Over the past year Jay has been in development as a new open-source shader compiler for Intel GPUs. Alyssa Rosenzweig has been leading its development since joining Intel in 2025. For Mesa 26.3 the Jay shader compiler is being enabled by default for modern Intel graphics hardware and is around ~10% faster than the existing BRW compiler code. With this merge over the weekend, Intel Jay is now used by default for Xe2 and Xe3 graphics platforms. Intel Battlemage discrete GPUs and then Lunar Lake and Panther Lake integrated graphics are the immediate users of the Jay compiler. There are plans to potentially support Jay back to Skylake era Gen9 integrated graphics but for now the main focus is on Xe2 graphics and newer. Alyssa Rosenzweig presented on the Intel Jay shader compiler at this week's X.Org Developer's Conference (XDC 2026) in Toronto. There ~10% is around the generalized performance gain of Jay over BRW. For Fossil compile times it's 55% faster than BRW and 66% faster than the Windows driver. Alyssa's presentation is embedded above and there is also the PDF slides.
Mesa 26.3 Enables Intel's New Jay Shader Compiler By Default, Compile Times 55% Faster Or 66% Faster Than Windows
Full Article
Original Source
Read the full article at Phoronix →KhanList aggregates and links to publicly available news content. We do not host full articles from third-party sources. Always verify important information with original sources.