Skip to content

  • 体验新版
    • 正在加载...
  • 登录
  • PaddlePaddle
  • Paddle
  • Issue
  • #6896

P
Paddle
  • 项目概览

PaddlePaddle / Paddle
大约 2 年 前同步成功

通知 2325
Star 20933
Fork 5424
  • 代码
    • 文件
    • 提交
    • 分支
    • Tags
    • 贡献者
    • 分支图
    • Diff
  • Issue 1423
    • 列表
    • 看板
    • 标记
    • 里程碑
  • 合并请求 543
  • Wiki 0
    • Wiki
  • 分析
    • 仓库
    • DevOps
  • 项目成员
  • Pages
P
Paddle
  • 项目概览
    • 项目概览
    • 详情
    • 发布
  • 仓库
    • 仓库
    • 文件
    • 提交
    • 分支
    • 标签
    • 贡献者
    • 分支图
    • 比较
  • Issue 1,423
    • Issue 1,423
    • 列表
    • 看板
    • 标记
    • 里程碑
  • 合并请求 543
    • 合并请求 543
  • Pages
  • 分析
    • 分析
    • 仓库分析
    • DevOps
  • Wiki 0
    • Wiki
  • 成员
    • 成员
  • 收起侧边栏
  • 动态
  • 分支图
  • 创建新Issue
  • 提交
  • Issue看板
已关闭
开放中
Opened 12月 22, 2017 by saxon_zh@saxon_zhGuest

test_machine_translation.py gets a crash dump in GPU mode

Created by: lcy-seso

I run this example: https://github.com/PaddlePaddle/Paddle/blob/develop/python/paddle/v2/fluid/tests/book/test_machine_translation.py

I only change this line: place = core.CPUPlace() into:

place = core.GPUPlace(0)

The error information is as follows:

terminate called after throwing an instance of 'thrust::system::system_error'
  what():  an illegal memory access was encountered
Aborted (core dumped)

Backtrace in gdb :

#0  0x000000318ae328a5 in raise () from /lib64/libc.so.6
#1  0x000000318ae34085 in abort () from /lib64/libc.so.6
#2  0x00007faad387d705 in __gnu_cxx::__verbose_terminate_handler() () from /home/caoying/.jumbo/opt/gcc48/lib64/libstdc++.so.6
#3  0x00007faad387b8a6 in __cxxabiv1::__terminate(void (*)()) () from /home/caoying/.jumbo/opt/gcc48/lib64/libstdc++.so.6
#4  0x00007faad387a969 in __cxa_call_terminate () from /home/caoying/.jumbo/opt/gcc48/lib64/libstdc++.so.6
#5  0x00007faad387b51a in __gxx_personality_v0 () from /home/caoying/.jumbo/opt/gcc48/lib64/libstdc++.so.6
#6  0x00007faaec4df813 in _Unwind_RaiseException_Phase2 (exc=exc@entry=0x122b4f20, context=context@entry=0x7fff1bb73680) at /home/caoying/.jumbo/var/jumbo/tmp/jumbo.KVlu7OwtI/gcc48/src/gcc-4.8.3/libgcc/unwind.inc:62
#7  0x00007faaec4dfbab in _Unwind_RaiseException (exc=0x122b4f20) at /home/caoying/.jumbo/var/jumbo/tmp/jumbo.KVlu7OwtI/gcc48/src/gcc-4.8.3/libgcc/unwind.inc:131
#8  0x00007faad387bae5 in __cxa_throw () from /home/caoying/.jumbo/opt/gcc48/lib64/libstdc++.so.6
#9  0x00007faabee98b95 in thrust::system::cuda::experimental::pinned_allocator<unsigned long>::deallocate (this=this@entry=0x4d, cnt=<optimized out>, p=<optimized out>)
    at /home/work/cuda-7.5/include/thrust/system/cuda/experimental/pinned_allocator.h:182
#10 thrust::detail::allocator_traits<thrust::system::cuda::experimental::pinned_allocator<unsigned long> >::deallocate(thrust::system::cuda::experimental::pinned_allocator<unsigned long>&, unsigned long*, unsigned long)::workaround_warnings::deallocate(thrust::system::cuda::experimental::pinned_allocator<unsigned long>&, unsigned long*, unsigned long) (a=..., n=<optimized out>, p=<optimized out>)
    at /home/work/cuda-7.5/include/thrust/detail/allocator/allocator_traits.inl:256
#11 thrust::detail::allocator_traits<thrust::system::cuda::experimental::pinned_allocator<unsigned long> >::deallocate (a=..., n=<optimized out>, p=<optimized out>)
    at /home/work/cuda-7.5/include/thrust/detail/allocator/allocator_traits.inl:260
#12 thrust::detail::contiguous_storage<unsigned long, thrust::system::cuda::experimental::pinned_allocator<unsigned long> >::deallocate (this=this@entry=0x7fff1bb73a00)
    at /home/work/cuda-7.5/include/thrust/detail/contiguous_storage.inl:172
#13 0x00007faabee9af4c in thrust::detail::contiguous_storage<unsigned long, thrust::system::cuda::experimental::pinned_allocator<unsigned long> >::deallocate (this=0x7fff1bb73a00)
    at /home/caoying/.jumbo/opt/gcc48/include/c++/4.8.3/ext/new_allocator.h:110
#14 thrust::detail::contiguous_storage<unsigned long, thrust::system::cuda::experimental::pinned_allocator<unsigned long> >::~contiguous_storage (this=0x7fff1bb73a00, __in_chrg=<optimized out>)
    at /home/work/cuda-7.5/include/thrust/detail/contiguous_storage.inl:64
#15 thrust::detail::vector_base<unsigned long, thrust::system::cuda::experimental::pinned_allocator<unsigned long> >::~vector_base (this=0x7fff1bb73a00, __in_chrg=<optimized out>)
    at /home/work/cuda-7.5/include/thrust/detail/vector_base.inl:475
#16 thrust::host_vector<unsigned long, thrust::system::cuda::experimental::pinned_allocator<unsigned long> >::~host_vector (this=0x7fff1bb73a00, __in_chrg=<optimized out>) at /home/work/cuda-7.5/include/thrust/host_vector.h:52
#17 paddle::framework::AppendLoD (lod=lod@entry=0x122ba268, lod_length=...) at /home/caoying/paddle_codes/paddle_github/paddle/framework/lod_tensor.cc:180
#18 0x00007faabee02c1b in paddle::operators::LoDTensorToArrayOp::Run (this=<optimized out>, scope=..., dev_ctx=...) at /home/caoying/paddle_codes/paddle_github/paddle/operators/lod_tensor_to_array_op.cc:65
#19 0x00007faabe8520ed in paddle::framework::Executor::Run (this=<optimized out>, pdesc=..., scope=0x395bee0, block_id=<optimized out>, create_local_scope=<optimized out>, create_vars=<optimized out>)
    at /home/caoying/paddle_codes/paddle_github/paddle/framework/executor.cc:110
#20 0x00007faabe7cfa6c in pybind11::cpp_function::cpp_function<void, paddle::framework::Executor, paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool, pybind11::name, pybind11::is_method, pybind11::sibling>(void (paddle::framework::Executor::*)(paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool), pybind11::name const&, pybind11::is_method const&, pybind11::sibling const&)::{lambda(paddle::framework::Executor*, paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool)#1}::operator()(paddle::framework::Executor*, paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool) const
    (args#4=<optimized out>, args#3=<optimized out>, args#2=<optimized out>, args#1=<optimized out>, args#0=..., c=<optimized out>, __closure=<optimized out>)
    at /home/caoying/paddle_codes/paddle_github/build/third_party/pybind/src/extern_pybind/include/pybind11/pybind11.h:69
#21 pybind11::detail::argument_loader<paddle::framework::Executor*, paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool>::call_impl<void, pybind11::cpp_function::cpp_function(Return (Class::*)(Arg ...), const Extra& ...) [with Return = void; Class = paddle::framework::Executor; Arg = {const paddle::framework::ProgramDesc&, paddle::framework::Scope*, int, bool, bool}; Extra = {pybind11::name, pybind11::is_method, pybind11::sibling}]::__lambda5&, 0ul, 1ul, 2ul, 3ul, 4ul, 5ul> (f=..., this=0x7fff1bb74120) at /home/caoying/paddle_codes/paddle_github/build/third_party/pybind/src/extern_pybind/include/pybind11/cast.h:1411
#22 pybind11::detail::argument_loader<paddle::framework::Executor*, paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool>::call<void, pybind11::cpp_function::cpp_function(Return (Class::*)(Arg ...), const Extra& ...) [with Return = void; Class = paddle::framework::Executor; Arg = {const paddle::framework::ProgramDesc&, paddle::framework::Scope*, int, bool, bool}; Extra = {pybind11::name, pybind11::is_method, pybind11::sibling}]::__lambda5&> (f=..., this=0x7fff1bb74120) at /home/caoying/paddle_codes/paddle_github/build/third_party/pybind/src/extern_pybind/include/pybind11/cast.h:1393
#23 void pybind11::cpp_function::initialize<pybind11::cpp_function::initialize<void, paddle::framework::Executor, paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool, pybind11::name, pybind11::is_method, pybind11::sibling>(void (paddle::framework::Executor::*)(paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool), pybind11::name const&, pybind11::is_method const&, pybind11::sibling const&)::{lambda(paddle::framework::Executor*, paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool)#1}, void, paddle::framework::Executor*, paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool, pybind11::name, pybind11::is_method, pybind11::sibling>(pybind11::cpp_function::initialize<void, paddle::framework::Executor, paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool, pybind11::name, pybind11::is_method, pybind11::sibling>(void (paddle::framework::Executor::*)(paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool), pybind11::name const&, pybind11::is_method const&, pybind11::sibling const&)::{lambda(paddle::framework::Executor*, paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool)#1}&&, void (*)(paddle::framework::Executor*, paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool), pybind11::name const&, pybind11::is_method const&, pybind11::sibling const&)::{lambda(pybind11::detail::function_call&)#3}::operator()(pybind11::detail::function_call) const (call=..., __closure=0x0)
    at /home/caoying/paddle_codes/paddle_github/build/third_party/pybind/src/extern_pybind/include/pybind11/pybind11.h:148
#24 void pybind11::cpp_function::initialize<pybind11::cpp_function::initialize<void, paddle::framework::Executor, paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool, pybind11::name, pybind11::is_metho

I compile fluid both in Debug and Release mode. The problem is the same.

指派人
分配到
无
里程碑
无
分配里程碑
工时统计
无
截止日期
无
标识: paddlepaddle/Paddle#6896
渝ICP备2023009037号

京公网安备11010502055752号

网络110报警服务 Powered by GitLab CE v13.7
开源知识
Git 入门 Pro Git 电子书 在线学 Git
Markdown 基础入门 IT 技术知识开源图谱
帮助
使用手册 反馈建议 博客
《GitCode 隐私声明》 《GitCode 服务条款》 关于GitCode
Powered by GitLab CE v13.7