Voltron is an extensible debugger UI toolkit written in Python. It aims to improve the user experience of various debuggers (LLDB, GDB, VDB and WinDbg) by enabling the attachment of utility views that can retrieve and display data from the debugger host. By running these views in other TTYs, you can build a customised debugger user interface to suit your needs.
Voltron does not aim to be everything to everyone. It's not a wholesale replacement for your debugger's CLI. Rather, it aims to complement your existing setup and allow you to extend your CLI debugger as much or as little as you like. If you just want a view of the register contents in a window alongside your debugger, you can do that. If you want to go all out and have something that looks more like OllyDbg, you can do that too.
Built-in views are provided for:
- Registers
- Disassembly
- Stack
- Memory
- Breakpoints
- Backtrace
The author's setup looks something like this:
More screenshots are here.
voltron
supports LLDB, GDB, VDB and WinDbg/CDB (via PyKD) and runs on OS X, Linux and Windows.
WinDbg support is new, please open an issue if you have problems.
The following architectures are supported:
lldb | gdb | vdb | windbg | |
---|---|---|---|---|
x86 | ✓ | ✓ | ✓ | ✓ |
x86_64 | ✓ | ✓ | ✓ | ✓ |
arm | ✓ | ✓ | ✓ | ✗ |
arm64 | ✓ | ✗ | ✗ | ✗ |
powerpc | ✗ | ✓ | ✗ | ✗ |
Releases are on PyPI. Install with pip
:
$ pip install voltron
If you want to be bleeding edge, clone this repo and install with setup.py
:
$ python setup.py install
-
If your debugger has an init script (
.lldbinit
for LLDB or.gdbinit
for GDB) configure it to load Voltron when it starts by sourcing theentry.py
entry point script. The full path will be inside thevoltron
package. For example, on OS X it might be /Library/Python/2.7/site-packages/voltron/entry.py. If you don't add this to your init script, you'll need to execute the commands after starting your debugger.LLDB:
command script import /path/to/voltron/entry.py
GDB:
source /path/to/voltron/entry.py voltron init set disassembly-flavor intel
-
Start your debugger and initialise Voltron manually if necessary.
On recent versions of LLDB you do not need to initialise Voltron manually:
$ lldb target_binary
On older versions of LLDB you need to call
voltron init
after you load the inferior:$ lldb target_binary (lldb) voltron init
GDB:
$ gdb target_binary
VDB:
$ ./vdbbin target_binary > script /path/to/voltron/entry.py
WinDbg/CDB (requires PyKD):
> cdb -c '.load C:\path\to\pykd.pyd ; !py --global C:\path\to\voltron\entry.py' target_binary
-
In another terminal (I use iTerm panes) start one of the UI views. On LLDB and WinDbg the views will update immediately. On GDB and VDB they will not update until the inferior stops (at a breakpoint, after a step, etc):
$ voltron view register $ voltron view stack $ voltron view disassembly $ voltron view backtrace
-
Set a breakpoint and run your inferior.
(*db) b main (*db) run
-
When the debugger hits the breakpoint, the views will be updated to reflect the current state of registers, stack, memory, etc. Views are updated after each command is executed in the debugger CLI, using the debugger's "stop hook" mechanism. So each time you step, or continue and hit a breakpoint, the views will update.
See the wiki on github.
See the issue tracker on github for more information or to submit issues.
-
GDB on some distros is built with Python 3, but the system's Python is version 2. If Voltron is installed into Python 2's
site-packages
it will not work with GDB. See this page on the wiki for installation instructions. -
There is no clean way to hook GDB's exit, only the inferior's exit, so the Voltron server is started and stopped along with the inferior. This results in views showing "Connection refused" before the inferior has been started.
-
Due to a limitation in the GDB API, the views are only updated each time the debugger is stopped (e.g. by hitting a breakpoint), so view contents are not populated immediately when the view is connected, only when the first breakpoint is hit.
-
If the stack view is causing GDB to hang then it must be launched after the debugger has been launched, the inferior started, and the debugger stopped (e.g. a breakpoint hit). This is due to a GDB bug that has not yet been resolved.
On older versions of LLDB, the voltron init
command must be run manually after loading the debug target, as a target must be loaded before Voltron's hooks can be installed. Voltron will attempt to automatically register its event handler, and it will inform the user if voltron init
is required.
More information about WinDbg/CDB support here.
-
The authors primarily use Voltron with the most recent version of LLDB on OS X. We will try to test everything on as many platforms and architectures as possible before releases, but LLDB/OS X/x64 is going to be by far the most frequently-used combination. Hopefully Voltron doesn't set your pets on fire, but YMMV.
-
Intel is the only disassembly flavour currently supported for syntax highlighting.
See the LICENSE file.
If you use this and don't hate it, buy me a beer at a conference some time. This license also extends to other contributors - richo definitely deserves a few beers for his contributions.
Thanks to Azimuth Security for letting me spend time working on this.
Props to richo for all his contributions to Voltron.
fG!'s gdbinit was the original inspiration for this project.
Thanks to Willi for implementing the VDB support.
Voltron now uses Capstone for disassembly as well as the debugger hosts' internal disassembly mechanism. Capstone is a powerful, open source, multi-architecture disassembler upon which the next generation of reverse engineering and debugging tools are being built. Check it out.