How to properly link the gflags?

[CNLHC]

Building the python wheel from master branch and run gcn model will get

python train_supervised.pyTraceback (most recent call last):
  File "train_supervised.py", line 22, in <module>
    import graphlearn as gl
  File "/home/cn/research/graph/venv/lib/python2.7/site-packages/graphlearn/__init__.py", line 16, in <module>
    from graphlearn import pywrap_graphlearn as pywrap
ImportError: /home/cn/research/graph/graph-learn/built/lib/libgraphlearn_shared.so: undefined symbol: _ZN6google14FlagRegistererC1IiEEPKcS3_S3_PT_S5_

It seems that symbol _ZN6google14FlagRegistererC1IiEEPKcS3_S3_PT_S5_ (which is the mangled FlagRegisterer from gflags package ) was not found.

So, I tried to modify the Makefile and link gflags library manually, then the problem was solved.(modified line: here)

so:protobuf grpc glog gtest proto common platform service core
	@mkdir -p $(INCLUDE_DIR)
	@mkdir -p $(LIB_DIR)
	@mkdir -p $(BIN_DIR)
	$(CXX) $(CXXFLAGS) -shared $(PROTO_OBJ) $(COMMON_OBJ) $(PLATFORM_OBJ) $(SERVICE_OBJ) $(CORE_OBJ) \
		-L$(ROOT) -L$(GLOG_LIB) -L$(PROTOBUF_LIB) -L$(GRPC_LIB) -L$(GFLAGS_LIB)\
		-lglog -lprotobuf -lgrpc++ -lgrpc -lgpr -lupb -lgflags\
		-o $(LIB_DIR)/libgraphlearn_shared.so

Is this a bug or one need special way not mentioned in doc so far to link gflags ?

[patricxu]

configure glags as shared library which is not the default option!

[CNLHC]

configure glags as shared library which is not the default option!

But it seems that the default configuration do treat gflags as shared library. Am i missing something ?🤔

I will try to reproduce this problem in a brand new environment later.

[patricxu]

@CNLHC see my notes
The output of the default config is .a, but we need .so.

[CNLHC]

@CNLHC see my notes
The output of the default config is .a, but we need .so.

Obviously. In fact, I have already noticed that and done a similar job in a build script here

But i do not think this is the key point of the problem. What I concern is that do we need to compile gflags manually or not?

After testing in a brand new environment and reading your notes, i can pretty sure that the answer is yes: we do need to build gflags manually.

I will try to submit a PR to automate this process.

Thanks for your communication!

[jackonan]

@CNLHC Great thanks for your PR. Would you mind show your building steps and commands that lead to the errors exactly? I did not reproduce the problem in my environment.

[CNLHC]

The reproducing steps

export GL_DIR=<Graph-Learn Directory>
cd $GL_DIR
git checkout  1d024e7
make cleanall
git reset --hard
git clean -dif
make
make python
source env.sh
python -m pip install -I ./dist/graphlearn-0.1-cp27-cp27mu-linux_x86_64.whl
cd examples/tf/gcn
python train_supervised.py

then , got error

Traceback (most recent call last):
  File "train_supervised.py", line 22, in <module>
    import graphlearn as gl
  File "/home/cn/research/graph/venv/lib/python2.7/site-packages/graphlearn/__init__.py", line 16, in <module>
    from graphlearn import pywrap_graphlearn as pywrap
ImportError: /home/cn/research/graph/graph-learn/built/lib/libgraphlearn_shared.so: undefined symbol: _ZN6google14FlagRegistererC1IiEEPKcS3_S3_PT_S5_

os

NAME="Ubuntu"
VERSION="18.04.3 LTS (Bionic Beaver)"
ID=ubuntu
ID_LIKE=debian
PRETTY_NAME="Ubuntu 18.04.3 LTS"
VERSION_ID="18.04"
HOME_URL="https://www.ubuntu.com/"
SUPPORT_URL="https://help.ubuntu.com/"
BUG_REPORT_URL="https://bugs.launchpad.net/ubuntu/"
PRIVACY_POLICY_URL="https://www.ubuntu.com/legal/terms-and-policies/privacy-policy"
VERSION_CODENAME=bionic
UBUNTU_CODENAME=bionic

and kernel-version

4.15.0-88-generic