversion/TorchToMhlo/DiscTorchToMhlo.cpp:1235
↓ 1 callersFunctiongather_loops
(outer_block)
pytorch_blade/torch_blade/tools/shape_inference.py:147
↓ 1 callersFunctiongating_init_
(weights)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/ops.py:77
↓ 1 callersFunctiongcd
Suppose m and n are non-negative integers.
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:157
↓ 1 callersFunctiongen_file_urls
(path)
scripts/python/bazel_snapshot_client.py:79
↓ 1 callersFunctiongenerateSignatureForFusion
tao_compiler/mlir/disc/transforms/fusion_utils.cc:306
↓ 1 callersFunctiongenerateUnrolledLoopMayInterleave
Generates unrolled copies of scf::ForOp 'loopBodyBlock', with associated 'forOpIV' by 'unrollFactor', calling 'ivRemapFn' to remap 'forOpIV' for each
tao_compiler/mlir/disc/transforms/codegen_utils.cc:392
↓ 1 callersFunctiongenerate_build_info
(file)
scripts/python/tao_build.py:596
↓ 1 callersFunctiongenerate_config
(enable_disc=True, enable_amp=True, enable_trace=True, dtype=torch.float, device="cuda:0")
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/optimize.py:62
↓ 1 callersFunctiongenerate_node
( n: tf.NodeDef, new_inputs: List[str], identity_map: Dict[str, str] )
tensorflow_blade/tf_blade/util/tf_util.py:146
↓ 1 callersMethodgetAliasingOpResults
tao_compiler/mlir/disc/tools/disc-transform/LinalgExt/LinalgExtOps.cc:977
↓ 1 callersMethodgetAllCandidateScheduleFactories
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:1483
↓ 1 callersFunctiongetAutomaticAllocationScope
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:618
↓ 1 callersFunctiongetBroadcastedDim
Get the resulting extent in a given dimension. This is computed with any number of extent tensors and shifted offsets into them.
tao_compiler/mlir/disc/transforms/disc_shape_to_std.cc:144
↓ 1 callersFunctiongetBufferizationOptions
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:106
↓ 1 callersFunctiongetColReductionScheduleHint
tao_compiler/mlir/disc/transforms/codegen_utils.cc:78
↓ 1 callersMethodgetCollapsedTileDim
tao_compiler/mlir/disc/transforms/fusion_utils.cc:864
↓ 1 callersFunctiongetConstProducer
tao_compiler/mlir/disc/transforms/disc_duplicate_computation_after_fusion.cc:34
↓ 1 callersMethodgetConstStore
tao_compiler/mlir/xla/ral/context/common_context_impl.cc:169
↓ 1 callersFunctiongetConstantElementsOfShapeTensor
tao_compiler/mlir/disc/transforms/disc_shape_optimization.cc:1620
↓ 1 callersFunctiongetContainerOrTensorTypeWithValueSemantics
pytorch_blade/pytorch_blade/torch-mlir/lib/Dialect/TorchConversion/Transforms/ApplyValueSemantics.cpp:33
↓ 1 callersFunctiongetConvDimensionNumbersAttr
tao_compiler/mlir/disc/transforms/lower_tf.cc:893
↓ 1 callersFunctiongetCopyRemovableResult
Return null is not a copy-removable copy-like ops, otherwise return the copied result value.
tao_compiler/mlir/disc/transforms/disc_lower_to_library_call.cc:430
↓ 1 callersFunctiongetDType
pytorch_blade/pytorch_blade/compiler/jit/torch/shape_analysis.cpp:312
↓ 1 callersFunctiongetDefaultLowerVectorsOptions
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:220
↓ 1 callersFunctiongetDependentOpsInBlock
Get ops that depends on the given op in the same block. It assumes the ops in the block do not have regions.
tao_compiler/mlir/disc/transforms/codegen_utils.cc:355
↓ 1 callersMethodgetDeviceConstByIndex
Returns nullptr if not found or supported.
tao_compiler/mlir/xla/ral/context/common_context_impl.h:100
↓ 1 callersFunctiongetDimSize
tao_compiler/mlir/disc/transforms/disc_shape_to_std.cc:52
↓ 1 callersFunctiongetDirectProducerOpsInFusionPattern
tao_compiler/mlir/disc/transforms/fusion_utils_dot_gpu.cc:75
↓ 1 callersMethodgetEdgePaddingHigh
Returns the `EdgePaddingHigh` for `axis`.
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.h:223
↓ 1 callersMethodgetEffects
tao_compiler/mlir/disc/tools/disc-transform/LinalgExt/LinalgExtOps.cc:217
↓ 1 callersFunctiongetEffectsImpl
tao_compiler/mlir/disc/tools/disc-transform/LinalgExt/LinalgExtOps.cc:56
↓ 1 callersFunctiongetEnvironmentSettings
tao_compiler/mlir/disc/tests/mlir_feature_test.cc:248
↓ 1 callersFunctiongetFirstOperandIndex
Return the operand index if 'value' is a operand of 'op' If there are multiple operands of 'op' that are all 'value', return the first index
tao_compiler/mlir/disc/transforms/revise_kernel_outlining.cc:123
↓ 1 callersMethodgetFusionOp
Returns the fusion op this descriptor holds.
tao_compiler/mlir/disc/transforms/disc_transform_schedule.h:61
↓ 1 callersFunctiongetGenericEffectsImpl
tao_compiler/mlir/disc/tools/disc-transform/LinalgExt/LinalgExtOps.cc:617
↓ 1 callersMethodgetHostConstByIndex
Returns nullptr if not found or supported.
tao_compiler/mlir/xla/ral/context/common_context_impl.h:87
↓ 1 callersFunctiongetInnermostForLoops
tao_compiler/mlir/disc/transforms/disc_for_loop_unroll_interleave.cc:32
↓ 1 callersMethodgetIterationDomain
Return the iteration domain range.
tao_compiler/mlir/disc/tools/disc-transform/LinalgExt/LinalgExtOps.cc:725
↓ 1 callersFunctiongetKernelName
tao_compiler/mlir/disc/transforms/disc_outline_cpu_kernel.cc:196
↓ 1 callersFunctiongetLargeConcatNumOperandsLimit
tao_compiler/mlir/disc/transforms/fusion_utils.cc:1339
↓ 1 callersMethodgetLastError
tao_compiler/mlir/xla/ral/ral_context.cc:162
↓ 1 callersFunctiongetLayout
tao_compiler/mlir/xla/ral/context/stream_executor_based_impl.cc:1031
↓ 1 callersFunctiongetMaxIndexFromItemOps
pytorch_blade/pytorch_blade/torch-mlir/lib/Dialect/TorchConversion/Transforms/DiscDecomposeComplexOps.cpp:87
↓ 1 callersFunctiongetMaxIndicesInDim
pytorch_blade/pytorch_blade/torch-mlir/lib/Conversion/TorchToMhlo/DiscTorchToMhlo.cpp:1046
↓ 1 callersFunctiongetMaxValueInDim
pytorch_blade/pytorch_blade/torch-mlir/lib/Conversion/TorchToMhlo/DiscTorchToMhlo.cpp:1003
↓ 1 callersFunctiongetMinNumUnitPerCore
tao_compiler/mlir/xla/ral/context/common_context_impl.cc:777
↓ 1 callersFunctiongetMultiDimIndex
tao_compiler/mlir/disc/transforms/revise_kernel_outlining.cc:58
↓ 1 callersFunctiongetMutableType
For any mutable type, map it to a type such that all other types which it can alias will be mapped to the same type. This function follows a similar l
pytorch_blade/pytorch_blade/compiler/jit/torch/alias_analysis.cpp:24
↓ 1 callersMethodgetName
tao_compiler/mlir/disc/transforms/fusion_utils.cc:1267
↓ 1 callersFunctiongetNextTestInstanceId
tao_compiler/mlir/disc/tests/mlir_feature_test.cc:84
↓ 1 callersMethodgetNextUniqueId
static */
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:1459
↓ 1 callersMethodgetNumUnits
tao_compiler/mlir/xla/ral/context/common_context_impl.cc:686
↓ 1 callersFunctiongetOrCreateGlobalString
Return a value representing an access into a global string with the given name, creating the string if necessary.
tao_compiler/mlir/disc/transforms/disc_to_llvm.cc:1265
↓ 1 callersMethodgetOrCreateSymbolicDimsForRankedValue
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:651
↓ 1 callersFunctiongetOrInsertPrintf
tao_compiler/mlir/disc/transforms/disc_to_llvm.cc:1243
↓ 1 callersFunctiongetOrderedSkeletonGroups
tao_compiler/mlir/disc/transforms/fusion_utils.cc:880
↓ 1 callersFunctiongetPaddingValues
tao_compiler/mlir/disc/transforms/lower_tf.cc:807
↓ 1 callersFunctiongetPredefinedPDLPatterns
add pre-defined pdll patterns here.
tao_compiler/mlir/disc/transforms/lower_tf.cc:1485
↓ 1 callersFunctiongetPromotedTypeForArithmeticOp
Promotes result types for arithmetic operations on Tensor operands using new type promotion logic. See tensor_attributes.rst for details. This doesn't
pytorch_blade/pytorch_blade/compiler/jit/torch/shape_analysis.cpp:334
↓ 1 callersFunctiongetRootMemRefIfSafe
tao_compiler/mlir/disc/transforms/disc_lower_to_library_call.cc:448
↓ 1 callersFunctiongetScalarTypePtr
pytorch_blade/pytorch_blade/ltc/disc_compiler/passes/disc_fuser.cpp:29
↓ 1 callersMethodgetScheduleFactoryWithHighestPriority
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:1474
↓ 1 callersFunctiongetShmemSizeBytesNotAffectOccupancy
tao_compiler/mlir/disc/disc_util.cc:369
↓ 1 callersFunctiongetSparseReshapePDLPattern
sparse_ids = sparse_ops.sparse_reshape(sparse_ids, [ math_ops.reduce_prod( array_ops.slice(original_shape, [0], [original_rank - 1])), array_ops.gathe
tao_compiler/mlir/disc/transforms/disc_sparse_op_rewriter.cc:44
↓ 1 callersFunctiongetSparseSegmentReductionPDLPattern
rewrite sparse_fill_empty_rows + sparse_segment_reduction to sparse_segment_reduction_with_empty_rows
tao_compiler/mlir/disc/transforms/disc_sparse_op_rewriter.cc:131
↓ 1 callersMethodgetSpecificLoader
tao_compiler/mlir/disc/transforms/lhlo_elemental_utils.cc:47
↓ 1 callersMethodgetSpecificStore
tao_compiler/mlir/disc/transforms/lhlo_elemental_utils.cc:67
↓ 1 callersMethodgetStartIndex
Returns the `StartIndex` for `axis`.
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.h:139
← previousnext →2,101–2,200 of 6,088, ranked by callers