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.

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

[Proposal] Refactor the remaining helpers out of the diplomacy module in rocket-chip. #3604

Open
lordspacehog opened this issue Mar 28, 2024 · 5 comments
Labels

Comments

@lordspacehog
Copy link

Type of issue: other enhancement

Impact: API modification

Development Phase: proposal

refactor freechips.rocketchip.diplomacy to move remaining old diplomacy helpers out to other modules or to the utils module.

@lordspacehog
Copy link
Author

@jerryrzhao @sequencer it seems like most of the remaining items in the diplomacy module dont really depend on diplomacy and are just helpers or utils. Any thoughts on moving these out to different areas so we can fully deprecate the diplomacy module in rocket-chip at some point?

@jerryz123
Copy link
Contributor

Do we need to deprecate it? I think for some things.. it might make sense to have a freechips.rocketchip.diplomacy that provides implementations of APIs from org.chipsalliance.diplomacy

Removing the alias could be done removing freechips.rocketchip.diplomacy

@lordspacehog
Copy link
Author

I think short term we'd still want to alias these objects to preserve functionality. It's mostly that the stuff that's left isn't generally diplomacy specific. it's a mix of things, like clock crossings (has a note that it should move to PRCI which seems reasonable), the "Resource" stuff, DTS builder, and some tools for things like address and id ranges that don't depend on anything in the broken out diplomacy. They're also not direct implementations of anything specifically diplomacy related.

mostly that the cleanup will help keep things a little more usefully organized.

@sequencer
Copy link
Member

DTS is harder then what I expected: at least blocked by:
chipsalliance/chisel#3939
This is a tricky issue for D/I integrations...

I personally would like to keep this refactoring slow, until we really achieve it.
I'm experimentally using it for T1, I'll take a look at it until I get thing right.

@lordspacehog
Copy link
Author

yeah, i dont think we should remove code, or really change functionality right now. I mostly just want to do a little organization to clarify what it's actually doing.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

3 participants