Materialize SceneEntityCfg ids as device tensors to avoid per-step CUDA syncs#1020
Closed
kevinzakka wants to merge 1 commit into
Closed
Materialize SceneEntityCfg ids as device tensors to avoid per-step CUDA syncs#1020kevinzakka wants to merge 1 commit into
kevinzakka wants to merge 1 commit into
Conversation
Collaborator
Author
|
cc @rdeits-bd |
kevinzakka
force-pushed
the
scene-entity-cfg-device-tensors
branch
from
a3b0a96 to
1a0a5be
Compare
Collaborator
|
Very cool. @bd-pdomanico is definitely more of an expert on this than I am. From what I've learned, the major benefit comes when we eliminate all the CUDA syncs, at which point the physics steps and MPD terms can start to interleave their execution where possible. Eliminating this class of syncs would be a necessary step in that direction. |
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Prompted by #1019, prototyping using torch tensors for
*_idsto avoid per-step CUDA syncs fromtensor[:, list]indexing.