Skip to content

SIMD implementations for particle to grid evaluation loops #233

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
Merged
w1th0utnam3 merged 41 commits into from
Sep 15, 2025
Merged

SIMD implementations for particle to grid evaluation loops #233

w1th0utnam3 merged 41 commits into from
Sep 15, 2025

Conversation

@w1th0utnam3
Copy link
Member

@w1th0utnam3 w1th0utnam3 commented Sep 15, 2025 •
edited
Loading

  • Implements AVX and NEON variants of cubic spline kernel
  • Implements AVX and NEON variants of particle to grid evaluation of levelset function in subdomain approach

@w1th0utnam3
Prepare for benchmark of grid loop
423a60f
@w1th0utnam3
Add benchmark
5bdb32e
@w1th0utnam3
Start implementing NEON variant of dense grid loop
445326f
@w1th0utnam3
Implement NEON kernel
3958326
@w1th0utnam3
Improve remainder processing
ece74e4
@w1th0utnam3
Small optimizations
0405f22
@w1th0utnam3
Fix limits of evaluation
58ea549
@w1th0utnam3
Add benchmark data
e954fcb
@w1th0utnam3
Update bounds
9783573
@w1th0utnam3
Reduce code duplication
788126b
@w1th0utnam3
Reformulate kernel
cdbd2a6
@w1th0utnam3
Improve performance
0963e75
@w1th0utnam3
Fix use of Neon features
8c94486
@w1th0utnam3
Fix errors with Canyon benchmarks
9b93f47
@w1th0utnam3
Add AVX implementation
398fcce
@w1th0utnam3
Use FMA
79e88a4
@w1th0utnam3
Auto dispatch for vectorization
05fb89d
@w1th0utnam3
Formatting
6b24b2e
@w1th0utnam3
Use vectorization in surface reconstruction
5c8d4ae
@w1th0utnam3
Add CLI flags to enable SIMD
d54ca42
@w1th0utnam3
Show vectorization type in timings
8d77b11
@w1th0utnam3
Refactor, enable vectorization on 32bit as well
64d8591
@w1th0utnam3
Refactoring, add some safety checks
11b61a4
@w1th0utnam3
Log detected SIMD features
f90b5fc
@w1th0utnam3
Move code for SIMD kernels
e9aaad8
@w1th0utnam3
Implement test for NEON cubic spline kernel
30ce6e6
@w1th0utnam3
Fixes on x86
0f2fbba
@w1th0utnam3
Rename CLI arg for SIMD
1f45728
@w1th0utnam3
Add AVX cubic kernel test
c17f8e6
@w1th0utnam3
Py: Add SIMD arguments
931b64a
@w1th0utnam3 w1th0utnam3 merged commit c98e4bb into main Sep 15, 2025
27 checks passed
@w1th0utnam3 w1th0utnam3 deleted the avx_grid_loop branch September 15, 2025 15:27
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.

2 participants

@w1th0utnam3