I feel like I must be doing something fundamentally wrong. I created a recipe based on a cmake project. Compiling the project using the toolchain yocto created is as simple as running cmake then make but it fails to compile using a recipe:
SUMMARY = "Opendnp3 is the de facto reference implementation of IEEE-1815 (DNP3)"
DESCRIPTION = "Opendnp3 is a portable, scalable, and rigorously tested implementation of the DNP3 (www.dnp.org) protocol stack written in C++11. The library is designed for high-performance applications like many concurrent TCP sessions or huge device simulations. It also embeds very nicely on Linux."
HOMEPAGE = "https://www.automatak.com/opendnp3"
SECTION = "libs"
DEPENDS = "asio"
LICENSE = "Apache-2.0"
LIC_FILES_CHKSUM = "file://NOTICE;md5=9788d3abe6c9401d72fdb3717de33e6a"
SRCREV = "e00ff31b837821064e5208c15866a9d46f8777b1"
SRC_URI = "git://github.com/automatak/dnp3;branch=2.0.x"
S = "${WORKDIR}/git"
inherit cmake
EXTRA_OECMAKE += ""
Problem is I think that the CXXFLAGS used for g++ don't appear to be consistent with the CXXFLAGS defined by the toolchain's environment. Mainly --sysroot is missing and g++ fails to find standard c++ headers (ex: cstdint).
I partially fixed the issue by overriding do_configure from cmake.bbclas simply by removing -DCMAKE_TOOLCHAIN_FILE=${WORKDIR}/toolchain.cmake. As a matter of fact CXXFLAGS are defined by toolchain.cmake as:
-march=armv7-a -mfpu=neon -mfloat-abi=hard -mcpu=cortex-a8 --sysroot=/home/star/moxa-build/tmp/work/cortexa8hf-neon-poky-linux-gnueabi/dnp3/2.2.0-r0/recipe-sysroot -O2 -pipe -g -feliminate-unused-debug-types -fdebug-prefix-map=/home/star/moxa-build/tmp/work/cortexa8hf-neon-poky-linux-gnueabi/dnp3/2.2.0-r0=/usr/src/debug/dnp3/2.2.0-r0 -fdebug-prefix-map=/home/star/moxa-build/tmp/work/cortexa8hf-neon-poky-linux-gnueabi/dnp3/2.2.0-r0/recipe-sysroot-native= -fdebug-prefix-map=/home/star/moxa-build/tmp/work/cortexa8hf-neon-poky-linux-gnueabi/dnp3/2.2.0-r0/recipe-sysroot= -fvisibility-inlines-hidden -march=armv7-a -mfpu=neon -mfloat-abi=hard -mcpu=cortex-a8 --sysroot=/home/star/moxa-build/tmp/work/cortexa8hf-neon-poky-linux-gnueabi/dnp3/2.2.0-r0/recipe-sysroot
So sysroot is defined (twice actually) as:
/home/star/moxa-build/tmp/work/cortexa8hf-neon-poky-linux-gnueabi/dnp3/2.2.0-r0/recipe-sysroot
but doesn't end up in the Makefile generated by cmake so I guess that something in toolchain.cmake makes the project's cmake configuration go south.
Turns out it was a cmake issue with the project itself. There was a merry mix-up between C and CXX flags in one of the included .cmake configuration file. I'll submit a pull request to the maintainer. CXXFLAGS being handled differently between the SDK (included directly in th CXX command) and by bitbake explained why it worked with the SDK (still not quite sure why it worked without toolchain.cmake though).
Thanks.
--sysroot is missing from CXXFLAGS because yocto stuffs it in the CXX variable itself. Note that alongside the sysroots directory yocto will have an environment-setup-XXX file which you should source. That's where all the sysroot stuff comes from.
Related
I have inherited a build system that uses autoconf and automake to build. I am not at all familiar with how this works except on a high level.
The build system itself works fine. However a lot of warnings are generated from system headers. In my case wxWidgets. The cause of this is that the include dir of wxWidgets is passed as -I instead of -isystem. I have searched online but I can't find how I can change this.
The following code is in the configure.ac file:
AM_OPTIONS_WXCONFIG
AM_PATH_WXCONFIG(3.0.2, wxWin=1, , ,[--debug=yes])
if test "$wxWin" != 1; then
AC_MSG_ERROR([
wxWidgets must be installed on your system
but no wx-config script could be found.
Please check that wx-config is in path, the directory
where wxWidgets libraries are installed as returned by
'wx-config --libs' is in the linker path (LD_LIBRARY_PATH
or equivalent variable) and wxWidgets version is 2.9.4 or
above.
])
fi
if test "$wxWin" != 0; then
AC_DEFINE(HAVE_WXWIDGETS,1,[define if the wxWidgets 3.0.2. or higher is available])
fi
Then there is a Makefile.am which does the following:
bin_PROGRAMS = project
project_SOURCES = {sources here}
project_CPPFLAGS = $(AM_CPPFLAGS) $(PROJECT_CPPFLAGS) $(WX_CPPFLAGS)
project_CFLAGS = $(AM_CFLAGS) $(PROJECT_CFLAGS) $(WX_CFLAGS)
project_CXXFLAGS = $(AM_CXXFLAGS) $(PROJECT_CXXFLAGS) $(WX_CXXFLAGS)
project_LDFLAGS = $(AM_LDFLAGS)
project_LDADD = $(AM_LIBADD) $(WX_LIBS)
There doesn't seem an option that I can specify it as a system header in some way.
Do anyone of you know how I can tackle this?
The problem has nothing to do with autoconf. It's the wx-config tool that is the culprit. That tool is used to obtain the include directories. For instance invoking wx-config --cxxflag will result in this:
-I/usr/lib/wx/include/gtk2-unicode-3.1 -I/usr/include/wx-3.1 -D_FILE_OFFSET_BITS=64 -DWXUSINGDLL -D__WXGTK__ -pthread
Which is passed to the compiler. There are a couple of ways of solving this. Either manually replace it in the output or patch wx-config.
I would like to compile and run my program in two different environments. The libraries in both environments are installed on slightly different places, resulting in different makefile-lines:
In makefile A:
CXXFLAGS=-I$(DIR) -flto -fopenmp -O3 -g -march=native -std=gnu++17 -c -I/opt/interp2d/include -std=c++17 -I/opt/splinter/include -I/usr/include/eigen3
In makefile B:
CXXFLAGS=-I$(DIR) -nostindc++ -I~/local_opt/eigen/include/eigen3/ -I~/local_opt/boost/include -I~/local_opt/armadillo/include -flto -fopenmp -O3 -g -march=native -std=gnu++17 -c -I~/local_opt/interp2d/include -std=c++17 -I~/local_opt/splinterp/include -I/usr/include/eigen3
My problem now is that I am developing the program on the first machine, using makefile A, but also deploying it on the second machine. The deployment is done using git.
Every time I do a git pull on the second machine, I have to fix all the paths in the makefile in order to compile the program properly. Nevertheless I still would like to include the makefile in the git repository in order to keep both makefiles at the same level regarding compiling flags and linked libraries.
Thus, is there an easier way to still sync the makefile via git, while using different paths for the libraries and includes?
I think you could solve your problem by conditionally setting the variable CXXFLAGS in a common file (e.g.: config.mk) and by including that file in your makefiles.
The value used for setting the CXXFLAGS variable could, for example, depend on the value of the environment variable HOST:
ifeq ($(HOST),A)
CXXFLAGS = ... # for machine A
else # B
CXXFLAGS = ... # for machine B
endif
Then, include this config.mk makefile in both makefileA and makefileB:
include config.mk
I like this answer, however, I thought I'd mention this for completeness: If you have a lot of different hosts you can do something to the effect of:
include HostConfig_$(HOST).mk
And then create HostConfig_A.mk and HostConfig_B.mk which set host specific flags (Be it directories, etc). This is useful if you are managing a large project with lots of different host-specific variables.
As well, (for smaller projects), you could do something to the effect of:
CXX_INCLUDES_A = ...
CXX_INCLUDES_B = ...
CXX_FLAGS := -I$(DIR) -flto -fopenmp -O3 -g -march=native -std=gnu++17
CXX_FLAGS += $(CXX_INCLUDES_$(HOST))
The traditional answer to this problem is a configure script (see automake, autoconf for widely used framework). After checking out the source you run ./configure --with-eigen=~/local_opt/eigen/include/eigen3/ and it will adjust your Makefiles accordingly (usually generates Makefile from Makefile.in and only Makefile.in is in git).
Note: Properly done you only need to run configure on the first checkout, not on updates. make can generate Makefile again automatially as needed.
I'm trying to compile the crypto++ library to run for the armhf architecture. I'm following the method provided in this answer. I tweaked the setenv-embed.sh to match my system's configuration. The output of running . ./setenv-embed.sh is
CPP: /usr/bin/arm-linux-gnueabihf-cpp
CXX: /usr/bin/arm-linux-gnueabihf-g++
AR: /usr/bin/arm-linux-gnueabihf-ar
LD: /usr/bin/arm-linux-gnueabihf-ld
RANLIB: /usr/bin/arm-linux-gnueabihf-gcc-ranlib-4.8
ARM_EMBEDDED_TOOLCHAIN: /usr/bin
ARM_EMBEDDED_CXX_HEADERS: /usr/arm-linux-gnueabihf/include/c++/4.8.2
ARM_EMBEDDED_FLAGS: -march=armv7-a mfloat-abi=hard -mfpu=neon -I/usr/arm-linux-gnueabihf/include/c++/4.8.2 -I/usr/arm-linux-gnueabihf/include/c++/4.8.2/arm-linux-gnueabihf
ARM_EMBEDDED_SYSROOT: /usr/arm-linux-gnueabihf
which indicates that the correct compilers have been found. However, when I build the library using make I run into the following error
/usr/lib/gcc-cross/arm-linux-gnueabihf/4.8/../../../../arm-linux-gnueabihf/bin/ld: cannot find /usr/arm-linux-gnueabihf/lib/libc.so.6 inside /usr/arm-linux-gnueabihf
/usr/lib/gcc-cross/arm-linux-gnueabihf/4.8/../../../../arm-linux-gnueabihf/bin/ld: cannot find /usr/arm-linux-gnueabihf/lib/libc_nonshared.a inside /usr/arm-linux-gnueabihf
/usr/lib/gcc-cross/arm-linux-gnueabihf/4.8/../../../../arm-linux-gnueabihf/bin/ld: cannot find /usr/arm-linux-gnueabihf/lib/ld-linux-armhf.so.3 inside /usr/arm-linux-gnueabihf
But when I open the location /usr/arm-linux-gnueabihf/lib I can find all the three error files mentioned above ie libc.so.6, libc_nonshared.a and ld-linux-armhf.so.3
I'm trying to compile the library for Beaglebone, if that helps.
Update 1:
The results of running make -f GNUmakefile-cross system after doing a fresh git pull
hassan#hassan-Inspiron-7537:~/cryptopp-armhf$ make -f GNUmakefile-cross system
CXX: /usr/bin/arm-linux-gnueabihf-g++
CXXFLAGS: -DNDEBUG -g2 -Os -Wall -Wextra -DCRYPTOPP_DISABLE_ASM -march=armv7-a -mfloat-abi=hard -mfpu=neon -mthumb -I/usr/arm-linux-gnueabihf/include/c++/4.8.2 -I/usr/arm-linux-gnueabihf/include/c++/4.8.2/arm-linux-gnueabihf --sysroot=/usr/arm-linux-gnueabihf -Wno-type-limits -Wno-unknown-pragmas
LDLIBS:
GCC_COMPILER: 1
CLANG_COMPILER: 0
INTEL_COMPILER: 0
UNALIGNED_ACCESS:
UNAME: Linux hassan-Inspiron-7537 3.13.0-35-generic #62-Ubuntu SMP Fri Aug 15 01:58:42 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux
MACHINE:
SYSTEM:
RELEASE:
make: Nothing to be done for `system'.
The problem is simple. It is in the --sysroot option. The value of this option is /usr/arm-linux-gnueabihf/ and it is used by the linker and the resulting library folder becomes
/usr/arm-linux-gnueabihf/usr/arm-linux-gnueabihf/lib/
I removed the --sysroot option from line 68 in the file GNUmakefile-cross and everything compiled and linked OK.
However, I couldn't run the example on my BeagleBone Black because of mismatch of some shared libraries versions. But this wasn't a real problem for me, because in my application I link crypto++ statically, not dynamically.
Based on Crosswalking's research I think I can explain what is going on. I don't think I agree with the assessment "The problem is simple. It is in the --sysroot option" since the Crypto++ environment script and makefile are doing things as expected.
I think Crosswalking's answer could be how to work around it; but see open questions below. The following is from Crypto++ Issue 134: setenv-embedded.sh and GNUmakefile-cross:
I think this another distro problem, similar to g++-arm-linux-gnueabi
cannot compile a C++ program with
--sysroot.
It might be a Ubuntu problem or a Debian problem if it is coming from
upstream.
When cross-compiling, we expect the following (using ARMHF):
SYSROOT is /usr/arm-linux-gnueabihf
INCLUDEDIR is /usr/arm-linux-gnueabihf/include
LIBDIR is /usr/arm-linux-gnueabihf/lib
BINDIR is /usr/arm-linux-gnueabihf/bin
How LIBDIR morphed into into
/usr/arm-linux-gnueabihf/usr/arm-linux-gnueabihf/lib/ (i.e.,
$SYSROOT/$SYSROOT/lib) is a mystery. But in all fairness, building
GCC is not a trivial task.
You should probably file a bug report with Debian or Ubuntu (or
whomever provides the toolchain).
The open question for me is, since $SYSROOT/lib is messed up, then is $SYSROOT/include messed up, too?
If the include directory is also messed up, then the cross compile is using the host's include files, and not the target include files. That will create hard to diagnose problems later.
If both $SYSROOT/include and $SYSROOT/lib are messed up, then its not enough to simply remove --sysroot. Effectively, this is what has to be done:
# Exported by setenv-embedded
export=ARM_EMBEDDED_SYSROOT=/usr/arm-linux-gnueabihf
# Used by the makefile
-I $ARM_EMBEDDED_SYSROOT/$ARM_EMBEDDED_SYSROOT/include
-L $ARM_EMBEDDED_SYSROOT/$ARM_EMBEDDED_SYSROOT/lib
Which means we should be able to do the following:
# Exported by setenv-embedded
export=ARM_EMBEDDED_SYSROOT=/usr/arm-linux-gnueabihf/usr/arm-linux-gnueabihf
# Used by the makefile
--sysroot="$ARM_EMBEDDED_SYSROOT"
Finally, this looks a lot like Ubuntu's Bug 1375071: g++-arm-linux-gnueabi cannot compile a C++ program with --sysroot. The bug report specifically calls out ... the built-in paths use an extra "/usr/arm-linux-gnueabi".
We need the paths:
A) /usr/arm-linux-gnueabi/include/c++/4.7.3 B)
/usr/arm-linux-gnueabi/include/c++/4.7.3/arm-linux-gnueabi
But the built-in paths tries to use:
C) /usr/arm-linux-gnueabi/usr/arm-linux-gnueabi/include/c++/4.7.3
D)
/usr/arm-linux-gnueabi/usr/arm-linux-gnueabi/include/c++/4.7.3/arm-linux-gnueabi/sf
E)
/usr/arm-linux-gnueabi/usr/arm-linux-gnueabi/include/c++/4.7.3/backward
Notice the built-in paths use an extra "/usr/arm-linux-gnueabi"
I am trying to link a compiled research experiment project, built in C/C++.
The project is dependant on HyperNEAT and robot simulation software WeBots.
I have cloned and built the HyperNEAT project successfully (in that project there are other dependancies such as Boost, TinyXML, JGTL (custom library) and other unrelated subprojects).
I have made a makefile including all neccesary header search paths and library paths, and compiling the two main .cpp files:
/ModHyperNEAT/mod_ctrler7.cpp
/ModSupervisor/mod_supervisor.cpp
works, giving me 2 .o files.
However, in the make link step, when I want to create (separate) executables of both files, I am getting the 'undefined symbols for architecture x86_64' error (see pastebin here: http://pastebin.com/kiwwCcUf). It seems that C++ standard datatypes and functions such as
std::string::end() const cannot be found.
I have googled and searched SO for answers regarding this, and it seems that either libraries are missing or binary incompatible if i understand correctly, but the libraries are there and both projects have been compiled with the -lstdc++ flag.
This is the make link step (and the used macro's from the makefile) :
CC = gcc
CFLAGS = -v -g -lstdc++ -Wall -Wno-error -ferror-limit=100 -fmessage-length=0
DEFINES = -DHCUBE_NOGUI -DTIXML_USE_STL
FLAGS = $(CFLAGS) $(DEFINES)
LIB_TINYXML = -L/Users/michahell/Documents/projects_c++/HyperNEAT/tinyxmldll/out
LIB_HYPERNEAT = -L/Users/michahell/Documents/projects_c++/HyperNEAT/NE/HyperNEAT/out
LIB_BOOST = -L/usr/local/Cellar/boost/1.57.0/lib
LIB_WEBOTS = -I/Applications/Webots/lib
LIBS = $(LIB_TINYXML) $(LIB_HYPERNEAT) $(LIB_BOOST) $(LIB_WEBOTS)
LIBFLAGS = -ltinyxmlpluslib -lboost_filesystem-mt -lboost_random-mt -lboost_system-mt -lNEATLib_d -lHypercube_NEAT_Base_d
WEBOTS_DYLIB = -dylib_file /Applications/Webots/lib/libController.dylib:/Applications/Webots/lib/libController.dylib
$(CC) $(FLAGS) $(LIBS) ./mod_ctrler7.o $(WEBOTS_DYLIB) $(LIBFLAGS)
I found out that to link to .dylib's I had to use a specific flag and specify the full path, hence the $(WEBOTS_DYLIB) macro.
I am using the -lstdc++ flag because in the HyperNEAT project that flag was also used for building that library. If i exclude this flag, i get a lot of errors during compilation (libc++ and libstdc++ incompatibility as I now understand). All of the library paths check out, and .a and/or .dylib files are present.
My knowledge of C/C++ and GCC tooling is very limited, as I have never had to use it before.
I think it might have to do with the fact that the HyperNEAT project contains a Boost 1.57.0 distribution which is used for their build, and that i have a separate (using homebrew) Boost version installed on my system, which is the same version:
$ brew info boost
boost: stable 1.57.0 (bottled), HEAD
http://www.boost.org
/usr/local/Cellar/boost/1.57.0 (10572 files, 439M) *
What could be the cause of this error failing my link step? Anyone should be able to reproduce my linker errors if both HyperNEAT and my project are cloned and put their root dirs in the same location. WeBots should be downloaded but only for the header includes and libraries. And of course my makefile paths should be modified.
If anyone can give me tips on how to solve this problem, i would GREATLY appreciate it!
It turns out that, for some reason, I had to include the lstdc++ flag to the library link flags and not as a compiler flag, AND the stdlib=libstdc++ as compiler flag.
Gooday everyone
I'm fairly new to ubuntu C programing although I'm
rather experienced in C programing in windows.
I have recently come accross a number of codes written
in 2005 which I'm interested in learning how they work.
Those codes needs BOOST library to compile, however they won't
compile on the newest BOOST version present on my ubuntu 12.04.
I set the gcc compiler on lenient so that it ignores all those error
messages. The code did compile and ran afterwards.
However, when I used GDB debugger to watch how the program flows
I noticed that there are likely errors in the way the program runs
due to using a different BOOST version rather than it's original. Hence
I like to install the BOOST version corresponding to the code I downloaded.
To do that, I installed Ubuntu 5.04 and BOOST 1.33.0 which seemed to have been created in late 2005. I downloaded it
but I didnt found any detailed instruction on how to install it.
Only vague description on using BOOST jam, I played around with BOOST
jam for quite awhile without success.
And this old BOOST does not have installation commands like
"sudo apt-install boost-dev" style option
Thus I like to ask if anyone can give a easy to understand step by step instruction
on how to install the BOOST library downloaded from the above link.
like.....
step1: download boost jam from boost webpage
step2: unpack it in home/boost/ then type make configure
...and so on...
Big thanks for any useful info.
New Contents appended here
in response to the comments given
Hi, I went through the info given by your link and
managed to run the boost library examples given by your link.
That is, I can compile a single cpp file with the command
g++ -I boost_1_33_0 test.cpp -o test
(I'm keeping the boost library and the cpp file to be compiled in the
same folder)
However, the program package I'm interested in is build with make (not cmake).
I have some experience writting cmake files but not make files.
And I do not see any link to boost library command in the make file of the
program package. The readme file only has one sentence that says I
need to have boost installed without explaining what that meant.
I assume it means that either I have to build and do makeinstall the boost or
I could add some lines in the makefile for a link. I thought
maybe you can quickly point out whats missing in the makefile.
The readme file:
To compile, go into the moses directory and do 'make'. You'll need the
latest boost libraries. If compilation still fails for weird reasons,
you could try g++ with the -fpermissive (newer versions reject lots of
code that was ok with older ones). If you are going to be making
changes and recompiling frequently you'll probably want to disable -O3
in the makefile (I use templates liberally, so -O3 really speeds up
the code, but really slows down compilation).
And the makefile:
CC = g++
PROJ_NAME = moses
LINK_FLAGS = -Wall -Iutils/ -Itrees/ -Irewrite -I./ -Imodeling/ -Ifitness/ \
-Ialignment/ -Isim/ -Ilocal/ -O3
COMP_FLAGS = -Wall -Wno-sign-compare -Iutils/ -Itrees/ -Irewrite -I./ \
-Imodeling/ -Ifitness/ -Ialignment/ -Isim/ -Ilocal/ -O3
src := $(wildcard *.cc) $(wildcard utils/*.cc) $(wildcard trees/*.cc) $(wildcard modeling/*.cc) $(wildcard fitness/*.cc) $(wildcard alignment/*.cc) $(wildcard main/*.cc) $(wildcard rewrite/*.cc) $(wildcard sim/*.cc) $(wildcard local/*.cc)
obj := $(patsubst %.cc,%.o,$(src))
all: $(PROJ_NAME)
%.o: %.cc
$(CC) $(COMP_FLAGS) $< -c -o $#
$(PROJ_NAME): $(obj)
$(CC) $(LINK_FLAGS) $^ -o $(PROJ_NAME)
run:
$(PROJ_NAME)
clean:
find -regex ".*~\|.*\.o"|xargs rm -f
rm -f $(PROJ_NAME)
rm -f $(PROJ_NAME).exe*
depend:
makedepend -Y -- $(COMP_FLAGS) -- $(src)
utils/exceptions.o: utils/exceptions.h utils/utils.h
utils/io_util.o: utils/io_util.h utils/tree.h utils/basic_types.h
# ......lots more lines like that.........
I have an old instruction flying around here for Boost 1.34.1, which reads like this (project-specific stuff cut away):
unpack boost sources
cd into tools/jam/src
run ./build.sh to build bjam
cd into the main source directory
tools/jam/src/bin.linux/bjam threading=multi --layout=system --toolset=gcc --without-python variant=release --prefix=/usr/local install
The --without-python was necessary as the target system didn't have Python installed, which caused the build to fail messily.
Obviously you can / need to fiddle with the individual settings (like threading support, release vs. debug variant) to suit your needs, but it should be a good starting point.
If you need ICU support (for Boost.Regex and Boost.Locale), it gets more complicated...
Note that the build process has changed over the years; you shouldn't use the same procedure for more up-to-date boost versions. It's just what I used back then.
Edit:
As for the second part of your question, the Makefile doesn't need to refer to Boost explicitly if boost is installed in the standard system directories.
You do not have to state -I /usr/include for compilation as that is searched automatically; the same goes for -L /usr/lib during linkage.
The fact that the author of the Makefile copied the compiler flags into the linker flags verbatim doesn't really help intuitivity either... ;-)
If you have Boost in a custom directory (either the headers only, or by stating a custom directory in the --prefix option of my build instructions), you need to make the following modifications (look for "boost"):
LINK_FLAGS = -Wall -Iutils/ -Itrees/ -Irewrite -I./ -Imodeling/ -Ifitness/ \
-Ialignment/ -Isim/ -Ilocal/ -L /path/to/boost/libs -O3
COMP_FLAGS = -Wall -Wno-sign-compare -Iutils/ -Itrees/ -Irewrite -I./ \
-Imodeling/ -Ifitness/ -Ialignment/ -Isim/ -Ilocal/ \
-I /path/to/boost/includes -O3
That should do the trick. As the Makefile does not link any of the Boost binaries (e.g. -l boost_program_options or somesuch), it seems that it makes use of the Boost headers only, which would make the -L /path/to/boost/libs part (and, actually, the whole compilation step detailed above) superfluous. You should be able to get away with simply unpacking the sources and giving the header directory as additional include directory using -I /path/to/boost/headers.