Squash the main verkle PR ahead of rebase
don't call Bytes() in GetTreeKey (#137)
trie: avoid endianness conversion in GetTreeKey (#140)
* trie/utils: add concrete expected value in trie key generation test
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* mod: update to latest go-verkle
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* trie/utils: avoid endianness conversions
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* apply review changes & update to official go-verkle version
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
upgrade go-verkle to CoW version and get TestProcessVerkle to build (#138)
updating ci to use self-hosted machine (#143)
fix: storage offset in non-header group + reuse of value buffer (#145)
dedup call to ChunkifyCode, same as replay branch (#156)
* dedup call to ChunkifyCode, same as replay branch
* fix some linter issues
fix code offset in tree update (#157)
fix REVERT in state processor test execution (#158)
* fix code offset in tree update
* fix REVERT in test execution
save on key hashing: lump code size update with first code chunk group (#159)
fix code chunk key calculation and storage key calculation (#161)
* fix codeKey calculation
* Remove
* fix storageOffset
* fix the fix to the fix to the offset fix
* Remove copy/pasted, unused code in test
* fix linter
---------
Co-authored-by: Guillaume Ballet <3272758+gballet@users.noreply.github.com>
fix: infinite loop when calling extcodecopy on empty code (#151)
upgrade to latest go-verkle
fix: only update code in the tree if it's dirty (#174)
fix: read-touch the code size and Keccak of the origin (#175)
List of changes for converting a sepolia database (#182)
* naive conversion rebased on top of beverly hills
* changes for the sepolia shadow fork conversion
* fixes to please the linter
* fixes to please the linter
Unified point cache (#180)
* Unified point cache
* Use cache for Try*Account
* alter Trie interface to use caching for slots (#181)
* alter Trie interface to use caching for slots
* fix: use a lock to protect the point cache (#185)
* use fastest non-master go-verkle version & pull trie/Verkle.go changes to use new api (#184)
* mod: update to fastest go-verkle version today
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* trie/verkle: use new batch serialization api
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
---------
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
---------
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
Co-authored-by: Ignacio Hagopian <jsign.uy@gmail.com>
* fix: TryDelete signature in unit tests
---------
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
Co-authored-by: Ignacio Hagopian <jsign.uy@gmail.com>
trie/utils: fix potential overflow (#191)
* trie/utils: fix potential overflow
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* trie/utils: receive storage key as a byte slice
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* revert formatter changes
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* trie/utils: fix mod 256
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
---------
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
trie/utils: fix incorrect bigint assignment (#193)
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
upgrade precomp link to fix CI
fix: add missing code size&keccak leaves in empty accounts (#192)
fixes to use the latest go-verkle@master (#197)
* fixes to use the latest go-verkle@master
* linter fixes
* linter fixes for tests
* fix: use jsign's go-verkle fix
refactor: remove unused (*StateDB).GetXLittleEndian methods (#204)
fix gas accounting issue in state_processor_test.go (#207)
update go-verkle not to use StatelessNode anymore (#206)
* update go-verkle not to use StatelessNode anymore
* update go-verkle to latest
refactor: move verkle gas accounting to its own block in TransitionDB (#208)
fix a panic in deserializeVerkleProof if GetProofItems returns a nil ProofElements
use the cachingDB instead of a custom VerkleDB (#209)
* use the cachingDB instead of a custom VerkleDB
* fix stack trace in LES
remove holiman from CODEOWNERS as he gets too many emails
read from tree in state object if the snapshot is nil (#205)
add missing error checks for the root node type (#214)
implement OpenStorageTrie for verkle trees (#210)
* implement OpenStorageTrie for verkle trees
* add a few comments for future maintenance
* fix linter issue
fix: copy balance leaf to new buffer in TryGetAccount (#217)
implement some heretofore unimplemented iterator methods (#219)
params: move verkle params to their own file (#228)
fix: proper number of chunk evals (#215)
overlay transition (#244)
* overlay transition
Fix some bugs identified in the code review
Co-authored-by: Ignacio Hagopian <jsign.uy@gmail.com>
Include base -> overlay key-values migration logic (#199)
* mod: add go-verkle version with key-value migration new apis
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* core/stateprocessor: use constant for max number of migrated key-values
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* core: add base->overlay key-values migration logic
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* core: fix some compiler errors
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* trie: consider removing transition trie api in the future
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* mod: use latest go-verkle
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
---------
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
fix some unit tests errors
get convresion block from file
fix compilation issues
fix initialization issue in migrator
fix: changes needed to run the first 28 blocks
important sutff: fix the banner
fix: use nonce instead of balance in nonce leaf (#202)
fixes for performing the overlay transition (#203)
* fixes for performing the overlay transition
* fixes for the full replay
* fix: deletion-and-recreation of EoA
* fixes to replay 2M+ blocks
* upgrade to go-verkle@master
* fix: proper number of chunk evals
* rewrite conversion loop to fix known issues
changes to make replay work with the overlay method (#216)
* fixes for performing the overlay transition
fixes for the full replay
fix: deletion-and-recreation of EoA
fixes to replay 2M+ blocks
upgrade to go-verkle@master
fix: proper number of chunk evals
rewrite conversion loop to fix known issues
changes to make replay work with the overlay method
fixes to replay 2M+ blocks
update to latest go-verkle@master
* use a PBSS-like scheme for internal nodes (#221)
* use a PBSS-like scheme for internal nodes
* a couple of fixes coming from debugging replay
* fix: use an error to notify the transition tree that a deleted account was found in the overlay tree (#222)
* fixes for pbss replay (#227)
* fixes for pbss replay
* trie/verkle: use capped batch size (#229)
* trie/verkle: use capped batch size
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* trie/verkle: avoid path variable allocation per db.Put
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* don't keep more than 32 state root conversions in RAM (#230)
---------
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
Co-authored-by: Guillaume Ballet <3272758+gballet@users.noreply.github.com>
* cleanup some code
* mod: update go-verkle
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* re-enable snapshot (#231)
* re-enable cancun block / snapshot (#226)
* clear storage conversion key upon translating account (#234)
* clear storage conversion key upon translating account
* mod: use latest go-verkle
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
---------
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
Co-authored-by: Ignacio Hagopian <jsign.uy@gmail.com>
* fix: self-deadlock with translated root map mutex (#236)
* return compressed commitment as root commitment (#237)
---------
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
Co-authored-by: Ignacio Hagopian <jsign.uy@gmail.com>
---------
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
Co-authored-by: Ignacio Hagopian <jsign.uy@gmail.com>
---------
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
Co-authored-by: Ignacio Hagopian <jsign.uy@gmail.com>
fix first panic in *TransitionTrie.Copy()
upgrade go-verkle to latest master
mod: update go-verkle (#239)
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
core: print state root every 100 blocks (#240)
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
fix: only Commit the account trie (#242)
fixes to get TestProcessVerkle to work with the overlay branch (#238)
* fixes to get TestProcessVerkle to work with the overlay branch
* fix all panics in verkle state processor test
* fix proof verification
move transition management to cachingDB
* fix: mark the verkle transition as started if it's ended without being started
* fix the verkle state processing test
* fix linter errors
* Add a function to clear verkle params for replay
* fix: handle TransitionTrie in OpenStorageTrie
* fix linter issue
* fix the deleted account error (#247)
* code cleanup (#248)
* fix: don't error on a missing conversion.txt (#249)
* Overlay Tree preimages exporting and usage (#246)
* export overlay preimages tool
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* use preimages flat file in overlay tree migration logic
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* cmd/geth: add --roothash to overlay tree preimage exporting command
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* cleanup
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* review feedback
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
---------
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
* fix: reduce the PR footprint (#250)
* fix: don't fail when preimages.bin is missing (#251)
* fix: don't fail when preimages.bin is missing
* fix: don't open the preimages file when outside of transition
---------
Signed-off-by: Ignacio Hagopian <jsign.uy@gmail.com>
Co-authored-by: Ignacio Hagopian <jsign.uy@gmail.com>
review changes
remove replay-specific code
* all: implement path-based state scheme
* all: edits from review
* core/rawdb, trie/triedb/pathdb: review changes
* core, light, trie, eth, tests: reimplement pbss history
* core, trie/triedb/pathdb: track block number in state history
* trie/triedb/pathdb: add history documentation
* core, trie/triedb/pathdb: address comments from Peter's review
Important changes to list:
- Cache trie nodes by path in clean cache
- Remove root->id mappings when history is truncated
* trie/triedb/pathdb: fallback to disk if unexpect node in clean cache
* core/rawdb: fix tests
* trie/triedb/pathdb: rename metrics, change clean cache key
* trie/triedb: manage the clean cache inside of disk layer
* trie/triedb/pathdb: move journal function
* trie/triedb/path: fix tests
* trie/triedb/pathdb: fix journal
* trie/triedb/pathdb: fix history
* trie/triedb/pathdb: try to fix tests on windows
* core, trie: address comments
* trie/triedb/pathdb: fix test issues
---------
Co-authored-by: Felix Lange <fjl@twurst.com>
Co-authored-by: Martin Holst Swende <martin@swende.se>
* core: check excessBlobGas in front
Signed-off-by: jsvisa <delweng@gmail.com>
* core: no need to manual panic
Signed-off-by: jsvisa <delweng@gmail.com>
* core: no comment
Signed-off-by: jsvisa <delweng@gmail.com>
---------
Signed-off-by: jsvisa <delweng@gmail.com>
* core/types: add data gas fields in Receipt
* core/types: use BlobGas method of tx
* core: fix test
* core/types: fix receipt tests, add data gas used field test
---------
Co-authored-by: Péter Szilágyi <peterke@gmail.com>
* core/blobpool: implement txpool for blob txs
* core/txpool: track address reservations to notice any weird bugs
* core/txpool/blobpool: add support for in-memory operation for tests
* core/txpool/blobpool: fix heap updating after SetGasTip if account is evicted
* core/txpool/blobpool: fix eviction order if cheap leading txs are included
* core/txpool/blobpool: add note as to why the eviction fields are not inited in reinject
* go.mod: pull in inmem billy form upstream
* core/txpool/blobpool: fix review commens
* core/txpool/blobpool: make heap and heap test deterministic
* core/txpool/blobpool: luv u linter
* core/txpool: limit blob transactions to 16 per account
* core/txpool/blobpool: fix rebase errors
* core/txpool/blobpool: luv you linter
* go.mod: revert some strange crypto package dep updates
EIP-6780: SELFDESTRUCT only in same transaction
> SELFDESTRUCT will recover all funds to the caller but not delete the account, except when called in the same transaction as creation
---------
Co-authored-by: Martin Holst Swende <martin@swende.se>
This updates the reference tests to the latest version and also adds logic
to process EIP-4844 blob transactions into the state transition. We are now
passing most Cancun fork tests.
Co-authored-by: Marius van der Wijden <m.vanderwijden@live.de>
Co-authored-by: Felix Lange <fjl@twurst.com>
This change adds the ability to perform reads from freezer without size limitation. This can be useful in cases where callers are certain that out-of-memory will not happen (e.g. reading only a few elements).
The previous API was designed to behave both optimally and secure while servicing a request from a peer, whereas this change should _not_ be used when an untrusted peer can influence the query size.
This change makes the StateDB track the state key value diff of a block transition.
We already tracked current account and storage values for the purpose of updating
the state snapshot. With this PR, we now also track the original (pre-transition) values
of accounts and storage slots.
Implements [EIP 5656](https://eips.ethereum.org/EIPS/eip-5656), MCOPY instruction, and enables it for Cancun.
---------
Co-authored-by: Martin Holst Swende <martin@swende.se>
Back before #27178 , we spun up a number of ethash verifiers to verify headers. So we also had tests to ensure that we were indeed able to abort verification even if we had multiple workers running.
With PR #27178, we removed the parallelism in verification, and these tests are now failing, since we now just sequentially fire away the results as fast as possible on one routine.
This change removes the (sometimes failing) tests
This change adds back the 'geth --dev' mode of operation, using a cl-mocker.
---------
Co-authored-by: Martin Holst Swende <martin@swende.se>
Co-authored-by: rjl493456442 <garyrong0905@gmail.com>
Co-authored-by: lightclient <14004106+lightclient@users.noreply.github.com>
The clean trie cache is persisted periodically, therefore Geth can
quickly warmup the cache in next restart.
However it will reduce the robustness of system. The assumption is
held in Geth that if the parent trie node is present, then the entire
sub-trie associated with the parent are all prensent.
Imagine the scenario that Geth rewinds itself to a past block and
restart, but Geth finds the root node of "future state" in clean
cache then regard this state is present in disk, while is not in fact.
Another example is offline pruning tool. Whenever an offline pruning
is performed, the clean cache file has to be removed to aviod hitting
the root node of "deleted states" in clean cache.
All in all, compare with the minor performance gain, system robustness
is something we care more.
* core/state, light, les: make signature of ContractCode hash-independent
* push current state for feedback
* les: fix unit test
* core, les, light: fix les unittests
* core/state, trie, les, light: fix state iterator
* core, les: address comments
* les: fix lint
---------
Co-authored-by: Gary Rong <garyrong0905@gmail.com>
Verkle trees store the code inside the trie. This PR changes the interface to pass the code, as well as the dirty flag to tell the trie package if the code is dirty and needs to be updated. This is a no-op for the MPT and the odr trie.
The state availability is checked during the creation of a state reader.
- In hash-based database, if the specified root node does not exist on disk disk, then
the state reader won't be created and an error will be returned.
- In path-based database, if the specified state layer is not available, then the
state reader won't be created and an error will be returned.
This change also contains a stricter semantics regarding the `Commit` operation: once it has been performed, the trie is no longer usable, and certain operations will return an error.
This removes the feature where top nodes of the proof can be elided.
It was intended to be used by the LES server, to save bandwidth
when the client had already fetched parts of the state and only needed
some extra nodes to complete the proof. Alas, it never got implemented
in the client.
* go.mod: update kzg libraries to use big-endian
* go.sum: ran go mod tidy
* core/testdata/precompiles: fix blob verification test
* core/testdata/precompiles: fix blob verification test
This change ensures Reheap will be called even before the London fork activates.
Since Reheap would otherwise only be called through `SetBaseFee` after London,
the list would just keep growing if the fork was not enabled or not reached yet.
* all: move main transaction pool into a subpool
* go.mod: remove superfluous updates
* core/txpool: review fixes, handle txs rejected by all subpools
* core/txpool: typos
The logs in this function are pulled straight from disk in rawdb.ReadRawReceipts and
also modified in receipts.DeriveFields, so removing the copy should be fine.
* core/txpool: abstraction prep work for secondary pools (blob pool)
* core/txpool: leave subpool concepts to a followup pr
* les: fix tests using hard coded errors
* core/txpool: use bitmaps instead of maps for tx type filtering
This changes the journal logic to mark the state object dirty immediately when it
is reset.
We're mostly adding this change to appease the fuzzer. Marking it dirty immediately
makes no difference in practice because accounts will always be modified by EVM
right after creation.
Continuing with a series of PRs to make the Trie interface more generic, this PR moves
the RLP encoding of storage slots inside the StateTrie and light.Trie implementations,
as other types of tries don't use RLP.
This PR adds a staleness-check to AccountRLP, before checking the bloom-filter and potentially going directly into the disklayer.
---------
Co-authored-by: rjl493456442 <garyrong0905@gmail.com>
Regenerate receipt json code to remove omit empty. Previously, there was a discrepancy between the generated code and the source.
---------
Co-authored-by: lightclient@protonmail.com <lightclient@protonmail.com>
Co-authored-by: Martin Holst Swende <martin@swende.se>
* cmd/utils, node: switch to Pebble as the default db if none exists
* node: fall back to LevelDB on platforms not supporting Pebble
* core/rawdb, node: default to Pebble at the node level
* cmd/geth: fix some tests explicitly using leveldb
* ethdb/pebble: allow double closes, makes tests simpler
In this PR, all TryXXX(e.g. TryGet) APIs of trie are renamed to XXX(e.g. Get) with an error returned.
The original XXX(e.g. Get) APIs are renamed to MustXXX(e.g. MustGet) and does not return any error -- they print a log output. A future PR will change the behaviour to panic on errorrs.
The EIP150Hash was an idea where, after the fork, we hardcoded the forked hash as an extra defensive mechanism. It wasn't really used, since forks weren't contentious and for all the various testnets and private networks it's been a hassle to have around.
This change removes that config field.
---------
Signed-off-by: jsvisa <delweng@gmail.com>
