1. Computers
  2. Display Drivers
  3. Graphics Cards
  4. Memory
  5. Motherboards
  6. Processors
  7. Software
  8. Storage
  9. Operating Systems


Facebook RSS Twitter Twitter Google Plus


Phoronix Test Suite

OpenBenchmarking.org

Open-Source ATI Driver Achieves Major R500 3D Success

Michael Larabel

Published on 27 May 2008
Written by Michael Larabel
Page 1 of 1 - 70 Comments

While the RadeonHD developers have been busy working on Radeon HD 3200 / 780 Series support and other features for this open-source ATI R500/600+ driver, the DRM (Direct Rendering Manager) support has been lagging behind. Earlier this month Matthias Hopf was successful in getting DRM working on an RS690 GPU and he has published RadeonHD DRM code into his personal development tree, but no code has yet to reach master. Meanwhile, as the xf86-video-ati driver is using AtomBIOS, they are able to spend more time working on the 3D features and other areas and less time "banging on registers" or even waiting on register documentation to arrive. David Airlie has been working on the R500 3D support along with Alex Deucher and Corbin Simpson. The trio has been making some great headway towards open-source 3D goodness for Radeon X1000 and HD 2000/3000 GPUs. Their most recent efforts have focused around the R500 fragment program code and today they have reached a monumental milestone.

After some recent work today, Compiz is now working on R500 graphics cards using the open-source xf86-video-ati and the latest Mesa git. In addition, the Unreal Tournament 2004 demo is even running! There is now parity between the open-source R500 3D support and the level of Mesa support for the older R200/300/400 generations! This is one hell of a milestone.

We are, of course, celebrating this major open-source achievement and we are in the midst of trying out this latest code on a horde of different R500 graphics cards. We will be delivering benchmarks comparing the Radeon X1000 performance using the open-source 3D driver to the fglrx driver in the coming days with different graphics cards and we will compare its performance to the earlier R400 parts. If you have any other requests, let us know in the Phoronix Forums.

It was only in late March that the open-source community had achieved hardware-accelerated glxgears on R500 GPUs. To look back, it has been just short of nine months since we shared with you AMD's open-source strategy and the first 900 pages of documentation being opened up. On top of that, it has only been 74 days since AMD publicly released their R500 3D register reference guide.

David Airlie mentions on his blog that he's very close to merging this code to master and pushing this R500/600 3D support into Fedora 9 as an update. While on the RadeonHD side there isn't this same level of support today, both parties have vowed for greater cooperation and we will hopefully see some fruits of that in the near future with a single DRM implementation.

Another celebration for today is the DRI support being merged to master for the xf86-video-radeonhd driver. Matthias Hopf has pushed 23 patches into the xf86-video-radeonhd driver that adds a trivial memory manager, rudimentary code for R600 DRI initialization (but it's not working yet), cleaning up of the existing DRI code that has been available later, and a lot of other work. The DRI support is officially working on the RS690 IGP, but if using the latest Mesa it should be working fine on other R500 GPUs.

Cheers to the open-source ATI driver teams for these accomplishments and AMD for continuing to support these open efforts. Hopefully in the near future NVIDIA will join in. Stay tuned for more information on these major open-source developments. Next up: open-source R600 3D support...

More information on the development activities can be found in the IRC channels, which can be viewed at RadeonHD.org (also owned by Phoronix Media).

About The Author
Michael Larabel is the principal author of Phoronix.com and founded the web-site in 2004 with a focus on enriching the Linux hardware experience and being the largest web-site devoted to Linux hardware reviews, particularly for products relevant to Linux gamers and enthusiasts but also commonly reviewing servers/workstations and embedded Linux devices. Michael has written more than 10,000 articles covering the state of Linux hardware support, Linux performance, graphics hardware drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated testing software. He can be followed via and or contacted via .
Latest Linux Hardware Reviews
  1. A Walkthrough Of The New 32 System Open-Source Linux Benchmarking Test Farm
  2. Habey MITX-6771: Mini-ITX Board With Quad-Core J1900 Bay Trail
  3. OCZ Vector 150 SSD On Linux
  4. Noctua i4 CPU Cooler: Great For Cooling High-End LGA-2011v3 CPUs
Latest Linux Articles
  1. AMD Kaveri: Open-Source Radeon Gallium3D vs. Catalyst 14.12 Omega Driver
  2. 12-Way AMD Catalyst 14.12 vs. NVIDIA 346 Series Linux GPU Comparison
  3. AMD Catalyst 14.12 Omega Driver Brings Mixed Results For Linux Users
  4. 6-Way Winter 2014 Linux Distribution Comparison
Latest Linux News
  1. Linux 3.19-rc1 Kernel Released Ahead Of Schedule
  2. Civilization: Beyond Earth Linux GPU/Driver Benchmarks
  3. X.Org Server 1.16.3 Released To Fix Security Issues
  4. Linux 3.19 Merge Window Closes Ahead Of Schedule
  5. MIPS R6 Architecture Now Supported By GCC
  6. LowRISC To Feature Tagged Memory & Minion Cores
  7. Intel Skylake Audio Support For Linux 3.19
  8. After 10+ Years, NetworkManager Reaches v1.0
  9. VDPAU Updated To v0.9
  10. An Open Hardware Random Number Generator Proposed
Latest Forum Discussions
  1. FPS capped on Linux (AMD fglrx drivers)
  2. Need some hand holding with upgrading xserver
  3. Are there an app using HSA ?
  4. The New SuperTuxKart Looks Better, But Can Cause GPU/Driver Problems
  5. XLennart: A Game For Systemd Haters With Nothing Better To Do
  6. Updated and Optimized Ubuntu Free Graphics Drivers
  7. Debian init discussion in Phoenix Wright format
  8. Bench specific mount point