[Question / Feature Request] Parallel RunAhead
Nobody has claimed this yet.
- Dominant language
- C
- Stars
- 14.1k
- Forks
- 2.2k
- Avg merge
- 7h 35m
- Merged PRs (30d)
- 51
Description
I was wondering, how hard would it be to distribute runahead workload over multiple CPU cores? I know there's a setting to run two instances of the core in runahead, which now seems to be the default but I think it is still effectively single-threaded.
For N frames of runahead, would it be possible to have N+1 instances of cores running in parallel? One would run at 2x speed, the next one at 3x, and so on, to make N frames into the future available at the actual next frame?
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by locating RetroArch's runahead implementation and reviewing how its existing two-instance mode schedules core execution. Determine whether N+1 parallel core instances can produce the requested future frames safely, then define completion as a working multi-core runahead mode with verified frame ordering and performance.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- c
- Domain
- performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100