INSTALL

ORPSoCv3 can be run directly from a downloaded tar ball release or from a
checked out repository. In these cases the main program is available as
bin/orpsoc.in. This is handy if you want to work on ORPSoC itself.

Unless you want to development on ORPSoC itself, it is better to do a system
installation. Follow these steps to do a system-wide installation

(If you install ORPSoC from from a GIT repository instead of using a released
tar ball, you first need to run "autoreconf -i")

Run "./configure && make" to build ORPSoC.
Install it with "make install" and make sure you have write permissions to the
target directory

You most certainly want the default collection of cores to get started as well.
These are currently hosted at github.com/openrisc/orpsoc-cores. Download the
orpsoc-cores package and put the directories systems and cores somewhere on
your machine.

The next step is to create a directory that will host the build directories and
caches of downloaded cores. Create a file named orpsoc.conf in this directory
and add paths to the systems and cores directories like this:

[main]
cores_root=/path/to/cores
systems_root=/path/to/systems

The default is to use a directory called orpsoc-cores for systems and cores that
 is a sibling to your orpsoc run directory. In that case your orpsoc.conf will
look like this:

[main]
cores_root   =../orpsoc-cores/cores
systems_root =../orpsoc-cores/systems

If orpsoc is in your path, you can now try to do a test simulation. Run
"orpsoc sim wb_bfm" to run through the self tests of the Wishbone BFM that is
provided by ORPSoC. This should download all the needed cores to a cache
directory, set up a build structure in the build directory and run a simulation
with Icarus Verilog