This repository was archived by the owner on Dec 5, 2021. It is now read-only.
[pull] develop from ethereum-optimism:develop #573
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
pull
bot
added
merge-conflict
…19076) * chore: return migrator address on ReadImplementationAddresses output (#819) * chore: return migrator address on ReadImplementationAddresses output * chore: use fixed solidity version * chore: just pr ready * test: add zero length code test for opcm * chore: update tests on migrator (#822) * refactor: replace contract mocks with vm.mock * fix: read impl typo * fix: remove nonexistent delayedWETHPermissionedGameProxy from test The test referenced input_.delayedWETHPermissionedGameProxy which doesn't exist in the ReadImplementationAddresses.Input struct. The delayedWETH implementation is read from OPCM's implementations() return value, not from an input proxy address. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * fix: use common test * fix: handle OPCM V2 feature flag in ReadImplementationAddresses tests The test was using the `opcm` variable directly, which is only set when OPCM_V2 feature is disabled. Added `_opcm()` helper to return the correct instance based on the feature flag. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> --------- Co-authored-by: IamFlux <175354924+0xiamflux@users.noreply.github.com> Co-authored-by: niha <205694301+0xniha@users.noreply.github.com> Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
* docs: add TODO checker workflow to docs/ai/ci-ops.md Adds detailed operational instructions for resolving TODO checker CI failures in a dedicated file following the docs/ai/ pattern. The workflow includes: - CircleCI API commands to find the latest scheduled pipeline - Parsing job output to identify closed issues - GitHub API commands to find who closed the issue - Proper reopening comment format with attribution and traceability This enables natural language commands like "fix the latest TODO checker failure" to work automatically via AGENTS.md/CLAUDE.md guidance that Claude Code reads. References the new ci-ops.md file from AGENTS.md alongside other topic-specific documentation. * Simplify docs * docs: fix TODO checker workflow to work without CIRCLECI_TOKEN Updates the TODO checker workflow with two key fixes: 1. Remove CIRCLECI_TOKEN requirement - CircleCI API is publicly accessible for this repository, so no authentication needed 2. Fix "who closed the issue" detection - Use GraphQL to find the most recent person who closed the issue from the timeline, not just the closing PR author. This correctly handles cases where an issue was closed via PR, then reopened, then closed directly by a different user. Also improves Step 2 to search through recent scheduled pipelines to find one with the actual "scheduled-todo-issues" workflow, since the latest may only contain a "setup" workflow. Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> * Add claude skill for fixing todo check job. * Put detail in the skill file and simple more human readable instructions in ci-ops.md --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
* feat(op-deployer): build contracts in docker * fix(tests): comments * fix(tests): comments * fix(tests): comments * fix(tests): comments
* op-acceptance: Change to creating super cannon kona games * op-acceptance: Use the interop prestate dir for kona * Fix lint * Enable super cannon kona game type in challenger.
#19077) * fix: only allow enabling permissioned game types in opcmv2 deployment (#835) * fix: only allow permissioned game types in opcmv2 deployment * docs: add comment for permissionless games config in deployopchain * feat: add isInitialDeployment argument to std validator script * fix: undo OPCMv1 changes * fix: old test content * fix: expect validator errors for disabled games in deploy test CANNON and CANNON_KONA are intentionally disabled during initial deployment, so no implementations are registered. The validator correctly flags this as PLDG-10 and CKDG-10 errors. Update test to expect these errors. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * fix: revert on non-PERMISSIONED_CANNON game type in DeployOPChain Address review feedback: raise an error instead of silently overwriting the disputeGameType input. Only PERMISSIONED_CANNON is valid for initial deployment since no prestate exists for permissionless games. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: niha <205694301+0xniha@users.noreply.github.com> Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
…elated code (#19080) * Revert "Revert "op-service: rip out deprecated blob sidecars client and relat…" This reverts commit aaa41a0. * Remove unused GetBlobs method from MockBlobsFetcher * Add NewBlobHint and NewLegacyBlobHint methods * Refactor blob hint generation in PreimageOracle * Add Test for Blob Hints in Prefetcher This commit adds comprehensive testing for blob hints in the prefetcher, covering both legacy and current blob hint types. The test verifies that the prefetcher can correctly handle and retrieve blob data based on different hint implementations. * Refactor Blob Hint Parsing with Improved Validation Add parseBlobHint function to handle legacy and current blob hint formats, improving error handling and hint parsing for blob-related operations * Remove trailing newline in prefetcher test file
Add descriptions for the `ai/` and `handbook/` directories in the docs README. Refine the CODEOWNERS pattern for security-reviews to be more specific, and fix the trailing slash formatting for consistency. Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>
* Update RewindFinalizedHeadBackward test to expect panic * Fix rewind test finalized block reference * Use error handling instead of panic
* mise: upgrade semgrep from 1.90.0 to 1.131.0 Semgrep 1.90.0 has a transitive dependency on opentelemetry-instrumentation which imports pkg_resources from setuptools. Python 3.12 does not include setuptools by default in venvs, so when the mise cache is invalidated (by any change to mise.toml), a fresh pipx install of semgrep 1.90.0 fails with: ModuleNotFoundError: No module named 'pkg_resources' This was reported as semgrep/semgrep#11069 and fixed in later versions. Upgrading to 1.131.0 resolves the issue. Co-Authored-By: Kelvin Fichter <kelvinfichter@gmail.com> * mise: upgrade semgrep from 1.131.0 to 1.137.0 v1.137.0 is the first version that actually bumps the opentelemetry packages (PR semgrep/semgrep#11180), fixing the pkg_resources ModuleNotFoundError on Python 3.12 without setuptools. Co-Authored-By: Kelvin Fichter <kelvinfichter@gmail.com> --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Co-authored-by: Kelvin Fichter <kelvinfichter@gmail.com>
* chore(rust): merge all the rust workspaces * chore(ci): formatting, fixing dockerfile and ci * chore: more ci fixes * Update .circleci/continue/rust-ci.yml Co-authored-by: Sebastian Stammler <seb@oplabs.co> * Update .circleci/continue/rust-e2e.yml Co-authored-by: Sebastian Stammler <seb@oplabs.co> * Update .circleci/continue/rust-e2e.yml Co-authored-by: Sebastian Stammler <seb@oplabs.co> * chore: pr comments * nit: fix kona path in op-challenger * chore: fix prestate artifacts path --------- Co-authored-by: Sebastian Stammler <seb@oplabs.co>
#18866) * fix: alt da: handle no commitments case when finalized head is updated * add tests * update fix
…block (#19114) * Update L1 genesis handling for genesis L2 block Clarify that for the genesis L2 block, we always return L1 block 0 without relying on the original configuration's L1 genesis. This allows for more flexibility in dispute game scenarios involving earlier L1 blocks. * empty commit to trigger CI checks
…19136) The rust workspace unification moved op-alloy under rust/, but the semgrepignore pattern was not updated, causing semgrep to flag the mdbook theme's JavaScript try/catch blocks as Solidity violations. Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Co-authored-by: Kelvin Fichter <kelvinfichter@gmail.com>
The [alias] section in mise.toml has been deprecated in favor of [tool_alias]. This fixes the deprecation warning on mise startup.
* Merge main and acceptance-test workflows. Avoids performing duplicate work. * Switch where anchor references are defined.
* op-acceptor: Update to 3.8.3 * chore(op-acceptor): update to 3.8.3 --------- Co-authored-by: Stefano Charissis <stefano@oplabs.co>
Erroneous spacing lead to Error: bad file '.circleci/continue/rust-ci.yml': yaml: line 8: mapping values are not allowed in this context
…bled (#19138) * Add metrics recording before sending to throttling loop This means we can still track the metric when throttling is disabled. * Remove unnecessary conditional for throttling threshold if we can't send on the channel, we drop the update anyway * Add test for disabled throttling in op-batcher This commit adds a test case to verify the behavior of the batch submitter when throttling is disabled (LowerThreshold = 0). The test ensures that: - The throttling loop does not start - No throttle updates are sent to the endpoint - Unsafe data availability bytes metric is still recorded - The driver can process blocks and update metrics without running the throttling loop * Add test helper method to simulate block loading * dedupe test cases * Add comment explaining channel send behavior in sendToThrottlingLoop
Rename all pipeline parameters in CircleCI continuation configs (main.yml, rust-ci.yml, rust-e2e.yml) to use a `c-` prefix, eliminating naming overlap with the setup config (config.yml). This fixes the "Conflicting pipeline parameters" error that occurs when API-triggered pipelines (e.g. from opgitgovernance bot) pass non-default parameter values. The path-filtering mapping in config.yml bridges between the setup parameter names (used by API triggers) and the new continuation parameter names.
* op-supernode: Add block invalidation and deny list to chain container - Add DenyList using bbolt for persistent storage of invalid block hashes - Add InvalidateBlock method to ChainContainer interface - InvalidateBlock adds hash to denylist and triggers rewind if current block matches - Add IsDenied helper to check if a block is on the deny list - Add acceptance test skeleton for block replacement flow * op-supernode: Add unit tests for block invalidation - Add table-driven tests for DenyList (Add, Contains, Persistence, GetDeniedHashes) - Add tests for InvalidateBlock (rewind triggers, no-rewind cases, timestamp calculation) - Add tests for IsDenied helper - Fix OpenDenyList to create parent directories - Add InvalidateBlock/IsDenied stubs to mock ChainContainers in existing tests * Wire up block invalidation from interop activity to chain container When the interop activity detects an invalid executing message, it now calls InvalidateBlock on the chain container. This: 1. Adds the block to the chain's denylist (persisted via bbolt) 2. Checks if the chain is currently using that block 3. Triggers a rewind to the prior timestamp if so The acceptance test is renamed to TestSupernodeInteropInvalidMessageReset and temporarily skipped due to pre-existing logsDB consistency issues blocking interop progression. * Fix block invalidation: use eth.Unsafe label and improve test resilience - Use eth.Unsafe label in BlockAtTimestamp call (was using empty string) - Make acceptance test resilient to block not existing after rewind - Test now passes: detects reset when block 8 is rewound away * op-node: add SuperAuthority interface for payload denial Introduce SuperAuthority interface that allows external authority (like op-supernode) to deny payloads before they are inserted into the engine. - Define SuperAuthority interface in op-node/node/node.go with IsDenied method - Add SuperAuthority to InitializationOverrides for injection - Wire SuperAuthority through Driver to EngineController - Check IsDenied before NewPayload in payload_process.go - If denied during Holocene derivation, request deposits-only replacement - Update tests and op-program to pass nil for SuperAuthority * op-supernode: add ResetOn method to Activity interface Add ResetOn(chainID, timestamp) to Activity interface so activities can clean up cached state when a chain container rewinds. - Activity.ResetOn called by Supernode when any chain resets - Supernode.onChainReset distributes reset to all activities - Heartbeat and Superroot implement no-op ResetOn (no cached state) - Update test mocks to implement ResetOn * op-supernode: add SetResetCallback to ChainContainer Add reset callback mechanism to ChainContainer so it can notify the supernode when a chain rewinds due to block invalidation. - Add ResetCallback type and onReset field to simpleChainContainer - Add SetResetCallback to ChainContainer interface - Call onReset after successful RewindEngine in InvalidateBlock - Update test mocks to implement SetResetCallback * op-supernode: implement ResetOn in Interop activity When a chain rewinds, the Interop activity must clean up its cached state: - Rewind logsDB for the chain to before the reset timestamp - Rewind verifiedDB to remove entries at or after reset timestamp - Log error if verified results were deleted (unexpected) Also adds: - Rewind/Clear methods to LogsDB interface - RewindTo method to VerifiedDB - noopInvalidator for calling logsDB.Rewind - Update test mocks for new interface methods * op-acceptance-tests: add replacement block assertions Add assertions to verify block replacement behavior: - Replacement block hash differs from invalid block hash - Invalid transaction is NOT present in replacement block These checks confirm the deposits-only replacement mechanism works correctly after an invalid executing message is detected. * fill in missing unit tests Add comprehensive unit tests for block invalidation sub-features: - TestDenyList_ConcurrentAccess: Verify concurrent Add/Contains operations - TestInvalidateBlock: Test invalidateBlock wiring from Interop to ChainContainer - TestResetOn: Test Interop activity reset behavior on chain rewind - TestVerifiedDB_RewindTo: Test VerifiedDB rewind functionality - TestSuperAuthority_*: Test payload denial in op-node engine controller Also includes BlockInvalidation_Feature.md diary documenting all sub-features. * address review feedback: genesis block guard and docs - Add guard check rejecting InvalidateBlock for height=0 (genesis block) - Document SetResetCallback must only be called during initialization - Add test case for genesis block invalidation error * op-acceptance-tests: rename halt package to reorg, consolidate tests - Rename halt/ package to reorg/ (reflects actual behavior) - Move invalid_message_replacement_test.go into reorg package - Delete obsolete invalid_message_halt_test.go (superseded by reorg test) The halt test tested old behavior where invalid messages caused the chain to halt. With block invalidation, the chain now rewinds and replaces the invalid block with a deposits-only block. * sub-feature 6: interop test control (PauseInterop/ResumeInterop) Add test-only control methods for the interop activity to support precise timing control in acceptance tests. Specification: - PauseInterop(ts): When called, interop activity returns early when it would process the given timestamp, without making progress - ResumeInterop(): Clears the pause, allowing normal processing - Zero value indicates 'not paused' (always process all values) - Atomic read/write for concurrent safety Implementation layers: - Interop: pauseAtTimestamp atomic.Uint64 field + check in progressInterop() - Supernode: delegates to interop activity by type assertion - sysgo.SuperNode: exposes methods, delegates to supernode.Supernode - stack.InteropTestControl: interface defining the test control contract - Orchestrator: InteropTestControl(id) method to get control for a supernode - DSL Supernode: NewSupernodeWithTestControl() + wrapper methods - Preset: wires up test control in NewTwoL2SupernodeInterop() This enables acceptance tests to: sys.Supernode.PauseInterop(targetTimestamp + 1) // ... perform test setup ... sys.Supernode.ResumeInterop() * Updates from Self Review - Consolidate SuperAuthority interface to op-node/rollup/iface.go (remove duplicate definitions from node.go, engine_controller.go, resources/) - Move onReset callback invocation from InvalidateBlock to RewindEngine (fires on any engine reset, not just block invalidation) - Move helper methods (SetResetCallback, blockNumberToTimestamp, IsDenied) from invalidation.go to chain_container.go - Delete unused op-supernode/supernode/resources/super_authority.go * lint * datadir for unit tests * Add functions to DSL * PR comments: Rename Functions ; Remove non-holocene replacement logic * refactor: cleanup DSL, tests, and interop activity DSL changes: - Add TimestampForBlockNum helper to L2Network - Simplify SendInvalidExecMessage to just increment log index Test changes: - invalid_message_reorg_test: use TimestampForBlockNum, check eth.NotFound - Move SuperAuthority tests to super_authority_deny_test.go - Fix denied payload test to expect DepositsOnlyPayloadAttributesRequestEvent - Make denyBlock private in test mock Interop activity: - Extract resetLogsDB and resetVerifiedDB helper functions * Refacor Tests to Test Cases * Delete AI Diary * Final PR Comments * More tests to sub-cases
) The split preimage uploader previously chose the “large preimage proposal” path purely by preimage size. That path is keccak-only on-chain (it finalizes by writing a 0x02-prefixed key), so routing large sha256/blob/precompile preimages there could never satisfy the requested oracle key/offset. To fix this we update routing to only use the large uploader for non-local Keccak256KeyType preimages. Co-authored-by: inphi <mlaw2501@gmail.com>
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
20 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
See Commits and Changes for more details.
Created by
pull[bot]
Can you help keep this open source service alive? 💖 Please sponsor : )