Skip to content

Conversation

@adust09
Copy link
Contributor

@adust09 adust09 commented Oct 28, 2025

optimized the process of evaluating execution tables in Grand Product's sumcheck, ensuring that the weights of the polynomial equations are calculated only once and reused for each column.

@TomWambsgans
Copy link
Collaborator

Hm good point indeed, thanks. Originally this takes 5ms on my laptop. I have added parallelism + SIMD on your PR (otherwise it was > 150ms). But even with this we are still above 10ms, i.e. slower than the current implem. My guess is that there are some inefficiencies in dot_product_ef_packed_par, but not sure.

@adust09
Copy link
Contributor Author

adust09 commented Oct 29, 2025

umm same as my laptop(Macbook), I'll check it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants