Error in building the project

[cn-tu]

Error in building the project

[HenryChang213]

The error log looks like this: (only a part of the log due to the length)

1976/2019] Linking �[33mbuild/utils/ns3.30.1-test-runner-optimized�[0m
[1978/2019] Linking �[33mbuild/utils/ns3.30.1-print-introspected-doxygen-optimized�[0m
/usr/bin/ld: examples/traffic-control/red-vs-fengadaptive.cc.3.o: in function `c10::UndefinedTensorImpl::singleton()':
red-vs-fengadaptive.cc:(.text._ZN3c1019UndefinedTensorImpl9singletonEv[_ZN3c1019UndefinedTensorImpl9singletonEv]+0xb): undefined reference to `c10::UndefinedTensorImpl::_singleton'
/usr/bin/ld: examples/traffic-control/red-vs-fengadaptive.cc.3.o: in function `c10::IValue::toDouble() const':
red-vs-fengadaptive.cc:(.text._ZNK3c106IValue8toDoubleEv[_ZNK3c106IValue8toDoubleEv]+0xd8): undefined reference to `c10::Error::Error(c10::SourceLocation, std::string const&)'
/usr/bin/ld: examples/traffic-control/red-vs-fengadaptive.cc.3.o: in function `c10::IValue::toInt() const':
red-vs-fengadaptive.cc:(.text._ZNK3c106IValue5toIntEv[_ZNK3c106IValue5toIntEv]+0xd8): undefined reference to `c10::Error::Error(c10::SourceLocation, std::string const&)'
/usr/bin/ld: examples/traffic-control/red-vs-fengadaptive.cc.3.o: in function `c10::IValue::toTensor() const &':
red-vs-fengadaptive.cc:(.text._ZNKR3c106IValue8toTensorEv[_ZNKR3c106IValue8toTensorEv]+0xfa): undefined reference to `c10::Error::Error(c10::SourceLocation, std::string const&)'
/usr/bin/ld: examples/traffic-control/red-vs-fengadaptive.cc.3.o: in function `c10::intrusive_ptr<c10::TensorImpl, c10::UndefinedTensorImpl>::retain_()':
red-vs-fengadaptive.cc:(.text._ZN3c1013intrusive_ptrINS_10TensorImplENS_19UndefinedTensorImplEE7retain_Ev[_ZN3c1013intrusive_ptrINS_10TensorImplENS_19UndefinedTensorImplEE7retain_Ev]+0xff): undefined reference to `c10::Error::Error(c10::SourceLocation, std::string const&)'
/usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `vtable for c10::AutogradMetaInterface'
/usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `torch::nn::Module::is_serializable() const'
/usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `typeinfo for c10::AutogradMetaInterface'
/usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `torch::nn::Module::zero_grad()'
/usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `torch::nn::Module::to(c10::ScalarType, bool)'
/usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `at::GradMode::set_enabled(bool)'
/usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `caffe2::detail::TypeMetaData const* caffe2::TypeMeta::_typeMetaDataInstance<c10::BFloat16>()'

[muxamilian]

Can you try to prefix all waf commands with this: LD_LIBRARY_PATH=../../libtorch/lib:$LD_LIBRARY_PATH. Like this: LD_LIBRARY_PATH=../../libtorch/lib:$LD_LIBRARY_PATH ./waf build

…

On Wed, Nov 23, 2022 at 12:02 AM Hengrui Zhang ***@***.***> wrote: The error log looks like this: (only a part of the log due to the length) 1976/2019] Linking �[33mbuild/utils/ns3.30.1-test-runner-optimized�[0m [1978/2019] Linking �[33mbuild/utils/ns3.30.1-print-introspected-doxygen-optimized�[0m /usr/bin/ld: examples/traffic-control/red-vs-fengadaptive.cc.3.o: in function `c10::UndefinedTensorImpl::singleton()': red-vs-fengadaptive.cc:(.text._ZN3c1019UndefinedTensorImpl9singletonEv[_ZN3c1019UndefinedTensorImpl9singletonEv]+0xb): undefined reference to `c10::UndefinedTensorImpl::_singleton' /usr/bin/ld: examples/traffic-control/red-vs-fengadaptive.cc.3.o: in function `c10::IValue::toDouble() const': red-vs-fengadaptive.cc:(.text._ZNK3c106IValue8toDoubleEv[_ZNK3c106IValue8toDoubleEv]+0xd8): undefined reference to `c10::Error::Error(c10::SourceLocation, std::string const&)' /usr/bin/ld: examples/traffic-control/red-vs-fengadaptive.cc.3.o: in function `c10::IValue::toInt() const': red-vs-fengadaptive.cc:(.text._ZNK3c106IValue5toIntEv[_ZNK3c106IValue5toIntEv]+0xd8): undefined reference to `c10::Error::Error(c10::SourceLocation, std::string const&)' /usr/bin/ld: examples/traffic-control/red-vs-fengadaptive.cc.3.o: in function `c10::IValue::toTensor() const &': red-vs-fengadaptive.cc:(.text._ZNKR3c106IValue8toTensorEv[_ZNKR3c106IValue8toTensorEv]+0xfa): undefined reference to `c10::Error::Error(c10::SourceLocation, std::string const&)' /usr/bin/ld: examples/traffic-control/red-vs-fengadaptive.cc.3.o: in function `c10::intrusive_ptr<c10::TensorImpl, c10::UndefinedTensorImpl>::retain_()': red-vs-fengadaptive.cc:(.text._ZN3c1013intrusive_ptrINS_10TensorImplENS_19UndefinedTensorImplEE7retain_Ev[_ZN3c1013intrusive_ptrINS_10TensorImplENS_19UndefinedTensorImplEE7retain_Ev]+0xff): undefined reference to `c10::Error::Error(c10::SourceLocation, std::string const&)' /usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `vtable for c10::AutogradMetaInterface' /usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `torch::nn::Module::is_serializable() const' /usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `typeinfo for c10::AutogradMetaInterface' /usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `torch::nn::Module::zero_grad()' /usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `torch::nn::Module::to(c10::ScalarType, bool)' /usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `at::GradMode::set_enabled(bool)' /usr/bin/ld: lib/libns3.30.1-traffic-control-optimized.so: undefined reference to `caffe2::detail::TypeMetaData const* caffe2::TypeMeta::_typeMetaDataInstance<c10::BFloat16>()' — Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAO2RJZHYECMCC77HGWU5BTWJVGGZANCNFSM6AAAAAASIKYVNU> . You are receiving this because you are subscribed to this thread.Message ID: <CN-TU/reinforcement-learning-for-per-flow-buffer-sizing/issues/1/1324346334 @github.com>

[HenryChang213]

Still has the same error.

[muxamilian]

I’ll try to reproduce the error using a similar environment and let you know.

[HenryChang213]

Many thanks!

[muxamilian]

I tried with all kinds of different compiler versions and libtorch versions but with no success. However, using Docker, I got it to run pretty easily using the same Debian version that I used originally. I added a section to the readme.

[HenryChang213]

Thanks! I can now run successfully!