Skip to content

Latest commit

 

History

History
353 lines (268 loc) · 13.4 KB

README.md

File metadata and controls

353 lines (268 loc) · 13.4 KB
Raw

GDB ImageWatch: The advanced Image Debugger

GDB ImageWatch is a tool for visualizing in-memory buffers during debug sessions using GDB. It works out of the box with instances of the OpenCV Mat class, but can also be customized to work with any arbitrary data structure.

Features

  • GUI interactivity:
    • Scroll to zoom, left click+drag to move the buffer around;
    • Rotate buffers 90° clockwise or counterclockwise;
    • Go-to widget that quickly takes you to any arbitrary pixel location;
  • Buffer values: Zoom in close enough to see the numerical values of the buffer.
  • Auto update: Whenever a breakpoint is hit, the buffer view is automatically updated.
  • Auto contrast: The entire range of values present in the buffer can be automatically mapped to the visualization range [0, 1], where 0 represents black and 1 represents white.
  • The contrast range can be manually adjusted, which is useful for inspecting buffers with extreme values (e.g. infinity, nan and other outliers).
  • Link views together, moving all watched buffers simultaneously when any single buffer is moved on the screen
  • Supported buffer types: uint8_t, int16_t, uint16_t, int32_t, uint32_t, float and double
  • Supported buffer channels: Up to four channels (Grayscale, two-channels, RGB and RGBA)
  • GPU accelerated
  • Supports large buffers whose dimensions exceed GL_MAX_TEXTURE_SIZE.
  • Supports data structures that map to a ROI of a larger buffer.
  • Exports buffers as png images (with auto contrast) or octave/matlab matrix files (unprocessed).
  • Auto-load buffers being visualized in the previous debug session
  • Designed to scale well for HighDPI displays

Requirements

  • An OpenGL 2.1+ compliant GPU
  • A C++11 compliant compiler (gcc-5 or later is recommended)
  • GDB 7.10+ compiled with python 3 support
  • Qt 5.6+ (required due to the HighDPI display support - download it here)
  • Python 3+ with its development packages

Installation

Dependencies

On Ubuntu, you can install most of the dependencies with the following command:

sudo apt-get install libpython3-dev python3-dev

Check GDB version

Before installing the gdb-imagewatch plugin, you need to first check if your GDB version is >= 7.10:

$ gdb --version

On Ubuntu 16.04.3, this will print GNU gdb (Ubuntu 7.11.1-0ubuntu1~16.5) 7.11.1, which is good enough.

You also need to make sure that your GDB was compiled with Python 3 support. For that, run the gdb command, and inside the gdb console, run the following command:

(gdb) python import sys; print(sys.version)

On Ubuntu 16.04.3, this will print 3.5.2 (default, Nov 23 2017, 16:37:01), which meets the requirements.

GDB 7.10+ with Python 3

If your GDB version is not supported, you need to download and install a compatible version of GDB with python3 support. Here are the commands for GDB 7.10:

$ wget http://ftp.gnu.org/gnu/gdb/gdb-7.10.tar.gz
$ tar -zxvf gdb-7.10.tar.gz
$ cd gdb-7.10
$ ./configure --with-python=python3 --disable-werror
$ make -j8

Notice that if you already have an older version of GDB, you will need to either reconfigure your environment running update-alternatives or reconfigure your IDE to use the updated version (which, by default, will be installed on /usr/local/bin/gdb).

After the installation, you can remove both the file gdb-7.10.tar.gz and the folder gdb-7.10.

Downloading GDB ImageWatch

Clone the GDB ImageWatch plugin to any folder you prefer and initialize the submodules:

$ git clone https://github.com/csantosbh/gdb-imagewatch
$ cd gdb-imagewatch
$ git submodule init
$ git submodule update

Ubuntu 16.04 manual installation with QtCreator

Ubuntu 16.04 comes with qt4, which is not compatible with GDB ImageWatch. In order to compile it, you need to install qt5 (get it here) and use its corresponding qmake during the compilation step.

If you are using QtCreator, you can change your Qt version under Tools->Options->Build & Run->Kits, by setting Qt version to any Qt version >= 5.

Build plugin and configure GDB

To build this plugin, run:

$ mkdir build && cd build
$ qmake .. BUILD_MODE=release PREFIX=/path/to/installation/folder
$ make -j4
$ make install

The installation step is optional; you can simply use the plugin from the build folder instead. If you choose to install the plugin, it will be placed under /path/to/installation/folder/bin/gdb-imagewatch/.

By default, the PREFIX variable is /usr/local.

libGL troubleshooting

Some users might experience a linking error if the libGL.so is not found by qmake, especially when using a nvidia graphics card. This issue will usually present itself with the message cannot find -lGL.

To fix that, you need to find the location for your libGL.so file. The following commands should help you find it:

$ sudo updatedb
$ locate -i libgl.so

If you have installed the proprietary drivers, you don't want to use the mesa folder. For example, running the command above could result in the following output:

/usr/lib/nvidia-384/libGL.so
/usr/lib/nvidia-384/libGL.so.1
/usr/lib/nvidia-384/libGL.so.384.90
/usr/lib/x86_64-linux-gnu/mesa/libGL.so.1
/usr/lib/x86_64-linux-gnu/mesa/libGL.so.1.2.0
/usr/lib32/nvidia-384/libGL.so
/usr/lib32/nvidia-384/libGL.so.1
/usr/lib32/nvidia-384/libGL.so.384.90

In this case, since I'm using the proprietary nvidia drivers, I'll choose the folder /usr/lib/nvidia-384.

Copy the name of the folder you found, and paste it in the file gdb-imagewatch.pro under the definition of the QMAKE_LFLAGS variable. In my case, this variable now looks like this:

QMAKE_LFLAGS += \
  # If you have an error "cannot find -lGL", uncomment the following line and
  # replace the folder by the location of your libGL.so
  -L/usr/lib/nvidia-384/ \
  -Wl,--exclude-libs,ALL

Now, just run the build steps again, including the qmake command, as described above.

Loading pugin

In order to load the GDB-ImageWatch plugin, simply edit the ~/.gdbinit file (create it if it doesn't exist) and append the following line:

source /path/to/gdb-imagewatch/gdb-imagewatch.py

This way, GDB will automatically load the GDB imagewatch plugin every time it starts.

Ubuntu 16.04 Automated Installation without QtCreator

The script configure_ubuntu_16.sh automates some of the installation steps for GDB ImageWatch on Ubuntu 16.04. From the root directory of this project, execute the following:

bash configure_ubuntu_16.sh

Note that this script will not instal Qt 5.6+ and GDB with python 3 support. If you don't have these packages, please follow the instructions above.

When you are done, follow the instructions below in the section Testing your installation.

Testing your installation

After compiling the plugin, you can test it by opening a console in the installation folder and running the following command from the root project directory:

python3 /path/to/gdb-imagewatch/gdb-imagewatch.py --test

If the installation was succesful, you should see the GDB ImageWatch window with the same sample_buffer_1 and sample_buffer_2 buffers from the image on the header of this page.

Using plugin

When GDB hits a breakpoint, the GDB ImageWatch window will be opened. You only need to type the name of the buffer to be watched in the "add symbols" input, and press <enter>.

Alternatively, you can also invoke the GDB ImageWatch window directly from GDB with the following command:

plot variable_name

Auto-contrast and manual contrast

The (min) and (max) fields on top of the buffer view can be changed to control autocontrast settings. By default, GDB ImageWatch will automatically fill these fields with the mininum and maximum values inside the entire buffer, and the channel values will be normalized from these values to the range [0, 1] inside the renderer.

Sometimes, your buffer may contain trash, uninitialized values that are either too large or too small, making the entire image look flat because of this normalization. If you know the expected range for your image, you can manually change the (min) and (max) values to focus on the range that you are interested.

Locking buffers

Sometimes you want to compare two buffers being visualized, and need to zoom in different places of these buffers. If they are large enough, this can become a very hard task, especially if you are comparing pixel values. This task is made easier by the lock buffers tool (which is toggled by the button with a chain icon).

When it is activated, all buffers are moved/zoomed simultaneously by the same amount. This means you only need to align the buffers being compared once; after activating the lock buffers mode, you can zoom in anywhere you wish in one buffer that all other buffers will be zoomed in the same location.

Quickly moving to arbitrary coordinates

If you need to quickly move to any pixel location, then the go to functionality is what you are looking for. Press Ctrl+L and two input fields corresponding to the target destination in format <x, y> will appear at the bottom right corner of the buffer screen. Type the desired location, then press enter to quickly zoom into that location.

Exporting bufers

Sometimes you may want to export your buffers to be able to process them in an external tool. In order to do that, right click the thumbnail corresponding to the buffer you wish to export on the left pane and select "export buffer".

GDB ImageWatch supports two export modes. You can save your buffer as a PNG (which may result in loss of data if your buffer type is not uint8_t) or as a binary file that can be opened with any tool.

Loading exported buffers on Octave/Matlab

Buffers exported in the Octave matrix format can be loaded with the function giw_load.m, which is available in the matlab folder. To use it, add this folder to Octave/Matlab path variable and call giw_load('/path/to/buffer.dump').

Configure your IDE to use GDB 7.10

If you're not using gdb from the command line, make sure that your IDE is correctly configured to use GDB 7.10. On QtCreator, go to Tools->Options->Build & Run->Debuggers and make sure that the configured path references a compatible GDB version.

Basic configuration

The settings file for the plugin can be located under $HOME/.config/gdbimagewatch.ini. You can change the following settings:

  • Rendering
    • maximum_framerate Determines the maximum framerate for the buffer rendering backend. Must be greater than 0.

Advanced configuration

By default, the plugin works with several data types, including OpenCV's Mat and CvMat and Eigen's Matrix.

If you use a different buffer type, you can create a python parser inside the folder resources/giwscripts/giwtypes. This is actually pretty simple and only involves implementing a class with the methods get_buffer_metadata() and is_symbol_observable().

The function get_buffer_metadata() must return a dictionary with the following fields:

  • display_name Name of the buffer as it must appear in the ImageWatch window. Can be customized to also show its typename, for instance.

  • pointer Pointer to the buffer

  • width Width of the ROI

  • height Height of the ROI

  • channels Number of color channels (Must be in the range 1 <= channels <= 3)

  • type Identifier for the type of the underlying buffer. The supported values, defined under resources/giwscripts/symbols.py, are:

    • GIW_TYPES_UINT8 = 0
    • GIW_TYPES_UINT16 = 2
    • GIW_TYPES_INT16 = 3
    • GIW_TYPES_INT32 = 4
    • GIW_TYPES_FLOAT32 = 5
    • GIW_TYPES_FLOAT64 = 6

  • row_stride Number of pixels you have to skip in order to reach the pixel right below any arbitrary pixel. In other words, this can be thought of as the width, in pixels, of the underlying containing buffer. If the ROI is the total buffer size, this is the same of the buffer width.

  • pixel_layout String describing how internal channels should be ordered for display purposes. The default value for buffers of 3 and 4 channels is 'bgra', and 'rgba' for images of 1 and 2 channels. This string must contain exactly four characters, and each one must be one of 'r', 'g', 'b' or 'a'. Repeated channels, such as 'rrgg' are also valid.

  • transpose_buffer Boolean indicating whether or not to transpose the buffer in the interface. Can be very useful if your data structure represents transposition with an internal metadata.

The function is_symbol_observable() receives a gdb symbol and a string containing the variable name, and must only return True if that symbol is of the observable type (the buffer you are dealing with).

It is possible to debug your custom inspector methods by using the python decorators @interface.debug_buffer_metadata and @interface.debug_symbol_observable in the methods get_buffer_metadata and is_symbol_observable, respectively. This will print information about all analyzed symbols in the debugger console every time a breakpoint is hit.

For more information on how to customize this file, check out this more detailed blog post.