Repository navigation
Check grid narrow phase distances 8 pairs at a time - #20
Merged
Merged
Conversation
The narrow phase distance pass now transposes 8 candidate pairs into one simplex with `SIMD.Vec` lanes and runs the generic `dist_sqr` on it, for every pair type. The kernels do the same operations on every lane, so the result is bit identical to the scalar check. The candidate buffer has `GRID_SIMD_WIDTH - 1` slots of padding, and the last group is padded with copies of the last pair, so there is no scalar tail loop. The distances go to a `d2` scratch buffer in their own loop, followed by a scalar cutoff check and compaction. Doing the cutoff check in the distance loop made LLVM load each group's indexes as a vector and extract them with `vpextrq` before the position loads, about twice as slow. Add a test that `dist_sqr` on SIMD lanes is bit identical to the scalar version for the kernels the grid uses. Ring benchmark, single thread: 0.95 ms -> 0.79 ms. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The narrow phase distance pass now transposes 8 candidate pairs into one simplex with
SIMD.Veclanes and runs the genericdist_sqron it, for every pair type. The kernels do the same operations on every lane, so the result is bit identical to the scalar check.The candidate buffer has
GRID_SIMD_WIDTH - 1slots of padding, and the last group is padded with copies of the last pair, so there is no scalar tail loop. The distances go to ad2scratch buffer in their own loop, followed by a scalar cutoff check and compaction. Doing the cutoff check in the distance loop made LLVM load each group's indexes as a vector and extract them withvpextrqbefore the position loads, about twice as slow.Add a test that
dist_sqron SIMD lanes is bit identical to the scalar version for the kernels the grid uses.Ring benchmark, single thread: 0.95 ms -> 0.79 ms.