NVIDIA / NVIDIA/stdexec

Improve on timers in `io_uring_context`

Open
#1,139 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C++
Stars
2.4k
Forks
270
Avg merge
3d 6h
Merged PRs (30d)
39

Description

The current implementation submits an operation for each timer that is active. This makes it impossible to have a large number of timers at once with io_uring_context.

This issue is a reminder that timers should be handled differently. For example, we could build an intrusive heap of timers and only submit the root timer. Or we copy how libunifex is doing it.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reading the existing timer handling in io_uring_context and trace how each active timer leads to a submitted operation. Compare the design with libunifex as suggested, then determine how to support many timers while submitting only the root timer; done means the timer strategy is implemented and its scalability behavior is verified.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp, linux
Domain
backend-api-design, operating-systems
Issue type
Refactor
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.