Dockerized Environment for CodeTransformer

Build scripts modified from https://github.com/AlexMarginean/dockerizedROSE

Requires access to https://hub.docker.com/r/naomilwx/dockerrosebuild/ for base image

The source for the CodeTransformer is downloaded into project/development

About Docker

Docker is a relatively recent technology that developers and sysadmins use to build, ship, and run distributed applications. Compared to virtual machines, docker is more lightweight and efficient, since it provides an additional layer of abstraction of operating-system-level virtualization. You can read more about it at https://www.docker.com/ .

Requirements

The current version support Linux and Mac OS X systems. We automatically detect your operating system, and check some of the hardware requirements. You should have at least 6GB RAM memory and 50GB available disk space on the phisic location of the docker container. By default the compilation of ROSE is done on 8 threads. Acording to the number of cores in your system you can increase / decrease this number: in ''Dockerfile'' find and replace ''-j8'' with ''-jx'', where x is the number of threads to be used by Make in building ROSE. This should be maximum the number of cores existing on your system.

Windows

The current scripts and configuration should work on any Linux or Mac OS X machine. The image by itself may be used also on Windows, but the scripts were not tested. If runned from boot2docker you may try to run ./dockerizeRetypeMacOS.sh and ./runMacOS.sh for building and running. However the GUI will not work in this case.

Install

• Run:

   sudo ./buildRoseContainer
  

This may take a while, because we have to build ROSE here. After it you will have a full instalation of ROSE compiler infrastructure. The folder exampleProject is the folder shared between the filesystem and Docker. Only the modifications in the container done in shared paths will be saved after the container is closed. The folder exampleProject is mounted at /root folder in the container. Thus, any modifications done in container, in the path /root will be saved on the disk, including configurations of eclipse.

Run

• For developing / testing retype you have to run the script ./run.sh . If you want to run just eclipse, you can run: ./run.sh eclipse . If you run: ./run.sh , then you will enter in an interactive shell. From the interactive shell you can also start eclipse, by running: eclipse . The project is already configured and tested.
• For using your own ssh key, run the following on the local file system: sudo cp /path/to/ssh/id_rsa dockerizedretype_full/.ssh/ || sudo cp /path/to/ssh/id_rsa .ssh/
• Note: for running ROSE on Mac OS X, you have to run all the scripts from the boot2docker terminal.

First time usage of eclipse:

• Once entered in eclipse you can import the project: File -> New -> Makefile Project with Existing Code and follow the wizzard; after this is finished: File -> New -> Convert to a C/C++ Autotools Project . After the project is configured right click on the project; Properties; expand Autotools; Configure Settings; expand configure; Advanced; copy and paste in the box near Additional command-line options:

  _ CPPFLAGS="-I/home/ROSE/RoseInstallTree/include/rose -I/home/ROSE/BoostInstallTree/include" LDFLAGS="-L/home/ROSE/RoseInstallTree/lib -L/home/ROSE/BoostInstallTree/lib -L/usr/lib/jdk1.8.0_25/jre/lib/amd64/server" CFLAGS="-O0 -g" CXXFLAGS="-O0 -g" Run Project -> Build All, and then you will be able to use eclipse for retype.

• After the first configuration the settings will be persistent.

GUI on Mac OS X

• On linux machines the GUI (for eclipse) should run by default. However, on Mac OS X systems there are some additional requirements for running GUI program. A way for this is:

  1. Install socat and xquartz:

     brew install socat
     brew cask install xquartz
     open -a XQuartz
     

  2. Before running ./run.sh open a different terminal window, and open socat:

     socat TCP-LISTEN:6000,reuseaddr,fork UNIX-CLIENT:\"$DISPLAY\"
     

  3. Run ./run.sh eclipse and the GUI should work

• We assume that you use the default ip address for boot2docker. If not, check the ip address in boot2docker (boot2docker conf), and replace: 192.168.59.3 with the identified IP address .

Change the Default Location of the Docker Images

If you want to change the default location of the docker images (for saving space on / ), on Linux systems, then:

• Edit with a text editor the configurations of docker:
  • sudo vim /etc/default/docker.io : for Ubuntu 14.04
  • sudo vim /etc/default/docker : for any supported system, excepting Ubuntu 14.04
• Add / replace the line containing DOCKER_OPTS with:
  • DOCKER_OPTS="--dns 8.8.8.8 --dns 8.8.4.4 --storage-driver=devicemapper --storage-opt dm.basesize=50G -g PATH_CONTAINERS"
• Restart the docker service:
  • sudo services docker.io restart : for Ubuntu 14.04
  • sudo services docker restart: for any supported system, excepting Ubuntu 14.04
• Check the size available for containers. It should be at least 50GB:
  • sudo docker run -P -e --rm ubuntu:14.04 df -h / : check that the size of the root partition is 50G. If it is not then: *

     sudo docker stop $(sudo docker ps -a -q)
             sudo docker rm $(sudo docker ps -a -q)
             sudo docker rmi $(sudo docker images -a)
             sudo services docker.io stop / sudo services docker stop           
             sudo rm -r /var/lib/docker
             sudo services docker.io start / sudo services docker start
    

Security

Docker has been used and trusted by different companies, such as Gilt Groupe Inc., Yelp, and Baidu Inc, so we believe it does not contain any trojans or malware.

We have used Docker version 1.2.0 and Klee is built on Ubuntu 14.04.

The docker daemon always runs as the root user, and since Docker version 0.5.2, the docker daemon binds to a Unix socket instead of a TCP port. By default that Unix socket is owned by the user root, and so, by default, you need to access it with sudo.

Starting in version 0.5.3, if you (or your Docker installer) create a Unix group called docker and add users to it, then the docker daemon will make the ownership of the Unix socket read/writable by the docker group when the daemon starts. The docker daemon must always run as the root user, but if you run the docker client as a user in the docker group then you don't need to add sudo to all the client commands. As of 0.9.0, you can specify that a group other than docker should own the Unix socket with the -G option. However, the docker group (or the group specified with -G) is root-equivalent.

You can verify that the script does nothing more by inspection. The secure hash of the Dockerfile and two patches can be found in the checksum file.