Intel Arc Graphics Driver Change Leads To A Big Speed-Up Under Linux

With the latest Mesa 23.2 code as of Friday there is now a rather significant performance optimization for Intel's graphics driver stack that really helps out Intel Arc Graphics DG2/Alchemist along with upcoming Meteor Lake graphics. Counter-Strike: Global Offensive, for example, was found to be 11% faster now with this single driver change and other Vulkan apps/games benefiting as well.

The change that was merged yesterday enables L3 partial write merging for compressible surfaces and other suitable cases for DG2 and moving forward with platforms like Meteor Lake. In turn this change of just a few lines of code is leading to measurable speed-ups for Intel's Arc Graphics discrete GPUs and helping to narrow the gap between the Windows and Linux graphics driver performance.

Longtime Intel open-source Linux graphics driver engineer Francisco Jerez explained in the merge request:

"This enables L3 partial write merging for a number of cases that seem to be getting accidentally disabled by the kernel, which was causing a serious performance bottleneck on DG2 and MTL platforms. The "Compressible Partial Write Merge Enable", "Coherent Partial Write Merge Enable" and "Cross-Tile Partial Write Merge Enable" bits in L3SQCREG5 were expected to be enabled by default (and confusingly, they even read off as enabled if you ran 'intel_reg read 0xb158' on an idle system), but they are getting clobbered during 3D context initialization by an i915 workaround.

Enabling L3 partial write merging of compressible surfaces in particular seems to increase rendering fillrate by over 3x in some cases (e.g. the "VulkanFillRate/FillRateGPU/resolution:1[0-3]/format:*/blend:0" fillrate-bound microbenchmarks). Significant improvements can also be reproduced in most real-world workloads we've tested so far, e.g. Counter Strike GO improves by ~11%, Shadow Of the Tomb Raider improves by ~5.5%, and AztecRuins-VK improves by ~6.5% on DG2-512 -- Thanks a lot to Caleb Callaway for these figures. No regressions have been observed so far.

Even though this patch might strike as surprisingly simple for such a large payoff, it's the result of @fjdegroo and I trying to root-cause the rendering performance gap of DG2 on Linux vs Windows on and off during the last year, and some of the OA statistics captured by Felix early this month were greatly helpful for me to connect the last few dots, so Felix deserves a big chunk of the credit for this work."

Seeing measurable wins for various Linux games is exciting and needless to say I'm going to be firing up some Arc Graphics A750/A770 Linux benchmarks this weekend to look at the benefits of this change across more games on Linux and Steam Play titles.

This change is currently in the Mesa 23.2-devel Git code, which will be out as stable around late August or September, while it remains to be seen yet if the few lines of code are deemed reasonable enough for back-porting to the current Mesa 23.1 stable series in a future point release.