This repository has been archived by the owner on Jun 6, 2024. It is now read-only.
Share code between Verifier and executor wrapper #14
Comments
JoshOrndorff
added a commit
that referenced
this issue
Nov 3, 2021
See #14 about de-duplicating this code
JoshOrndorff
added a commit
that referenced
this issue
Dec 8, 2021
* checkpoint. Gonna try switching to preruntime digest first * pallet allows preruntime digests and validates at end of block * remove incorrect comment * factor slot prediction into helper function, start sketching manual seal consensus data provider * better message for future self * preruntime digest interface * `create_digest` compiles * service compiles * whoe service checks * allow running from cli * actually install our new manual seal consensus data provider * Handle no-key scenario * Preruntime digest an no memoizing author account * remove wrong comment. Better to follow aura * prune commented inherent-related code * Hack in fake signature in manual seal. * unused import * verifier supports both digest types * duplicate code from verifier to executor wrapper See #14 about de-duplicating this code * author inherent no longer required * sign blocks in manual seal !! * Hack import queue so it works in instant seal although it is now broken in parachain mode. * prune some debugging logs * NimbusBlockImport with optional parachain rules * we can use std on the client-side * remove unnecessary generic * indentation and comment * use constant instead of hard-coded nmbs * code style: map instead of if let * assume relay chain slot beacon in manual seal for now * Actually checkvalidity and send notification when using preruntime digest * Typos from code review Co-authored-by: Amar Singh <[email protected]> Co-authored-by: girazoki <[email protected]> * Proper index Co-authored-by: Amar Singh <[email protected]> * Switch consensus worker from inherent to preruntime digest. Was it really that easy!? * map from author id to account id in FindAuthor * more toward parachain consensus worker * Update nimbus-consensus/src/import_queue.rs * inherent is a no-op * Revert "more toward parachain consensus worker" This reverts commit 807e72b. * validators don't need backward compatability * Fix incorrect merge conflict resolution. Still detect NimbusApi * Add copyright to manual seal file. Hell Yeah CI! * Typo found by @girazoki * stray space * Single author in dev spec so that `run-instant-seal --dev` works with one node * warnings * Apply suggestions from code review Co-authored-by: girazoki <[email protected]> Co-authored-by: Éloïs <[email protected]> * better weight in `on_initialize` * thoughts about the legacy author inherent * sketch kick-off inherent in pallet * fees * Update the inherent extrinsics Co-authored-by: Amar Singh <[email protected]> Co-authored-by: girazoki <[email protected]> Co-authored-by: Éloïs <[email protected]>
tgmichel
pushed a commit
that referenced
this issue
Dec 16, 2021
* checkpoint. Gonna try switching to preruntime digest first * pallet allows preruntime digests and validates at end of block * remove incorrect comment * factor slot prediction into helper function, start sketching manual seal consensus data provider * better message for future self * preruntime digest interface * `create_digest` compiles * service compiles * whoe service checks * allow running from cli * actually install our new manual seal consensus data provider * Handle no-key scenario * Preruntime digest an no memoizing author account * remove wrong comment. Better to follow aura * prune commented inherent-related code * Hack in fake signature in manual seal. * unused import * verifier supports both digest types * duplicate code from verifier to executor wrapper See #14 about de-duplicating this code * author inherent no longer required * sign blocks in manual seal !! * Hack import queue so it works in instant seal although it is now broken in parachain mode. * prune some debugging logs * NimbusBlockImport with optional parachain rules * we can use std on the client-side * remove unnecessary generic * indentation and comment * use constant instead of hard-coded nmbs * code style: map instead of if let * assume relay chain slot beacon in manual seal for now * Actually checkvalidity and send notification when using preruntime digest * Typos from code review Co-authored-by: Amar Singh <[email protected]> Co-authored-by: girazoki <[email protected]> * Proper index Co-authored-by: Amar Singh <[email protected]> * Switch consensus worker from inherent to preruntime digest. Was it really that easy!? * map from author id to account id in FindAuthor * more toward parachain consensus worker * Update nimbus-consensus/src/import_queue.rs * inherent is a no-op * Revert "more toward parachain consensus worker" This reverts commit 807e72b. * validators don't need backward compatability * Fix incorrect merge conflict resolution. Still detect NimbusApi * Add copyright to manual seal file. Hell Yeah CI! * Typo found by @girazoki * stray space * Single author in dev spec so that `run-instant-seal --dev` works with one node * warnings * Apply suggestions from code review Co-authored-by: girazoki <[email protected]> Co-authored-by: Éloïs <[email protected]> * better weight in `on_initialize` * thoughts about the legacy author inherent * sketch kick-off inherent in pallet * fees * Update the inherent extrinsics Co-authored-by: Amar Singh <[email protected]> Co-authored-by: girazoki <[email protected]> Co-authored-by: Éloïs <[email protected]> (cherry picked from commit d610b68) # Conflicts: # Cargo.lock # nimbus-consensus/Cargo.toml # nimbus-consensus/src/lib.rs # pallets/author-inherent/src/lib.rs # parachain-template/node/Cargo.toml
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
The verifier is used to check block signatures when importing blocks in a normal node. It lives in
nimbus-consensus/src/lib/import_queue.rsThe executor wrapper is used to check block signatures when validating a parachain block as a validator entirely in wasm. It lives in
pallets/author-inherent/src/exec.rs.While the context is different, the job of these two pieces of code is nearly identical. Therefore we should share code between them to the extent reasonable.
The text was updated successfully, but these errors were encountered: