<!-- Thanks for sending a PR! Before submitting:
1. If this is your first PR, check out our contribution guide here
https://docs.prylabs.network/docs/contribute/contribution-guidelines
You will then need to sign our Contributor License Agreement (CLA),
which will show up as a comment from a bot in this pull request after
you open it. We cannot review code without a signed CLA.
2. Please file an associated tracking issue if this pull request is
non-trivial and requires context for our team to understand. All
features and most bug fixes should have
an associated issue with a design discussed and decided upon. Small bug
fixes and documentation improvements don't need issues.
3. New features and bug fixes must have tests. Documentation may need to
be updated. If you're unsure what to update, send the PR, and we'll
discuss
in review.
4. Note that PRs updating dependencies and new Go versions are not
accepted.
Please file an issue instead.
5. A changelog entry is required for user facing issues.
-->
**What type of PR is this?**
Feature
**What does this PR do? Why is it needed?**
This PR replaces the previous PR
https://github.com/OffchainLabs/prysm/pull/16121, which built the entire
Merkle tree and generated proofs only after the tree was complete. In
this PR, the Merkle proof is produced by collecting hashes while the
Merkle tree is being built. This approach has proven to be more
efficient than the one in
https://github.com/OffchainLabs/prysm/pull/16121.
- **ProofCollector**:
- New `ProofCollector` type in `encoding/ssz/query/proof_collector.go`:
Collects sibling hashes and leaves needed for Merkle proofs during
merkleization.
- Multiproof-ready design with `requiredSiblings`/`requiredLeaves` maps
for registering target gindices before merkleization.
- Thread-safe: read-only required maps during merkleization,
mutex-protected writes to `siblings`/`leaves`.
- `AddTarget(gindex)` registers a target leaf and computes all required
sibling gindices along the path to root.
- `toProof()` converts collected data into `fastssz.Proof` structure.
- Parallel execution in `merkleizeVectorBody` for composite elements
with worker pool pattern.
- Optimized container hashing: Generalized
`stateutil.OptimizedValidatorRoots` pattern for any SSZ container type:
- `optimizedContainerRoots`: Parallelized field root computation +
level-by-level vectorized hashing via `VectorizedSha256`.
- `hashContainerHelper`: Worker goroutine for processing container
subsets.
- `containerFieldRoots`: Computes field roots for a single container
using reflection and SszInfo metadata.
- **`Prove(gindex)` method** in `encoding/ssz/query/merkle_proof.go`:
Entry point for generating SSZ Merkle proofs for a given generalized
index.
- **Testing**
- Added `merkle_proof_test.go` and `proof_collector_test.go` to test and
benchmark this feature.
The main outcomes of the optimizations are here:
```
❯ go test ./encoding/ssz/query -run=^$ -bench='Benchmark(OptimizedContainerRoots|OptimizedValidatorRoots|ProofCollectorMerkleize)$' -benchmem
goos: darwin
goarch: arm64
pkg: github.com/OffchainLabs/prysm/v7/encoding/ssz/query
cpu: Apple M2 Pro
BenchmarkOptimizedValidatorRoots-10 3237 361029 ns/op 956858 B/op 6024 allocs/op
BenchmarkOptimizedContainerRoots-10 1138 969002 ns/op 3245223 B/op 11024 allocs/op
BenchmarkProofCollectorMerkleize-10 522 2262066 ns/op 3216000 B/op 19000 allocs/op
PASS
ok github.com/OffchainLabs/prysm/v7/encoding/ssz/query 4.619s
```
Knowing that `OptimizedValidatorRoots` implements very effective
optimizations, `OptimizedContainerRoots` mimics them.
In the benchmark we can see that `OptimizedValidatorRoots` remain as the
most performant and tit the baseline here:
- `ProofCollectorMerkleize` is **~6.3× slower**, uses **~3.4× more
memory** (B/op), and performs **~3.2× more allocations**.
- `OptimizedContainerRoots` sits in between: it’s **~2.7× slower** than
`OptimizedValidatorRoots` (and **~3.4× higher B/op**, **~1.8× more
allocations**), but it is a clear win over `ProofCollectorMerkleize` for
lists/vectors: **~2.3× faster** with **~1.7× fewer allocations** (and
essentially the same memory footprint).
The main drawback is that `OptimizedContainerRoots` can only be applied
to vector/list subtrees where we don’t need to collect any sibling/leaf
data (i.e., no proof targets within that subtree); integrating it into
the recursive merkleize(...) flow when targets are outside the subtree
is expected to land in a follow-up PR.
**Which issues(s) does this PR fix?**
Partially https://github.com/OffchainLabs/prysm/issues/15598
**Other notes for review**
In this [write-up](https://hackmd.io/@fernantho/BJbZ1xmmbg), I depict
the process to come up with this solution.
Future improvements:
- Defensive check that the gindex is not too big, depicted [here](
https://github.com/OffchainLabs/prysm/pull/16177#discussion_r2671684100).
- Integrate optimizedContainerRoots into the recursive merkleize(...)
flow when proof targets are not within the subtree (skip full traversal
for container lists).
- Add multiproofs.
- Connect `proofCollector` to SSZ-QL endpoints (direct integration of
`proofCollector` for BeaconBlock endpoint and "hybrid" approach for
BeaconState endpoint).
**Acknowledgements**
- [x] I have read
[CONTRIBUTING.md](https://github.com/prysmaticlabs/prysm/blob/develop/CONTRIBUTING.md).
- [x] I have included a uniquely named [changelog fragment
file](https://github.com/prysmaticlabs/prysm/blob/develop/CONTRIBUTING.md#maintaining-changelogmd).
- [x] I have added a description with sufficient context for reviewers
to understand this PR.
- [x] I have tested that my changes work as expected and I added a
testing plan to the PR description (if applicable).
---------
Co-authored-by: Radosław Kapka <radoslaw.kapka@gmail.com>
Co-authored-by: Jun Song <87601811+syjn99@users.noreply.github.com>
* updated path processing data types, refactored ParsePath and fixed tests
* updated generalized index accordingly, changed input parameter path type from []PathElemen to Path
* updated query.go accordingly, changed input parameter path type from []PathElemen to Path
* added descriptive changelog
* Update encoding/ssz/query/path.go
Co-authored-by: Jun Song <87601811+syjn99@users.noreply.github.com>
* Added documentation for Path struct and renamed to for clarity
* Update encoding/ssz/query/path.go
Co-authored-by: Radosław Kapka <radoslaw.kapka@gmail.com>
* updated changelog to its correct type: Changed
* updated outdated comment in generalized_index.go and removed test in generalized_index_test.go as this one belongs in path_test.go
* Added validateRawPath with strict raw-path validation only - no raw-path fixing is added. Added test suite covering
* added extra tests for wrongly formated paths
---------
Co-authored-by: Jun Song <87601811+syjn99@users.noreply.github.com>
Co-authored-by: Radosław Kapka <radoslaw.kapka@gmail.com>
Co-authored-by: Radosław Kapka <rkapka@wp.pl>
* added tests for calculating generalized indices
* added first version of GI calculation walking the specified path with no recursion. Extended test coverage for bitlist and bitvectors.
vectors need more testing
* refactored code. Detached PathElement processing, currently done at the beginning. Swap to regex to gain flexibility.
* added an updateRoot function with the GI formula. more refactoring
* added changelog
* replaced TODO tag
* udpated some comments
* simplified code - removed duplicated code in processingLengthField function
* run gazelle
* merging all input path processing into path.go
* reviewed Jun's feedback
* removed unnecessary idx pointer var + fixed error with length data type (uint64 instead of uint8)
* refactored path.go after merging path elements from generalized_indices.go
* re-computed GIs for tests as VariableTestContainer added a new field.
* added minor comment - rawPath MUST be snake case
removed extractFieldName func.
* fixed vector GI calculation - updated tests GIs
* removed updateRoot function in favor of inline code
* path input data enforced to be snake case
* added sanity checks for accessing outbound element indices - checked against vector.length/list.limit
* fixed issues triggered after merging develop
* Removed redundant comment
Co-authored-by: Jun Song <87601811+syjn99@users.noreply.github.com>
* removed unreachable condition as `strings.Split` always return a slice with length >= 1
If s does not contain sep and sep is not empty, Split returns a slice of
length 1 whose only element is s.
* added tests to cover edge cases + cleaned code (toLower is no longer needed in extractFieldName function
* added Jun's feedback + more testing
* postponed snake case conversion to do it on a per-element-basis. Added more testing focused mainly in snake case conversion
* addressed several Jun's comments.
* added sanity check to prevent length of a multi-dimensional array. added more tests with extended paths
* Update encoding/ssz/query/generalized_index.go
Co-authored-by: Radosław Kapka <radoslaw.kapka@gmail.com>
* Update encoding/ssz/query/generalized_index.go
Co-authored-by: Radosław Kapka <radoslaw.kapka@gmail.com>
* Update encoding/ssz/query/generalized_index.go
Co-authored-by: Radosław Kapka <radoslaw.kapka@gmail.com>
* placed constant bitsPerChunk in the right place. Exported BitsPerChunk and BytesPerChunk and updated code that use them
* added helpers for computing GI of each data type
* changed %q in favor of %s
* Update encoding/ssz/query/path.go
Co-authored-by: Jun Song <87601811+syjn99@users.noreply.github.com>
* removed the least restrictive condition isBasicType
* replaced length of containerInfo.order for containerInfo.fields for clarity
* removed outdated comment
* removed toSnakeCase conversion.
* moved isBasicType func to its natural place, SSZType
* cosmetic refactor
- renamed itemLengthFromInfo to itemLength (same name is in spec).
- arranged all SSZ helpers.
* cleaned tests
* renamed "root" to "index"
* removed unnecessary check for negative integers. Replaced %q for %s.
* refactored regex variables and prevented re-assignation
* added length regex explanation
* added more testing for stressing regex for path processing
* renamed currentIndex to parentIndex for clarity and documented the returns from calculate<Type>GeneralizedIndex functions
* Update encoding/ssz/query/generalized_index.go
Co-authored-by: Radosław Kapka <radoslaw.kapka@gmail.com>
* run gazelle
* fixed never asserted error. Updated error message
---------
Co-authored-by: Jun Song <87601811+syjn99@users.noreply.github.com>
Co-authored-by: Radosław Kapka <radoslaw.kapka@gmail.com>
Co-authored-by: Radosław Kapka <rkapka@wp.pl>
* Move ssz_query objects into testing folder (ensuring test objects only used in test environment)
* Add containers for response
* Export sszInfo
* Add QueryBeaconState/Block
* Add comments and few refactor
* Fix merge conflict issues
* Return 500 when calculate offset fails
* Add test for QueryBeaconState
* Add test for QueryBeaconBlock
* Changelog :)
* Rename `QuerySSZRequest` to `SSZQueryRequest`
* Fix middleware hooks for RPC to accept JSON from client and return SSZ
* Convert to `SSZObject` directly from proto
* Move marshalling/calculating hash tree root part after `CalculateOffsetAndLength`
* Make nogo happy
* Add informing comment for using proto unsafe conversion
---------
Co-authored-by: Radosław Kapka <rkapka@wp.pl>
* Add basic parsing feature for accessing by index
* Add more tests for 2d byte vector
* Add List case for access indexing
* Handle 2D bytes List example
* Fix misleading cases for CalculateOffsetAndLength
* Use elementSizes[index] if it is the last path element
* Add variable_container_list field for mocking attester_slashings in BeaconBlockBody
* Remove redundant protobuf message
* Better documentation
* Changelog
* Fix `expectedSize` of `VariableTestContainer`: as we added `variable_container_list` here
* Apply reviews from Radek
* Add SizeSSZ as a member of SSZObject
* Temporarily rename dereferencePointer function
* Fix analyzeType: use reflect.Value for analyzing
* Fix PopulateVariableLengthInfo: change function signature & reset pointer
* Remove Container arm for Size function as it'll be handled in the previous branch
* Remove OffsetBytes function in listInfo
* Refactor and document codes
* Remove misleading "fixedSize" concept & Add Uint8...64 SSZTypes
* Add size testing
* Move TestSSZObject_Batch and rename it as TestHashTreeRoot
* Changelog :)
* Rename endOffset to fixedOffset
---------
Co-authored-by: Radosław Kapka <rkapka@wp.pl>
* stored CL object to enable the usage Fastssz's HashTreeRoot(). added basic test
* refactorization - using interfaces instead of storing original object
* added tests covering ssz custom types
* renamed hash_tree_root to ssz_interface as it contains MarshalSSZ and UnmarshalSSZ functions
* run gazelle
* renamed test and improved comments
* refactored test and extend to marshalSSZ and UnmarshalSSZ
* added changelog
* updated comment
* Changed SSZIface name to SSZObject. Removed MarshalSSZ and UnmarshalSSZ function signatures from interface as they are not used still. Refactored tests.
* renamed file ssz_interface.go to ssz_object.go. merge test from ssz_interface_test.go into query_test.go.
reordered source SSZObject field from sszInfo struct
* sticked SSZObject interface to HashTreeRoot() function, the only one needed so far
* run gazelle :)
---------
Co-authored-by: Radosław Kapka <rkapka@wp.pl>
* Add bitvector field for FixedTestContainer
* Handle Bitvector type using isBitfield flag
* Add Bitvector case for Stringify
* Add bitlist field for VariableTestContainer
* Add bitlistInfo
* Changelog
* Add bitvectorInfo
* Remove analyzeBit* functions and just inline them
* Fix misleading comments
* Add comments for bitlistInfo's Size
* Apply reviews from Radek
* Add vectorInfo
* Add 2D bytes field for test
* Add tag_parser for parsing SSZ tags
* Integrate tag parser with analyzer
* Add ByteList test case
* Changelog
* Better printing feature with Stringer implementation
* Return error for non-determined case without printing other values
* Update tag_parser.go: handle Vector and List mutually exclusive (inspired by OffchainLabs/fastssz)
* Make linter happy
---------
Co-authored-by: Radosław Kapka <rkapka@wp.pl>
* Add VariableTestContainer in ssz_query.proto
* Add listInfo
* Use errors.New for making an error with a static string literal
* Add listInfo field when analyzing the List type
* Persist the field order in the container
* Add actualOffset and goFieldName at fieldInfo
* Add PopulateFromValue function & update test runner
* Handle slice of ssz object for marshalling
* Add CalculateOffsetAndLength test
* Add comments for better doc
* Changelog :)
* Apply reviews from Radek
* Remove actualOffset and update offset field instead
* Add Nested container of variable-sized for testing nested path
* Fix offset adding logics: for variable-sized field, always add 4 instead of its fixed size
* Fix multiple import issue
---------
Co-authored-by: Radosław Kapka <rkapka@wp.pl>
* Add basic PathElement
* Add ssz_type.go
* Add basic sszInfo
* Add containerInfo
* Add basic analyzer without analyzing list/vector
* Add analyzer for homogeneous collection types
* Add offset/length calculator
* Add testutil package in encoding/ssz/query
* Add first round trip test for IndexedAttestationElectra
* Go mod tidy
* Add Print function for debugging purpose
* Add changelog
* Add testonly flag for testutil package & Nit for nogo
* Apply reviews from Radek
* Replace fastssz with prysmaticlabs one
* Add proto/ssz_query package for testing purpose
* Update encoding/ssz/query tests to decouple with beacon types
* Use require.* instead of assert.*
* Fix import name for proto ssz_query package
* Remove uint8/uint16 and some byte arrays in FixedTestContainer
* Add newline for files
* Fix comment about byte array in ssz_query.proto
---------
Co-authored-by: Radosław Kapka <rkapka@wp.pl>
* refactor: use auto-generated HashTreeRoot functions in htrutil.go
* refactor: use type alias for Transaction & use SliceRoot for TransactionsRoot
* changelog
* fix: TransactionsRoot receives raw 2d bytes as an argument
* fix: handle nil argument
* test: add nil test for fork and checkpoint
---------
Co-authored-by: james-prysm <90280386+james-prysm@users.noreply.github.com>
* initialize genesis data asap at node start
* add genesis validation tests with embedded state verification
* Add test for hardcoded mainnet genesis validator root and time from init() function
* Add test for UnmarshalState in encoding/ssz/detect/configfork.go
* Add tests for genesis.Initialize
* Move genesis/embedded to genesis/internal/embedded
* Gazelle / BUILD fix
* James feedback
* Fix lint
* Revert lock
---------
Co-authored-by: Kasey <kasey@users.noreply.github.com>
Co-authored-by: terence tsao <terence@prysmaticlabs.com>
Co-authored-by: Preston Van Loon <preston@pvl.dev>
* Add the new Fulu state with the new field
* fix the hasher for the fulu state
* Fix ToProto() and ToProtoUnsafe()
* Add the fields as shared
* Add epoch transition code
* short circuit the proposer cache to use the state
* Marshal the state JSON
* update spectests to 1.6.0-alpha.1
* Remove deneb and electra entries from blob schedule
This was cherry picked from PR #15364
and edited to remove the minimal cases
* Fix minimal tests
* Increase deadling for processing blocks in spectests
* Preston's review
* review
---------
Co-authored-by: terence tsao <terence@prysmaticlabs.com>
* Migrate Prysm repo to Offchain Labs organization ahead of Pectra upgrade v6
* Replace prysmaticlabs with OffchainLabs on general markdowns
* Update mock
* Gazelle and add mock.go to excluded generated mock file
* Implement static analysis to prevent panics
* Add nopanic to nogo
* Fix violations and add exclusions
Fix violations and add exclusions for all
* Changelog fragment
* Use pass.Report instead of pass.Reportf
* Remove strings.ToLower for checking init method name
* Add exclusion for herumi init
* Move api/client/beacon template function to init and its own file
* Fix nopanic testcase
* Add feature flag to start from any beacon block in db
The new feature flag called --sync-from takes a string that can take
values:
- `head` or
- a 0x-prefixed hex encoded beacon block root.
The beacon block root or the head block root has to be known in db and
has to be a descendant of the current justified checkpoint.
* Fix Bugs In Sync From Head (#15006)
* Fix Bugs
* Remove log
* missing save
* add tests
* Kasey review #1
* Kasey's review #2
* Kasey's review #3
---------
Co-authored-by: Nishant Das <nishdas93@gmail.com>
* change field IDs in `AggregateAttestationAndProofElectra`
* fix typo in `validator.proto`
* correct slashing indices length and shashings length
* check length in indexed attestation's `ToConsensus` method
* use `fieldparams.BLSSignatureLength`
* Add length checks for execution request
* fix typo in `beacon_state.proto`
* fix typo in `ssz_proto_library.bzl`
* fix error messages about incorrect types in block factory
* add Electra case to `BeaconBlockContainerToSignedBeaconBlock`
* move PeerDAS config items to PeerDAS section
* remove redundant params
* rename `PendingDepositLimit` to `PendingDepositsLimit`
* improve requests unmarshaling errors
* rename `get_validator_max_effective_balance` to `get_max_effective_balance`
* fix typo in `consolidations.go`
* rename `index` to `validator_index` in `PendingPartialWithdrawal`
* rename `randomByte` to `randomBytes` in `validators.go`
* fix for version in a comment in `validator.go`
* changelog <3
* Revert "rename `index` to `validator_index` in `PendingPartialWithdrawal`"
This reverts commit 87e4da0ea2.
* Prepare for future fork boilerplate.
* Implement the Fulu fork boilerplate.
* `Upgraded state to <fork> log`: Move from debug to info.
Rationale:
This log is the only one notifying the user a new fork happened.
A new fork is always a little bit stressful for a node operator.
Having at least one log indicating the client switched fork is something useful.
* Update testing/util/helpers.go
Co-authored-by: Radosław Kapka <rkapka@wp.pl>
* Fix Radek's comment.
* Fix Radek's comment.
* Update beacon-chain/state/state-native/state_trie.go
Co-authored-by: Radosław Kapka <rkapka@wp.pl>
* Update beacon-chain/state/state-native/state_trie.go
Co-authored-by: Radosław Kapka <rkapka@wp.pl>
* Fix Radek's comment.
* Fix Radek's comment.
* Fix Radek's comment.
* Remove Electra struct type aliasing.
---------
Co-authored-by: Radosław Kapka <rkapka@wp.pl>
* Cast to String Without Allocating
* Make it its own method
* Changelog
* Gosec
* Add benchmark, fuzz test, and @kasey's implementation.
* Gosec
* Fix benchmark test names
* Kasey's Suggestion
* Radek's Suggestion
---------
Co-authored-by: Preston Van Loon <preston@pvl.dev>
* renaming functions and fields based on consensus changes
* execution api rename
* fixing test
* reverting spectests changes, it should be changed with new version
* reverting temporarily
* revert exclusions
* Electra: Beacon State
* Electra: Beacon state fixes from PR 13919
* Add missing tests - part 1
* Split eip_7251_root.go into different files and reuse/share code with historical state summaries root. It's identical!
* Add missing tests - part 2
* deposit receipts start index getters and setters (#13947)
* adding in getters and setters for deposit receipts start index
* adding tests
* gaz
* Add missing tests - part 3 of 3
Update the electra withdrawal example with a ssz state containing pending partial withdrawals
* add tests for beacon-chain/state/state-native/getters_balance_deposits.go
* Add electra field to testing/util/block.go execution payload
* godoc commentary on public methods
* Fix failing test
* Add balances index out of bounds check and relevant tests.
* Revert switch case electra
* Instead of copying spectest data into testdata, use the spectest dependency
* Deepsource fixes
* Address @rkapka PR feedback
* s/MaxPendingPartialsPerWithdrawalSweep/MaxPendingPartialsPerWithdrawalsSweep/
* Use multivalue slice compatible accessors for validator and balance in ActiveBalanceAtIndex
* More @rkapka feedback. What a great reviewer!
* More tests for branching logic in ExitEpochAndUpdateChurn
* fix build
---------
Co-authored-by: james-prysm <90280386+james-prysm@users.noreply.github.com>
* First take at updating everything to v5
* Patch gRPC gateway to use prysm v5
Fix patch
* Update go ssz
---------
Co-authored-by: Preston Van Loon <pvanloon@offchainlabs.com>
