Lldbvis is a debug visualizer. It is made for lldb, but can be extended in the future to also be used with gdb or other debuggers. It has only be tested on C++ programs, but should support any language that is supported by lldb.

Because lldb is only supported for Python 2, lldbvis has also be built on Python 2. Therefore you require Python 2 (can be installed here) before you can install lldbvis.

To be able to run lldbvis you are required to install lldb. There is an instruction manual on how to build the sources over here. You can also install lldb directly from the command line using:

• On Ubuntu: sudo apt-get install lldb

• On Arch (requires e.g. yaourt): yaourt -S lldb Next, you need to install all the requirements of lldbvis. This is as easy as running:

pip install -r requirements.txt

There is one requirement that can't be installed using pip though. Lldbvis uses Qt4, as PyQt5 is not supported for Python 2, hence PyQt4 needs to be installed. PyQt4 can only be downloaded from the PyQt4 download page. Note that in order to build PyQt4, you also need to install SIP from the SIP download page.

Use the package manager pip to install lldbvis in the following way:

pip install lldbvis

When using multiple Python versions, make sure you can differentiate the pip installments. Usually the following command makes use of Python 2's pip:

pip2 install lldbvis

You can now easily run lldbvis as a GUI desktop program by running the following from a shell:

lldbvis

To debug a program, simply set the 'Run Target' from the menu. This should be an executable, such as a C++ compiled program. The left window provides more information on the running threads and their frames. In the bottom, there are multiple windows/tabs. One of these is used for processing the execution, one to display console output and the last one can be used by people that are more acquainted to debugging and can be used to run lldb commands and view the output. Furthermore, the right window displays information about selected objects in the center view. This center view gives a graphical representation of the debug data as a 3d tree layout. This way, members of objects can be traversed in an easy manner.

More explanatory usage instructions can be found at the wiki page.

MIT