I have Slackware 13-64 bit and I would like to cross-compile some Fortran and C programs to windows 32 and 64 bit applications. The Linux versions utilize X11 for the graphical users interface. I have tried MINWG but I could not get it to work, either in pre-built binary or source code form. Is there a binary version that works on Slackware 13? Alternatively, is there another method available?
View 1 Repliesi'm currently using the gnu compiler "gfortran" , but i don't think the executable files it creates are compatible with windows XP (though i may be mistaken). my boss needs a few simple programs that i have written, but i will not be able to compile the code on his pc. are there any compilers out there that i can use to create such executables?
View 3 Replies View RelatedCommand to compile n execute a Fortran program on a Linux machine
View 6 Replies View Relatedi have fedora 11(64 bit).i want to do the cross compilation for 32 bit arm processor.i install all required thing but any of the make command(make,make mrproper,defconfig) is not working.
View 8 Replies View RelatedI want to compile a program with the source I already have but I have to add an option. When I run the command ./autogen.sh the terminal returns aclocal not found. Is it a package or something i have to install?
View 6 Replies View RelatedI'm trying to cross compile the GNU make for Alpha Architecture on my i686 PC and the GNU make i compiled would be placed in my virtual hard disk which is a Alpha based linux simulated system. My question is now , I'm able to cross compile the GNU make on my i686 real PC machine , but when i let the make program run in my virtual machine , it pops out the error ..
make: /lib/libc.so.6.1: version `GLIBC_2.4' not found (required by make)
After that, I try another alternative , and I read through the file "INSTALL" on the GNU make directory that I downloaded from the internet . In this case , I downloaded make-3.81. On the sub section "Compiling For Multiple Architectures" It says that
"You can compile the package for more than one kind of computer at the same time, by placing the object files for each architecture in their own directory. To do this, you must use a version of `make' that supports the `VPATH' variable, such as GNU `make'. `cd' to the
directory where you want the object files and executables to go and run the `configure' script.`configure' automatically checks for the
source code in the directory that `configure' is in and in `..'. If you have to use a `make' that does not support the `VPATH'
variable, you have to compile the package for one architecture at a time in the source code directory. After you have installed the
package for one architecture, use `make distclean' before reconfiguring for another architecture." And I do not understand the line "by placing the object files for each architecture in their own directory" . What object files that I should put ?
First, I'm trying to cross compile mono 2.6.4 so that it will run on Freescale Embedded Linux for a PowerPC CPU. My host machine is running Fedora 13 on an x86. Freescale provides a copy of the Linux Target Image Builder (LTIB) that has been pre-configured for the particular board I'm using, and LTIB seems to be able to help with the cross compilation of other stuff--you can add in your own packages to be built and included in your newly-built Linux image.
Mono depends on pkg-config and glib-2, so I have selected them in the LTIB package selection config. I've also added a new package for mono that builds mono-2.6.4 from the source tar (after the other deps have also been built).
I'm having a problem getting the glib-2.24.0 package libraries created. Basically they appear to build and link ok, but then libtool runs and errors out claiming it has a syntax error! (numerous wths followed...)
Here is LTIB's temporary build script for only the glib2 package:
Code:
Here is the output when building the glib2 package (configure + make):
Code:
Build path taken because: directory build, build key set, no prebuilt rpm,
I am trying to build cross-toolchain for arm. I dont want ready-to-use solution. I want my own . I want to build toolchain manually, but I cannot compile gcc. When I invoke "make" I get the following error:
Code:
-c ../../../../src/gcc-4.4.2/libgcc/../gcc/config/arm/lib1funcs.asm -include _arm_addsubsf3.vis
/home/ecco/Progs/cross-compilers/build/arm-linux/build/build-gcc/./gcc/xgcc -B/home/ecco/Progs/cross-compilers/build/arm-linux/build/build-gcc/./gcc/ -B/home/ecco/Progs
[code]....
Then, I try to build gcc:
Code:
../../src/gcc-4.4.2/configure --target=arm-linux --enable-languages=c -- prefix=/home/ecco/Progs/cross-compilers/result/arm-linux --with-headers=../../src/kernel_headers_2.6.33.4/include/
Make fails...my arm-linux-gcc is actually can be built. When I do the following:
Code:
make all-gcc
make install
I get my arm-linux-gcc in ~/Progs/cross-compilers/result/arm-linux. So I can compile kernel , but I want user-space , so I do:
Code:
make all-target-libgcc
Make fails when it tries to build libgcc (AFAIK I need libgcc to build glibc). But why does it fails? I have header files like "stdio.h" on my gentoo box. Maybe compiler somehow wants files like "stdio.h" for my TARGET (arm-linux)? But AFAIK these are files are Standart C Library header files, and glibc did not even compiled yet, so I dont have these headers for my TARGET...
I found description of this error here:[URL](gcc part). But when I add additional flags -Dinhibit_libc -D__gthr_posix_h, then make fails when it tries to compile ../gcc-4.4.2/gcc/crtstuff.c with the same errors.
Actually I had a folder called Lib, in which I had a few libraries installed and configured.(MPI, PETSc, SLEPc) I accidently deleted (by rm) the contents of that folder.
Then I reinstalled MPI,PETSC,SLEPC using the same tarballs as earlier(thus, the version etc. is the same).
I already had a fortran program with many modules and subroutine files, and the corresponding makefile which used to compile and run fine. But now when I type "make" in the same directory as before, I get this: (shortened, many similar errors)
Code:
Firstly, why is mpif90 compiling my program when I didnt tell it to? In my makefile, I have specified gfortran as my compiler....nowhere have I even mentioned mpif90. Such a thing never used to occur before.
Also, if I rename file_variable.f to file_variable.f90, these errors dissappear (I understand why, but that is not the problem), but a new error comes:
Why is the new MPI installation interfering with make ? I want to go back to how things were before I stupidly broke all installations. The errors themselves are not the problem, the problem is why the new mpi installation is interfering with make.
It seems as if mpif90 has taken over make and gfortran. I suppose I didnt install it correctly, or probably I didnt uninstall the earlier one correctly.
Everything that worked earlier doesnt work suddenly. I use ubuntu 11.04 MPICH2-1.3.2
I am trying to nuild executables to run on Windoze under Fedora 12. I have install all of the mingw32 RPMs. Now I just need a URL that discribes how to compile. I am using autoconf and automake. I have done it before but that was installing from Mingw32 source.
View 2 Replies View RelatedI want to cross compile gtk program from linux to arm. I am using ubuntu10.04. How can I cross compile it for arm. I have tried this but this is my output:
root@habesh-laptop:~/habesh/gtk# /root/CodeSourcery/Sourcery_G++_Lite/bin/arm-none-linux-gnueabi-gcc -o trail1 base1.c 'pkg-config --cflags --libs gtk+-2.0'
cc1: warning: include location "/usr/include/gtk-2.0" is unsafe for cross-compilation
cc1: warning: include location "/usr/include/atk-1.0" is unsafe for cross-compilation
cc1: warning: include location "/usr/include/cairo" is unsafe for cross-compilation .....
I'm trying to cross-compile "sudo" source for Power-PC platform using Montavista tool chain. I'm getting the following error message during configuration: checking host system type... Invalid configuration `ppc_82xx': machine `ppc_82xx' not recognized It is clear that it has found the cross-compiler and configure knows that we are cross-compiling but it fails to recognize the machine. The complete dump follows:
[vhn@localhost sudo-1.7.2p2]$ ./configure --host=ppc_82xx
configure: WARNING: If you wanted to set the --build type, don't use --host.
If a cross compiler is detected then cross compile mode will be used.
configure: Configuring Sudo version 1.7.2p2
checking whether to lecture users the first time they run sudo... yes
checking whether sudo should log via syslog or to a file by default... syslog
[Code]...
I am trying to get openswan installed on cygwin. Having issues with that, their mailing list has suggested to cross compile it from linux, and I've had some gcc related errors. So I was wondering what I can use that will install on ubuntu to compile this for cygwin. As am I supposed to run this configure on the gcc package? [URL]
View 1 Replies View RelatedI try to cross compile httpd for power pc.It seems that configure doesn't detect cross compilation. The first error was that sizeof (void*) is less than (long), so I have commented out that check (I think it is a hard work around). I've got another error ./dftables: cannoct execute binary file. Why it tries to execute binary for ppc on x86 machine?This is my build script.
#!/bin/sh
NAME=httpd-2.2.17
PWD=$(pwd)
[code]....
how to cross compile the perl script for arm_v5t.le.gcc??downloaded perl script unable to compile from the cross folder as per the readme.txt instruction
View 1 Replies View Relatedhow to cross compile Mplayer into ARM target board , my target board details are URL...and i have successfully compiled Mplayer in intel pc.
View 2 Replies View RelatedI need to cross-compile some libs for using in a ARM11 board running Linux (at this time, 2.6.28, but I'll try update it... and for costs reason, maybe it turns on a ARM9 board). The libs are libfprint, libusb-1.0 (that is a requirement of libfprint), libSDL (and some of its extensions) and maybe libsqlite3. How I can make this, with the development files in the host machine (an Ubuntu 10.04 32bits machine) and the runtime files at the host ARM11 Board.
View 6 Replies View RelatedI'm trying to cross-compile glib-2.24.0 for Linux running on a Freescale PPC CPU and running into one problem after another. My host is running Fedora 13, and there I can compile the same source for the host machine with no problems. For the cross compilation, I'm currently using LTIB (provided already configured from Freescale for my board/CPU).
View 3 Replies View RelatedI'm trying to cross compile wvdial to ARM, wvdial needs wvstreams and wvstreams need openssl and zlib. I've managed to cross compile openssl 1.0.0d;
View 3 Replies View RelatedI wanted to compile a program for my openwrt router from source i already have but to add an option. I ran ./autogen.sh in the terminal but it returned aclocal not found.
View 5 Replies View Relatedhow compile our own application using busy box
View 4 Replies View RelatedI am going to compile and install my first linux application from the command line. The program is the port scanner program NMAP, and I am looking at the documentation to learn how to do it. I saw this thread [URL], and I just want to make sure I understand what it is I'm doing.
bzip2 -cd nmap-5.35DC1.tar.bz2 | tar xvf -
cd nmap-5.35DC1
./configure
make
su root
make install
"bzip2 -cd nmap-5.35DC1.tar.bz2 | tar xvf -"
Here I am calling the bzip2 program to work with a specified file. look for the file nmap-5.35DC1.tar.bz2 in the current diretory (-cd) and to pipe (redirect) something, this is where I get lost, I think. And I know what x & v mean but not f, or what that has to do with the previous commands.
cd nmap-5.35DC1
Chage to the named directory.
./configure
I assume this is a command?
make
Builds an executable
su root
make install
not sure what the difference is between this and the above.
I am new to the embedded domain.I am learning about cross compilation. Just i want to compile an application and an driver file for ARM platform. I have downloaded the toolchain binaries from [URL]..
View 3 Replies View RelatedI am trying to cross compile php for "arm" target using "arm-angstrom-linux-gnueabi" tool chain. Apache compilation was successful after using a few ac_cv configuration options(and a few others to get rid of compile errors). But was unable to compile php for the target, as the 'apxs' file used was not executable in the build machine.
I tried creating an apxs file for the build machine architecture(intel 686), and using it. It was able to compile then, but still php page is not getting displayed on the web server. (Have done the procedures like, loading the libphp5 module on httpd conffile and all)
I'm looking for a way to create a cross-platform GUI application. The result must be able to run on linux, windows and OSX. And it must be a point-and-click GUI.For development I should be able to rely on Open-Source tools on linux only (that means no access to Windows or OSX)The target should be able to install the result relatively easily, that means any dependencies must be freely available, and the setup steps must be very minimal (probably means no installing development tools or running compilers)
My first thought was java, but the standard Swing GUI can look a bit ugly on some platforms. So I'm wondering if there's anything else. My next thought was C++/Qt, but I don't think I can cross-compile this from Debian for Windows or OSX, can I? Next I thought of python and PyQt, but it looks like PyQt isn't available for OSX. And finally I even thought of making some kind of web application running on a tiny web server of some kind, then accessing it with a native browser, but I'm pretty sure this doesn't meet the "easy to install" criteria. I'm finding this so tricky, do all the existing cross-platform applications use natively-compiled C++ for this? Or is there an obvious alternative that I'm overlooking?
I am looking for a cross-platform LAN IM software for use on my home network, i've got some windows vista & Debian boxes. Is there any such software? I have looked around sourceforge and freshmeat. I tried J-Lan Communicator, it didnt work
View 1 Replies View RelatedA friend backed up an in accessible Windows 7 partition to a big (close to 1TB) external USB drive using KNOPIX. Unfortunately the umlauts and accents in the file names of his extensive music collection now appear garbled (UTF8 vs ISO xxx, I guess) when he accesses that drive from Windows. My guess is he needs to mount the two drives in a 'special way,' that takes the diacritics translation into account, on KNOPIX? Or do you have another idea what went wrong? This is the second part of a windows/linux problem I'm having. See how-does-one-mount-a-partition-in-windows-7 for the other part
View 1 Replies View RelatedI am trying to cross compile the "libtorrent-rasterbar-0.14.8" i got the error
Code:
checking for boostlib >= 1.35... yes
checking whether the Boost::System library is available... yes
[code]....
i am getting segmentation fault while porting project having c,c++,fortran files which has been developed in vc++ windows to linux.this project is working fine in windows but not in linux.when i runned in linux using makefile it is going through some output but again am getting segmentation fault.
View 1 Replies View RelatedI am using embedded linux (xlinux) and i need to compile the programs on desktop pc.i am currently using eclipse.can someone suggest me how I can configure, to make a project thats compile the programs for the embedded linix where I need to run them.
View 1 Replies View Related