Skip to content

intel/nir: Lower array-deref-of-vector UBO and SSBO loads

Faith Ekstrand requested to merge gfxstrand/mesa:review/nir-lower-array-of-vec into master

This fixes a serious performance issue with DXVK:

https://github.com/doitsujin/dxvk/issues/937

This was caused by a recent change that to improve performance on RADV which back-fired on ANV and killed performance for some apps:

https://github.com/doitsujin/dxvk/commit/e5a06d3f4a103a54cd4eb51970fedee405d1d698

Throwing in this bit of lowering lets us come along and CSE those UBO loads (or copy-prop for SSBO load) and get one load where we previously would have gotten several.

VkPipeline-db results on Kaby Lake:

total instructions in shared programs: 5115361 -> 5073185 (-0.82%)
instructions in affected programs: 1754333 -> 1712157 (-2.40%)
helped: 5331
HURT: 63

total cycles in shared programs: 2544501169 -> 2481144545 (-2.49%)
cycles in affected programs: 2531058653 -> 2467702029 (-2.50%)
helped: 9202
HURT: 4323

total loops in shared programs: 3340 -> 3331 (-0.27%)
loops in affected programs: 9 -> 0
helped: 9
HURT: 0

total spills in shared programs: 3246 -> 3053 (-5.95%)
spills in affected programs: 384 -> 191 (-50.26%)
helped: 10
HURT: 5

total fills in shared programs: 4626 -> 4452 (-3.76%)
fills in affected programs: 439 -> 265 (-39.64%)
helped: 10
HURT: 5

All of the shaders with hurt spilling were in Rise of the Tomb Raider which also had shaders solidly helped in the spilling department. There were no shader-db changes. Apparently, this just isn't a pattern that happens in OpenGL.

I've CCd the series to stable because the perf drop in DXVK is significant and I think our users would like this sooner rather than later in a shipping release.

Merge request reports