I have been using gcc to compile C programs.
I want to know what exactly is the meaning of Compiling a C program.
I checked
cc -S prog.c
will give an .S file and if you analyze the hexdump I saw some assembly instructions.
So what does the compiler do which is not required in case of interpreted languages?
How can I compile a C++ application using g++ with x86 platform settings on a x86_64 machine ? What are the possible negative side-effects of compiling a C++ application with x86 settings on a x86_64 platform ?
I had developed a C program on linux (x86). Now I need to port it to HP-UX OS on IA/PA 64 bit arch. What are the options to be used with cc compiler, on linux (x86), so that this could be done, if at all.
I am trying to compile systemc. Configuration is done OK and Makefiles are created. As soon as the "make" command i issue, recursively reaches the "utils" directory, errors are produced -declaration and include errors.
The files referenced by the errors DO exist and i hava managed to give g++ "-I ${LD_INCLUDE_PATH}, ie the variable, where all header dirs are listed. Libraries and compilers for gcc3.3 and g++3.3, i believe, are installed OK. code...
Compiling a C++ program including libpq-fe.h by the command Code: g++ -I/usr/include -L/lib -lpq my_program.cpp for access to a PostgreSQL database results in the error Code: undefined reference to '_PQconnectdb'
I've got an error with compiling a program: relocation R_X86_64_32 against `a local symbol' can not be used when making a shared object; recompile with -fPIC /usr/lib/python2.5/config/libpython2.5.a: could not read symbols: Bad value collect2: ld a retourn 1 code dtat d'excution make[2]: *** [bindings/_yafqt.so] Erreur 1.
I have 3 c++ files, classdef.h (header file with class definition), methods.cpp (class methods) and program.cpp - the program itself. Both .cpp files have
Code:
#include "classdef.h"
in files. How I can link the files together and compile them in one executable program? I am using Geany IDE for coding.
I've been requested to modify a perl backup script we use which mounts a networked NAS machine. The actual mount command is called inside a system() command. In testing, I can do a test of the return value, so I'm able to tell whether or not I've mounted properly, but I can't make heads or tails out of the return on deliberate errors. (Such as mounting a non-existent box.) I'm displaying the returned value, which is 256 in my tests. The man page for mount listed error messages with values from 1 to 64, but no higher. Am I drawing this error value from the system command somehow? And what would be the best way to determine what this error means?
I was reading the following forum guide:openSUSE software installation hints So I decide to install from the souce code the following webcam application:
GTK+ UVC Viewer In order to avoid problems ( it is my first installation from souce code) I downloaded and unpacked everything in an USB key and I worked from, as the guide said, from the Linux terminal. Once I was in the right directory and I compute the command:
I am trying to compile a simple remote procedure call program. I am getting an error: /tmp/ccy0M5rT.o: In function 'main': rpchighlayer.c.text+0x5c): undefined reference to 'rnusers' collect2: ld returned 1 exit status
I wrote a program in c++ on fedora 11. When i compiled it using g++. it displays the following error:
g++ fibnocci.cpp fibnocci.cpp:2:22: error: iostream.h: No such file or directory fibnocci.cpp: In function int main(): fibnocci.cpp:8: error: cout was not declared in this scope fibnocci.cpp:8: error: endl was not declared in this scope
I'm sometimes compiling a piece of software on my Fedora 13 machine and then transferring it to an Ubuntu box. Recently I noticed this would yield issues with Fedora's .so files:
Given I had a program's binary and an .so file the program would attempt to load, both compiled on Fedora, the program would run fine on Ubuntu 8.04 LTS but would be unable to load to open the Fedora .so with an "ELF file OS ABI invalid" error.
Now any program I natively compile on Ubuntu which I run on Fedora instead, will throw the *same* error with Fedora .so's - so basically neither a Fedora bin on Ubuntu, nor an Ubuntu bin on Fedora can load the Fedora .so files, only a Fedora bin natively on Fedora can.
That means Fedora .so files are always a problem on Ubuntu, and Ubuntu bins are always a problem on Fedora (unable to load system libs there).
This is a bit odd for e.g. binary redistributions of programs (which are rare I know) and I also found out that e.g. Tibia's linux client suffers from this issue (throws "ELF file OS ABI invalid" on my Fedora 13 when attempting to load my system's libGl).
And I started to wonder whether the only way to work-around to get something that works on all platforms is a Fedora binary shipped with .so files compiled on Ubuntu - which seems a bit stupid and ridiculously complicated just to get something that works flawlessly everywhere. Isn't there e.g. a gcc switch that will allow me to compile the .so files on Fedora in such a way that they can also be used on an Ubuntu system so just compiling everything on Fedora, yet getting it to work on Ubuntu aswell would be possible?
I decided I was going to compile XChat from source today for "fun." What I ended up doing was spending a few hours getting it to compile, then find out everything didn't work the way I thought it did. I was under the impression that after I ran "sudo make install" I was "upgrading", when in reality I'm just installing a seperate version alongside the old one. Apparently I have to keep the old version so every program that was compiled with it will continue to work, so how do I:
1) Force a program to use a specific version of GTK when compiling. 2) Find where my version of GTK installed to.
I've asked some people and I've just been told to "learn LD_PRELOAD". I've googled and can't figure how I could even apply that to my current problem
Having problems compiling a program on a new server. It compiles just fine on Ubuntu 8.0.4 server but a new install of the latest Ubuntu Server edition it throws out a bunch of warnings on configure. I am assuming I am missing some lib's or something of the sort.
Code: checking for sys/select.h... yes checking for sys/types.h... (cached) no checking for unistd.h... (cached) no checking for memory.h... (cached) no checking crypt.h usability... no checking crypt.h presence... yes configure: WARNING: crypt.h: present but cannot be compiled configure: WARNING: crypt.h: check for missing prerequisite headers? configure: WARNING: crypt.h: see the Autoconf documentation configure: WARNING: crypt.h: section "Present But Cannot Be Compiled" configure: WARNING: crypt.h: proceeding with the preprocessor's result configure: WARNING: crypt.h: in the future, the compiler will take precedence checking for crypt.h... yes checking assert.h usability... no checking assert.h presence... yes configure: WARNING: assert.h: present but cannot be compiled configure: WARNING: assert.h: check for missing prerequisite headers? configure: WARNING: assert.h: see the Autoconf documentation configure: WARNING: assert.h: section "Present But Cannot Be Compiled" configure: WARNING: assert.h: proceeding with the preprocessor's result configure: WARNING: assert.h: in the future, the compiler will take precedence checking for assert.h... yes checking arpa/telnet.h usability... no checking arpa/telnet.h presence... yes configure: WARNING: arpa/telnet.h: present but cannot be compiled configure: WARNING: arpa/telnet.h: check for missing prerequisite headers? configure: WARNING: arpa/telnet.h: see the Autoconf documentation configure: WARNING: arpa/telnet.h: section "Present But Cannot Be Compiled" configure: WARNING: arpa/telnet.h: proceeding with the preprocessor's result configure: WARNING: arpa/telnet.h: in the future, the compiler will take precede nce checking for arpa/telnet.h... yes checking arpa/inet.h usability... no checking arpa/inet.h presence... yes configure: WARNING: arpa/inet.h: present but cannot be compiled configure: WARNING: arpa/inet.h: check for missing prerequisite headers? configure: WARNING: arpa/inet.h: see the Autoconf documentation configure: WARNING: arpa/inet.h: section "Present But Cannot Be Compiled" configure: WARNING: arpa/inet.h: proceeding with the preprocessor's result configure: WARNING: arpa/inet.h: in the future, the compiler will take precedenc
I have Fedoara 11 install on my PC. I want to compile program as 32 by g++ option -m32. I have 32 bit library install on my machine which present in path
When i compile my sample program with -c & -m32 option it compile correctly
I am trying to compile a program with mpicc. To compile I use:
CC=/usr/bin/mpicc ./configure
However mpicc's default C compiler is gcc 4.4.2 and I need to use 4.0.0 which I have installed in another directory. I do not (nor can get) root privileges so I cannot edit mpicc's default compiler. Is there an option I can set while configuring?
I've been trying to compile a small program on my 64bit Centos 5.3 but I need to compile it in 32bit mode. I've tried editing the make file and changing CC=gcc to CC='gcc -m32" and that doesn't work. I have also tried to use "make CC='gcc -m32' and that doesn't work either. It wont compile. Do I need to install something with yum to get 32bit compiler to work?
I'm compiling a project that uses boost libraries (1.45.0) for a MIPS target, and when trying to compile with bjam I get the following error:
Code: libboost_system-mt.a(error_code.o): In function `(anonymous namespace)::generic_error_category::message(int) const': error_code.cpp:(.text+0x55c): undefined reference to `__glibc_strerror_r' collect2: ld returned 1 exit status
A C program which compiled well in Dev CPP compiler is not compiling in Linux-Fedora.It shows the following error message/tmp/ccy02C6e.o.eh_frame+0x12):undefined reference to '__gxx_personality_v0'collect2:ld returned 1 exit status