I am currently trying to build a gcc 4.4 on a newly updated ubuntu 11.10. However building the compiler always fails with some problems, when trying to bootstrap.
First crti.o and crtn.o could not be found. I symlinked them into another directory within the search path, and so I could get a bootstrap compiler. However now it fails with the message:
configure: error: C compiler cannot create executables
I tried out to use the intermediate compiler on a simple test program, and I get the message:
$ ./host-x86_64-unknown-linux-gnu/prev-gcc/xgcc test.c
xgcc: error trying to exec 'cc1': execvp: File or directory not found
So something is seriously messed up with the paths of my intermediate compiler it seems. When I did the same thing on ubuntu 11.04 everything just worked fine.
I used the following command line to configure:
./configure --prefix=/opt --program-suffix=44
I also tried both a VPATH-built, as well as an in source build, but both have the same problem. I still have the gcc44 from before the update lying around within /opt, but it has become unusable since the update.
Is there any easy way to get gcc 4.4 working under Ubuntu 11.10? I does not have to be a self build compiler, if I can just download a package and install that, it would be fine as well.
This is caused by Ubuntu's switch to multi-arch. Basically all the "standard" paths have been changed so you can install libraries for multiple architectures on the same machine without any clashes.
Ubuntu GCC has been patched to fix the problem, of course, but the upstream compiler still wasn't fixed last time I checked.
See here: https://askubuntu.com/questions/69365/cant-build-gcc-anymore-since-upgrade-to-11-10
Related
I am trying to recreate an environment for a C++ program that uses google's protobuf 3.5.1.1. I have built protobuf using MinGW, and all seems well. When I run the make check command, I get 5 passes and 1 fail (protobuf-test.exe fails).
The commands that I run in MSYS2 are:
./autogen.sh
./configure
make
make check
When I try to build the program that uses protobuf, I get a massive wall of linker errors.
Any suggestions on how to fix this issue is appreciated.
Thank you!
----- Edit -----
To solve this issue, I needed to update make protobuf in the MSYS2 MinGW 32-bit terminal.
I also needed to update MSYS2 by following these directions
Try cross compiling in a linux environment then running your make install through msys2 on windows platform.
./configure --help will display a list of flags you can include. Specifically --target="your desired target"
I am a non-administrator user of a Linux (CentOS 6.6) server at work. I log in through a terminal program on a windows computer. My problem is that the IT does not feel comfortable upgrading GCC/gfortran for me so I want to just run it from my personal folders. They claim the latest yum (devtoolset-3 for this OS version) will downgrade some other feature they have. It's not the latest version of GCC anyway.
I have downloaded the latest GCC 5.3 binaries and prerequisites from gfortran.com and can almost get my test code to compile. Actually, when I do the following it will compile with -c but will not not link. That folder is where I put the prerequisites and also I copied stuff from the /usr/lib64 directory into there as well.
LD_LIBRARY_PATH=/users/home/me/me/gcc53mark/my_lib
export LD_LIBRARY_PATH
./gfortran test.f90
The error message is as follows:
collect2: fatal error: cannot find 'ld'
But ld does exist
-bash-4.1$ whereis ld
ld: /usr/bin/ld /usr/share/man/man1/ld.1.gz
After much effort I have answered my own question! To install GCC to my own personal Linux account as a non-administrator, I ended up having to compile GCC and not use the binaries I originally downloaded. My friend was make install which knows where everything needs to go, which I did not have with the binaries.
The key was to install to my user root directory /GCC with the following command provided with the build-it-yourself method (-prefix):
../gcc-5.3.0/configure --prefix$HOME/gcc-5.3.0 --enable-languages=c,c++,fortran --disable-multilib
For me I had to disable multilib because I guess my system only has 64 bit libraries (I think this was causing my original problem).
Before I got to that point, I had to also download the prerequisites manually since my workplace I suppose blocks the automatic prerequisite downloader as referenced in these instructions.
https://gcc.gnu.org/wiki/InstallingGCC
Prerequisites need to be taken from here, placed into the root of the directory that gets created when you unzip GCC. Then unzip them and link them as is done in the batch file you have already unzipped ./contrib/download_prerequisites.
ftp://gcc.gnu.org/pub/gcc/infrastructure/
Finally I need to run the following command, after it is all up so it looks at the new libraries. I will add this to my .profile when I am ready to fully switch to the local newer version.
LD_LIBRARY_PATH=/users/home/myself/gcc-5.3.0/lib64
export LD_LIBRARY_PATH
For now I am running gfortran with the following command but also I will add this to my .profile later.
/users/home/myself/gcc-5.3.0/bin/gfortran Test.f90
It works! Latest version of GCC running from my local Linux user non-administrator account!
Edited to add how we resolved this for the network group:
IT did not want to overwrite the original installation so we installed to some network folder /gcc-5.3.0 . Then we modified the group's .profile to add the library and binary paths to that, before the standard path.
I am trying to compile "xz-5.2.1" in MinGW/MSYS environment. I see the following errors:
#error UINT32_C is not defined and unsigned int is not 32-bit.
error: #error size_t is not 32-bit or 64-bit
I am not familiar with MinGW, could anyone shed some light on this? It looks like some macro definition are missing. Some header file missing?
ADD 1
The commands I used to compile the xz-5.2.1 are:
./configure
./make
The error screenshot:
Some background, I am following this link to compile the Tesseract-OCR library. And this is just one of the steps.
ADD 2
Based on the error message, I checked the sysdefs.h file. It contains the following content:
#ifdef HAVE_CONFIG_H
# include <config.h>
#endif
The above make output contains the -DHAVE_CONFIG_H, so I think the system header file <config.h> should be included.
But strange enough, I searched the C:\MinGW\include, there's no such file. So I GUESS this may have caused the undefined UINT_MAX warning at line 57. And then the UINT32_C is not defined error at line 58.
But I don't know why the system header file config.h is missing and where to get it.
ADD 3
I dig a bit about the GNU autotools. And luckily enough I find that the following commands can carry on my build process: (Though I am still not very sure why it works. All I know is that it may be related to portability.)
autoheader (this generates the config.h.in file)
./Configure (this generates the config.h file from the config.h.in file)
And now, my build process is blocked by another issue as below:
It seems this is a known issue. And another thread has addressed it.
(I will continue update with my progress.)
If you care for an easier way to handle this kind of dependency management or a general update on toolchain functionality, I strongly suggest switching to MSYS2 with MinGW-w64.
Both projects aim (and succeed) in bringing a better version of the original. MSYS2 comes with a large number of 3rd party libraries that you can easily install. MinGW-w64 allows for GCC with full C++11/14/... support and extended Windows API availability, along with some useful extensions and more up to date headers. You'll notice that most problems originating from system headers will have already been solved, either by the MinGW-Packages scripts below, or upstream (of either MinGW-w64 or the projects themselves).
For you specifically, I suggest the following steps:
Install and update MSYS2.
Open an MSYS2 command prompt (or the 32-bit or 64-bit command prompts if you plan on building 32-bit or 64-bit things) from the start menu entries. Install {32-bit,64-bit} MinGW-w64 GCC:
pacman -S mingw-w64-{i686,x86_64}-gcc
Install tesseract-OCR:
pacman -S mingw-w64-{i686,x86_64}-tesseract-ocr
and optionally the data files:
pacman -S mingw-w64-tesseract-ocr-osd mingw-w64-{i686,x86_64}-tesseract-ocr-eng
And you're done. Of course, you can still compile the various dependencies yourself, but why bother? If you really want to, you can start from the build scripts for the packages you can install in MSYS2, which are located here:
https://github.com/Alexpux/MINGW-packages
Just open the PKGBUILD files and you can see the build steps required. Note that all these scripts assume the dependencies have been installed within MSYS2.
Also note that the installed packages and compilers are all independent of MSYS2 as you'd expect: you can use it only as a tool to keep your development tree up to date, and build from any other Windows environment.
I'm on mac 10.7.5, using xcode 4.6.2 and working with the OpenCV 2.4.3 library. I went through the process of making the build directory with the cmake files in terminal and did the download.
I added the .dylib files in Xcode and changed the header path, changed C++ Library to libstdc++, but when I compiled I got this error :
ld: library not found for -lopencv_calib3d.2.4.3
clang: error: linker command failed with exit code 1
I have libopencv_calib3d.2.4.3.dylib added in the project so I have no idea what else it needs.Any ideas?
If your "make back-end" is Cmake you should stick to using it. Local config changes in Xcode can "secretly" be overwritten when the cmake is rerun (which for example happens after you make changes to it) creating weird build errors and forcing you to remember all the manual changes you made.
Your problem seems to be that the generated project doesn't seem to know where to look for opencv.
Assuming you installed opencv using macports you should add a line saying
link_directories(/opt/local/lib)
to your CMakeLists.txt. (if you installed it using brew, or compiled it manually just replace /opt/local/lib for /usr/local/lib or the path to your compiled libraries)
Also make sure to link against opencv_calib3d instead of opencv_calib3d.2.4.3 (unless you have a very particular reason for bypassing this, but that usually means that something else is weird in the setup :) )
Final pointer that you might already know of: As you are already using Cmake you should add the libraries to link against using TARGET_LINK_LIBRARIES(...) in Cmake rather than manually adding them in Xcode (referring to my previous argument).
I spent the whole day setting up GCC as per
http://cygwin.wikia.com/wiki/How_to_install_GCC_4.3.0
On the gcc make step, it told me it needed a c++ compiler to continue for whatever reason,
So I updated Cygwin in the usual fashion and got gcc-c++.
After that, cgywin did me the courtesy of deleting the 3 prereq libraries that I had installed previously,
So I went about remaking them. GMP worked fine, and then I got to MPFR. For whatever reason MPFR is throwing this error:
configure: error: libgmp not found or uses a different ABI.
The resources mention this error here.
I’m pretty sure its finding GMP alright as cygcheck shows it up no problem, and the config even mentions it finds it.
When I try:
./configure --with-gmp-build=/gmp/,
it throws:
error: header files gmp-impl.h and longlong.h not found
So far I have not been able to get MPFR working, despite googling and trying for hours.
MPC also fails to find GMP in the usual manner.
I've also tried just handing the libraries to GCC in a subdirectory and letting it compile them.
When I try that, it gets to the make step, then throws a stage1-bubble error and fails. No description of the error or anything.
I have no idea what else to try.
After compiling and installing gmp, I had the same problem building mpfr until I set correctly the library search path:
export LD_LIBRARY_PATH=/usr/local/lib
export LIBRARY_PATH=/usr/local/lib
export CPATH=/usr/local/include
as it is explained in the http://cygwin.wikia.com/wiki/How_to_install_GCC_4.3.0
Instead of this:
./configure --with-gmp-build=/gmp/
USE THIS:
./configure --with-gmp-build=/gmp
It doesn't like having the "/" at the end.
I know this post is a bit dated, but I was looking for an answer or results to a similar situation but with much more current versions of the same libraries. While reading this post as for some suggestions as well as other online material I was able to piece everything together. So for anyone who may be working with the newest releases here it goes:
First you will probably want to configure, make and install GMP first for both C and C++ as static libs; also make note of the architecture x86 or x64. Once you run make, make check, make install; and you are confidant that you have no errors, then when you go to configure MPFR you will want to run it as such:
./configure --with-gmp-build=/path/to/gmp --enable-static --disable-shared
and if the paths are set properly this should give you the needed generated make file.
I'm brand new to Unix-Posix environments and commands and new to Cygwin. I've mostly used Windows platforms and very few Macs or Apples. Matter of fact I've only been working at this for a little less than a week as I'm trying to do the same thing. So now I'm venturing into a whole new world of Unix... but it's sort of simulated through Cygwin as I'm still running windows.
I'm trying to build GMP, MPIR, MPFR & MPFRC++ all from the latest original distributions and eventually will try to integrate them into Visual Studio 2017.
I had actually found this post because I was able to build MPIR through visual studio with the help of python and vyasm, but as for GMP I needed Cygwin with GCC. It took me a little while to figure out how to get GMP to build properly but once I did and ran the tests I was able to have all pass with 0 failures. I then started to work with MPFR the same as you.
My similar issue was after I ran ./configure as default, it was giving me the message that the libgmp wasn't found. Also if I tried to call make, it would do nothing as the make file didn't exist. This led me back to doing google searches... While searching for tips, suggestions etc. I came across this Q/A. While I was reading this Q/A, the install files, some other FAQ and even another website, I was kind of able to piece everything together and what I have found that works is what I suggested above.
Not only do you need to set the flag --with-gmp-build but you also need to set the appropriate library types that gmp was built as and you need to make sure that the architecture types match too. They need to have the same matching ABIs.
Ah, just as I was about to finish up typing this answer, make check just finished running through Cygwin and I have a result of:
Testsuite summary for MPFR 4.0.1
============================================================================
# TOTAL: 180
# PASS: 180
# SKIP: 0
# XFAIL: 0
# FAIL: 0
# XPASS: 0
# ERROR: 0
============================================================================
I hope this helps you out. The only thing is for myself; I'll have to go back and probably run all of this again from the beginning including GMP only because I believe I only did GCC and not G++ and my next step is to try and build MPFRC++.
Then I might try to do some related libraries from here afterwards. My first phase is to try to build them all through Cygwin and GCC/G++. The next step will be to try and port them over to Visual Studio 2017.
You might want to use --with-gmp-lib=/path/to/gmp/lib/dir