NVIDIA / NVIDIA/cuCascade

Create a reserve and callback method

Open
#13 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

rapidsmpf request
Dominant language
C++
Stars
59
Forks
35
Avg merge
17h 38m
Merged PRs (30d)
3

Description

We want to provide a non blocking way of being able to provide reservations at the point in time at which they might be available without having to block the reservation requesting thread.

The api would be something like

reservation_id handle = 
    memory_reservation_manager.reserve_and_callback(
    callback call_me_when_done,
   size_t how_much_memory, 
   reservation_strategy strategy
 )

callback function signatures would be something like

void callback(reservation res, reservation_id handle, status_type status){
   if(status == ok){

        //user better be sure that this is lightweight
       signal that the reservation is ready and that the user can proceed with it
   }else{
     signal failure, and apologize profusely
}
}

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

The issue names memory_reservation_manager.reserve_and_callback and gives a proposed callback signature; begin by reviewing those API concepts and the existing reservation lifecycle. No files, tests, or completion criteria are provided, so confirm the callback, status, and non-blocking semantics with maintainers before implementation.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
hpc
Issue type
Feature
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.