Show Your Support: This site is primarily supported by advertisements. Ads are what have allowed this site to be maintained on a daily basis for the past 18+ years. We do our best to ensure only clean, relevant ads are shown, when any nasty ads are detected, we work to remove them ASAP. If you would like to view the site without ads while still supporting our work, please consider our ad-free Phoronix Premium.
Zink OpenGL-Over-Vulkan Still Has ~200 Patches To Land, More Performance Work
With supporting OpenGL 4 already and continuing to squeeze out more performance, Zink is on a solid footing but there still is even more work to land to further enhance this OpenGL-on-Vulkan implementation. Mike noted in the latest post that his work-in-progress "zink-wip" branch is down to around 200 patches left to clean-up and see through the review process for merging. Two hundred or so patches isn't bad at all when considering a month or so ago they were around 600 patches.
With working through the zink-wip queue for mainlining, one recent addition that was merged for Mesa 21.1 is descriptor caching support. Mike noted that this should yield a 80~100% performance improvement nearly across the board for CPU bound scenarios. As another performance win to happen on mainline, there is less than 100 patches to go before seeing threaded context support with Mesa Git.
Zink has also expanded its continuous integration (CI) work to help fend off future regressions. The Lavapipe software-based Vulkan implementation nearing Vulkan 1.1 support will also allow Zink CI testing through OpenGL 4.5.
The latest Zink status details can be found over on Mike's blog.