Compiling Tachyon From Source Code

In order to use Tachyon, particularly on distributed memory parallel computers, you may need to compile it from source code. Compilation of Tachyon is a fairly straightforward process. Once the Tachyon distribution has been downloaded from the web site, unpack the distribution using gunzip and tar. Once the distribution is unpacked, change directory into the 'tachyon' subdirectory, and then into the 'unix' directory. Once in the 'unix' directory, type make to see the list of configurations that are currently supported. To compile Tachyon for most platform targets, one need only run make platform-name-string. The resulting Tachyon executable and library files are contained within a subdirectory of the 'tachyon/compile' directory, named to match the compilation target.

Hopefully once you've run 'make' to build Tachyon, everything went well and you now have a binary to run. If you are building a version that uses MPI, OpenGL, or other special features, you may need to edit the Make-config file to edit the locations of libraries and header files needed for compilation. If you have trouble, for now the best way to go is to send me email, at john.stone@gmail.com As I have time, I plan to improve this document and provide more detailed instructions on compilation.

Figure 1: Example Tachyon compilation steps.
\begin{figure}\begin{center}
\lstset{numbers=none,
frame=lines,
basicstyle=\sm...
...ns:/tmp/tachyon/unix % make linux-thr
\end{lstlisting}\end{center}
\end{figure}
john.stone@gmail.com