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

Design of graph support #114

Open
leofang opened this issue Sep 19, 2024 · 2 comments
Open

Design of graph support #114

leofang opened this issue Sep 19, 2024 · 2 comments
Assignees
@vzhurba01
Labels
cuda.core Everything related to the cuda.core module feature New feature or request P0 High priority - Must do!
Milestone
cuda.core beta 3

Comments

@leofang
Copy link
Member

leofang commented Sep 19, 2024

No description provided.

@github-actions github-actions bot added the triage Needs the team's attention label Sep 19, 2024
@leofang leofang added this to the cuda.core beta 2 milestone Oct 9, 2024
@leofang leofang added P1 Medium priority - Should do feature New feature or request cuda.core Everything related to the cuda.core module and removed cuda.core beta2 triage Needs the team's attention labels Oct 9, 2024
@leofang
Copy link
Member Author

leofang commented Nov 12, 2024 •
edited
Loading

A few pointers to consider when we design this:

@leofang leofang added P0 High priority - Must do! and removed P1 Medium priority - Should do labels Nov 14, 2024
@leofang leofang removed their assignment Nov 14, 2024
@leofang
Copy link
Member Author

leofang commented Jan 23, 2025

Discussed internally. With all things considered will take a multi-phase approach to iteratively enhance the CUDA graph coverage. Below is the phase-1 design considerations:

  • Only cover stream capture (no explicit graph construction)
  • Exclude memory allocation/deallocation steps, and assume when entering the capturing context all needed memory are already allocated by the user
  • Exclude host callback operations
  • Basic coverage for conditional nodes
  • The resulting graph should be replay-able, meaning
    • user objects' lifetimes are properly managed
    • ...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@vzhurba01 vzhurba01

Labels
cuda.core Everything related to the cuda.core module feature New feature or request P0 High priority - Must do!
Projects
None yet
Milestone
cuda.core beta 3
Development

No branches or pull requests
2 participants
@leofang @vzhurba01