Case1: paddle.linalg.matrix_power

Reproduced Case：

import paddle
import numpy as np
from paddle.linalg import matrix_power
x = paddle.to_tensor(np.random.uniform(-10, 10, [1, 1, 0, 0]).astype(np.float32))
print(x)
matrix_power(x, -1)

Stack Trace Message：

AddressSanitizer:DEADLYSIGNAL
=================================================================
==172907==ERROR: AddressSanitizer: FPE on unknown address 0x7f66837b29c1 (pc 0x7f66837b29c1 bp 0x7ffc685552b0 sp 0x7ffc68555080 T0)
    #0 0x7f66837b29c1 in void phi::funcs::ComputeInverseEigen<phi::CPUContext, float>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*) /home/work/yakun/paddle-2.4.0/Paddle/paddle/phi/kernels/funcs/matrix_inverse.h:39:41
    #1 0x7f66806f6b47 in void phi::MatrixPowerFunction<phi::CPUContext, float>(phi::DenseTensor const*, int, phi::DenseTensor*, phi::CPUContext const&) /home/work/yakun/paddle-2.4.0/Paddle/paddle/phi/kernels/impl/matrix_power_kernel_impl.h:68:5
    #2 0x7f66806f5c17 in void phi::MatrixPowerKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, int, phi::DenseTensor*) /home/work/yakun/paddle-2.4.0/Paddle/paddle/phi/kernels/impl/matrix_power_kernel_impl.h:200:3
    #3 0x7f667d72b18c in paddle::experimental::matrix_power(paddle::experimental::Tensor const&, int) /home/work/yakun/paddle-2.4.0/Paddle/paddle/phi/api/lib/api.cc:15321:5
    #4 0x7f6673350d24 in matrix_power_ad_func(paddle::experimental::Tensor const&, int) /home/work/yakun/paddle-2.4.0/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:25890:21
    #5 0x7f667011f00a in paddle::pybind::eager_api_matrix_power(_object*, _object*, _object*) /home/work/yakun/paddle-2.4.0/Paddle/paddle/fluid/pybind/eager_op_function.cc:11352:49
    #6 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #7 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #8 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #9 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #10 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #11 0x560136 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x560136) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #12 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #13 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #14 0x62a1ef  (/usr/bin/python3.8+0x62a1ef) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #15 0x62a179  (/usr/bin/python3.8+0x62a179) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #16 0x47a7f2  (/usr/bin/python3.8+0x47a7f2) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #17 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #18 0x4247dc in _init (/usr/bin/python3.8+0x4247dc) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #19 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
    #20 0x7f67596b983f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #21 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8) (BuildId: 69b06f9a4b2e8428d7e32aa682c34a91dc0b961e)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE /home/work/yakun/paddle-2.4.0/Paddle/paddle/phi/kernels/funcs/matrix_inverse.h:39:41 in void phi::funcs::ComputeInverseEigen<phi::CPUContext, float>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*)
==172907==ABORTING

Case2: paddle.static.nn.sparse_embedding

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
input = paddle.to_tensor(np.reshape(array, [0]), dtype='int64')
paddle.static.nn.sparse_embedding(input, [2097152, 2097152, 2097152, 2097152], padding_idx=2097152)

Stack Trace Message：

Traceback (most recent call last):
  File "bug.py", line 6, in <module>
    paddle.static.nn.sparse_embedding(input, [2097152, 2097152, 2097152, 2097152], padding_idx=2097152)
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/fluid/contrib/layers/nn.py", line 1205, in sparse_embedding
    w = helper.create_parameter(
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/fluid/layer_helper_base.py", line 432, in create_parameter
    return self.main_program.global_block().create_parameter(
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/fluid/framework.py", line 3963, in create_parameter
    initializer(param, self)
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/fluid/initializer.py", line 72, in __call__
    return self.forward(param, block)
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/fluid/initializer.py", line 712, in forward
    limit = math.sqrt(6.0 / float(fan_in + fan_out))
ZeroDivisionError: float division by zero

Case3: paddle.static.nn.deform_conv2d

Reproduced Case：

import paddle
import numpy as np
paddle.enable_static()
x = paddle.static.data(name='x', shape=[1, 1], dtype='float32')
mask = paddle.static.data(name='mask', shape=[1], dtype='float32') 
offset = paddle.static.data(name='offset', shape=[1, 1], dtype='float32')
paddle.static.nn.deform_conv2d(x, offset, mask, 1, 1, padding=1, groups=0)

Stack Trace Message：

Traceback (most recent call last):
  File "bug.py", line 9, in <module>
    paddle.static.nn.deform_conv2d(x, offset, mask, 1, 1, padding=1, groups=0)
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/decorator.py", line 232, in fun
    return caller(func, *(extras + args), **kw)
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/fluid/wrapped_decorator.py", line 25, in __impl__
    return wrapped_func(*args, **kwargs)
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/fluid/framework.py", line 553, in __impl__
    return func(*args, **kwargs)
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/static/nn/common.py", line 578, in deform_conv2d
    return deformable_conv(
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/static/nn/common.py", line 343, in deformable_conv
    if num_channels % groups != 0:
ZeroDivisionError: integer division or modulo by zero

Case4: paddle.static.nn.sequence_concat

Reproduced Case：

import paddle
paddle.enable_static()
x = paddle.static.data(name='x', shape=[0], dtype='float32', lod_level=1)
y = paddle.static.data(name='y', shape=[0], dtype='float32', lod_level=1)
out = paddle.static.nn.sequence_concat(input=[x, y])

Stack Trace Message：

==170412==ERROR: AddressSanitizer: FPE on unknown address 0x7f9a97c1250a (pc 0x7f9a97c1250a bp 0x7fffeceb70b0 sp 0x7fffeceb63a0 T0)
    #0 0x7f9a97c1250a in paddle::operators::SequenceConcatOp::InferShape(paddle::framework::InferShapeContext*) const /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/paddle/fluid/operators/sequence_ops/sequence_concat_op.cc:73:44
    #1 0x7f9a8fa8f9b8 in std::function<void (paddle::framework::InferShapeContext*)>::operator()(paddle::framework::InferShapeContext*) const /usr/bin/../lib/gcc/x86_64-linux-gnu/8/../../../../include/c++/8/bits/std_function.h:687:14
    #2 0x7f9a8fa8f9b8 in paddle::framework::OpDesc::InferShape(paddle::framework::BlockDesc const&) /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/paddle/fluid/framework/op_desc.cc:1076:5
    #3 0x7f9a8e2dda81 in pybind11::cpp_function::cpp_function<void, paddle::framework::OpDesc, paddle::framework::BlockDesc const&, pybind11::name, pybind11::is_method, pybind11::sibling>(void (paddle::framework::OpDesc::*)(paddle::framework::BlockDesc const&), pybind11::name const&, pybind11::is_method const&, pybind11::sibling const&)::'lambda'(paddle::framework::OpDesc*, paddle::framework::BlockDesc const&)::operator()(paddle::framework::OpDesc*, paddle::framework::BlockDesc const&) const /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/build/third_party/pybind/src/extern_pybind/include/pybind11/pybind11.h:78:66

Case5: paddle.vision.ops.psroi_pool

Reproduced Case：

import paddle
import numpy as np
array = np.array([1], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [1, 1, 1, 1]), dtype='float32')
boxes = paddle.to_tensor(np.reshape(array, [1, 1]), dtype='float32')
boxes_num = paddle.to_tensor(np.reshape(array, [1, 1]), dtype='float32') 
paddle.vision.ops.psroi_pool(x, boxes, boxes_num, 0, spatial_scale=0)

Stack Trace Message：

Traceback (most recent call last):
  File "bug.py", line 9, in <module>
    paddle.vision.ops.psroi_pool(x, boxes, boxes_num, 0, spatial_scale=0)
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/vision/ops.py", line 1589, in psroi_pool
    output_channels = int(x.shape[1] / (pooled_height * pooled_width))
ZeroDivisionError: division by zero

NOTE：https://github.com/PaddlePaddle/Paddle/blob/4c8c5b4b6e897be3297fbff8aa29ebe2f32bd005/python/paddle/vision/ops.py#L1423-L1427

原因：paddle.vision.ops.psroi_pool 传入 output_size = 0，会导致 pooled_height 和 pooled_width 都为0，从而触发除0错误。
Reason： The input parameter output_size of paddle.vision.ops.psroi_pool = 0, then, pooled_height and pooled_width are both 0 and trigger the divide by 0 error.

Case6: paddle.nn.functional.softmax_with_cross_entropy

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
logits = paddle.to_tensor(np.reshape(array, [1, 0]), dtype='float32')
array = np.array([1], dtype=np.float32)
label = paddle.to_tensor(np.reshape(array, [1]), dtype='float32')
paddle.nn.functional.softmax_with_cross_entropy(logits, label, soft_label=True, numeric_stable_mode=True, axis=0)

Stack Trace Message：

==206963==ERROR: AddressSanitizer: FPE on unknown address 0x7f3da0602cd7 (pc 0x7f3da0602cd7 bp 0x7ffef5a33690 sp 0x7ffef5a33628 T0)
    #0 0x7f3da0602cd7 in __udivti3 (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7)
    #1 0x7f3ce737daf0 in Eigen::internal::(anonymous namespace)::DividerHelper<64, long>::computeMultiplier(int, long) /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128:82
    #2 0x7f3ce737daf0 in Eigen::internal::TensorIntDivisor<long, false>::TensorIntDivisor(long) /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:165:18

NOTE：

paddle.nn.functional.softmax_with_cross_entropy 可触发 third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128 的除0问题，原因可能是没有对传入参数 logits 的 shape 做严格的检查。
paddle.nn.functional.softmax_with_cross_entropy could trigger the divide by 0 problems in third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128, may be caused by lacking enough input parameter check for the shape of logits.

  struct DividerHelper<64, T> {
    static EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE uint64_t computeMultiplier(const int log_div, const T divider) {
#if EIGEN_HAS_BUILTIN_INT128 && !defined(EIGEN_GPU_COMPILE_PHASE) && !defined(SYCL_DEVICE_ONLY)
      return static_cast<uint64_t>((static_cast<__uint128_t>(1) << (64+log_div)) / static_cast<__uint128_t>(divider) - (static_cast<__uint128_t>(1) << 64) + 1); // <-- 128行
#else

Case7: paddle.nn.functional.pixel_shuffle

Reproduced Case：

import paddle
import numpy as np
array = np.array([1], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [1, 1, 1, 1]), dtype='float32')
paddle.nn.functional.pixel_shuffle(x, 0)

Stack Trace Message：

==65621==ERROR: AddressSanitizer: FPE on unknown address 0x7f1828c86e8b (pc 0x7f1828c86e8b bp 0x7ffe8af791d0 sp 0x7ffe8af789c0 T0)
, std::allocator    #0 0x7f1828c86e8b in phi::PixelShuffleInferMeta(phi::MetaTensor const&, int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, phi::MetaTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/paddle/phi/infermeta/unary.cc:2278:5
, std::allocator    #1 0x7f1822051f17 in paddle::experimental::pixel_shuffle(paddle::experimental::Tensor const&, int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/paddle/phi/api/lib/api.cc:17938:3
, std::allocator    #2 0x7f1816f09564 in pixel_shuffle_ad_func(paddle::experimental::Tensor const&, int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >) /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:22800:21
    #3 0x7f181346f6ef in paddle::pybind::eager_api_pixel_shuffle(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/paddle/fluid/pybind/eager_op_function.cc:13238:75

NOTE：https://github.com/PaddlePaddle/Paddle/blob/60d1199a8ca26958ed44058987689aa06a169e9d/paddle/phi/infermeta/unary.cc#L2539-L2547

原因：paddle.nn.functional.pixel_shuffle 中传入 upscale_factor = 0，触发了2540行的除0。
Reason: The input parameter upscale_factor of paddle.nn.functional.pixel_shuffle = 0, thus triggering the divide by 0 problems of line 2540.

Case8: paddle.nn.functional.npair_loss

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
anchor = paddle.to_tensor(np.reshape(array, [0, 0, 0]), dtype='float32')
positive = paddle.to_tensor(np.reshape(array, [0]), dtype='float32')
array = np.array([1, 2, 3, 4], dtype=np.float32)
labels = paddle.to_tensor(np.reshape(array, [4]), dtype='float32')
paddle.nn.functional.npair_loss(anchor, positive, labels)

Stack Trace Message：

==54596==ERROR: AddressSanitizer: FPE on unknown address 0x7f2bab9f1cd7 (pc 0x7f2bab9f1cd7 bp 0x7ffd0111a5f0 sp 0x7ffd0111a588 T0)
    #0 0x7f2bab9f1cd7 in __udivti3 (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7)
    #1 0x7f2af5ee4c65 in Eigen::internal::(anonymous namespace)::DividerHelper<64, long>::computeMultiplier(int, long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_pan3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128:82
    #2 0x7f2af5ee4c65 in Eigen::internal::TensorIntDivisor<long, false>::TensorIntDivisor(long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_esupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:165:18
    #3 0x7f2af5ee4c65 in Eigen::TensorReductionEvaluatorBase<Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, longgen::MakePointer> const, Eigen::MakePointer> const, Eigen::DefaultDevice>::TensorReductionEvaluatorBase(Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::Tensgen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/thy/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h:584:36
    #4 0x7f2af5ee25ee in Eigen::TensorEvaluator<Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::ter> const, Eigen::MakePointer> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float , 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/exter/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h:990:115

Case9: paddle.nn.functional.normalize

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [1, 1, 0]), dtype='float32')
paddle.nn.functional.normalize(x)

Stack Trace Message：

==103897==ERROR: AddressSanitizer: FPE on unknown address 0x7f4d42de8cd7 (pc 0x7f4d42de8cd7 bp 0x7ffee7c3aff0 sp 0x7ffee7c3af88 T0)
    #0 0x7f4d42de8cd7 in __udivti3 (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7)
    #1 0x7f4c8f5c3e63 in Eigen::internal::(anonymous namespace)::DividerHelper<64, long>::computeMultiplier(int, long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128:82
    #2 0x7f4c8f5c3e63 in Eigen::internal::TensorIntDivisor<long, false>::TensorIntDivisor(long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:165:18
    #3 0x7f4c8f5c3e63 in Eigen::TensorReductionEvaluatorBase<Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const, Eigen::DefaultDevice>::TensorReductionEvaluatorBase(Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h:584:36

Case10: paddle.nn.functional.max_pool2d/max_pool3d

Reproduced Case：

import paddle
import numpy as np
array = np.array([1], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [1, 1, 1]), dtype='float32')
paddle.nn.functional.max_pool1d(x, 1, stride=(0))
paddle.nn.functional.max_pool2d
import paddle
import numpy as np
array = np.array([1], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [1, 1, 1, 1]), dtype='float32')
paddle.nn.functional.max_pool2d(x, 1, stride=(0, 0), return_mask=False, data_format='NHWC')
import paddle
import numpy as np
array = np.array([1], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [1, 1, 1, 1, 1]), dtype='float32')
paddle.nn.functional.max_pool3d(x, 1, stride=(0, 0, 0), ceil_mode=False)

Stack Trace Message：

==90343==ERROR: AddressSanitizer: FPE on unknown address 0x7f5378bfa156 (pc 0x7f5378bfa156 bp 0x7ffc6e68ff70 sp 0x7ffc6e68fc60 T0)
    #0 0x7f5378bfa156 in phi::funcs::PoolOutputSize(int, int, int, int, int, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/pooling.h
    #1 0x7f5378bf56c6 in phi::PoolInferMeta(phi::MetaTensor const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, phi::MetaTensor*, phi::MetaConfig) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/infermeta/unary.cc:2783:32
    #2 0x7f5378bf331b in phi::Pool2DInferMeta(phi::MetaTensor const&, paddle::experimental::IntArrayBase<phi::DenseTensor> const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, phi::MetaTensor*, phi::MetaConfig) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/infermeta/unary.cc:2679:5
    #3 0x7f5371d89d91 in paddle::experimental::pool2d(paddle::experimental::Tensor const&, paddle::experimental::IntArrayBase<paddle::experimental::Tensor> const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:20065:3
    #4 0x7f5365c45151 in pool2d_ad_func(paddle::experimental::Tensor const&, paddle::experimental::IntArrayBase<paddle::experimental::Tensor>, std::vector<int, std::allocator<int> >, std::vector<int, std::allocator<int> >, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:32933:21

Case11: paddle.nn.functional.max_pool1d/max_pool2d/max_pool3d

Reproduced Case：

import paddle
import numpy as np
array = np.array([1], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [1, 1, 1]), dtype='float32')
paddle.nn.functional.max_pool1d(x, 1, stride=0, padding=1, return_mask=True, ceil_mode=True) 
import paddle
import numpy as np
array = np.array([1], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [1, 1, 1, 1]), dtype='float32')
paddle.nn.functional.max_pool2d(x, 1, stride=0, padding=1, return_mask=True, ceil_mode=True) 
import paddle
import numpy as np
array = np.array([1], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [1, 1, 1, 1, 1]), dtype='float32')
paddle.nn.functional.max_pool3d(x, 1, stride=0, padding=1, return_mask=True, ceil_mode=True)

Stack Trace Message：

==162650==ERROR: AddressSanitizer: FPE on unknown address 0x7f7b6150b768 (pc 0x7f7b6150b768 bp 0x7fff298f6e90 sp 0x7fff298f6500 T0)
    #0 0x7f7b6150b768 in phi::funcs::MaxPoolOutputSize(int, int, int, int) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/pooling.h:405:62
    #1 0x7f7b6150b768 in phi::MaxPoolWithIndexInferMeta(phi::MetaTensor const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, bool, bool, phi::MetaTensor*, phi::MetaTensor*, phi::MetaConfig) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/infermeta/unary.cc:2011:32
    #2 0x7f7b5a66587f in paddle::experimental::max_pool2d_with_index(paddle::experimental::Tensor const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, bool, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:17439:3
    #3 0x7f7b4e4be84a in max_pool2d_with_index_ad_func(paddle::experimental::Tensor const&, std::vector<int, std::allocator<int> >, std::vector<int, std::allocator<int> >, std::vector<int, std::allocator<int> >, bool, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:28983:21
    #4 0x7f7b4a96cea5 in paddle::pybind::eager_api_max_pool2d_with_index(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:12684:109

Case12: paddle.linalg.svd

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0, 0]), dtype='float32')
paddle.linalg.svd(x, full_matrices=False)

Stack Trace Message：

AddressSanitizer:DEADLYSIGNAL
=================================================================
==144309==ERROR: AddressSanitizer: FPE on unknown address 0x7fdefd49ddf5 (pc 0x7fdefd49ddf5 bp 0x7fffa988d050 sp 0x7fffa988cae0 T0)
    #0 0x7fdefd49ddf5 in void phi::SvdKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/svd_kernel.cc:108:23
    #1 0x7fdefc4ea66d in paddle::experimental::svd(paddle::experimental::Tensor const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:24421:5
    #2 0x7fdef0452dab in svd_ad_func(paddle::experimental::Tensor const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:39772:21
    #3 0x7fdeec7d90c1 in paddle::pybind::eager_api_svd(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:17605:52
    #4 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #5 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #6 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #7 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #8 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #9 0x5610ed in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x5610ed)
    #10 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #11 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #12 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #13 0x62a179  (/usr/bin/python3.8+0x62a179)
    #14 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #15 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #16 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #17 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #18 0x7fdfb3ab683f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #19 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/svd_kernel.cc:108:23 in void phi::SvdKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*)
==144309==ABORTING

Case13: paddle.linalg.norm

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0, 0]), dtype='float32')
paddle.linalg.norm(x, axis=0)

Stack Trace Message：

AddressSanitizer:DEADLYSIGNAL
=================================================================
==182546==ERROR: AddressSanitizer: FPE on unknown address 0x7f8bcccdccd7 (pc 0x7f8bcccdccd7 bp 0x7ffd70455d10 sp 0x7ffd70455ca8 T0)
    #0 0x7f8bcccdccd7 in __udivti3 (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7)
    #1 0x7f8b194b7e63 in Eigen::internal::(anonymous namespace)::DividerHelper<64, long>::computeMultiplier(int, long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128:82
    #2 0x7f8b194b7e63 in Eigen::internal::TensorIntDivisor<long, false>::TensorIntDivisor(long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:165:18
    #3 0x7f8b194b7e63 in Eigen::TensorReductionEvaluatorBase<Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const, Eigen::DefaultDevice>::TensorReductionEvaluatorBase(Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h:584:36
    #4 0x7f8b194a315d in Eigen::TensorEvaluator<Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h:990:115
    #5 0x7f8b194a315d in Eigen::TensorEvaluator<Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorEvaluator.h:461:7
    #6 0x7f8b194a315d in Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorReshapingOp<Eigen::DSizes<int, 2> const, Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 0, Eigen::MakePointer> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const> const> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorAssignOp<Eigen::TensorReshapingOp<Eigen::DSizes<int, 2> const, Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 0, Eigen::MakePointer> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h:129:7
    #7 0x7f8b194a315d in Eigen::internal::TensorExecutor<Eigen::TensorAssignOp<Eigen::TensorReshapingOp<Eigen::DSizes<int, 2> const, Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 0, Eigen::MakePointer> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const> const> const, Eigen::DefaultDevice, false, (Eigen::internal::TiledEvaluation)0>::run(Eigen::TensorAssignOp<Eigen::TensorReshapingOp<Eigen::DSizes<int, 2> const, Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 0, Eigen::MakePointer> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h:98:41
    #8 0x7f8b194a315d in Eigen::TensorDevice<Eigen::TensorReshapingOp<Eigen::DSizes<int, 2> const, Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 0, Eigen::MakePointer> >, Eigen::DefaultDevice>& Eigen::TensorDevice<Eigen::TensorReshapingOp<Eigen::DSizes<int, 2> const, Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 0, Eigen::MakePointer> >, Eigen::DefaultDevice>::operator=<Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const> >(Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, Eigen::DSizes<int, 1> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::bind2nd_op<Eigen::internal::scalar_pow_op<float, float> >, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_abs_op<float const>, Eigen::TensorReshapingOp<Eigen::DSizes<int, 3> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::MakePointer> const> const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h:37:7
    #9 0x7f8b194a315d in void phi::PNormKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, float, int, float, bool, bool, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/p_norm_kernel.cc:87:25
    #10 0x7f8b15c74faa in paddle::experimental::p_norm(paddle::experimental::Tensor const&, float, int, float, bool, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:19808:5
    #11 0x7f8b09b1f72f in p_norm_ad_func(paddle::experimental::Tensor const&, float, int, float, bool, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:32436:21
    #12 0x7f8b05f6c4b8 in paddle::pybind::eager_api_p_norm(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:14334:78
    #13 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #14 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #15 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #16 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #17 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #18 0x5610ed in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x5610ed)
    #19 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #20 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #21 0x5610ed in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x5610ed)
    #22 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #23 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #24 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #25 0x62a179  (/usr/bin/python3.8+0x62a179)
    #26 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #27 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #28 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #29 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #30 0x7f8bcd2f383f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #31 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7) in __udivti3
==182546==ABORTING

Case14: paddle.linalg.pinv

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0, 0]), dtype='float32')
paddle.linalg.pinv(x)

Stack Trace Message：

AddressSanitizer:DEADLYSIGNAL
=================================================================
==48351==ERROR: AddressSanitizer: FPE on unknown address 0x7fd60b795df5 (pc 0x7fd60b795df5 bp 0x7fff82282970 sp 0x7fff82282400 T0)
    #0 0x7fd60b795df5 in void phi::SvdKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/svd_kernel.cc:108:23
    #1 0x7fd60a7e266d in paddle::experimental::svd(paddle::experimental::Tensor const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:24421:5
    #2 0x7fd5fe74adab in svd_ad_func(paddle::experimental::Tensor const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:39772:21
    #3 0x7fd5faad10c1 in paddle::pybind::eager_api_svd(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:17605:52
    #4 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #5 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #6 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #7 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #8 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #9 0x564f2d in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x564f2d)
    #10 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #11 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #12 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #13 0x62a179  (/usr/bin/python3.8+0x62a179)
    #14 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #15 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #16 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #17 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #18 0x7fd6c1dae83f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #19 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/svd_kernel.cc:108:23 in void phi::SvdKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*)
==48351==ABORTING

Case15: paddle.linalg.eig

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0, 0]), dtype='float32')
paddle.linalg.eig(x)

Stack Trace Message：

AddressSanitizer:DEADLYSIGNAL
=================================================================
==137096==ERROR: AddressSanitizer: FPE on unknown address 0x7f0fd5be2cd7 (pc 0x7f0fd5be2cd7 bp 0x7fff9ff6af70 sp 0x7fff9ff6af08 T0)
    #0 0x7f0fd5be2cd7 in __udivti3 (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7)
    #1 0x7f0f25fd54bf in Eigen::internal::(anonymous namespace)::DividerHelper<64, long>::computeMultiplier(int, long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128:82
    #2 0x7f0f25fd54bf in Eigen::internal::TensorIntDivisor<long, false>::TensorIntDivisor(long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:165:18
    #3 0x7f0f25fd54bf in Eigen::TensorEvaluator<Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorShuffling.h:155:34
    #4 0x7f0f25fd54bf in Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h:129:7
    #5 0x7f0f25fd39d0 in Eigen::internal::TensorExecutor<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const> const, Eigen::DefaultDevice, true, (Eigen::internal::TiledEvaluation)1>::run(Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h:186:15
    #6 0x7f0f25efb76d in Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>& Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>::operator=<Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> >(Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h:37:7
    #7 0x7f0f25efb76d in phi::funcs::Transpose<phi::CPUContext, float, 2>::operator()(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<int, std::allocator<int> > const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/math_function_impl.h:73:28
    #8 0x7f0f1f89a1b3 in void phi::funcs::TransCompute<phi::CPUContext, float>(int, phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<int, std::allocator<int> > const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/math_function.h:147:7
    #9 0x7f0f24120d41 in void phi::TransposeTwoAxis<float, phi::CPUContext>(phi::DenseTensor const&, phi::DenseTensor*, int, int, phi::CPUContext const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/eig.h:118:3
    #10 0x7f0f240fa527 in void phi::ApplyEigKernel<float, phi::CPUContext>(phi::DenseTensor const&, phi::DenseTensor*, phi::DenseTensor*, phi::CPUContext const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/eig.h:222:3
    #11 0x7f0f240f8ce8 in void phi::EigKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/eig_kernel.cc:51:5
    #12 0x7f0f1e8de561 in paddle::experimental::eig(paddle::experimental::Tensor const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:1986:5
    #13 0x7f0f124853df in eig_ad_func(paddle::experimental::Tensor const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:4626:21
    #14 0x7f0f0ebe282c in paddle::pybind::eager_api_eig(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:1463:38
    #15 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #16 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #17 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #18 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #19 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #20 0x564f2d in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x564f2d)
    #21 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #22 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #23 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #24 0x62a179  (/usr/bin/python3.8+0x62a179)
    #25 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #26 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #27 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #28 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #29 0x7f0fd61f983f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #30 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7) in __udivti3
==137096==ABORTING

Case16: paddle.linalg.lu

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0, 0, 0]), dtype='float32')
paddle.linalg.lu(x, get_infos=True)

Stack Trace Message：

AddressSanitizer:DEADLYSIGNAL
=================================================================
==144245==ERROR: AddressSanitizer: FPE on unknown address 0x7fbaf4625cd7 (pc 0x7fbaf4625cd7 bp 0x7ffd35110b50 sp 0x7ffd35110ab8 T0)
    #0 0x7fbaf4625cd7 in __udivti3 (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7)
    #1 0x7fba44a53490 in Eigen::internal::(anonymous namespace)::DividerHelper<64, long>::computeMultiplier(int, long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128:82
    #2 0x7fba44a53490 in Eigen::internal::TensorIntDivisor<long, false>::TensorIntDivisor(long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:165:18
    #3 0x7fba44a53490 in Eigen::TensorEvaluator<Eigen::TensorShufflingOp<std::array<int, 3ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorShufflingOp<std::array<int, 3ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorShuffling.h:155:34
    #4 0x7fba44a5156b in Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 3, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorShufflingOp<std::array<int, 3ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const> const> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 3, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorShufflingOp<std::array<int, 3ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h:129:7
    #5 0x7fba44a5156b in Eigen::internal::TensorExecutor<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 3, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorShufflingOp<std::array<int, 3ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const> const> const, Eigen::DefaultDevice, true, (Eigen::internal::TiledEvaluation)1>::run(Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 3, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorShufflingOp<std::array<int, 3ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h:186:15
    #6 0x7fba4494698a in Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<float, 3, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>& Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<float, 3, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>::operator=<Eigen::TensorShufflingOp<std::array<int, 3ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const> >(Eigen::TensorShufflingOp<std::array<int, 3ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const> const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h:37:7
    #7 0x7fba4494698a in phi::funcs::Transpose<phi::CPUContext, float, 3>::operator()(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<int, std::allocator<int> > const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/math_function_impl.h:73:28
    #8 0x7fba4135bed4 in phi::DenseTensor phi::Transpose2DTo6D<phi::CPUContext, float>(phi::CPUContext const&, phi::DenseTensor const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/impl/lu_kernel_impl.h:536:7
    #9 0x7fba4135a11f in void phi::LUKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/lu_kernel.cc:37:10
    #10 0x7fba3d54755a in paddle::experimental::lu(paddle::experimental::Tensor const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:16737:5
    #11 0x7fba31374772 in lu_ad_func(paddle::experimental::Tensor const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:27729:21
    #12 0x7fba2d84bd95 in paddle::pybind::eager_api_lu(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:12122:43
    #13 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #14 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #15 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #16 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #17 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #18 0x5610ed in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x5610ed)
    #19 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #20 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #21 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #22 0x62a179  (/usr/bin/python3.8+0x62a179)
    #23 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #24 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #25 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #26 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #27 0x7fbaf4c3c83f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #28 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7) in __udivti3
==144245==ABORTING

Case17: paddle.incubate.graph_reindex

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0]), dtype='int32')
layer = paddle.incubate.graph_reindex(x=x, neighbors=x, count=x, value_buffer=x, index_buffer=x, flag_buffer_hashtable=False)

Stack Trace Message：

AddressSanitizer:DEADLYSIGNAL
=================================================================
==107590==ERROR: AddressSanitizer: FPE on unknown address 0x7ff6916a2599 (pc 0x7ff6916a2599 bp 0x7ffef2be5210 sp 0x7ffef2be4ec0 T0)
    #0 0x7ff6916a2599 in void phi::GraphReindexKernel<int, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/graph_reindex_kernel.cc:63:40
    #1 0x7ff6916af8d4 in void phi::KernelImpl<void (*)(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*), &(void phi::GraphReindexKernel<int, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*))>::KernelCallHelper<phi::TypeTag<int> >::Compute<1, 5, 1, 3>(phi::KernelContext*, phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool&, phi::DenseTensor*&, phi::DenseTensor*&, phi::DenseTensor*&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/core/kernel_utils.h:334:14
    #2 0x7ff6916af8d4 in void phi::KernelImpl<void (*)(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*), &(void phi::GraphReindexKernel<int, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*))>::KernelCallHelper<phi::DenseTensor*, phi::TypeTag<int> >::Compute<1, 5, 1, 2, phi::CPUContext const, phi::DenseTensor const, phi::DenseTensor const, phi::DenseTensor const, paddle::optional<phi::DenseTensor>, paddle::optional<phi::DenseTensor>, bool, phi::DenseTensor*, phi::DenseTensor*>(phi::KernelContext*, phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor>&, paddle::optional<phi::DenseTensor>&, bool&, phi::DenseTensor*&, phi::DenseTensor*&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/core/kernel_utils.h:312:3
    #3 0x7ff6916af8d4 in void phi::KernelImpl<void (*)(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*), &(void phi::GraphReindexKernel<int, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*))>::KernelCallHelper<phi::DenseTensor*, phi::DenseTensor*, phi::TypeTag<int> >::Compute<1, 5, 1, 1, phi::CPUContext const, phi::DenseTensor const, phi::DenseTensor const, phi::DenseTensor const, paddle::optional<phi::DenseTensor>, paddle::optional<phi::DenseTensor>, bool, phi::DenseTensor*>(phi::KernelContext*, phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor>&, paddle::optional<phi::DenseTensor>&, bool&, phi::DenseTensor*&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/core/kernel_utils.h:312:3
    #4 0x7ff6916af8d4 in void phi::KernelImpl<void (*)(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*), &(void phi::GraphReindexKernel<int, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*))>::KernelCallHelper<phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*, phi::TypeTag<int> >::Compute<1, 5, 1, 0, phi::CPUContext const, phi::DenseTensor const, phi::DenseTensor const, phi::DenseTensor const, paddle::optional<phi::DenseTensor>, paddle::optional<phi::DenseTensor>, bool>(phi::KernelContext*, phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor>&, paddle::optional<phi::DenseTensor>&, bool&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/core/kernel_utils.h:312:3
    #5 0x7ff6916af8d4 in void phi::KernelImpl<void (*)(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*), &(void phi::GraphReindexKernel<int, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*))>::KernelCallHelper<bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*, phi::TypeTag<int> >::Compute<1, 5, 0, 0, phi::CPUContext const, phi::DenseTensor const, phi::DenseTensor const, phi::DenseTensor const, paddle::optional<phi::DenseTensor>, paddle::optional<phi::DenseTensor> >(phi::KernelContext*, phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor>&, paddle::optional<phi::DenseTensor>&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/core/kernel_utils.h:289:3
    #6 0x7ff6916af8d4 in void phi::KernelImpl<void (*)(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*), &(void phi::GraphReindexKernel<int, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*))>::KernelCallHelper<paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*, phi::TypeTag<int> >::Compute<1, 4, 0, 0, phi::CPUContext const, phi::DenseTensor const, phi::DenseTensor const, phi::DenseTensor const, paddle::optional<phi::DenseTensor> >(phi::KernelContext*, phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor>&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/core/kernel_utils.h:264:3
    #7 0x7ff6916af36b in void phi::KernelImpl<void (*)(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*), &(void phi::GraphReindexKernel<int, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*))>::KernelCallHelper<paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*, phi::TypeTag<int> >::Compute<1, 3, 0, 0, phi::CPUContext const, phi::DenseTensor const, phi::DenseTensor const, phi::DenseTensor const>(phi::KernelContext*, phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/core/kernel_utils.h:264:3
    #8 0x7ff6916af36b in void phi::KernelImpl<void (*)(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*), &(void phi::GraphReindexKernel<int, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*))>::KernelCallHelper<phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*, phi::TypeTag<int> >::Compute<1, 2, 0, 0, phi::CPUContext const, phi::DenseTensor const, phi::DenseTensor const>(phi::KernelContext*, phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/core/kernel_utils.h:263:3
    #9 0x7ff68c6fa504 in std::function<void (phi::KernelContext*)>::operator()(phi::KernelContext*) const /usr/bin/../lib/gcc/x86_64-linux-gnu/8/../../../../include/c++/8/bits/std_function.h:687:14
    #10 0x7ff68c6fa504 in phi::Kernel::operator()(phi::KernelContext*) const /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/core/kernel_factory.h:218:47
    #11 0x7ff68c6fa504 in void paddle::imperative::PreparedOpRunPtImpl<egr::EagerVariable>(paddle::framework::OperatorBase const&, paddle::framework::OpKernelType const&, std::function<phi::KernelSignature (phi::ArgumentMappingContext const&)> const*, phi::KernelSignature const*, phi::KernelSignature const&, phi::Kernel const&, phi::DeviceContext*, paddle::imperative::details::NameVarMapTrait<egr::EagerVariable>::Type const&, paddle::imperative::details::NameVarMapTrait<egr::EagerVariable>::Type const&, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double>, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double> > > > const&, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double>, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double> > > > const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/imperative/prepared_operator.cc:696:5
    #12 0x7ff68c6fa504 in paddle::imperative::PreparedOp::Run(std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<std::shared_ptr<egr::EagerVariable>, std::allocator<std::shared_ptr<egr::EagerVariable> > >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::vector<std::shared_ptr<egr::EagerVariable>, std::allocator<std::shared_ptr<egr::EagerVariable> > > > > > const&, std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<std::shared_ptr<egr::EagerVariable>, std::allocator<std::shared_ptr<egr::EagerVariable> > >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::vector<std::shared_ptr<egr::EagerVariable>, std::allocator<std::shared_ptr<egr::EagerVariable> > > > > > const&, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double>, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double> > > > const&, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double>, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double> > > > const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/imperative/prepared_operator.cc:784:5
    #13 0x7ff68012f7f8 in void paddle::imperative::OpBaseRunImpl<egr::EagerVariable>(paddle::framework::OperatorBase const&, paddle::imperative::details::NameVarMapTrait<egr::EagerVariable>::Type const&, paddle::imperative::details::NameVarMapTrait<egr::EagerVariable>::Type const&, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double>, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double> > > > const&, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double>, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double> > > > const&, phi::Place const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/imperative/layer.cc:525:17
    #14 0x7ff68012f7f8 in paddle::imperative::OpBase::Run(paddle::framework::OperatorBase const&, std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<std::shared_ptr<egr::EagerVariable>, std::allocator<std::shared_ptr<egr::EagerVariable> > >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::vector<std::shared_ptr<egr::EagerVariable>, std::allocator<std::shared_ptr<egr::EagerVariable> > > > > > const&, std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<std::shared_ptr<egr::EagerVariable>, std::allocator<std::shared_ptr<egr::EagerVariable> > >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::vector<std::shared_ptr<egr::EagerVariable>, std::allocator<std::shared_ptr<egr::EagerVariable> > > > > > const&, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double>, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double> > > > const&, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double>, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double> > > > const&, phi::Place const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/imperative/layer.cc:560:3
    #15 0x7ff680184d56 in void paddle::imperative::Tracer::TraceOpImpl<egr::EagerVariable>(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, paddle::imperative::details::NameVarMapTrait<egr::EagerVariable>::Type const&, paddle::imperative::details::NameVarMapTrait<egr::EagerVariable>::Type const&, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double>, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double> > > >&, phi::Place const&, bool, std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > > const&, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double>, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double> > > >*, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/imperative/tracer.cc:340:7
    #16 0x7ff68018d627 in paddle::imperative::Tracer::TraceOp(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<std::shared_ptr<egr::EagerVariable>, std::allocator<std::shared_ptr<egr::EagerVariable> > >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::vector<std::shared_ptr<egr::EagerVariable>, std::allocator<std::shared_ptr<egr::EagerVariable> > > > > > const&, std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<std::shared_ptr<egr::EagerVariable>, std::allocator<std::shared_ptr<egr::EagerVariable> > >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::vector<std::shared_ptr<egr::EagerVariable>, std::allocator<std::shared_ptr<egr::EagerVariable> > > > > > const&, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double>, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double> > > >&, phi::Place const&, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double>, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double> > > >*, bool, std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > > const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/imperative/tracer.cc:432:3
    #17 0x7ff68223128d in graph_reindex_dygraph_function(paddle::experimental::Tensor const&, paddle::experimental::Tensor const&, paddle::experimental::Tensor const&, paddle::experimental::Tensor const&, paddle::experimental::Tensor const&, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double>, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, paddle::variant<paddle::blank, int, float, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, bool, std::vector<bool, std::allocator<bool> >, paddle::framework::BlockDesc*, long, std::vector<paddle::framework::BlockDesc*, std::allocator<paddle::framework::BlockDesc*> >, std::vector<long, std::allocator<long> >, std::vector<double, std::allocator<double> >, paddle::framework::VarDesc*, std::vector<paddle::framework::VarDesc*, std::allocator<paddle::framework::VarDesc*> >, double> > > > const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/fluid_generated/forwards/dygraph_forward_functions3.cc:4506:51
    #18 0x7ff67d610319 in paddle::pybind::eager_legacy_api_graph_reindex(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_legacy_op_function.cc:5178:16
    #19 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #20 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #21 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #22 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #23 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #24 0x5610ed in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x5610ed)
    #25 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #26 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #27 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #28 0x62a179  (/usr/bin/python3.8+0x62a179)
    #29 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #30 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #31 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #32 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #33 0x7ff74466283f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #34 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/graph_reindex_kernel.cc:63:40 in void phi::GraphReindexKernel<int, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, paddle::optional<phi::DenseTensor> const&, paddle::optional<phi::DenseTensor> const&, bool, phi::DenseTensor*, phi::DenseTensor*, phi::DenseTensor*)
==107590==ABORTING

Case18: paddle.logsumexp

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0] * 3), dtype='float32')
paddle.logsumexp(x, axis=0)

Stack Trace Message：

AddressSanitizer:DEADLYSIGNAL
=================================================================
==9616==ERROR: AddressSanitizer: FPE on unknown address 0x7fd5f349ccd7 (pc 0x7fd5f349ccd7 bp 0x7ffcef21ee30 sp 0x7ffcef21edc8 T0)
    #0 0x7fd5f349ccd7 in __udivti3 (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7)
    #1 0x7fd53f31db05 in Eigen::internal::(anonymous namespace)::DividerHelper<64, long>::computeMultiplier(int, long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128:82
    #2 0x7fd53f31db05 in Eigen::internal::TensorIntDivisor<long, false>::TensorIntDivisor(long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:165:18
    #3 0x7fd53f31db05 in Eigen::TensorReductionEvaluatorBase<Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::DefaultDevice>::TensorReductionEvaluatorBase(Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h:584:36
    #4 0x7fd5402479d8 in Eigen::TensorEvaluator<Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h:990:115
    #5 0x7fd5402479d8 in Eigen::TensorEvaluator<Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_log_op<float>, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_exp_op<float>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReshapingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const> const> const> const, Eigen::MakePointer> const> const> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_log_op<float>, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_exp_op<float>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReshapingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const> const> const> const, Eigen::MakePointer> const> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorEvaluator.h:577:7
    #6 0x7fd5402479d8 in Eigen::TensorEvaluator<Eigen::TensorReshapingOp<Eigen::DSizes<long, 2> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_log_op<float>, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_exp_op<float>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReshapingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const> const> const> const, Eigen::MakePointer> const> const> const> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorReshapingOp<Eigen::DSizes<long, 2> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_log_op<float>, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_exp_op<float>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReshapingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const> const> const> const, Eigen::MakePointer> const> const> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorMorphing.h:146:9
    #7 0x7fd5402479d8 in Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorReshapingOp<Eigen::DSizes<long, 2> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_log_op<float>, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_exp_op<float>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReshapingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const> const> const> const, Eigen::MakePointer> const> const> const> const> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorReshapingOp<Eigen::DSizes<long, 2> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_log_op<float>, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_exp_op<float>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReshapingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const> const> const> const, Eigen::MakePointer> const> const> const> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h:129:7
    #8 0x7fd5402479d8 in Eigen::internal::TensorExecutor<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorReshapingOp<Eigen::DSizes<long, 2> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_log_op<float>, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_exp_op<float>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReshapingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const> const> const> const, Eigen::MakePointer> const> const> const> const> const, Eigen::DefaultDevice, true, (Eigen::internal::TiledEvaluation)0>::run(Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorReshapingOp<Eigen::DSizes<long, 2> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_log_op<float>, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_exp_op<float>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReshapingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const> const> const> const, Eigen::MakePointer> const> const> const> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h:130:48
    #9 0x7fd54021f111 in Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>& Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>::operator=<Eigen::TensorReshapingOp<Eigen::DSizes<long, 2> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_log_op<float>, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_exp_op<float>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReshapingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const> const> const> const, Eigen::MakePointer> const> const> const> >(Eigen::TensorReshapingOp<Eigen::DSizes<long, 2> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_log_op<float>, Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, std::array<int, 1ul> const, Eigen::TensorCwiseUnaryOp<Eigen::internal::scalar_exp_op<float>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReshapingOp<Eigen::DSizes<long, 3> const, Eigen::TensorReductionOp<Eigen::internal::MaxReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const> const> const> const, Eigen::MakePointer> const> const> const> const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h:37:7
    #10 0x7fd54021f111 in void phi::LogsumexpFunctor<float>::operator()<Eigen::DefaultDevice, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, std::array<int, 1ul> >(Eigen::DefaultDevice const&, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer>*, Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>*, std::array<int, 1ul> const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/impl/logsumexp_kernel_impl.h:55:22
    #11 0x7fd54021f111 in void phi::funcs::ReduceFunctor<phi::CPUContext, float, 3ul, 1ul, phi::LogsumexpFunctor<float> >(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<long, std::allocator<long> > const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/reduce_function.h:1246:5
    #12 0x7fd5402160f3 in void phi::LogsumexpKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, std::vector<long, std::allocator<long> > const&, bool, bool, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/impl/logsumexp_kernel_impl.h:106:5
    #13 0x7fd53c3b936e in paddle::experimental::logsumexp(paddle::experimental::Tensor const&, std::vector<long, std::allocator<long> > const&, bool, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:16587:5
    #14 0x7fd5301de5fa in logsumexp_ad_func(paddle::experimental::Tensor const&, std::vector<long, std::allocator<long> >, bool, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:27492:21
    #15 0x7fd52c6bdc40 in paddle::pybind::eager_api_logsumexp(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:12017:68
    #16 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #17 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #18 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #19 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #20 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #21 0x5610ed in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x5610ed)
    #22 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #23 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #24 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #25 0x62a179  (/usr/bin/python3.8+0x62a179)
    #26 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #27 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #28 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #29 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #30 0x7fd5f3ab383f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #31 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7) in __udivti3
==9616==ABORTING

Case19: paddle.min

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0, 0, 0]), dtype='float32')
paddle.min(x, axis=0)

Stack Trace Message：

AddressSanitizer:DEADLYSIGNAL
=================================================================
==194480==ERROR: AddressSanitizer: FPE on unknown address 0x7fd1b1773cd7 (pc 0x7fd1b1773cd7 bp 0x7ffe666c4f10 sp 0x7ffe666c4ea8 T0)
    #0 0x7fd1b1773cd7 in __udivti3 (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7)
    #1 0x7fd0fcf98c15 in Eigen::internal::(anonymous namespace)::DividerHelper<64, long>::computeMultiplier(int, long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128:82
    #2 0x7fd0fcf98c15 in Eigen::internal::TensorIntDivisor<long, false>::TensorIntDivisor(long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:165:18
    #3 0x7fd0fcf98c15 in Eigen::TensorReductionEvaluatorBase<Eigen::TensorReductionOp<Eigen::internal::MinReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::DefaultDevice>::TensorReductionEvaluatorBase(Eigen::TensorReductionOp<Eigen::internal::MinReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h:584:36
    #4 0x7fd0fcf9633e in Eigen::TensorEvaluator<Eigen::TensorReductionOp<Eigen::internal::MinReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorReductionOp<Eigen::internal::MinReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h:990:115
    #5 0x7fd0fcf9633e in Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorReductionOp<Eigen::internal::MinReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorReductionOp<Eigen::internal::MinReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h:129:7
    #6 0x7fd0fcf9633e in Eigen::internal::TensorExecutor<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorReductionOp<Eigen::internal::MinReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const, Eigen::DefaultDevice, true, (Eigen::internal::TiledEvaluation)0>::run(Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorReductionOp<Eigen::internal::MinReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h:130:48
    #7 0x7fd0fcf68494 in Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>& Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>::operator=<Eigen::TensorReductionOp<Eigen::internal::MinReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> >(Eigen::TensorReductionOp<Eigen::internal::MinReducer<float, 0>, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h:37:7
    #8 0x7fd0fcf68494 in void phi::funcs::MinFunctor::operator()<Eigen::DefaultDevice, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, std::array<int, 1ul> >(Eigen::DefaultDevice const&, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, long>, 0, Eigen::MakePointer>*, Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>*, std::array<int, 1ul> const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/reduce_functor.h:87:22
    #9 0x7fd0fcf68494 in void phi::funcs::ReduceFunctor<phi::CPUContext, float, 3ul, 1ul, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<long, std::allocator<long> > const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/reduce_function.h:1246:5
    #10 0x7fd0fcf51780 in void phi::funcs::ReduceKernelImpl<phi::CPUContext, float, float, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<long, std::allocator<long> > const&, bool, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/reduce_function.h:1367:7
    #11 0x7fd0fcda72ce in void phi::Reduce<phi::CPUContext, float, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, bool, std::vector<long, std::allocator<long> > const&, bool, paddle::experimental::DataType, phi::DenseTensor*)::'lambda'()::operator()() const::'lambda7'()::operator()() const /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/reduce.h:49:5
    #12 0x7fd0fcda72ce in void phi::Reduce<phi::CPUContext, float, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, bool, std::vector<long, std::allocator<long> > const&, bool, paddle::experimental::DataType, phi::DenseTensor*)::'lambda'()::operator()() const /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/reduce.h:49:5
    #13 0x7fd0fcd99a86 in void phi::Reduce<phi::CPUContext, float, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, bool, std::vector<long, std::allocator<long> > const&, bool, paddle::experimental::DataType, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/reduce.h:49:5
    #14 0x7fd0fcd992f3 in void phi::MinRawKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, paddle::experimental::IntArrayBase<phi::DenseTensor> const&, bool, bool, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/reduce_min_kernel.cc:32:3
    #15 0x7fd0fa6d82de in paddle::experimental::min(paddle::experimental::Tensor const&, paddle::experimental::IntArrayBase<paddle::experimental::Tensor> const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:18393:5
    #16 0x7fd0ee53f6de in min_ad_func(paddle::experimental::Tensor const&, paddle::experimental::IntArrayBase<paddle::experimental::Tensor>, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:30228:21
    #17 0x7fd0ea9cfe05 in paddle::pybind::eager_api_min(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:13248:51
    #18 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #19 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #20 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #21 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #22 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #23 0x5610ed in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x5610ed)
    #24 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #25 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #26 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #27 0x62a179  (/usr/bin/python3.8+0x62a179)
    #28 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #29 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #30 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #31 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #32 0x7fd1b1d8a83f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #33 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7) in __udivti3
==194480==ABORTING

Case20: paddle.min

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0, 0, 0, 0, 0, 0, 0]), dtype='float32')
paddle.min(x, axis=0)

Stack Trace Message：

AddressSanitizer:DEADLYSIGNAL
=================================================================
==108797==ERROR: AddressSanitizer: FPE on unknown address 0x7f40f66a4be3 (pc 0x7f40f66a4be3 bp 0x7ffd361ea110 sp 0x7ffd361e9da0 T0)
    #0 0x7f40f66a4be3 in void phi::funcs::HandleLargeDim<phi::CPUContext, float, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<long, std::allocator<long> > const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/reduce_function.h:1317:50
    #1 0x7f40f66a4394 in void phi::funcs::ReduceKernelImpl<phi::CPUContext, float, float, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<long, std::allocator<long> > const&, bool, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/reduce_function.h:1350:7
    #2 0x7f40f64fa2ce in void phi::Reduce<phi::CPUContext, float, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, bool, std::vector<long, std::allocator<long> > const&, bool, paddle::experimental::DataType, phi::DenseTensor*)::'lambda'()::operator()() const::'lambda7'()::operator()() const /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/reduce.h:49:5
    #3 0x7f40f64fa2ce in void phi::Reduce<phi::CPUContext, float, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, bool, std::vector<long, std::allocator<long> > const&, bool, paddle::experimental::DataType, phi::DenseTensor*)::'lambda'()::operator()() const /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/reduce.h:49:5
    #4 0x7f40f64eca86 in void phi::Reduce<phi::CPUContext, float, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, bool, std::vector<long, std::allocator<long> > const&, bool, paddle::experimental::DataType, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/reduce.h:49:5
    #5 0x7f40f64ec2f3 in void phi::MinRawKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, paddle::experimental::IntArrayBase<phi::DenseTensor> const&, bool, bool, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/reduce_min_kernel.cc:32:3
    #6 0x7f40f3e2b2de in paddle::experimental::min(paddle::experimental::Tensor const&, paddle::experimental::IntArrayBase<paddle::experimental::Tensor> const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:18393:5
    #7 0x7f40e7c926de in min_ad_func(paddle::experimental::Tensor const&, paddle::experimental::IntArrayBase<paddle::experimental::Tensor>, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:30228:21
    #8 0x7f40e4122e05 in paddle::pybind::eager_api_min(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:13248:51
    #9 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #10 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #11 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #12 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #13 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #14 0x5610ed in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x5610ed)
    #15 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #16 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #17 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #18 0x62a179  (/usr/bin/python3.8+0x62a179)
    #19 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #20 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #21 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #22 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #23 0x7f41ab4dd83f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #24 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/reduce_function.h:1317:50 in void phi::funcs::HandleLargeDim<phi::CPUContext, float, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<long, std::allocator<long> > const&, bool)
==108797==ABORTING

Case21: paddle.mode

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0, 0]), dtype='float32')
paddle.mode(x, axis=0)

Stack Trace Message：

AddressSanitizer:DEADLYSIGNAL
=================================================================
==163314==ERROR: AddressSanitizer: FPE on unknown address 0x7f55be0eecd7 (pc 0x7f55be0eecd7 bp 0x7fff22135930 sp 0x7fff221358c8 T0)
    #0 0x7f55be0eecd7 in __udivti3 (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7)
    #1 0x7f550e4e14bf in Eigen::internal::(anonymous namespace)::DividerHelper<64, long>::computeMultiplier(int, long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128:82
    #2 0x7f550e4e14bf in Eigen::internal::TensorIntDivisor<long, false>::TensorIntDivisor(long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:165:18
    #3 0x7f550e4e14bf in Eigen::TensorEvaluator<Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorShuffling.h:155:34
    #4 0x7f550e4e14bf in Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h:129:7
    #5 0x7f550e4df9d0 in Eigen::internal::TensorExecutor<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const> const, Eigen::DefaultDevice, true, (Eigen::internal::TiledEvaluation)1>::run(Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h:186:15
    #6 0x7f550e40776d in Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>& Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>::operator=<Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> >(Eigen::TensorShufflingOp<std::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, long>, 0, Eigen::MakePointer> const> const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h:37:7
    #7 0x7f550e40776d in phi::funcs::Transpose<phi::CPUContext, float, 2>::operator()(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<int, std::allocator<int> > const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/math_function_impl.h:73:28
    #8 0x7f5507da61b3 in void phi::funcs::TransCompute<phi::CPUContext, float>(int, phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<int, std::allocator<int> > const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/math_function.h:147:7
    #9 0x7f550aafd5b0 in void phi::ModeKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, int, bool, phi::DenseTensor*, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/mode_kernel.cc:90:5
    #10 0x7f550705a5f8 in paddle::experimental::mode(paddle::experimental::Tensor const&, int, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:18599:5
    #11 0x7f54faece05d in mode_ad_func(paddle::experimental::Tensor const&, int, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:30610:21
    #12 0x7f54f7352432 in paddle::pybind::eager_api_mode(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:13402:52
    #13 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #14 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #15 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #16 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #17 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #18 0x5610ed in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x5610ed)
    #19 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #20 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #21 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #22 0x62a179  (/usr/bin/python3.8+0x62a179)
    #23 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #24 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #25 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #26 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #27 0x7f55be70583f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #28 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7) in __udivti3
==163314==ABORTING

Case22: paddle.lerp

Reproduced Case：

import paddle
import numpy as np
x = paddle.to_tensor([83.92], dtype='float32')
array = np.array([], dtype=np.float32)
y = paddle.to_tensor(np.reshape(array, [0]), dtype='float32')
weight = paddle.to_tensor(np.reshape(array, [0]), dtype='float32')
paddle.lerp(x, y, weight)

Stack Trace Message：

AddressSanitizer:DEADLYSIGNAL
=================================================================
==210993==ERROR: AddressSanitizer: FPE on unknown address 0x7f36fca923d4 (pc 0x7f36fca923d4 bp 0x7ffd06475280 sp 0x7ffd064750a0 T0)
    #0 0x7f36fca923d4 in Eigen::TensorEvaluator<Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::indexRowMajor(long) const /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h:312:30
    #1 0x7f36fca923d4 in Eigen::TensorEvaluator<Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::BroadcastBlock(Eigen::DSizes<long, 1> const&, Eigen::DSizes<long, 1> const&, Eigen::DSizes<long, 2> const&, Eigen::DSizes<long, 2> const&, Eigen::DSizes<long, 2> const&, long, long, Eigen::internal::TensorBlockScratchAllocator<Eigen::DefaultDevice>&, float*, float**, unsigned long*) const /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h:1030:52
    #2 0x7f36fca91949 in Eigen::TensorEvaluator<Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::BroadcastBlockAlongBcastDim(Eigen::TensorEvaluator<Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::BlockBroadcastingParams, long, Eigen::internal::TensorBlockScratchAllocator<Eigen::DefaultDevice>&, float*, float**, unsigned long*) const /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h
    #3 0x7f36fca90d2c in Eigen::TensorEvaluator<Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::block(Eigen::internal::TensorBlockDescriptor<1, long>&, Eigen::internal::TensorBlockScratchAllocator<Eigen::DefaultDevice>&, bool) const /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h:674:28
    #4 0x7f370541d443 in Eigen::TensorEvaluator<Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_product_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const, Eigen::DefaultDevice>::block(Eigen::internal::TensorBlockDescriptor<1, long>&, Eigen::internal::TensorBlockScratchAllocator<Eigen::DefaultDevice>&, bool) const /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorEvaluator.h:670:35
    #5 0x7f370541d443 in Eigen::TensorEvaluator<Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_product_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const> const, Eigen::DefaultDevice>::block(Eigen::internal::TensorBlockDescriptor<1, long>&, Eigen::internal::TensorBlockScratchAllocator<Eigen::DefaultDevice>&, bool) const /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorEvaluator.h:671:38
    #6 0x7f370541d443 in Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_product_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const> const> const, Eigen::DefaultDevice>::evalBlock(Eigen::internal::TensorBlockDescriptor<1, long>&, Eigen::internal::TensorBlockScratchAllocator<Eigen::DefaultDevice>&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h:221:42
    #7 0x7f370541c9a8 in Eigen::internal::TensorExecutor<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_product_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const> const> const, Eigen::DefaultDevice, true, (Eigen::internal::TiledEvaluation)1>::run(Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_product_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h:206:19
    #8 0x7f370540ac08 in Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>& Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>::operator=<Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_product_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const> >(Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_product_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_difference_op<float const, float const>, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const> const> const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h:37:7
    #9 0x7f370540ac08 in void phi::LerpFunction<phi::CPUContext, float, 1ul>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/impl/lerp_kernel_impl.h:48:27
    #10 0x7f370540ac08 in void phi::LerpKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/impl/lerp_kernel_impl.h:77:7
    #11 0x7f37013bb838 in paddle::experimental::lerp(paddle::experimental::Tensor const&, paddle::experimental::Tensor const&, paddle::experimental::Tensor const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:15644:5
    #12 0x7f36f51af9ca in lerp_ad_func(paddle::experimental::Tensor const&, paddle::experimental::Tensor const&, paddle::experimental::Tensor const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:26067:21
    #13 0x7f36f16bea7e in paddle::pybind::eager_api_lerp(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:11325:48
    #14 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #15 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #16 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #17 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #18 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #19 0x564f2d in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x564f2d)
    #20 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #21 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #22 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #23 0x62a179  (/usr/bin/python3.8+0x62a179)
    #24 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #25 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #26 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #27 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #28 0x7f37b8adb83f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #29 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h:312:30 in Eigen::TensorEvaluator<Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::indexRowMajor(long) const
==210993==ABORTING

Case23: paddle.floor_mod

Reproduced Case：

import paddle
import numpy as np
x = paddle.to_tensor([59], dtype='int32')
y = paddle.to_tensor([0], dtype='int32')
paddle.floor_mod(x, y)

Stack Trace Message：

==155145==ERROR: AddressSanitizer: FPE on unknown address 0x7f85fcaa53c4 (pc 0x7f85fcaa53c4 bp 0x7ffdf3c3d450 sp 0x7ffdf3c3cd20 T0)
    #0 0x7f85fcaa53c4 in phi::funcs::RemainderFunctor<int, void>::operator()(int, int) const /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/elementwise_functor.h:504:15
    #1 0x7f85fcaa53c4 in int* std::transform<int const*, int const*, int*, phi::funcs::RemainderFunctor<int, void> >(int const*, int const*, int const*, int*, phi::funcs::RemainderFunctor<int, void>) /usr/bin/../lib/gcc/x86_64-linux-gnu/8/../../../../include/c++/8/bits/stl_algo.h:4343:14
    #2 0x7f85fcaa53c4 in void paddle::platform::Transform<phi::CPUContext>::operator()<int const*, int const*, int*, phi::funcs::RemainderFunctor<int, void> >(phi::CPUContext const&, int const*, int const*, int const*, int*, phi::funcs::RemainderFunctor<int, void>) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/platform/transform.h:94:5
    #3 0x7f85fcaa53c4 in phi::funcs::TransformFunctor<phi::funcs::RemainderFunctor<int, void>, int, phi::CPUContext, int>::Run() const /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/elementwise_base.h:224:5
    #4 0x7f85fcaa53c4 in void phi::funcs::ElementwiseCompute<phi::funcs::RemainderFunctor<int, void>, int, int>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, int, phi::funcs::RemainderFunctor<int, void>, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/elementwise_base.h:386:13
    #5 0x7f85fcaa4de2 in void phi::RemainderRawKernel<int, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, int, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/elementwise_kernel.cc:59:5
    #6 0x7f85f7aa7572 in paddle::experimental::remainder(paddle::experimental::Tensor const&, paddle::experimental::Tensor const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:21029:5
    #7 0x7f85eb996a7c in remainder_ad_func(paddle::experimental::Tensor const&, paddle::experimental::Tensor const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:34567:21
    #8 0x7f85e7da8994 in paddle::pybind::eager_api_remainder(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:15350:46
    #9 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #10 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #11 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #12 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #13 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #14 0x564f2d in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x564f2d)
    #15 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #16 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #17 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #18 0x62a179  (/usr/bin/python3.8+0x62a179)
    #19 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #20 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #21 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #22 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #23 0x7f86af0fb83f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #24 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/elementwise_functor.h:504:15 in phi::funcs::RemainderFunctor<int, void>::operator()(int, int) const
==155145==ABORTING

Case24: paddle.chunk

Reproduced Case：

import paddle
import numpy as np
array = np.array([5.421010862427522e-18, 0.0, 0.0, 0.0], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [4]), dtype='float32')
paddle.chunk(x, chunks=0)

Stack Trace Message：

==202701==ERROR: AddressSanitizer: FPE on unknown address 0x7f26506d521c (pc 0x7f26506d521c bp 0x7ffca86775d0 sp 0x7ffca8676ee0 T0)
    #0 0x7f26506d521c in phi::SplitWithNumInferMeta(phi::MetaTensor const&, int, paddle::experimental::ScalarBase<phi::DenseTensor> const&, std::vector<phi::MetaTensor*, std::allocator<phi::MetaTensor*> >, phi::MetaConfig) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/infermeta/unary.cc:3532:5
    #1 0x7f26498cf7b0 in paddle::experimental::split_with_num(paddle::experimental::Tensor const&, int, paddle::experimental::ScalarBase<paddle::experimental::Tensor> const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:23771:3
    #2 0x7f263d81dd27 in split_with_num_ad_func(paddle::experimental::Tensor const&, int, paddle::experimental::ScalarBase<paddle::experimental::Tensor>) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:38631:21
    #3 0x7f2639bc1903 in paddle::pybind::eager_api_split_with_num(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:17138:58
    #4 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #5 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #6 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #7 0x55fa97 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55fa97)
    #8 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #9 0x5610ed in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x5610ed)
    #10 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #11 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #12 0x5610ed in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x5610ed)
    #13 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #14 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #15 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #16 0x62a179  (/usr/bin/python3.8+0x62a179)
    #17 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #18 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #19 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #20 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #21 0x7f2700eb383f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #22 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)

Case25: paddle.dot

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0]), dtype='float32')
y = paddle.to_tensor(np.reshape(array, [0]), dtype='float32')
paddle.dot(x, y)

Stack Trace Message：

==115998==ERROR: AddressSanitizer: FPE on unknown address 0x7fc3c30398ca (pc 0x7fc3c30398ca bp 0x7ffe65586ed0 sp 0x7ffe65586df0 T0)
    #0 0x7fc3c30398ca in void phi::DotKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/dot_kernel.cc:40:25
    #1 0x7fc3c30398ca in phi::KernelImpl<void (*)(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor*), &(void phi::DotKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor*))>::VariadicCompute(phi::DeviceContext const&, phi::DenseTensor const&, phi::DenseTensor const&, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/core/kernel_utils.h:239:12
    #2 0x7fc3bd6f9a32 in paddle::experimental::dot(paddle::experimental::Tensor const&, paddle::experimental::Tensor const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:1917:5
    #3 0x7fc3b129c433 in dot_ad_func(paddle::experimental::Tensor const&, paddle::experimental::Tensor const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:4496:21
    #4 0x7fc3ad9fdc16 in paddle::pybind::eager_api_dot(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:1414:40
    #5 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #6 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #7 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #8 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #9 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #10 0x564f2d in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x564f2d)
    #11 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #12 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #13 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #14 0x62a179  (/usr/bin/python3.8+0x62a179)
    #15 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #16 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #17 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #18 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #19 0x7fc47501783f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #20 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)

Case26: paddle.all

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.int32)
array = array.astype(bool)
x = paddle.to_tensor(np.reshape(array, (0, 0, 0)), dtype='bool')
paddle.all(x, axis=0, keepdim=True)

Stack Trace Message：

==92180==ERROR: AddressSanitizer: FPE on unknown address 0x7f73d5902cd7 (pc 0x7f73d5902cd7 bp 0x7ffd495de730 sp 0x7ffd495de6c8 T0)
    #0 0x7f73d5902cd7 in __udivti3 (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7)
    #1 0x7f7321ab00f5 in Eigen::internal::(anonymous namespace)::DividerHelper<64, long>::computeMultiplier(int, long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128:82
    #2 0x7f7321ab00f5 in Eigen::internal::TensorIntDivisor<long, false>::TensorIntDivisor(long) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:165:18
    #3 0x7f7321ab00f5 in Eigen::TensorReductionEvaluatorBase<Eigen::TensorReductionOp<Eigen::internal::AndReducer, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<bool const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::DefaultDevice>::TensorReductionEvaluatorBase(Eigen::TensorReductionOp<Eigen::internal::AndReducer, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<bool const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h:584:36
    #4 0x7f7321aaf6b0 in Eigen::TensorEvaluator<Eigen::TensorReductionOp<Eigen::internal::AndReducer, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<bool const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorReductionOp<Eigen::internal::AndReducer, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<bool const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h:990:115
    #5 0x7f7321aaf6b0 in Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<bool, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorReductionOp<Eigen::internal::AndReducer, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<bool const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::TensorEvaluator(Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<bool, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorReductionOp<Eigen::internal::AndReducer, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<bool const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h:129:7
    #6 0x7f7321aaf6b0 in Eigen::internal::TensorExecutor<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<bool, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorReductionOp<Eigen::internal::AndReducer, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<bool const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const, Eigen::DefaultDevice, false, (Eigen::internal::TiledEvaluation)0>::run(Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<bool, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorReductionOp<Eigen::internal::AndReducer, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<bool const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const> const&, Eigen::DefaultDevice const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h:98:41
    #7 0x7f7321aaf6b0 in Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<bool, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>& Eigen::TensorDevice<Eigen::TensorMap<Eigen::Tensor<bool, 2, 1, long>, 0, Eigen::MakePointer>, Eigen::DefaultDevice>::operator=<Eigen::TensorReductionOp<Eigen::internal::AndReducer, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<bool const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> >(Eigen::TensorReductionOp<Eigen::internal::AndReducer, std::array<int, 1ul> const, Eigen::TensorMap<Eigen::Tensor<bool const, 3, 1, long>, 0, Eigen::MakePointer> const, Eigen::MakePointer> const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h:37:7
    #8 0x7f7321aaf6b0 in void phi::funcs::AllFunctor::operator()<Eigen::DefaultDevice, Eigen::TensorMap<Eigen::Tensor<bool const, 3, 1, long>, 0, Eigen::MakePointer>, Eigen::TensorMap<Eigen::Tensor<bool, 2, 1, long>, 0, Eigen::MakePointer>, std::array<int, 1ul> >(Eigen::DefaultDevice const&, Eigen::TensorMap<Eigen::Tensor<bool const, 3, 1, long>, 0, Eigen::MakePointer>*, Eigen::TensorMap<Eigen::Tensor<bool, 2, 1, long>, 0, Eigen::MakePointer>*, std::array<int, 1ul> const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/reduce_functor.h:95:22
    #9 0x7f7321a99285 in void phi::funcs::ReduceFunctor<phi::CPUContext, bool, 3ul, 1ul, phi::funcs::AllFunctor>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<long, std::allocator<long> > const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/reduce_function.h:1246:5
    #10 0x7f7321a81fba in void phi::funcs::ReduceKernelImpl<phi::CPUContext, bool, bool, phi::funcs::AllFunctor>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<long, std::allocator<long> > const&, bool, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/funcs/reduce_function.h:1367:7
    #11 0x7f7321a80044 in void phi::BoolReduceKernel<phi::CPUContext, bool, phi::funcs::AllFunctor>(phi::CPUContext const&, phi::DenseTensor const&, std::vector<long, std::allocator<long> > const&, bool, bool, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/cpu/reduce.h:88:3
    #12 0x7f731e6b4145 in paddle::experimental::all(paddle::experimental::Tensor const&, std::vector<long, std::allocator<long> > const&, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:7076:5
    #13 0x7f7312318d1f in all_ad_func(paddle::experimental::Tensor const&, std::vector<long, std::allocator<long> >, bool) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:12904:21
    #14 0x7f730e9ae7ec in paddle::pybind::eager_api_all(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:5051:51
    #15 0x5025f3 in PyCFunction_Call (/usr/bin/python3.8+0x5025f3)
    #16 0x500db4 in _PyObject_MakeTpCall (/usr/bin/python3.8+0x500db4)
    #17 0x566223 in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x566223)
    #18 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #19 0x5016c5 in _PyFunction_Vectorcall (/usr/bin/python3.8+0x5016c5)
    #20 0x5610ed in _PyEval_EvalFrameDefault (/usr/bin/python3.8+0x5610ed)
    #21 0x55f470 in _PyEval_EvalCodeWithName (/usr/bin/python3.8+0x55f470)
    #22 0x55f102 in PyEval_EvalCode (/usr/bin/python3.8+0x55f102)
    #23 0x62a1ef  (/usr/bin/python3.8+0x62a1ef)
    #24 0x62a179  (/usr/bin/python3.8+0x62a179)
    #25 0x47a7f2  (/usr/bin/python3.8+0x47a7f2)
    #26 0x47a5cb in PyRun_SimpleFileExFlags (/usr/bin/python3.8+0x47a5cb)
    #27 0x4247dc in _init (/usr/bin/python3.8+0x4247dc)
    #28 0x5fb9b8 in Py_BytesMain (/usr/bin/python3.8+0x5fb9b8)
    #29 0x7f73d5f1983f in __libc_start_main /build/glibc-S7Ft5T/glibc-2.23/csu/../csu/libc-start.c:291
    #30 0x5fb8b8 in _start (/usr/bin/python3.8+0x5fb8b8)

Case27: paddle.nn.functional.label_smooth

Reproduced Case：

import paddle
import numpy as np
array = np.array([1, 2, 3, 4], dtype=np.float32)
label = paddle.to_tensor(np.reshape(array, [4]), dtype='float32')
array = np.array([], dtype=np.float32)
prior_dist = paddle.to_tensor(np.reshape(array, [0, 4]), dtype='float32') 
paddle.nn.functional.label_smooth(label, prior_dist = prior_dist)

Stack Trace Message：

==183037==ERROR: AddressSanitizer: FPE on unknown address 0x7fd2bbea6e9a (pc 0x7fd2bbea6e9a bp 0x7ffffc7add00 sp 0x7ffffc7adb20 T0)
 const, Eigen::TensorMap::BroadcastBlockAlongBcastDim(Eigen::TensorEvaluator const, Eigen::TensorMap::BlockBroadcastingParams, long, Eigen::internal::TensorBlockScratchAllocator    #0 0x7fd2bbea6e9a in Eigen::TensorEvaluator<Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::BroadcastBlockAlongBcastDim(Eigen::TensorEvaluator<Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::BlockBroadcastingParams, long, Eigen::internal::TensorBlockScratchAllocator<Eigen::DefaultDevice>&, float*, float**, unsigned long*) const /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h:914:68
 const, Eigen::TensorMap::block(Eigen::internal::TensorBlockDescriptor&, Eigen::internal::TensorBlockScratchAllocator    #1 0x7fd2bbea63fc in Eigen::TensorEvaluator<Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const, Eigen::DefaultDevice>::block(Eigen::internal::TensorBlockDescriptor<1, long>&, Eigen::internal::TensorBlockScratchAllocator<Eigen::DefaultDevice>&, bool) const /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h:674:28
, Eigen::TensorBroadcastingOp const, Eigen::TensorMap::block(Eigen::internal::TensorBlockDescriptor&, Eigen::internal::TensorBlockScratchAllocator    #2 0x7fd2c3b21cb4 in Eigen::TensorEvaluator<Eigen::TensorCwiseUnaryOp<Eigen::internal::bind1st_op<Eigen::internal::scalar_product_op<float const, float const> >, Eigen::TensorBroadcastingOp<Eigen::DSizes<int, 1> const, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 0, Eigen::MakePointer> const> const> const, Eigen::DefaultDevice>::block(Eigen::internal::TensorBlockDescriptor<1, long>&, Eigen::internal::TensorBlockScratchAllocator<Eigen::DefaultDevice>&, bool) const /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorEvaluator.h:532:34
    #3 0x7fd2c3b21cb4 in Eigen::TensorEvaluator<Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>,

Case28: paddle.nn.functional.conv3d/conv2d

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0, 0, 0, 0, 0]), dtype='float32')
weight = paddle.to_tensor(np.reshape(array, [1, 0, 0, 0, 0]), dtype='float32')
paddle.nn.functional.conv3d(x, weight)
import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0, 0, 0, 0]), dtype='float32')
weight = paddle.to_tensor(np.reshape(array, [1, 0, 0, 0]), dtype='float32')
paddle.nn.functional.conv2d(x, weight)
import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0, 0, 0]), dtype='float32')
weight = paddle.to_tensor(np.reshape(array, [1, 0, 0]), dtype='float32')
paddle.nn.functional.conv1d(x, weight)

Stack Trace Message：

==99849==ERROR: AddressSanitizer: FPE on unknown address 0x7fab91f70117 (pc 0x7fab91f70117 bp 0x7fffe8ceb070 sp 0x7fffe8ce9a20 T0)
(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, std::vector const&, std::vector const&, std::__cxx11::basic_string, std::allocator const&, int, std::vector const&, std::__cxx11::basic_string, std::allocator    #0 0x7fab91f70117 in void phi::ConvKernelImpl<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, int, std::vector<int, std::allocator<int> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/kernels/impl/conv_kernel_impl.h:129:34
 const&, std::vector const&, std::__cxx11::basic_string, std::allocator const&, int, std::vector const&, std::__cxx11::basic_string, std::allocator    #1 0x7fab8c42b598 in paddle::experimental::conv3d(paddle::experimental::Tensor const&, paddle::experimental::Tensor const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, int, std::vector<int, std::allocator<int> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:9716:5
, std::vector, std::__cxx11::basic_string, std::allocator, int, std::vector, std::__cxx11::basic_string, std::allocator    #2 0x7fab80107b7c in conv3d_ad_func(paddle::experimental::Tensor const&, paddle::experimental::Tensor const&, std::vector<int, std::allocator<int> >, std::vector<int, std::allocator<int> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, int, std::vector<int, std::allocator<int> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:16762:21

NOTE：https://github.com/PaddlePaddle/Paddle/blob/60d1199a8ca26958ed44058987689aa06a169e9d/paddle/phi/kernels/impl/conv_kernel_impl.h#L127-L130

paddle.nn.functional.conv3d, paddle.nn.functional.conv2d, paddle.nn.functional.conv1d 都可以触发 paddle/phi/kernels/impl/conv_kernel_impl.h第129行的除0问题
All of paddle.nn.functional.conv3d, paddle.nn.functional.conv2d, paddle.nn.functional.conv1d could trigger the divide by 0 problems on line 129 of paddle/phi/kernels/impl/conv_kernel_impl.h.

Case29: paddle.nn.functional.conv1d_transpose/conv3d_transpose

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [6, 0, 6]), dtype='float32')
weight = paddle.to_tensor(np.reshape(array, [6, 0, 0]), dtype='float32')
paddle.nn.functional.conv1d_transpose(x, weight, dilation=0, data_format='NLC')
import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [2, 7, 0, 7, 2]), dtype='float32')
weight = paddle.to_tensor(np.reshape(array, [2, 7, 0, 2, 2]), dtype='float32')
paddle.nn.functional.conv3d_transpose(x, weight, stride=2, data_format='NDHWC')

Stack Trace Message：

==61914==ERROR: AddressSanitizer: FPE on unknown address 0x7fe08a814002 (pc 0x7fe08a814002 bp 0x7ffd589fbeb0 sp 0x7ffd589fbd40 T0)
::operator()(phi::CPUContext const&, std::vector    #0 0x7fe08a814002 in phi::funcs::ConcatFunctor<phi::CPUContext, float>::operator()(phi::CPUContext const&, std::vector<phi::DenseTensor, std::allocator<phi::DenseTensor> > const&, int, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/paddle/phi/kernels/funcs/concat_and_split_functor.cc:42:41
(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, std::vector const&, std::vector const&, std::__cxx11::basic_string, std::allocator const&, int, std::vector const&, std::__cxx11::basic_string, std::allocator    #1 0x7fe088a66f57 in void phi::ConvTransposeRawKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, int, std::vector<int, std::allocator<int> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/paddle/phi/kernels/impl/conv_transpose_kernel_impl.h:222:7
(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, std::vector const&, std::vector const&, std::vector const&, paddle::experimental::IntArrayBase const&, std::__cxx11::basic_string, std::allocator const&, int, std::vector const&, std::__cxx11::basic_string, std::allocator    #2 0x7fe088a78d4e in void phi::Conv2dTransposeKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, paddle::experimental::IntArrayBase<phi::DenseTensor> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, int, std::vector<int, std::allocator<int> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, phi::DenseTensor*) /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/paddle/phi/kernels/impl/conv_transpose_kernel_impl.h:240:3

NOTE：https://github.com/PaddlePaddle/Paddle/blob/60d1199a8ca26958ed44058987689aa06a169e9d/paddle/phi/kernels/funcs/concat_and_split_functor.cc#L34-L45

paddle.nn.functional.conv1d_transpose 和 paddle.nn.functional.conv3d_transpose 都可触发paddle/phi/kernels/funcs/concat_and_split_functor.cc 第42行的除0问题。原因是36行 dim_0中有的值为0，导致rows == 0。
Both paddle.nn.functional.conv1d_transpose and paddle.nn.functional.conv3d_transpose could trigger the divide by 0 problems on line 42 of paddle/phi/kernels/funcs/concat_and_split_functor.cc. The reason is that some value of dim_0 = 0 in line 36, which triggers rows == 0 in line 42.

Case30: DataNormKernel

Reproduced Case：

import paddle
import paddle.fluid as fluid
import numpy as np
paddle.enable_static()
x = fluid.data(name='x', shape=[0, 3], dtype='float32')
out = paddle.static.nn.data_norm(x, slot_dim=1)
cpu=fluid.core.CPUPlace()
exe=fluid.Executor(cpu)
exe.run(fluid.default_startup_program())
test_program = fluid.default_main_program().clone(for_test=True)
[out_array] = exe.run(test_program, fetch_list=out, feed={'x': np.ones([0, 3]).astype('float32')})

Stack Trace Message：

Traceback (most recent call last):
--------------------------------------
0   paddle::framework::StandaloneExecutor::Run(std::vector<std::string, std::allocator<std::string > > const&, std::vector<std::string, std::allocator<std::string > > const&)
1   paddle::framework::InterpreterCore::Run(std::vector<std::string, std::allocator<std::string > > const&)
2   paddle::framework::interpreter::build_op_func_list(phi::Place const&, paddle::framework::BlockDesc const&, std::vector<paddle::framework::OpFuncNode, std::allocator<paddle::framework::OpFuncNode> >*, paddle::framework::VariableScope*, bool)
3   std::function<void (paddle::framework::ExecutionContext const&)>::operator()(paddle::framework::ExecutionContext const&) const
4   std::_Function_handler<void (paddle::framework::ExecutionContext const&), paddle::framework::OpKernelRegistrarFunctor<phi::CPUPlace, false, 0ul, paddle::operators::DataNormKernel<paddle::platform::CPUDeviceContext, float>, paddle::operators::DataNormKernel<paddle::platform::CPUDeviceContext, double> >::operator()(char const*, char const*, int) const::{lambda(paddle::framework::ExecutionContext const&)#1}>::_M_invoke(std::_Any_data const&, paddle::framework::ExecutionContext const&)
5   paddle::framework::OpKernelRegistrarFunctor<phi::CPUPlace, false, 0ul, paddle::operators::DataNormKernel<paddle::platform::CPUDeviceContext, float>, paddle::operators::DataNormKernel<paddle::platform::CPUDeviceContext, double> >::operator()(char const*, char const*, int) const::{lambda(paddle::framework::ExecutionContext const&)#1}::operator()(paddle::framework::ExecutionContext const&) const
6   paddle::operators::DataNormKernel<paddle::platform::CPUDeviceContext, float>::Compute(paddle::framework::ExecutionContext const&) const
----------------------
Error Message Summary:
----------------------
FatalError: `Erroneous arithmetic operation` is detected by the operating system.
  [TimeInfo: *** Aborted at 1655865526 (unix time) try "date -d @1655865526" if you are using GNU date ***]
  [SignalInfo: *** SIGFPE (@0x7f4df5a5c026) received by PID 12082 (TID 0x7f4e64f8e740) from PID 18446744073535864870 ***]
Floating point exception (core dumped)

Case31: AffineChannelKernel

Reproduced Case：

import paddle
import paddle.fluid as fluid
import numpy as np
paddle.enable_static()
X = fluid.data(name="X", shape=[0, 1, 2, 2], dtype='float32')
Scale = fluid.layers.create_parameter(shape=[1], dtype="float32",default_initializer=fluid.initializer.Constant(2.0))
Bias = fluid.layers.create_parameter(shape=[1], dtype="float32",default_initializer=fluid.initializer.Constant(0.5))
out = fluid.layers.affine_channel(X,scale=Scale,bias=Bias, data_layout='NCHW')
cpu=fluid.core.CPUPlace()
exe=fluid.Executor(cpu)
exe.run(fluid.default_startup_program())
test_program = fluid.default_main_program().clone(for_test=True)
[out_array] = exe.run(test_program, fetch_list=out, feed={'X': np.ones([0,1,2,2]).astype('float32')})

Stack Trace Message：

(gdb) bt 5
#0  0x00007fff886c757b in paddle::operators::AffineChannelKernel<paddle::platform::CPUDeviceContext, float>::Compute
    (this=0x7fffffffc168, ctx=...) at /root/Paddle/paddle/fluid/operators/affine_channel_op.cc:194
#1  0x00007fff886c5cf8 in paddle::framework::OpKernelRegistrarFunctor<phi::CPUPlace, false, 0ul, paddle::operators::AffineChannelKernel<paddle::platform::CPUDeviceContext, float>, paddle::operators::AffineChannelKernel<paddle::platform::CPUDeviceContext, double> >::operator()(char const*, char const*, int) const::{lambda(paddle::framework::ExecutionContext const&)#1}::operator()(paddle::framework::ExecutionContext const&) const (this=0x7fffffffc5e8, ctx=...)
    at /root/Paddle/paddle/fluid/framework/op_registry.h:182
#2  0x00007fff886ce963 in std::_Function_handler<void (paddle::framework::ExecutionContext const&), paddle::framework::OpKernelRegistrarFunctor<phi::CPUPlace, false, 0ul, paddle::operators::AffineChannelKernel<paddle::platform::CPUDeviceContext, float>, paddle::operators::AffineChannelKernel<paddle::platform::CPUDeviceContext, double> >::operator()(char const*, char const*, int) const::{lambda(paddle::framework::ExecutionContext const&)#1}>::_M_invoke(std::_Any_data const&, paddle::framework::ExecutionContext const&) (__functor=..., __args#0=...)
    at /usr/local/gcc-8.2/include/c++/8.2.0/bits/std_function.h:297
#3  0x00007fff89ba61e1 in std::function<void (paddle::framework::ExecutionContext const&)>::operator()(paddle::framework::ExecutionContext const&) const (this=0x7fffffffc5e8, __args#0=...)
    at /usr/local/gcc-8.2/include/c++/8.2.0/bits/std_function.h:687
#4  0x00007fff89bb6e45 in paddle::framework::interpreter::build_op_func_list (place=..., block=..., 
    vec_func_list=0x7fffffffd490, var_scope=0x555557b987d8, use_local_scope=true)
    at /root/Paddle/paddle/fluid/framework/new_executor/interpretercore_util.cc:512

Case32: paddle.split

Reproduced Case：

import paddle
import numpy as np
array = np.array([2], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [1, 1, 1]), dtype='int32')
paddle.split(x, num_or_sections=0)

Stack Trace Message：

==162174==ERROR: AddressSanitizer: FPE on unknown address 0x7fe4822f121c (pc 0x7fe4822f121c bp 0x7ffc30830b50 sp 0x7ffc30830460 T0)
 const&, std::vector    #0 0x7fe4822f121c in phi::SplitWithNumInferMeta(phi::MetaTensor const&, int, paddle::experimental::ScalarBase<phi::DenseTensor> const&, std::vector<phi::MetaTensor*, std::allocator<phi::MetaTensor*> >, phi::MetaConfig) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/infermeta/unary.cc:3532:5
    #1 0x7fe47b4eb7b0 in paddle::experimental::split_with_num(paddle::experimental::Tensor const&, int, paddle::experimental::ScalarBase<paddle::experimental::Tensor> const&) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/phi/api/lib/api.cc:23771:3
    #2 0x7fe46f439d27 in split_with_num_ad_func(paddle::experimental::Tensor const&, int, paddle::experimental::ScalarBase<paddle::experimental::Tensor>) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.cc:38631:21
    #3 0x7fe46b7dd903 in paddle::pybind::eager_api_split_with_num(_object*, _object*, _object*) /home/work/wangying/paddlepaddle/pdpd_source_code_new/pdpd_1109/Paddle/paddle/fluid/pybind/eager_op_function.cc:17138:58

NOTE：https://github.com/PaddlePaddle/Paddle/blob/f9043c78e55b182e05d1d1efa7da930d8abc28d2/paddle/phi/infermeta/unary.cc#L3600-L3612

原因：没有对参数 num_or_sections 做严格的检查。
Reason: lack enough parameter checking of num_or_sections.

Case33: paddle.fft.rfftfreq

Reproduced Case：

import paddle
paddle.fft.rfftfreq(0)

Stack Trace Message：

Traceback (most recent call last):
  File "test_rfftfreq.py", line 3, in <module>
    paddle.fft.rfftfreq(0)
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/fft.py", line 1319, in rfftfreq
    val = 1.0 / (n * d)
ZeroDivisionError: float division by zero

Case34: paddle.fft.fftfreq

Reproduced Case：

import paddle
paddle.fft.fftfreq(1, 0)

Stack Trace Message：

Traceback (most recent call last):
  File "test_fftfreq.py", line 3, in <module>
    paddle.fft.fftfreq(1, 0)
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/fft.py", line 1267, in fftfreq
    val = 1.0 / (n * d)
ZeroDivisionError: float division by zero

Case35: paddle.optimizer.lr.NoamDecay

Reproduced Case：

import paddle
schedular = paddle.optimizer.lr.NoamDecay(d_model=0, warmup_steps=0)

Stack Trace Message：

Traceback (most recent call last):
  File "NoamDecay.py", line 3, in <module>
    schedular = paddle.optimizer.lr.NoamDecay(d_model=0, warmup_steps=0)
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/optimizer/lr.py", line 300, in __init__
    super().__init__(learning_rate, last_epoch, verbose)
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/optimizer/lr.py", line 106, in __init__
    self.step()
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/optimizer/lr.py", line 129, in step
    self.last_lr = self.get_lr()
  File "/home/work/wangying/paddlepaddle/pdpd_fuzz_protobuf/pdpd-env/clang_pd_fuzz_env/lib/python3.8/site-packages/paddle/optimizer/lr.py", line 307, in get_lr
    b = self.warmup_steps**-1.5 * self.last_epoch
ZeroDivisionError: 0.0 cannot be raised to a negative power

Case6: paddle.nn.functional.softmax_with_cross_entropy 在 cpu, python3.7 环境下复现失败。

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python3.7/dist-packages/paddle/nn/functional/loss.py", line 2320, in softmax_with_cross_entropy
    axis,
  File "/usr/local/lib/python3.7/dist-packages/paddle/nn/functional/loss.py", line 278, in fluid_softmax_with_cross_entropy
    axis,
ValueError: (InvalidArgument) Input(Logits) and Input(Label) should in same shape in dimensions except axis.
  [Hint: Expected logits_dims[i] == labels_dims[i], but received logits_dims[i]:0 != labels_dims[i]:140721759615927.] (at /paddle/paddle/phi/infermeta/binary.cc:915)

Case 19 和 Case 26 未复现，但有相近错误 (min 和 all 都能触发）

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
x = paddle.to_tensor(np.reshape(array, [0, 0, 0, 0]), dtype='float32')
paddle.min(x, axis=0)

发生条件是 reshape 的 shape 为 (0, 0, ... 0) 且 len(shape) in range(4, 7) （大于等于 7 的情况在 #50013 中已修复）

会发生

--------------------------------------
C++ Traceback (most recent call last):
--------------------------------------
0   min_ad_func(paddle::experimental::Tensor const&, paddle::experimental::IntArrayBase<paddle::experimental::Tensor>, bool)
1   paddle::experimental::min(paddle::experimental::Tensor const&, paddle::experimental::IntArrayBase<paddle::experimental::Tensor> const&, bool)
2   void phi::MinRawKernel<float, phi::CPUContext>(phi::CPUContext const&, phi::DenseTensor const&, paddle::experimental::IntArrayBase<phi::DenseTensor> const&, bool, bool, phi::DenseTensor*)
3   void phi::Reduce<phi::CPUContext, float, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, bool, std::vector<long, std::allocator<long> > const&, bool, paddle::experimental::DataType, phi::DenseTensor*)
4   void phi::funcs::ReduceKernelImpl<phi::CPUContext, float, float, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<long, std::allocator<long> > const&, bool, bool)
5   void phi::funcs::ReduceFunctor<phi::CPUContext, float, 4ul, 1ul, phi::funcs::MinFunctor>(phi::CPUContext const&, phi::DenseTensor const&, phi::DenseTensor*, std::vector<long, std::allocator<long> > const&, bool)

----------------------
Error Message Summary:
----------------------
FatalError: `Erroneous arithmetic operation` is detected by the operating system.
  [TimeInfo: *** Aborted at 1674474275 (unix time) try "date -d @1674474275" if you are using GNU date ***]
  [SignalInfo: *** SIGFPE (@0x7f850f02d0ef) received by PID 663826 (TID 0x7f8514469740) from PID 251842799 ***]

Floating point exception (core dumped)

update 2023/1/24 2:43 触发条件为 len(shape) - len(axis) >= 3 且 len(shape) < 7

Case6: paddle.nn.functional.softmax_with_cross_entropy

Reproduced Case：

import paddle
import numpy as np
array = np.array([], dtype=np.float32)
logits = paddle.to_tensor(np.reshape(array, [1, 0]), dtype='float32')
array = np.array([1], dtype=np.float32)
label = paddle.to_tensor(np.reshape(array, [1]), dtype='float32')
paddle.nn.functional.softmax_with_cross_entropy(logits, label, soft_label=True, numeric_stable_mode=True, axis=0)

Stack Trace Message：

==206963==ERROR: AddressSanitizer: FPE on unknown address 0x7f3da0602cd7 (pc 0x7f3da0602cd7 bp 0x7ffef5a33690 sp 0x7ffef5a33628 T0)
    #0 0x7f3da0602cd7 in __udivti3 (/lib/x86_64-linux-gnu/libgcc_s.so.1+0x5cd7)
    #1 0x7f3ce737daf0 in Eigen::internal::(anonymous namespace)::DividerHelper<64, long>::computeMultiplier(int, long) /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128:82
    #2 0x7f3ce737daf0 in Eigen::internal::TensorIntDivisor<long, false>::TensorIntDivisor(long) /home/work/wangying/paddlepaddle/pdpd_source_code/Paddle/build/third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:165:18

NOTE：

* `paddle.nn.functional.softmax_with_cross_entropy` 可触发
  `third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128` 的除0问题，原因可能是没有对传入参数 logits 的 shape 做严格的检查。

* `paddle.nn.functional.softmax_with_cross_entropy` could trigger the divide by 0 problems in  `third_party/eigen3/src/extern_eigen3/unsupported/Eigen/CXX11/src/Tensor/TensorIntDiv.h:128`, may be caused by lacking enough input parameter check for the `shape` of `logits`.

  struct DividerHelper<64, T> {
    static EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE uint64_t computeMultiplier(const int log_div, const T divider) {
#if EIGEN_HAS_BUILTIN_INT128 && !defined(EIGEN_GPU_COMPILE_PHASE) && !defined(SYCL_DEVICE_ONLY)
      return static_cast<uint64_t>((static_cast<__uint128_t>(1) << (64+log_div)) / static_cast<__uint128_t>(divider) - (static_cast<__uint128_t>(1) << 64) + 1); // <-- 128行
#else

使用aistudio develop环境测试是已经修改为抛出异常

这个我也复现了一下，在我的机器上已经不抱任何问题了。我把这个case关掉。非常感谢~

这个我也复现了一下，在我的机器上已经不抱任何问题了。我把这个case关掉。非常感谢~

在@co63oc的测试图可以看到一个小错误，检查shape的时候，labels_dim[i]越界访问了一个空指针~

PaddlePaddle / Paddle

➗0（divide by 0） #49919

Case1: paddle.linalg.matrix_power

Case2: paddle.static.nn.sparse_embedding

Case3: paddle.static.nn.deform_conv2d

Case4: paddle.static.nn.sequence_concat

Case5: paddle.vision.ops.psroi_pool

Case6: paddle.nn.functional.softmax_with_cross_entropy

Case7: paddle.nn.functional.pixel_shuffle

Case8: paddle.nn.functional.npair_loss

Case9: paddle.nn.functional.normalize

Case10: paddle.nn.functional.max_pool2d/max_pool3d

Case11: paddle.nn.functional.max_pool1d/max_pool2d/max_pool3d

Case12: paddle.linalg.svd

Case13: paddle.linalg.norm

Case14: paddle.linalg.pinv

Case15: paddle.linalg.eig

Case16: paddle.linalg.lu

Case17: paddle.incubate.graph_reindex

Case18: paddle.logsumexp

Case19: paddle.min

Case20: paddle.min

Case21: paddle.mode

Case22: paddle.lerp

Case23: paddle.floor_mod

Case24: paddle.chunk

Case25: paddle.dot

Case26: paddle.all

Case27: paddle.nn.functional.label_smooth

Case28: paddle.nn.functional.conv3d/conv2d

Case29: paddle.nn.functional.conv1d_transpose/conv3d_transpose

Case30: DataNormKernel

Case31: AffineChannelKernel

Case32: paddle.split

Case33: paddle.fft.rfftfreq

Case34: paddle.fft.fftfreq

Case35: paddle.optimizer.lr.NoamDecay

Case6: paddle.nn.functional.softmax_with_cross_entropy

51850