Skip to content

Conversation

@glyh
Copy link
Member

@glyh glyh commented May 24, 2025

This is the only type that would be passed across the network when snark worker returns. So I'll only remove spec in this. for other result it would be needed to still have spec in them so we can send them to network pool to further process.

Indeed stable types in those results need to go, will follow up in next PRs.

@glyh glyh requested a review from a team as a code owner May 24, 2025 06:54
@glyh glyh changed the title Drop spec in partitioned result Snark Work Lib: Drop spec in partitioned result May 24, 2025
@glyh
Copy link
Member Author

glyh commented May 24, 2025

!ci-build-me

@glyh glyh added the snark-worker-optim Snark worker optimization: make workers operate on the level of individual proofs instead of whole t label May 28, 2025
, Ledger_proof.Cached.t
, Sub_zkapp_spec.t
( unit
, unit
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I wonder, is it useful to have a version of this type with Ledger_proof.Cached.t?

Maybe we could just use Ledger_proof.t? It looks like this structure is scoped to a particular context, and it doesn't feel like we'll need cached proofs in that context.

We don't need to cache subzkapp proofs, I think. They are shortlived.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Indeed. I'll do that in next PR

@georgeee
Copy link
Member

!ci-bypass-changelog

@glyh glyh merged commit 66d53f7 into compatible May 29, 2025
49 checks passed
@glyh glyh deleted the corvo/drop-spec-in-partitioned-result branch May 29, 2025 14:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

snark-worker-optim Snark worker optimization: make workers operate on the level of individual proofs instead of whole t

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants