* op-program: Support agreed prestate hint in host

* op-program: Fix transition state serialization in test

* op-program: Fix prefetcher_test.go

* op-program: Prevent specifying both --l2.outputroot and --l2.agreed-prestate flags

Check config is consistent

* op-program: Return safe or local safe head from derivation

Avoids relying on the safe label which won't be updated on interop chains.

* op-program: Only stop derivation if local safe reaches the target block.

* op-program: Fix safe head trace extension

* op-program: Remove unused field.

* interop: dataflow refactor v2
Co-authored-by: Axel Kingsley <axel.kingsley@gmail.com>

* interop: many refactor-fixes, make action test pass

* op-e2e: fix lint

* op-service: RPC event-stream util, for subscription and polling flexibility

* op-node: update interop managed-mode RPC events to use RPC stream util

* op-supervisor: subscribe or poll for managed-node events

* op-service: stream fallback util

* op-node: fix interop managed L1 traversal

* op-supervisor: enable action tests to pull events

* op-e2e: fix interop action test

* lint

* fix unit test

* Fixes from E2E tests
- Reset Engine even when in Managed Mode
- Use correct wrapped subscription
- Export wrapped subscription type
- Close Controllers on exit

* Controller tests and fixes

---------
Co-authored-by: Axel Kingsley <axel.kingsley@gmail.com>

op-program: Run until all events are exhausted instead of running until the derivation is idle (#12804)

* op-e2e: Add action test for op-program trace extension behaviour.

* op-program: Run until all events are exhausted instead of running until the derivation is idle.

* update superchain-registry import

* replace all plasma with altda

* replace plasma with altda in filenames

* contracts: update snapshots

* go: update superchain import

* go: update superchain import after rebase

* go: run mod tidy

* op-batcher: fix altda env vars passed in devnet docker-compose

* go: update superchain import

* go: use main commit for superchain import

* op-node: sequencer / engine events refactor

incl sequencer events fixes

* op-node: distinguish block sealing error kinds

* op-node: review fixes, stashed tweaks

* op-node: events based sequencer chaos test

* op-node: fix missing DerivedFrom data in attributes test

* op-node: drop old wip debugging work log

* op-node: sequencer move OnEvent function

* op-node: update stale todo comment

* op-node: detect derivation block-building as sequencer, and avoid conflict

* op-node: clarify comments and rename PayloadSealTemporaryErrorEvent to PayloadSealExpiredErrorEvent to describe applicability better

* op-node: prevent temporary engine error from influencing inactive sequencer

* op-node: event-system with tracing

* op-node: fix global synchronous events queue execution

* op-node: bump rate limits on action tests

* op-node: add missing read-lock to synchronous event executor

* op-node: automatically attach emitters to derivers that need one

* Add new syncmode "force-execution-layer"

* Add tests

* Add l2.enginekind flag instead of ForceELSync

* Move l2EngineClientKind to EngineController

* Fix lint

* Rename engine.EngineClientKind to engine.Kind and added engineKind.SupportsPostFinalizationELSync()

* Refactor EL/CL tests in sync_test.go

* Refactor tests in sync_test.go

* Rename engine to ec

* Incorporate enginekind-specific flags into syncConfig, and hide engine kind from the engine controller logic

* Fix tests

---------
Co-authored-by: protolambda <proto@protolambda.com>

* op-node: fix driver step hot loop, improvement events and add utils

* op-node: fix verifier event processing liveness

* op-node: add events_rate_limited metric

* op-node: bump events rate limits

* op-program: refactor to use events derivers

* op-node: differentiate between invalid payload and invalid payload attributes

* op-node: remove engine queue

(squashed) remove debug line

* op-node: test VerifyNewL1Origin

* op-node: engine-queue removal review fixes

* op-node: refactor finality to encapsulate and simplify EngineQueue

* op-node: add lock to make concurrent use of Finalizer by plasma backend safe, and rename receiver-method var names

* op-node: reintroduce instant L2 finality check upon L1 signal, reintroduce extra check to handle contrived test

* op-node: fix plasma finalization test setup

* semgrep fix

* op-node: link TODO issue

* feat: plasma e2e

* feat: skip oversized inputs

* fix: bring back metrics

* feat: set usePlasma in e2e test params

* fix: lint

* fix: activate plasma flag in data source test

* fix: add DA contract proxy to deploy config

* more tests, fix leaky abstraction and refactor loadChallenges

* fix: cleanup type assertion

* support for l1 reorgs, proxy l1 finality signal and tests

* fix: plasma disabled

* add plasma specific e2e test run

* strongly typed commitment

* fix test

* fix sync lookback

* finalize with l1 signal events instead of derivation

* adjust pipeline errors

* fix batcher commitment encoding and invalid comm logging

* fix: adjust plasma state pruning and use bool for DA resetting flag

* fix: use l1 fetcher and check pq length

* op-challenger: Restrict the safe head in output provider BlockNumber instead of OutputRootProvider

This ensures the restriction is applied both to fetching output roots, and identifying the L2 block number the agreed output root comes from. The disputed block number remains restricted only by the claimed block number.

* go mod tidy cannon examples.

* op-e2e: Add test that proposals for future blocks are invalidated

* op-node: Add option to enable safe head history database.

Currently it just logs the information and has no actual database.

* op-node: Introduce pebble db to store safe head updates

* op-node: Reset the pipeline if safe head updates fail to be recorded

* go mod tidy

* op-node: Truncate when L1 head is reduced.

* op-node: Record accurate safe head data when restarting

* op-node: Ensure the latest safe head update is reset on a pipeline reset.

* op-node: Improve thread safety of safedb

* op-node: Add L2 block number to stored data

* op-node: Add API method to retrieve safe head at an L1 block number

* op-node: Tidy up key handling in safedb.

* op-node: Use an explicit reset event to clear entries invalidated by a pipeline reset

Add action test to confirm reorgs are correctly handled.

* op-node: Undo changes to start. We always step the safe head back at least one block.

* op-node: Simplify error message when requested record is prior to start of history

* op-node: Improve log message

* op-node: Tidy up

* op-node: Include L1 block number in response.

* op-node: Add missing AssertExpectations

* op-node: Verify key prefix

* feat: simple plasma DA storage

* feat: add verify-on-read flag

* feat: add daclient test coverage

* nit: err comments

* fix: propagate context to http requests

* fix: op-e2e missing plasma da src

* fix: golangcli-lint

* fix: docker builds

* fix: pass plasma client in any case

* fix: use rollup config da challenge address

* fix: review cleanup

* fix: export Storage field for mocking

* fix: switch log, use constructor, add comment

* Add blob preimage type

* 4844 blob fetcher

key type

* use params package constants

* Add l1 ref timestamp to commitment hint

* single preimage hint

add log

* Stub oracle

* Pass in blobs fetcher

* Encode hex in string hint route

* Panic if blob reconstruction was faulty

* @refcell review

* rebase

This moves the EngineController up to be able to use it without having
to intialize the Derivation Pipeline.

- Add new flag and sync.Config struct for engine p2p sync
- Fix EngineQueue to support engine p2p sync
- Add op-e2e test casees
- Fix related components to pass sync config
- Fix execution engine specs

Hooks up the L1 and L2 sources with the derivation pipeline so the program can run successfully in fetching mode.