May 23, 2013
The Raspberry Pi Foundation has taken the computing world by storm with its Raspberry Pi minicomputer. The strong educational aspirations of the Raspberry Pi are just a part of why Collabora is glad to contribute to the pages of computing history being written by this groundbreaking credit-card-sized, ARM-based board. The Raspberry Pi seeks to inspire the current generation of young learners to hardware and software literacy.
The Raspberry Pi is also suitable for many embedded applications beyond education, as many commercial implementations have already demonstrated. In particular, its graphic processing abilities offer a wealth of possibilities which are now available for everyone to take advantage of.
Over a period of four weeks, Collabora has been working with the Raspberry Pi Foundation to enhance the graphic software stack for the Raspberry Pi.
A high-definition version of the video can be viewed here.
Raspberry Pi has a high-performance, special-purpose graphic processing unit capable of handling overlays and compositing directly in the hardware, allowing the CPU to focus on other tasks. Although complete X.Org (X11) drivers exist, the limitations of the X stack are a serious performance impediment . The Raspberry Pi project was thus in need of a way to expose the true potential of the hardware through an improved software stack.
Collabora has been working with the Raspberry Pi Foundation to deliver an improved Weston backend, making full use of the Raspberry Pi VideoCore to offload as much graphics processing work as possible, reducing CPU and memory usage. Additionally, Collabora provided enhancements to further reduce the memory footprint of the system.
The overall result of Collabora's Wayland enablement initiative is a much faster and more fluid UI, consistently achieving 60 frames per second under typical load. The video case study shows the Raspberry Pi's performance whilst moving windows and displaying various UI animations—an area where X11 had proven to be unpredictably latent.
Wayland's protocol and architecture allows it to serve X11 clients, through an emulated server. Improvements made to Weston as part of this engagement with the Raspberry Pi Foundation by Collabora enabled X11 applications to run seamlessly, running faster than under the legacy X.Org server.
In addition to the various performance improvements, Collabora enhanced Weston's user interface by providing a window overview presentation mode, as well as various cosmetic enhancements, including fading animations.
Collabora also implemented a debug/benchmark mode to help evaluate the improvements made, especially those related to performance and memory consumption. Collabora performed quality assurance to validate both the software and Raspberry Pi hardware perform according to expectations.
All the work has been contributed upstream to the Wayland/Weston project. Improvements made as part of this project require no Raspberry Pi specific patches and will benefit the whole community, as the changes were confined to the compositor. Any improvements requiring changes to protocols were done in collaboration with the upstream community. Full source code has been released alongside Linux distribution packages.
"It's great to be working with Collabora to bring the hardware composition capabilities of BCM2835 to bear on both native Wayland and legacy X applications" said Eben Upton, a founder and trustee of the Raspberry Pi Foundation. "The resulting platform will form the centerpiece of the new Raspberry Pi user experience from the second half of 2013."
"It's exciting to see Collabora and the Raspberry Pi foundation work together on improving the Raspberry Pi user interface using Wayland and Weston. The result provides a great demonstration of how the Wayland architecture and the Weston implementation lets you mix rendering techniques and take advantage of special purpose 2D compositing hardware." — Kristian Høgsberg Kristensen, Wayland project lead.
Recently, we have been using Perfetto to successfully profile Apitrace traces in crosvm through VirGL renderer. We have now added perfetto…
As part of a project backed by INVEST-AI, a program managed by IVADO Labs, we have developed a multi-stage neural network-based solution…
Did you know you could run a permissively-licensed MTP implementation with minimal dependencies on an embedded device? Here's a step-by-step…
Earlier this year, the Rust compiler gained support for LLVM source-base code coverage. In this post we'll explain how to setup a CI job…
Over the past few months, I've been working on a side project to improve Meson sub-project support. The best stress test is to build projects…
The most complete automated testing and continuous integration tool for the Linux kernel continues to evolve at a rapid pace. Here's a look…