About

The fotonik3d benchmark in spec2017 is reasonably friendly for vectorization, performance gains relative to a single FPU scalar implementation should be on the order of 12%.   Verify the benchmark vectorizes and sees a comparable performance improvement on RISC-V.

Much like we've seen with cam4, fotonik seems to be vectorizing well and shows a 64% reduction in instruction counts on the k1, but the performance with vector is terrible at a 74% regression in cycle counts.  We believe this is due to a weak vector unit, possibly combined with additional weaknesses elsewhere in the design.  

Stakeholders/Partners

Dependencies

Status

Updates

May 29, 2024 

• Update with icount/cycle data from the k1.  

Mar 14, 2024 

• We're seeing roughly a 46% reduction in dynamic instruction counts.  This is far higher than the performance improvements seen on other targets

  • x86_64 sees a 17% performance improvement from vector

  • aarch64 sees a 10% performance improvement from vector

  • Conclusion: We're doing well on this benchmark with RVV 

Dec 29, 2023 

• Project reported as a priority for 1H2024