* cannon: Add some unit tests for LoadELF

* cannon: Fix off-by-one boundary check

* cannon: Adapt LoadELF addr check for 64-bit

* cannon: Handle zero-length segments

* cannon: Restrict virtual address space to 48-bits for MIPS64

* op-supervisor: experimental cross-safety, with hazard detection

* tweak: Add some errors/error returns in backend/cross.

* wip: Chain index <> ID mapping.

* fix: Check parent instead of re-checking hazardBlock.

* Remove Hazard Work

* Write missing DB Bindings OpenBlock, LocallyDerivedFrom, CrossDerivedFrom

* Configurable WorkFn for Workers

* op-supervisor: move chain-index <> chain ID translation into dependency set, fix some interfaces

* op-supervisor: update cross-safety worker routine

* op-supervisor: update more error handling

* op-supervisor: move errors to types package

* op-supervisor: check CanExecuteAt and CanInitiateAt

* op-supervisor: determine cross-safe candidate and L1 scope, and more fixes

* todo L1 scope increment

* op-supervisor: cross-safe L1 scope bump

* op-supervisor: dependency set getter

* op-supervisor: L1 scope increment fix

* op-supervisor: fix cross-safe updates typing

* op-node: signal L1 traversal of derivation to supervisor

* op-supervisor: fromda fixes and tests

* op-supervisor: fix OpenBlock, fix/add missing interface methods, hook up cross-safe worker routines

* OpenBlock to return map[uint32]ExecutingMessage

* Add Frontier Unit Tests

* fix WithParent panic

* op-node: register L1 traversal with op-supervisor

* op-node,op-supervisor: add logging, temp work around for interop local-safe updates

* Add safe_start_test, unsafe_start_test

* Add safe_update_test and unsafe_update_test

* add worker_test

* op-supervisor: fix cross-safe L1 scope bumping

* op-supervisor: fix logs DB test

* Add cycle.go and initial unit tests.

* fix: Use new execMsg log index info to complete graph.

* debug: Add helper to write out Mermaid of our graph.

* tests: Add more cycle tests.

* tests: Organize cycle tests.

* tests: Add NoCycle test.

* fix: Add edges between basic logs.

* tests: More comprehensive NoCycle tests.

* tests: Make tests not use index-0 exec msgs for now.

* fix: Disallow self-referencing messages.

* fix: Safely handle unknown chains.

* tests,fix: Remove unintended self-reference.

* tests,fix: Remove unintended self-reference.

* tests: Add test for the first log being exec.

* refactor: Create ErrFailedToOpenBlock for error.

* tests,refactor: Use table-driven tests to make it easier to iterate test vectors.

* cleanup: Change some comments and make Mermaid debugging easier to disable.

* tests: Add permuations of 2-cycles involving adjacency and first logs.

* fix: When adding edges for execMsgs, remove node from inDegree0.

* refactor: Split graph build from checks; abstract adding edges.

This allows us to more easily test specifics of our cycle checking.

Errors can be in either part, and only being able to test the entirety
makes things a little difficult.

Abstracting edge creation helps eliminate bugs like 118aeb6b8.

* tests: Test execMsg to base log which is right after an execMsg to the first execMsg.

* tests: Cycle across timestamps should not return an error.

* fix: Don't consider cycles that depend on out-of-timestamp nodes.

We do this by first collecting every execMsg from all chains, filtering
out the out-of-timestamp nodes, and putting into a lookup map.

When creating edges, we check to see if the target node is
uninteresting.

* fix: Remove debug mermaid logging.

* cleanup: Remove stray text in comment.

* tweak: Make presence check more idiomatic; remove possible nil map access.

* docs: Improve comments in cycle detection code.

* cleanup: Reorganize cycle detection code.

* cleanup: Move public types to top.

* cleanup: Change some names for consistency.

* cleanup: Better comment in cycle detection.

* docs: Explain graph representation.

* tests: Re-organize HazardCycleChecks tests.

* docs: Better naming and docs in cycle.go.

* tests: Add large graph cycle detection tests.

* tests: Add nil hazard map test.

* refactor: Slightly tweak graph sort code to nest less; add comments.

* tweak: Make self referencial messages fail with an ErrConflict.

* tests: Add tests for both older/younger timestamps, and document that younger init msgs shouldn't occur.

* tests: Add tests for 2-cycle with 3 chains.

* tweak: Check that an exec msg's init msg exists.

* tests: tweak test to use index 0 instead of 1.

* cleanup: Remove unused function.

* op-e2e: fix interop mock backend action test

CI try 2

---------
Co-authored-by: protolambda <proto@protolambda.com>
Co-authored-by: axelKingsley <axel.kingsley@gmail.com>

* cannon: Check --input and --output file formats in CLI

* cannon: Allow only binary state outputs in run cli

* golang lint

We're not getting a lot of value from hosted Semgrep. This PR removes that in favor of just adding rules to the `semgrep/` directory so they can be run locally in CI.

* chore: Remove obsolete cloudbuild file

* Remove coderabbit

* remove snyk

* fix abi.encodecall semgrep for tests

* fix safe call test

* rm unneccessary abi.encodeWithSignature, fix fuzz test

* use comment for semgrep...

* use comment for semgrep...

* use pattern-not for expectRevert cases

* fixes

* fixes

* try fix ci

* fix...

Old file, not used anymore.

* op-node: disable finality based on local-safe when interop is active

* op-e2e: fix interop action-test, relies on op-supervisor finality signal now

Not sure how I managed to commit that... removes a safe.json file
that was accidentally committed a few months ago.

Turns our GoReleaser also barfs if there's anything extraneous in the git directory.

CircleCI barfs if there's anything in the project folder.

* feat: add experimental L2 source flag

* Move L2 experimental client methods to eth client

* Fix missing return

* Improve config handling for experimental features

* Add l2 experimental test

`$(command -v geth)` is valid shell syntax, but it gets processed by
make first, so that this never gets correctly executed by the shell. As
a result, `make install-geth` and `make install-eth2-testnet-genesis`
are always executed and not, as intended, just when the commands are not
present.

The fix is easy: escape the dollar by duplicating it, so that make won't
do anything and the correct command reaches the shell.

The Docker release requires special args.

Signed-off-by: jsvisa <delweng@gmail.com>

Updates CI to check that the Kontrol summary dummy files are not
modified by any PR unless the change is deliberate. Committed files
are dummy files and aren't meant to change.

Alternative here would be to generate the dummy files with a
script that gets triggered before forge builds stuff. Main concern
with using an autogen script is that the script would need to be
added in front of build commands all over the place. Committing
the files and checking that they aren't changed is slightly more
janky but people rarely re-generate these files anyway so the
projected impact is minimal.

If this becomes a devx hassle we can consider autogen.

* optimize the example cannon command

* Update cannon/README.md
Co-authored-by: smartcontracts <kelvinfichter@gmail.com>

* Update cannon/README.md
Co-authored-by: smartcontracts <kelvinfichter@gmail.com>

---------
Co-authored-by: smartcontracts <kelvinfichter@gmail.com>

* use deploy utils over new* in tests

* replace new * with use of deployUtils

* fix failing test

* use deployutils for weth98 test file

* fix semgrep

* fixes

* fixes

* fixes...

* op-deployer: Add Docker builds

* fix job syntax

* specify dockerfile

* remove accidental whitespace changes

Replaces the golang package for natspec/semver matching with a
semgrep rule. Code reduction is good.

Bumps [github.com/minio/minio-go/v7](https://github.com/minio/minio-go) from 7.0.78 to 7.0.79.
- [Release notes](https://github.com/minio/minio-go/releases)
- [Commits](https://github.com/minio/minio-go/compare/v7.0.78...v7.0.79)

---
updated-dependencies:
- dependency-name: github.com/minio/minio-go/v7
  dependency-type: direct:production
  update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* fix failing interfaces

* fix crossl2inbox, revert some changes

* fix semgrep error

Removes the FeeVaultWithdrawal.s.sol script that was previously
used to trigger withdrawals manually. All of this is automated now
and you can just use Etherscan if you really want to do things
manually so the value of the script is minimal.

* add interfaces for safe contracts, use named imports for DeputyGuardianModule

* fix failing test

* op-supervisor: cross-safe-updates PR squashed

op-supervisor: experimental cross-safety, with hazard detection

tweak: Add some errors/error returns in backend/cross.

wip: Chain index <> ID mapping.

fix: Check parent instead of re-checking hazardBlock.

Remove Hazard Work

Write missing DB Bindings OpenBlock, LocallyDerivedFrom, CrossDerivedFrom

Configurable WorkFn for Workers

op-supervisor: move chain-index <> chain ID translation into dependency set, fix some interfaces

op-supervisor: update cross-safety worker routine

op-supervisor: update more error handling

op-supervisor: move errors to types package

op-supervisor: check CanExecuteAt and CanInitiateAt

op-supervisor: determine cross-safe candidate and L1 scope, and more fixes

todo L1 scope increment

op-supervisor: cross-safe L1 scope bump

op-supervisor: dependency set getter

op-supervisor: L1 scope increment fix

op-supervisor: fix cross-safe updates typing

op-node: signal L1 traversal of derivation to supervisor

op-supervisor: fromda fixes and tests

op-supervisor: fix OpenBlock, fix/add missing interface methods, hook up cross-safe worker routines

OpenBlock to return map[uint32]ExecutingMessage

Add Frontier Unit Tests

fix WithParent panic

op-node: register L1 traversal with op-supervisor

op-node,op-supervisor: add logging, temp work around for interop local-safe updates

Add safe_start_test, unsafe_start_test

Add safe_update_test and unsafe_update_test

add worker_test

op-supervisor: fix cross-safe L1 scope bumping

op-supervisor: fix logs DB test
Co-authored-by: axelKingsley <axel.kingsley@gmail.com>
Co-authored-by: Tyler Smith <mail@tcry.pt>

* op-node: fix interop deriver test

* op-e2e: fix interop action test

* op-supervisor: improve map init

* op-node: link interop TODO comment to issue, in engine events emitter

* op-supervisor: cleanup Worker instances of tests

---------
Co-authored-by: axelKingsley <axel.kingsley@gmail.com>
Co-authored-by: Tyler Smith <mail@tcry.pt>

* update test folder so that smegrep require and revert checks pass for it

* fix test

* feat(ct): add semgrep rule to use encodeCall

encodeCall is almost always better than encodeWithSelector because
it maintains type safety. Prefer encodeCall unless
encodeWithSelector is actually necessary.

* maint(ct): update contracts to use encodeCall

Updates a number of contracts to use encodeCall instead of
encodeWithSelector where possible.

* cannon: Simplify load/stores with helper functions

* use subword utils in MIPS.sol

* lint MIPS.sol

* add natspec to MIPSInstructions.sol

* use updateSubWord in MIPSInstructions.sol

* bump MIPS contract semver

* fix nits

* cannon: Remove memory.SetUint32

Remove uint32 word stores from the `mipsevm.memory` interface. `SetUint32` is inflexible
due to its word-alignment constraints. This prevents tests for 32 and 64-bit VMs from
using the same program counter values when writing instructions to memory.

Instead, tests should use the new `testutil.StoreInstruction` utility function to write instructions to any naturally aligned memory location.

* use arch.Word csats in go-ffi

Update WaitForBlock to maintain two timeouts: a no-change timeout, which fires if the chain's head does not change within a specified window, and an absolute timeout, which fires if the chain's head does not meet or exceed the specified block.

These changes should ideally reduce the number of test flakes we're seeing. Everything takes longer when test executors are under load; by maintaining these two timeouts we can provide longer-running tests with more buffer while retaining the ability to fail fast if the chain gets stuck.

As part of this PR I also refactored the wait method to use polling rather than WebSockets. I've found WebSockets to be unreliable in tests.