Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Prefix sum implementation #26

Draft
wants to merge 14 commits into
base: main
Choose a base branch
from
Draft

Prefix sum implementation #26

wants to merge 14 commits into from

Conversation

sudara
Copy link
Owner

@sudara sudara commented Nov 17, 2023 •
edited
Loading

Based off of implementation for Prefix Sum with SIMD - Algorithmica thanks to https://x.com/sasuke___420/status/1723297631953412243

  • Replace _mm_broadcast_ss with SSE2
  • Vertical pass happy
  • Try variable divisor vs. padding (also fixes edge bleed)
  • libdivide for SIMD integer division
  • NEON that’s faster than what the compiler does at the end of the pass
  • Add Benchmark target
  • ARGB, one channel at a time, using the same exact horizontal pass (but offset for the channel)
  • Try rotation vs. indexing into the rows

@sudara sudara changed the title Prefix sum Prefix sum implementation Nov 20, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@sudara