Skip to content

Conversation

@matschreyer
Copy link

Reworked memory access for reorderParticles*-functions to avoid uncoalesced writes. Uncoalesced writes are more expensive than uncoalesced reads on GPUs; the new access pattern uses coalesced writes. (Gather memory write instead of scatter memory write)

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.

1 participant