I use quad_sum to optimize the lighting grid and shadow filter performance.
Based on Metal Feature Set Tables, Apple Family 4 should support quad group operations like quad_sum and quad_max. However, on the iPhone X and iPhone 8, during creating pipeline states, we have the following error output: Encountered unlowered function call to air.quad_sum.f32.
It works perfectly for iPhone 11 and higher versions. Should I improve my feature-checking logic from Apple Family 4 to Apple Family 5, or do I have other options to fix this unexpected behavior?
Hello @Tianyu_Li,
quad_sum is a reduction operation. The row in the feature set tables that you reference is for permute operations.
Quad-scoped reduction operation are available in Apple7, please file an enhancement request using Feedback Assistant to request that this detail be clarified in the feature set tables.
Best regards,
Greg
