Raspberry Pi Userland

This repository is deprecated and contains proprietary APIs for interfacing with VideoCore firmware. For modern Raspberry Pi development, you should migrate to standard Linux APIs.

Recommended APIs:

• V4L2 (Video for Linux 2)
• DRM/KMS (Direct Rendering Manager / Kernel Mode Setting)
• Mesa

Utility Tools: If you need tools like dtoverlay, dtmerge, vcmailbox, or vcgencmd, they have been moved to the raspberrypi/utils repository.

The OpenVG 1.1 Reference Implementation (RI) provides sources for OpenVG, VGU, and EGL. The package is organized as follows:

• bin/win32/: Contains libOpenVG.dll (the Windows implementation) and tiger.exe (the Tiger sample executable).
• lib/: Contains libOpenVG.lib (the MSVC 6 import library).
• ri/: Contains the reference implementation source code and project files.
  • src/: Implementation files (.cpp, .h). Includes backends for Windows, Mac OS X, and a null backend.
  • include/: Public headers for development.
    • EGL/egl.h
    • VG/openvg.h
    • VG/vgu.h
• samples/: Contains the tiger sample application source and MSVC project files (samples.dsw, samples.dsp).

The documentation for the RaspiCam camera applications has been moved. For complete guides, usage instructions, and application details, refer to the official Raspberry Pi documentation site.

https://www.raspberrypi.org/documentation/raspbian/applications/camera.md

To build the ARM-side libraries (such as EGL, mmal, GLESv2, vcos, openmaxil, vchiq_arm, bcm_host, WFC, and OpenVG), use the buildme tool.

Prerequisites:

• cmake installed.
• An ARM cross compiler.
• For 32-bit cross compilation, use the arm-bcm2708 toolchain.

Building for 64-bit (aarch64): While 64-bit userspace is not officially supported, some libraries may work. To attempt a 64-bit cross-compile:

1. Install gcc-aarch64-linux-gnu and g++-aarch64-linux-gnu.
2. Use the --aarch64 flag with the buildme command.

To use the imv buffer post-processing tools, compile the C source files using gcc. The imv2pgm utility requires linking with the math library (-lm).

gcc imv2pgm.c -o imv2pgm -lm

gcc imv2txt.c -o imv2txt

To build the test applications directly on a Raspberry Pi, you must first build the required libraries (ilclient and vgfont).

1. Build the libraries:
   • make -C libs/ilclient
   • make -C libs/vgfont
2. Build the specific application:
   • Navigate to the application directory (e.g., hello_world).
   • Run make.
   • Execute the resulting binary (e.g., ./hello_world.bin).

Alternatively, you can rebuild all libraries and applications simultaneously by running ./rebuild.sh.

# Build libraries
make -C libs/ilclient
make -C libs/vgfont

# Build and run an app
cd hello_world
make
./hello_world.bin

# Or rebuild everything
./rebuild.sh

To cross-compile the sample applications on a machine other than the Raspberry Pi, you must provide the path to the firmware directory and the cross-compiler via environment variables before running make.

Required environment variables:

• SDKSTAGE: The path to the firmware directory.
• CC: The path to the cross-compiler.

Use ./imv2txt to extract motion data into a text file. Each line in the resulting file contains the center position (x, y) of the macro block, the velocities (u, v), and the sum of absolute differences (sad).

Usage: ./imv2txt <frame_file> <width> <height> <output_dat>

This data can be visualized using xmgrace.