backport #393 to nitro v3.3.x - fix flatCallTracer onexit out of bound #394
Conversation
This file contains 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
diegoximenes
approved these changes
Jan 6, 2025
joshuacolvin0
added a commit
to OffchainLabs/nitro
that referenced
this pull request
Jan 6, 2025
Sneh1999
added a commit
to EspressoSystems/nitro-espresso-integration
that referenced
this pull request
Jan 10, 2025
* Get's the bold_state_provider_test.go tests passing The big problem is still the end-to-end TestChallengeProtocolBOLD test. * Get TestChallengeProtocolBOLD passing * Fix the cache Now that the virtual padding is handled in the BoLD protocol itself and not in the creation of the hash leaves being fed into the history committer, the number of hashes read from the cache doesn't need to equal the number of leaves (including virtual leaves.) * Use geth's in-tree BLS again after v1.14.0 * handle race condition and address minor comments * update geth pin * StateBuildingLogFunction shouldn't take targetHeader as an arg * minor fix * fix linter * Revert "Do not use SetFinalizer to remove entries from preimageResolvers in a hostio machine" This reverts commit 5548069. * Uses preimageResolverRefCounter instead of relying on finalizers to remove preimage resolvers from global map * Reuse same global map to keep preimage resolver reference counter * TestEntriesAreDeletedFromPreimageResolversGlobalMap * Adds machine's finalizer back * Change machine destroy order in TestEntriesAreDeletedFromPreimageResolversGlobalMap * fix MockTransfer in trace * Add a test challenging the start step in BoLD * Update the pin for the bold submodule This is the current unify-req-meta branch which fixes a bug in one-step proof calculation. * Sets contextId to zero when destroying a machine * contextId as pointer to avoid worrying about what are the valid context ids * Simplifies if * WIP: Add the boldmach wrapper This *should* be able to wrap an arbitrator machine and do the special handling for the BoLD protocol to make it look like there is one more machine state at the front of processing a machine. * Fix the two obviously errored hasStepped bits This still doesn't get the test passing, but it's bound to be closer. * Fix TestChallengeProtocolBOLDStartStepChallenge * Handle SetPreimageResolver being called multiple times for the same machine * panic in case when resolver is not found, or when ref counter is negative * renames refCounterAfter to refCount * fix variable name * Fix the execution_run tests * Attempt at fixing the virtual leaves issue The CollectProof and CollectMachineHashes functions were both susceptible to challenges where it was possible that the rival would have committed to more messages than this validator. And, in that case, it would attempt to look up a message number which was greater than the highest messge number it had verified as part of the batch in which the challenge originated. * Begin work on TestChallengeProtocolBOLDVirtualBlocks * Update geth * fix build * Update geth * Changes based on PR comments * Changes based on PR comments * Update geth * disable blob posting in the test batch poster config * Complete TestChallengeProtocolBOLDVirtualBlocks * Handle challenges in the virtually padded part of the leaaves Before this change, if there was a block challenge at a height in any block above the batchLimit, then the validator was not correctly creating inclusion proofs because it was attempting to fetch execution results for blocks which didn't really exist. Now, the code detects that situation and simply returns the hash of an arbitrator machine in the FINISHED state (since the Virtual leaf hashes) are all in that state by virtue of their being repeated copies of the end state of a block. * common_test: remove race condition * Bump google.golang.org/grpc from 1.64.0 to 1.64.1 Bumps [google.golang.org/grpc](https://github.com/grpc/grpc-go) from 1.64.0 to 1.64.1. - [Release notes](https://github.com/grpc/grpc-go/releases) - [Commits](grpc/grpc-go@v1.64.0...v1.64.1) --- updated-dependencies: - dependency-name: google.golang.org/grpc dependency-type: indirect ... Signed-off-by: dependabot[bot] <[email protected]> * Get the test to compile * Split virtual blocks test into first and near last virtual block tests * Update test to better handle virtual blocks returned by honest impl * switch geth-hook to PostingGasHook * update geth pin for posting gas hook * rename to RPCPostingGasHook * Fix the CollectProof and CollectMachineHashes calls Previously, they couldn't produce correct inclusion proofs for block-level challenges on blocks in the virtual range. That is to say, if this validaotor processed n L2Blocks when creating an assertion and a rival processed >= n+1 L2Blocks, this validator would attempt to lookup a block index for which no real block existed. Now, the code properly catches this case and the last machine state for the last real block is used for all virtual L2Blocks. * Add dangerous batch poster options configurable * stylus test: infinite loop should create out-of-gas error * fix user.wat * update user.wat's module hash * Use exec for docker entrypoint * Get the branch building again This is probably not where we ultimately want to be. Too much boilerplate is escaping from the bold system. Maybe, before introducing a dependency injection framework, I should just introduce something manual that would instatiate all the instances that the challenge manager, watcher, assertion manager, etc. need and then wires them together in the "default" way using the constructors from each package. * Add gci to the golangci-lint config This change also runs the gci linter against the repo to give us a "hard reset" on all of the package imports. Editor integrations: https://golangci-lint.run/welcome/integrations/ * Make golangci-lint 1.62.0 happy The linter learned how to find missing checks for safe typecasts and we had some spots where they weren't being checked. * Also log an error if the backlogSegment type cannot be asserted * Panic in syncmap if the key cannot be converted to the type * change default scheme used in tests to HashScheme * Also panic in the other syncmap function on failed type assertions * TestHostioWithoutEVMEquivalentCosts * Renames test to TestGasUsageOfHostiosThatDontHaveGoodEVMEquivalents * expectedInc to expectedInk * Split TestHostioWithoutEVMEquivalentCosts in multiple tests, test hostios with multiple arguments * Adds missing t.Parallel() to gas usage hostio tests * TestPayForMemoryGrowGasUsage with zero pages * HOSTIO_INK const * Fix lint issues * Fix lint issues * Fix the wiring after the giant refactoring in bold repo * update batch-poster.max-size and batch-poster.max-4844-batch-size descriptions * Move ArbOS upgrade handling to a bit later in block production * block reexecution should requite init.then-quit * broadcastclient: check if compression was negotiated when connecting to feed server * use t.Run in broadcastclient tests * Merge branch 'master' into bold-review * Update to the same go-ethereum pin as master This version of the go-ethereum project is needed for some of the tests to behave correctly. * Fix the path since it was moved down a dir * update geth pin * update geth submodule * Update the bold pin and remove some logging * Fix 2 lint issues The first is to sort the imports in mock_machine_test.go. The second is to rearrange the bold machine to avoid the downcast. * update geth pin to master * Check batchProcessed > 0 to avoid underflow * Remove maxNumberOfBlocks from bold state provider The signature changed in the bold repo, and needed to be updated in Nitro as well. * Update the bold pin to use the tip of main * Update bold pin and adjust nitro code accordingly Specifically, the type of the 2 AssertionHash fields in the AssertionCreationInfo struct is now a protocol.AssertionHash instead of `[32]byte`. This change also updates nitro's `go.mod` and `go.sum` with `go mod tidy` to pick up some of the new indrect dependencies the bold machinary pulls in. * Get the BoLD challenge tests compiling again Note, not "passing", but "compiling." * Bump nitro-testnode to bold-upgrade branch for bold testing * Fix bold staker without block validation enabled * Fix multi protocol staker legacy staker wallet initialization * Sync in the latests changes from the bold repo This includes wiring the HeaderProvider through to the challenge stack and a slightly clunky backend wrapper around the ethereum client instances. * Validator wallet refactor for bold * Address PR comments * Add documentation for the EOA struct * Make bold state validated check more sturdy * Fix bold state provider test exiting loop early * bug fix * del * add back tag * edit bold submodule * bold submod * bold submod * Make previousGlobalState a value instead of pointer There's no longer a need to be able to pass in nil. * Update bold repo pin to main * remove nil opts from validation node creation * remove nil opts from validation node creation * rem print * Test an overflow assertion This test confirms that when there are more messages in the sequencer inbox than the block challenge level height, the next assertion is an "overflow" assertion and is not required to wait to minimumAssertionPeriod blocks before posting. Part of NIT-2794 * Make the finished machine actually pass the global state No reason to keep this in a separate API. * Add BoLD ascii art for the log * Rename maxInboxCount -> maxSeqInboxCount This makes it clear that the number is the maximum sequencer inbox count, and not the maximum delayed inbox count. * Move the Info logging into the staker implmentations * chore: update AEP link * commentary from lee addressed, part 1 * address more review comments * check stake token addr * move to test only file * lee feedback on testonly machine wrapper * nolint * feedback * more feedback from lee, contexts, no caching module root * address remaining bold commentary * add back * add back block validator set module root * rm log * assertion hash topic and path defaults * builds * gci * gci * build * test * Less useBoldMachine pointers * builds * builds * Don't double wrap bold machines * Bump nitro-testnode to latest master to fix CI * Update bold submodule to fix shutdown * Pull in geth tipReceipient fix * Include tx compression level in calldata units cache * Bump geth pin to fix lint * update geth: cherry-pick tracing fix * Pull in flatcalltracer fix Pulls in OffchainLabs/go-ethereum#394 * merge * update submodule * fix ci * update go-ethereum * fix transaction streamer * update go deps * Lint * update flake.nix * Safe cast * update ci.yml * Fix challenge test * Skip tests * Skip BOLD test * skip bold test TestChallengeProtocolBOLDFirstVirtualBlock --------- Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: Pepper Lebeck-Jobe <[email protected]> Co-authored-by: Lee Bousfield <[email protected]> Co-authored-by: amsanghi <[email protected]> Co-authored-by: Tsahi Zidenberg <[email protected]> Co-authored-by: Ganesh Vanahalli <[email protected]> Co-authored-by: Aman Sanghi <[email protected]> Co-authored-by: Diego Ximenes <[email protected]> Co-authored-by: Tsahi Zidenberg <[email protected]> Co-authored-by: Joshua Colvin <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: Maciej Kulawik <[email protected]> Co-authored-by: Maciej Kulawik <[email protected]> Co-authored-by: Harry Kalodner <[email protected]> Co-authored-by: Raul Jordan <[email protected]> Co-authored-by: Harry Kalodner <[email protected]> Co-authored-by: Jeremy <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Pulled in with OffchainLabs/nitro#2866