I found the only one simillar question but that's for linux and xcode only...
Is it possible to set up a Linux box with distcc to build my XCode projects?
im really interested to use OSX GCC for OPENBSD BOX. because my server is really slow, and all I need is file sharing and printing at home. I like to use ports very much, especially rebuilding all packages for slow cpu, it's really noticable difference, anyway, IMHO, is it possible?
Using distcc is not the issue here. What you need is a OS X -> OpenBSD crosscompiling toolchain, which is something that works well.
Some reading:
http://en.wikipedia.org/wiki/Cross_compiler
http://www.landley.net/ols/ols2007/tutorial.txt
http://wiki.osdev.org/GCC_Cross-Compiler
Related
I'm developing and building applications for a various amount of platforms (linux x86, x86_64, arm, aarch64, sparc64, mips, powerpc, macos x86_64, freebsd x86_64, solaris x86_64 and of course Windows) and I was using a very old linux box (2014 Ubuntu) for all this cross-compilation.
I've recently decided that it was more than time to move to a more updated build environment as many tools were obsolete and could not be updated, so I've moved everything to a Ubuntu 22.04. All worked fine but then I hit the "glibc version hell" when I tried to run that on other boxes as glibc on that buildbox is 2.35.
So I've tried to get older glibc to compile and link against these as I'd really like to avoid linking everything static. But now, all the gcc that are build with Ubuntu have been with a "--with-sysroot=/" which, AFAII means I can't do anything. The --sysroot option is ignored by gcc which uses / for sysroot, no matter what.
I've seen a few answers saying "use old box to build" and that seems really insane to me. On my Mac or Windows, I can chose minimum (old) target platform, even if I build on W11 or Monterey. And obviously, the reason WHY I move to a new buildbox is to NOT use an old one and be stuck with obsolete tools :-).
I can probably use again ct-ng and rebuild all compilers, including native ones, but that seems really an overkill. Anybody with a better solution?
Thanks!
Seems that there is really no solution for what I'm looking for. I ended up almost re-inventing the wheel while trying to manually installing glibc. It was a faster option to use ct-ng and install cross-compilers from there, not using the stock ones provides with my distro.
Hi stackoverflow community!
I am about to start developing patches for FreeBSD Ports Collection (pkgng utility) using C programming language.
The problem is that I am using Mac OS X and I am really do not want to switch to another operating system. I have installed Freebsd 9.0 on Parallels Desktop VM. Xcode 4.1 seems to be rather nice development tool for C.
Is it possible to implement development for FreeBSD via Xcode 4.1? How to set up project environment for such form of development and compile source for FreeBSD?
I am also opened for any other suggestions concerning cross-platform development using MacOS X to develop patches for FreeBSD. Which is the best way to organize all necessary stuff?
Since you want to develop patches for pkgng, I would strongly suggest that you compile and test the code on FreeBSD, because it is the only system that uses the ports and packages system that pkgng interfaces with. So unless it can cross-compile for FreeBSD, using any OS X IDE is probably not a good idea.
I'm not familiar with Parallels, but there is probably a file-sharing mechanism that you can set up between OS X and the FreeBSD running in the VM. That way you can edit your files on OS X and use them under FreeBSD.
X Code is really nice, and I would lean toward using it but then doing regular builds on a system actually running FreeBSD. If you have source for everything you're using (except standard libraries whose interfaces match), there is no reason not to work on OS X with Xcode. You can build your own libraries if need be. One thing writing cross platform does (provided you regularly build on the other platform) is make your code more portable. It's easier to avoid using platform specific "extensions".
I have done a little work on lazarus' free pascal. So when a client asked me to write an application for a mac, after the initial, "it can't be done" stage. (followed by an asp.net maybe stage) i thought about writing it using lazarus.
Question is. I have only a virtual machine running mac OSX, this means that i do not really want to develop on the mac. However, i just cannot seem to get the applications that i have written in lazarus on windows to work on the mac. I have tried the deployment using the Lazarus Wiki and the MACOS folder is empty and so when i put it on the mac it doesn't run the application.
What is the best way of doing this or am i barking up the wrong tree?
It seems you want to do cross-compiling, which is theoretically possible, but may not be practical, for the reasons mentioned by Marco above.
As an alternative, you could install XCode, FreePascal, and Lazarus on a MacOX machine. You could still do your development and some testing on Windows/Linux. When you hit a certain milestone, you can copy your source code to the Mac and compile your application to test and give to the user.
Even if it were possible to easily cross-compile, there some minor differences between platforms, so (especially if it's a GUI app), you would want to test it on an actual MacOS box before giving it to the client.
I've taken the route described by Noah - and I was incredibly surprised that after about three weeks development on Windows, it took about 10 minutes to get the application running on the Mac.
My route was to install Xcode 4.3 on an old Mac Mini running snow leopard, then install Lazarus using the fink version as described here. This took a while but was done in an evening.
Then I just copied my folder across to the Mac, opened the lpi on the Mac, compiled it. It failed so I removed a windows references, recompiled, and it was working. I was truly amazed.
What linker and assembler do you use to generate binaries? To my best knowledge the linker for recent OS X versions is not available in source.
Afaik what you want (crosscompiling to Mac) is not possible for recent versions (and I've done it for PowerPC myself in the past).
The easiest is to use the Unix "file" command on the binary to see what is generated, and make sure it reads something with "MachO" in it. Easiest is if you have a Linux install (where this command is pretty standard), but versions can be found for windows too (cygwin, mingw and 3rd party)
I am trying to compile lanshark for mac, but do not know how to compile on mac. I am guessing that it is possible to compile linux source code on mac. if not how can i get this program to run?
A mac is, under the hood, a BSD 'darwin' box.
Go and take a look at the MacPorts webpage. You'll find lots of interesting information (like where to get a compiler etc ) there.
Another place to go for an apple development environment is apple (xcode) ...
It should be possible to get that to run, but it will be a bit of work. The source may need a small amount of modification, depending on exactly how the protocol works (if they're using raw ethernet, that is done quite differently). Also, the OS X linker (ld) works completely differently to the Linux linker, and so the build system will need a bit of tinkering.
However, the compilers and build utilities are in the XCode bundle on your install disk, or at the download link in the other answer, so install that and give it a go. If you're lucky, just following the Linux instructions will build it.
I've been a Visual C++ user for quite some time now, and in the following weeks, I will be joining a (C++) Linux-only project. It should be noted that I have absolutely zero development experience on Linux.
Currently, Visual C++ 9 along with the Visual Assist X add-in make Windows-based development rather enjoyable. Having looked at various C++ IDEs on Linux left me a little disappointed with the coding experience, however.
I'd love to be able to use my current tool set (see above) for Linux development, so I have a couple of questions:
Can the VS editor convert between CR+LF <-> LF in a reliable manner?
Is it possible to set up a system where if I press "Build" in the IDE all of my work gets transferred to another box (e.g., a virtual machine) running Linux that compiles my project?
How would debugging work? Is it possible to remote-debug an application running in Linux via something like gdb? If so, would the "advanced" features of MS's debugger (e.g., tracepoints, break-if-cond, etc.) be lost?
Is this even a sane thing to think of? Does anyone use a Windows environment for Linux development? If not, I guess I should just suck it up and move to a Linux environment all together...?
Any input is appreciated, as I feel a little lost right now.
It's probable you'll eventually want to suck it up and work on Linux. microsoft, for obvious reasons, builds their tools to be very Windows-centric and Windows-specific; there are even a few ways in which Visual C++ isn't quite standard C++.
That said, though, you can work with VC++ to some extent.
There are easy tools on linux that will translate to UNIX line endings from Windows. (called something creative like 'dos2unix')
You can set up a remote file system, like with ExpanDrive for Windows, to let you keep your files there.
You can run ssh or rsh from Windows to run makes on the remote Linux box.
I've done windows/linux/freebsd cross-platform development with Windows as my primary OS... so it is possible.
Use VS as editor.
Access Linux Dev node via SSH with
putty.
Use revision control system
(subversion/git/bazaar/mercurial/cvs)
to check changes into repo & check
changes out on Linux.
Use a multi-platform build process. I
use SCons. It is python based
and works beautifully - though it is
a little slow with the default
settings.
Launch build process from shell using
putty.
Use additional shells as needed to
attach gdb & debug.
Now to answer your questions:
Yes. There is a VS setting to say use unix line endings. Many revision control systems can also be configured to correct this for you.
Maybe, but that's not how I did it.
Yes, you can remotely debug using GDB. GDB has many (all?) the features available with VS's debuger but you have to learn the GDB commands. I don't know of a way to use VS's debugger to debug processes compiled with GCC - though this is a frequent request.
I'd recommend that you dive in and try to learn the Linux way of doing things. What are the other people on your team using? If you use the same tool-set as other people, you'll be much better of when asking for help. If you're sitting on a Windows box cross-compiling using Cygwin, you're more or less on your own.
I work on a Linux team currently but use Visual Studio + Visual AssistX as my IDE. VS+VAX is the best environment for developing code in my opinion. Nothing like it exists anywhere else that I've found.
What you want to do is:
Install Cygwin openssh and configure it for shelling into your linux machine without a password.
Expose your development tree on your linux machine as a samba share and mount the share on your Windows machine.
Create a Visual Studio "makefile" project referencing the files on your linux machine.
Override the make command to ssh into your linux machine and execute the build.
Write a sed script on your linux machine that converts gcc/g++ output errors into visual studio format. I use something similar to this: sed -e 's/(.):([0-9]): (.*)/z:\path\to\code\\1(\2): \3/' \
I would really recommend you to give Linux a shot natively, not through a virtual machine. If you run it under a virtual machine you will probably end up doing 85% of your activities under Windows and this will prevent you from actually learning anything under linux.
Install linux in its own partition and work on it fulltime. Do this ASAP before the project start and get familiarized with the basics on how to use it first.
Regarding IDE's to use, find out what build system the project will use and what libraries you will be using. KDevelop is a great IDE although it might be a bit too cluttered. I really like Qt-Creator.
You might wanna give Eclipse a shot too.
As of what distribution to use, I would recommend you Archlinux if you feel like really learning something out of this and learning about linux along with learning to do development under linux. If you dont really want to learn too much about linux and go straight into developing then go with Kubuntu.
As for what Desktop Manager to use I would go with KDE.
There is no Visual Studio on Linux. Period. Try KDevelop or Code::Blocks. They should make development fun on Linux.
As for the compiler/executables you'd have to use the GNU GCC toolchain. There is no MS cl compiler on Linux either.
The debugger to use is gdb, again part of the GCC toolchain.
One more thing, try to pick a shell (like bash) and read up on it. It'll save quite some time in future.
I wrote a ftp script to move files from my pc to the linux box,
then set it up as a tool, so i can at least write the code on the pc, and have
easy access to source safe. I hit a tool button, and the files get transferred to
the linux box, where i use the utilities there to do the work
& 3.:
I use cygwin + Eclipse on Windows for developing and compiling.
With this combination I can also debug via gdbserver on the target machine (VM with x86 Linux or ARM "real" target).
Unfortunately you have to do some tiny but important settings (library paths in gdb etc...) but as soon as it works......
Sorry, no solution for debugging from within VS on a linux box...and I think there will none in the near future...
What you will find is that there are dialectic variations in the compiler and the libraries are different enough that it's unlikely to work well.
You're probably better off with a gcc based tool chain like MinGW if you don't want to switch your environment to Windows right off. MinGW uses native Win32 ports of standard open-source libraries such as getopt, and it is fairly easy to port software between Linux and MinGW32 (as long as you're not using Linux or Windows specific APIs).
However, porting software from MinGW to Visual C++ is quite a bit harder. On the few occasions that I've done this sort of thing I find that porting between Linux and MinGW is much easier than porting between MinGW and Visual C++.
As far as I'm aware I've never seen a MinGW plugin for Visual Studio, although one could theoretically write such a thing as Visual Studio will let you use third party compilers. Additionally, you will have to work with whatever build system the rest of the team uses, which is pretty much guaranteed not to be MSBuild.
So, I'd guess that it's fairly unlikely to work well. Still, as Hunt & Thomas would say getting used to a different way of doing things will make you a better programmer. While vi (or emacs if you're that way inclined), ctags and gdb work quite differently to Visual studio they're still quite a powerful development environment. With MinGW32 and Cygwin or MSYS you can use these without having to make a wholesale switch across to Linux.