[cpp-package]Wrong GLIBCXX version for cpp-package examples

[inglada]

Description

When running the cpp-package examples build from source, the following error occurs ./alexnet: /usr/lib/x86_64-linux-gnu/libstdc++.so.6: versionGLIBCXX_3.4.23' not found (required by ./alexnet)`

Environment info (Required)

----------Python Info----------
('Version      :', '2.7.13')
('Compiler     :', 'GCC 6.3.0 20170516')
('Build        :', ('default', 'Sep 26 2018 18:42:22'))
('Arch         :', ('64bit', ''))
------------Pip Info-----------
('Version      :', '18.0')
('Directory    :', '/usr/local/lib/python2.7/dist-packages/pip')
----------MXNet Info-----------
No MXNet installed.
----------System Info----------
('Platform     :', 'Linux-4.14.0-0.bpo.3-amd64-x86_64-with-debian-9.5')
('system       :', 'Linux')
('node         :', 'pc-117-162')
('release      :', '4.14.0-0.bpo.3-amd64')
('version      :', '#1 SMP Debian 4.14.13-1~bpo9+1 (2018-01-14)')
----------Hardware Info----------
('machine      :', 'x86_64')
('processor    :', '')
Architecture:          x86_64
CPU op-mode(s):        32-bit, 64-bit
Byte Order:            Little Endian
CPU(s):                4
On-line CPU(s) list:   0-3
Thread(s) per core:    2
Core(s) per socket:    2
Socket(s):             1
NUMA node(s):          1
Vendor ID:             GenuineIntel
CPU family:            6
Model:                 78
Model name:            Intel(R) Core(TM) i7-6600U CPU @ 2.60GHz
Stepping:              3
CPU MHz:               529.946
CPU max MHz:           3400.0000
CPU min MHz:           400.0000
BogoMIPS:              5616.00
Virtualization:        VT-x
L1d cache:             32K
L1i cache:             32K
L2 cache:              256K
L3 cache:              4096K
NUMA node0 CPU(s):     0-3
Flags:                 fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb invpcid_single pti intel_pt tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx rdseed adx smap clflushopt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp
----------Network Test----------
Setting timeout: 10
Timing for MXNet: https://github.com/apache/incubator-mxnet, DNS: 0.0072 sec, LOAD: 0.8300 sec.
Timing for PYPI: https://pypi.python.org/pypi/pip, DNS: 0.0308 sec, LOAD: 0.7744 sec.
Timing for FashionMNIST: https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/dataset/fashion-mnist/train-labels-idx1-ubyte.gz, DNS: 0.0537 sec, LOAD: 0.3521 sec.
Timing for Conda: https://repo.continuum.io/pkgs/free/, DNS: 0.0248 sec, LOAD: 0.1212 sec.
Timing for Gluon Tutorial(en): http://gluon.mxnet.io, DNS: 0.0909 sec, LOAD: 0.9658 sec.
Timing for Gluon Tutorial(cn): https://zh.gluon.ai, DNS: 0.2454 sec, LOAD: 0.9223 sec.

Package used (Python/R/Scala/Julia): I'm using C++

Build info (Required if built from source)

Compiler (gcc/clang/mingw/visual studio): g++ (GCC) 8.2.0

MXNet commit hash: b89a36d94b5b694b8fd926e6249f7490b38432f6

Build config:

# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements.  See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership.  The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License.  You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied.  See the License for the
# specific language governing permissions and limitations
# under the License.

#-------------------------------------------------------------------------------
#  Template configuration for compiling mxnet
#
#  If you want to change the configuration, please use the following
#  steps. Assume you are on the root directory of mxnet. First copy the this
#  file so that any local changes will be ignored by git
#
#  $ cp make/config.mk .
#
#  Next modify the according entries, and then compile by
#
#  $ make
#
#  or build in parallel with 8 threads
#
#  $ make -j8
#-------------------------------------------------------------------------------

#---------------------
# choice of compiler
#--------------------

ifndef CC
export CC = gcc
endif
ifndef CXX
export CXX = g++
endif
ifndef NVCC
export NVCC = nvcc
endif

# whether compile with options for MXNet developer
DEV = 0

# whether compile with debug
DEBUG = 0

# whether to turn on segfault signal handler to log the stack trace
USE_SIGNAL_HANDLER =

# the additional link flags you want to add
ADD_LDFLAGS =

# the additional compile flags you want to add
ADD_CFLAGS =

#---------------------------------------------
# matrix computation libraries for CPU/GPU
#---------------------------------------------

# whether use CUDA during compile
USE_CUDA = 0

# add the path to CUDA library to link and compile flag
# if you have already add them to environment variable, leave it as NONE
# USE_CUDA_PATH = /usr/local/cuda
USE_CUDA_PATH = NONE

# whether to enable CUDA runtime compilation
ENABLE_CUDA_RTC = 1

# whether use CuDNN R3 library
USE_CUDNN = 0

#whether to use NCCL library
USE_NCCL = 0
#add the path to NCCL library
USE_NCCL_PATH = NONE

# whether use opencv during compilation
# you can disable it, however, you will not able to use
# imbin iterator
USE_OPENCV = 1

#whether use libjpeg-turbo for image decode without OpenCV wrapper
USE_LIBJPEG_TURBO = 0
#add the path to libjpeg-turbo library
USE_LIBJPEG_TURBO_PATH = NONE

# use openmp for parallelization
USE_OPENMP = 1

# whether use MKL-DNN library
USE_MKLDNN = 0

# whether use NNPACK library
USE_NNPACK = 0

# choose the version of blas you want to use
# can be: mkl, blas, atlas, openblas
# in default use atlas for linux while apple for osx
UNAME_S := $(shell uname -s)
ifeq ($(UNAME_S), Darwin)
USE_BLAS = apple
else
USE_BLAS = atlas
endif

# whether use lapack during compilation
# only effective when compiled with blas versions openblas/apple/atlas/mkl
USE_LAPACK = 1

# path to lapack library in case of a non-standard installation
USE_LAPACK_PATH =

# add path to intel library, you may need it for MKL, if you did not add the path
# to environment variable
USE_INTEL_PATH = NONE

# If use MKL only for BLAS, choose static link automatically to allow python wrapper
ifeq ($(USE_BLAS), mkl)
USE_STATIC_MKL = 1
else
USE_STATIC_MKL = NONE
endif

#----------------------------
# Settings for power and arm arch
#----------------------------
ARCH := $(shell uname -a)
ifneq (,$(filter $(ARCH), armv6l armv7l powerpc64le ppc64le aarch64))
    USE_SSE=0
    USE_F16C=0
else
    USE_SSE=1
endif

#----------------------------
# F16C instruction support for faster arithmetic of fp16 on CPU
#----------------------------
# For distributed training with fp16, this helps even if training on GPUs
# If left empty, checks CPU support and turns it on.
# For cross compilation, please check support for F16C on target device and turn off if necessary.
USE_F16C =

#----------------------------
# distributed computing
#----------------------------

# whether or not to enable multi-machine supporting
USE_DIST_KVSTORE = 0

# whether or not allow to read and write HDFS directly. If yes, then hadoop is
# required
USE_HDFS = 0

# path to libjvm.so. required if USE_HDFS=1
LIBJVM=$(JAVA_HOME)/jre/lib/amd64/server

# whether or not allow to read and write AWS S3 directly. If yes, then
# libcurl4-openssl-dev is required, it can be installed on Ubuntu by
# sudo apt-get install -y libcurl4-openssl-dev
USE_S3 = 0

#----------------------------
# performance settings
#----------------------------
# Use operator tuning
USE_OPERATOR_TUNING = 1

# Use gperftools if found
USE_GPERFTOOLS = 1

# path to gperftools (tcmalloc) library in case of a non-standard installation
USE_GPERFTOOLS_PATH =

# Link gperftools statically
USE_GPERFTOOLS_STATIC =

# Use JEMalloc if found, and not using gperftools
USE_JEMALLOC = 1

# path to jemalloc library in case of a non-standard installation
USE_JEMALLOC_PATH =

# Link jemalloc statically
USE_JEMALLOC_STATIC =

#----------------------------
# additional operators
#----------------------------

# path to folders containing projects specific operators that you don't want to put in src/operators
EXTRA_OPERATORS =

#----------------------------
# other features
#----------------------------

# Create C++ interface package
USE_CPP_PACKAGE = 0

#----------------------------
# plugins
#----------------------------

# whether to use caffe integration. This requires installing caffe.
# You also need to add CAFFE_PATH/build/lib to your LD_LIBRARY_PATH
# CAFFE_PATH = $(HOME)/caffe
# MXNET_PLUGINS += plugin/caffe/caffe.mk

# WARPCTC_PATH = $(HOME)/warp-ctc
# MXNET_PLUGINS += plugin/warpctc/warpctc.mk

# whether to use sframe integration. This requires build sframe
# git@github.com:dato-code/SFrame.git
# SFRAME_PATH = $(HOME)/SFrame
# MXNET_PLUGINS += plugin/sframe/plugin.mk

Error Message:

./alexnet: /usr/lib/x86_64-linux-gnu/libstdc++.so.6: versionGLIBCXX_3.4.23' not found (required by ./alexnet)`

Minimum reproducible example

Steps to reproduce

(Paste the commands you ran that produced the error.)

1. git clone --recursive https://github.com/apache/incubator-mxnet mxnet
2. cd mxnet/
3. make -j4 USE_CPP_PACKAGE=1 USE_OPENCV=0 USE_BLAS=openblas USE_CUDA=0
4. cd cpp-package/example/
5. make all MXNET_USE_CPU=1
6. export LD_LIBRARY_PATH=../../lib/:$LD_LIBRARY_PATH

[lanking520]

@inglada thanks for filing issue. @leleamol here to help you out.

Usually GLIBC ship with your corresponding platform. Ideally, we have the best compatibility with Ubuntu 16.04 and most of the examples are built and tested there.

@mxnet-label-bot please add [c++, install, example]

[inglada]

Hi, Thanks for your answer. I helped me found some additional info. I am using a gcc version which is installed in /usr/local/ and made available through debian/ubuntu "alternatives" mechanism via symlinks in /usr/bin/. The problem is that libmxnet is using /usr/lib/x86_64-linux-gnu/libstdc++.so.6 instead of the one corresponding to the compiler used for building.

If I use the gcc 6.3 version in /usr/bin/ everything goes OK.

This is not a sane behaviour, since it means that something in mxnet build process is not using the compiler's default libstdc++; but the one which is in /usr/lib/. The "alternatives" mechanism works fine for many libraries I build from source.

There may be something hard coded in the provided Makefile. I can try to investigate further if you give me some hints.

Is CMake a supported build process?

Thanks.

[lanking520]

@inglada Thanks for pointing it out the issue and happy to see you get it to work. I am not surprised that MXNet is not using a gcc from the system but a fixed version because of compatibilities to most of the dependencies. But the inconsistency during the building process is a problem.

We do have a CMake process for Windows. I am not sure if it is also applicable to other platforms. https://github.com/apache/incubator-mxnet/blob/master/CMakeLists.txt @marcoabreu @lebeg any idea on Cmake?

[leleamol]

Hi @inglada Will you be able to share the use case that you were planning to address using C++ API?

[inglada]

@leleamol Do you mean the application field? I don't understand the question.

[marcoabreu]

@lanking520 cmake can also be used for other systems. We use it on Ubuntu, for example

[leleamol]

@inglada I would like to know how do you plan to use C++ API, whether in training workflow or inference workflow.

[inglada]

@leleamol I would like to use it for all steps of the ML process: building the model architecture (included new layers that I would implement), training and deployment over large amounts of data on a HPC cluster.

mxnet would be a dependency of http://www.orfeo-toolbox.org which is CMake based and has a "superbuild" process to build all its dependencies (boost, itk, gdal, opencv, libsvm, shark-ml, etc.).

We have now the means of integrating Tensorflow, but it is very difficult to build (bazel based!) and the c++ API is not exposed. Furthermore, I have found mxnet's API much more intuitive.

I am also evaluating PyTorch, since they are now exposing a C++ fontend. However, not using static graphs makes me wonder about performances.

[stu1130]

@inglada feel free to reopen it if the PR didn't resolve your issue or have any other question