This PR fixes an issues in the new simulated backend. The root cause is the fact that the transaction pool has an internal reset operation that runs on a background thread.
When a new transaction is added to the pool via the RPC, the transaction is added to a non-executable queue and will be moved to its final location on a background thread. If the machine is overloaded (or simply due to timing issues), it can happen that the simulated backend will try to produce the next block, whilst the pool has not yet marked the newly added transaction executable. This will cause the block to not contain the transaction. This is an issue because we want determinism from the simulator: add a tx, mine a block. It should be in there.
The PR fixes it by adding a Sync function to the txpool, which waits for the current reset operation (if any) to finish, and then runs an entire round of reset on top. The new round is needed because resets are only triggered by new head events, so newly added transactions will not trigger the outer resets that we can wait on. The transaction pool would eventually internally do a reset even on transaction addition, but there's no easy way to wait on that and there's no meaningful reason to bubble that across everything. A clean outer reset will at worse be a small noop goroutine.
This PR introduces a few changes with respect to payload verification in fcu and new payload requests:
* First of all, it undoes the `verifyPayloadAttributes(..)` simplification I attempted in #27872.
* Adds timestamp validation to fcu payload attributes [as required](https://github.com/ethereum/execution-apis/blob/main/src/engine/cancun.md#specification-1) (section 2) by the Engine API spec.
* For the new payload methods, I also update the verification of the executable data. For `newPayloadV2`, it does not currently ensure that cancun values are `nil`. Which could make it possible to submit cancun payloads through it.
* On `newPayloadV3` the same types of checks are added. All shanghai and cancun related fields in the executable data must be non-nil, with the addition that the timestamp is _only_ with cancun.
* Finally it updates a newly failing catalyst test to call the correct fcu and new payload methods depending on the fork.
This change makes use of uin256 to represent balance in state. It touches primarily upon statedb, stateobject and state processing, trying to avoid changes in transaction pools, core types, rpc and tracers.
This change simplifies the logic for indexing transactions and enhances the UX when transaction is not found by returning more information to users.
Transaction indexing is now considered as a part of the initial sync, and `eth.syncing` will thus be `true` if transaction indexing is not yet finished. API consumers can use the syncing status to determine if the node is ready to serve users.
The code to compute a versioned hash was duplicated a couple times, and also had a small
issue: if we ever change params.BlobTxHashVersion, it will most likely also cause changes
to the actual hash computation. So it's a bit useless to have this constant in params.
In the tracing tests, the base fee was generally set to nil. This commit changes this to pass the proper base instead, and fixes the few tests which become broken by the change.
* cmd, core: resolve scheme from a read-write database
* cmd, core, eth: move the scheme check in the ethereum constructor
* cmd/geth: dump should in ro mode
* cmd: reverts
This is a rewrite of the 'simulated backend', an implementation of the ethclient interfaces
which is backed by a simulated blockchain. It was getting annoying to maintain the old
version of the simulated backend feature because there was a lot of code duplication with
the main client.
The new version is built using parts that we already have: an in-memory geth node instance
running in developer mode provides the chain, while the Go API is provided by ethclient.
A backwards-compatibility wrapper is provided, but the simulated backend has also moved to
a more sensible import path: github.com/ethereum/go-ethereum/ethclient/simulated
---------
Co-authored-by: Felix Lange <fjl@twurst.com>
Co-authored-by: Gary Rong <garyrong0905@gmail.com>
This PR will allow a previously underpriced transaction back in after a timeout
of 5 minutes. This will block most transaction spam but allow for transactions to
be re-broadcasted on networks with less transaction flow.
---------
Co-authored-by: Felix Lange <fjl@twurst.com>
# Conflicts:
# eth/fetcher/tx_fetcher.go
This is a minor refactor in preparation of changes to range verifier. This PR contains no intentional functional changes but moves (and renames) the light.NodeSet
* cmd, les, tests: remove light client code
This commit removes the light client (LES) code.
Since the merge the light client has been broken and
it is hard to maintain it alongside the normal client.
We decided it would be best to remove it for now and
maybe rework and reintroduce it in the future.
* cmd, eth: remove some more mentions of light mode
* cmd: re-add flags and mark as deprecated
* cmd: warn the user about deprecated flags
* eth: better error message
* triedb/pathdb: fix async node buffer diskroot mismatches when journaling
* triedb: check whether the async flush is done
* fix: generate new eth config
---------
Co-authored-by: VM <arimas@foxmail.com>
This PR replaces Geth's logger package (a fork of [log15](https://github.com/inconshreveable/log15)) with an implementation using slog, a logging library included as part of the Go standard library as of Go1.21.
Main changes are as follows:
* removes any log handlers that were unused in the Geth codebase.
* Json, logfmt, and terminal formatters are now slog handlers.
* Verbosity level constants are changed to match slog constant values. Internal translation is done to make this opaque to the user and backwards compatible with existing `--verbosity` and `--vmodule` options.
* `--log.backtraceat` and `--log.debug` are removed.
The external-facing API is largely the same as the existing Geth logger. Logger method signatures remain unchanged.
A small semantic difference is that a `Handler` can only be set once per `Logger` and not changed dynamically. This just means that a new logger must be instantiated every time the handler of the root logger is changed.
----
For users of the `go-ethereum/log` module. If you were using this module for your own project, you will need to change the initialization. If you previously did
```golang
log.Root().SetHandler(log.LvlFilterHandler(log.LvlInfo, log.StreamHandler(os.Stderr, log.TerminalFormat(true))))
```
You now instead need to do
```golang
log.SetDefault(log.NewLogger(log.NewTerminalHandlerWithLevel(os.Stderr, log.LevelInfo, true)))
```
See more about reasoning here: https://github.com/ethereum/go-ethereum/issues/28558#issuecomment-1820606613
* eth/gasestimator: early exit for plain transfer and error allowance
* core, eth/gasestimator: hard guess at a possible required gas
* internal/ethapi: update estimation tests with the error ratio
* eth/gasestimator: I hate you linter
* graphql: fix gas estimation test
---------
Co-authored-by: Oren <orenyomtov@users.noreply.github.com>
This change fixes two type-inconsistencies in the JS tracer:
- In most places we return byte arrays as a `Uint8Array` to the tracer. However it seems we missed doing the conversion for `ctx` fields which are passed to the tracer during `result`. They are passed as simple arrays. I think Uint8Arrays are more suitable and we should change this inconsistency. Note: this will be a breaking-change. But I believe the effect is small. If we look at our tracers we see that these fields (`ctx.from`, `ctx.to`, etc.) are used in 2 ways. Passed to `toHex` which takes both array or buffer. Or the length was measured which is the same for both types.
- The `slice` taking in `int, int` params versus `memory.slice` taking `int64, int64` params. I suggest changing `slice` types to `int64`. This should have no effect almost in any case.
There were several problems related to dumping state.
- If a preimage was missing, even if we had set the `OnlyWithAddresses` to `false`, to export them anyway, the way the mapping was constructed (using `common.Address` as key) made the entries get lost anyway. Concerns both state- and blockchain tests.
- Blockchain test execution was not configured to store preimages.
This changes makes it so that the block test executor takes a callback, just like the state test executor already does. This callback can be used to examine the post-execution state, e.g. to aid debugging of test failures.
* cmd, les, tests: remove light client code
This commit removes the light client (LES) code.
Since the merge the light client has been broken and
it is hard to maintain it alongside the normal client.
We decided it would be best to remove it for now and
maybe rework and reintroduce it in the future.
* cmd, eth: remove some more mentions of light mode
* cmd: re-add flags and mark as deprecated
* cmd: warn the user about deprecated flags
* eth: better error message
geth --dev can be used with an existing data directory and genesis block. Since
dev mode only works with PoS, we need to verify that the merge has happened.
Co-authored-by: Felix Lange <fjl@twurst.com>
This PR moves our fuzzers from tests/fuzzers into whatever their respective 'native' package is.
The historical reason why they were placed in an external location, is that when they were based on go-fuzz, they could not be "hidden" via the _test.go prefix. So in order to shove them away from the go-ethereum "production code", they were put aside.
But now we've rewritten them to be based on golang testing, and thus can be brought back. I've left (in tests/) the ones that are not production (bls128381), require non-standard imports (secp requires btcec, bn256 requires gnark/google/cloudflare deps).
This PR also adds a fuzzer for precompiled contracts, because why not.
This PR utilizes a newly rewritten replacement for go-118-fuzz-build, namely gofuzz-shim, which utilises the inputs from the fuzzing engine better.
Currently, geth's will return `[]` for any `len(topics) > 4` log filter. The EVM only supports up to four logs, via LOG4 opcode, so larger criterias fail. This change makes the filter query exit early in those cases.
Currently, one can use the "withLogs" parameter to include logs in the
callTracer results, which allows the user to see at which trace level
was each log emitted.
This commit adds a position field to the logs which determine
the exact ordering of a call's logs and its subcalls. This would
be useful e.g. for explorers wishing to display the flow of execution.
Co-authored-by: jsvisa <delweng@gmail.com>
This change improves GenerateChain to support internal chain history access (ChainReader)
for the consensus engine and EVM.
GenerateChain takes a `parent` block and the number of blocks to create. With my changes,
the consensus engine and EVM can now access blocks from `parent` up to the block currently
being generated. This is required to make the BLOCKHASH instruction work, and also needed
to create real clique chains. Clique uses chain history to figure out if the current signer is in-turn,
for example.
I've also added some more accessors to BlockGen. These are helpful when creating transactions:
- g.Signer returns a signer instance for the current block
- g.Difficulty returns the current block difficulty
- g.Gas returns the remaining gas amount
Another fix in this commit concerns the receipts returned by GenerateChain. The receipts now
have properly derived fields (BlockHash, etc.) and should generally match what would be
returned by the RPC API.
* go.mod: upgrade prysm and the indrect dependency
prysm from v4.0.2 to v4.0.8, and run go mod tidy
* ci: upgrade go version from 1.19 to 1.20
* go-version: upgrade from v1.19 to v1.20
there is some dependency on go v1.20, such as go-libp2p v0.27.8
and also run go mod tidy
* dependency: upgrade docker version for security
it is not a big issue, since docker is only used for test purpose.
* rand: update the usage of math/rand after golang v1.20
2 APIs of math/rand module were deprecated since golang v1.20.
that is: rand.Seed() and rand.Read(), refer: ettps://pkg.go.dev/math/rand
"rand.Seed(seed int64)" has been replaced by: "r := rand.New(rand.NewSource(seed int64))",
need to initialize it with an instance before use
"rand.Read()" has been replaced by "crypto/rand.Read()"
* readme: need golang v1.20+ to build bsc
This change enhances the stacktrie constructor by introducing an option struct. It also simplifies the `Hash` and `Commit` operations, getting rid of the special handling round root node.
This changes fixes a bug in the fetcher, where the timeout for how long to remember underpriced transaction was erroneously compared, and the timeout never hit.
---------
Co-authored-by: Martin Holst Swende <martin@swende.se>
During snap-sync, we request ranges of values: either a range of accounts or a range of storage values. For any large trie, e.g. the main account trie or a large storage trie, we cannot fetch everything at once.
Short version; we split it up and request in multiple stages. To do so, we use an origin field, to say "Give me all storage key/values where key > 0x20000000000000000". When the server fulfils this, the server provides the first key after origin, let's say 0x2e030000000000000 -- never providing the exact origin. However, the client-side needs to be able to verify that the 0x2e03.. indeed is the first one after 0x2000.., and therefore the attached proof concerns the origin, not the first key.
So, short-short version: the left-hand side of the proof relates to the origin, and is free-standing from the first leaf.
On the other hand, (pun intended), the right-hand side, there's no such 'gap' between "along what path does the proof walk" and the last provided leaf. The proof must prove the last element (unless there are no elements).
Therefore, we can simplify the semantics for trie.VerifyRangeProof by removing an argument. This doesn't make much difference in practice, but makes it so that we can remove some tests. The reason I am raising this is that the upcoming stacktrie-based verifier does not support such fancy features as standalone right-hand borders.
This change addresses an issue in snap sync, specifically when the entire sync process can be halted due to an encountered empty storage range.
Currently, on the snap sync client side, the response to an empty (partial) storage range is discarded as a non-delivery. However, this response can be a valid response, when the particular range requested does not contain any slots.
For instance, consider a large contract where the entire key space is divided into 16 chunks, and there are no available slots in the last chunk [0xf] -> [end]. When the node receives a request for this particular range, the response includes:
The proof with origin [0xf]
A nil storage slot set
If we simply discard this response, the finalization of the last range will be skipped, halting the entire sync process indefinitely. The test case TestSyncWithUnevenStorage can reproduce the scenario described above.
In addition, this change also defines the common variables MaxAddress and MaxHash.
* cmd, core: resolve scheme from a read-write database
* cmd, core, eth: move the scheme check in the ethereum constructor
* cmd/geth: dump should in ro mode
* cmd: reverts
This change
- Removes the owner-notion from a stacktrie; the owner is only ever needed for comitting to the database, but the commit-function, the `writeFn` is provided by the caller, so the caller can just set the owner into the `writeFn` instead of having it passed through the stacktrie.
- Removes the `encoding.BinaryMarshaler`/`encoding.BinaryUnmarshaler` interface from stacktrie. We're not using it, and it is doubtful whether anyone downstream is either.
* eth: enforce announcement metadatas and drop peers violating the protocol
* eth/fetcher: relax eth/68 validation a bit for flakey clients
* tests/fuzzers/txfetcher: pull in suggestion from Marius
* eth/fetcher: add tests for peer dropping
* eth/fetcher: linter linter linter linter linter
This is a minor refactor in preparation of changes to range verifier. This PR contains no intentional functional changes but moves (and renames) the light.NodeSet
This PR will allow a previously underpriced transaction back in after a timeout
of 5 minutes. This will block most transaction spam but allow for transactions to
be re-broadcasted on networks with less transaction flow.
---------
Co-authored-by: Felix Lange <fjl@twurst.com>
* parlia: reject header with `WithdrawalsHash` before shanghai fork
* log: output chainconfig when start up
* eth: fix TestOptionMaxPeersPerIP failure of goroutine order
* fix: crash of highestVerifiedHeader
* fix: panic of blobpool
* fix: genesis set up
* 1. modify NewDatabaseWithNodeDB to upstream
2. fix race use of hasher in statedb
3. fix use wrong value when updateTrie
* fix dir legacypool
* fix dir blobpool
* fix dir vote
* remove diffsync related code
* fix core/state/snapshot
* disable pipeCommit for now
* fix applyTransaction for bloom setting
* CI: fast finality in gasprice test
* CI: diffFetcher was removed
* CI: downloader, remove beaconsync test
* CI: no beaconsync in downloader, remove a failed case
TestCheckpointChallenge was removed in:
https://github.com/ethereum/go-ethereum/pull/27147
since after merge, it is useless for ethereum, but might be useful for BSC.
disable the case right now, as it is not a big issue.
* CI: bsc protocol decHandlers
* CI: receipt Bloom process
* 1. skip CheckConfigForkOrder for non-parlia engine
2. all test cases in core work well now
cd core && go test ./... -v
* fix test cases in trie dir
* CI: no beaconsync in downloader, remove a failed case(redo)
* fix dir miner
* fix dir cmd/geth
* CI: filter test, BaseFee & Finality
* fix dir graphql
* remove diffStore
* fix ethclient
* fix TestRPCGetTransactionReceipt
* fix dir internal
* ut add dir ethstats and signer
* disable pipeCommit thoroughly; fix concurrent map iteration and map write in statedb
* CI: fix snap sync
it could be changed by mistake
* fix tests/Run to generate snapshot
* prepare for merge
* remove useless
* use common hasher in getDeletedStateObject, no race here
* an critical comment for state.Prepare
* do not copy nil accessList
* add omitempty tag for unused new fields of core.Genesis
* remove totalFees
* calculate fees before FinalizeAndAssemble
* revert interface Finalize of consensus
* do not double gas limit upon london block
* use Leveldb as default
* Revert "remove diffStore"
This reverts commit df343b137412b0beb25298a6ba9c3c19e47f20b1.
* Revert "remove diffsync related code"
This reverts commit 8d84b81feae5d794cb5d7fcfdb7f5f7da751941b.
* compile pass after revert
* remove diffsync
* fix dir eth/protocols/trust
* fix TestFastNode
* decHandlers for trust protocol
* keep persist diff in test
This PR makes EIP-4788 work in the engine API and miner. It also fixes some bugs related to
EIP-4844 block processing and mining. Changes in detail:
- Header.BeaconRoot has been renamed to ParentBeaconRoot.
- The engine API now implements forkchoiceUpdatedV3
- newPayloadV3 method has been updated with the parentBeaconBlockRoot parameter
- beacon root is now applied to new blocks in miner
- For EIP-4844, block creation now updates the blobGasUsed field of the header
ReadSkeletonHeader can return nil if the header is missing, so we should
not access fields on it. Note that calling .Hash() on a nil header is fine, so there
is no need to actually check for nil.
Co-authored-by: Martin Holst Swende <martin@swende.se>
This changes the forkID calculation to ignore time-based forks that occurred before the
genesis block. It's supposed to be done this way because the spec says:
> If a chain is configured to start with a non-Frontier ruleset already in its genesis, that is NOT considered a fork.
This PR removes the newly added txpool.Transaction wrapper type, and instead adds a way
of keeping the blob sidecar within types.Transaction. It's better this way because most
code in go-ethereum does not care about blob transactions, and probably never will. This
will start mattering especially on the client side of RPC, where all APIs are based on
types.Transaction. Users need to be able to use the same signing flows they already
have.
However, since blobs are only allowed in some places but not others, we will now need to
add checks to avoid creating invalid blocks. I'm still trying to figure out the best place
to do some of these. The way I have it currently is as follows:
- In block validation (import), txs are verified not to have a blob sidecar.
- In miner, we strip off the sidecar when committing the transaction into the block.
- In TxPool validation, txs must have a sidecar to be added into the blobpool.
- Note there is a special case here: when transactions are re-added because of a chain
reorg, we cannot use the transactions gathered from the old chain blocks as-is,
because they will be missing their blobs. This was previously handled by storing the
blobs into the 'blobpool limbo'. The code has now changed to store the full
transaction in the limbo instead, but it might be confusing for code readers why we're
not simply adding the types.Transaction we already have.
Code changes summary:
- txpool.Transaction removed and all uses replaced by types.Transaction again
- blobpool now stores types.Transaction instead of defining its own blobTx format for storage
- the blobpool limbo now stores types.Transaction instead of storing only the blobs
- checks to validate the presence/absence of the blob sidecar added in certain critical places
The Go authors updated golang/x/ext to change the function signature of the slices sort method.
It's an entire shitshow now because x/ext is not tagged, so everyone's codebase just
picked a new version that some other dep depends on, causing our code to fail building.
This PR updates the dep on our code too and does all the refactorings to follow upstream...
The Go authors updated golang/x/ext to change the function signature of the slices sort method.
It's an entire shitshow now because x/ext is not tagged, so everyone's codebase just
picked a new version that some other dep depends on, causing our code to fail building.
This PR updates the dep on our code too and does all the refactorings to follow upstream...
This change removes a chainconfig parameter passed into rawdb.ReadLogs, which is not used nor needed.
It also modifies the filter loop slightly, avoiding a labeled break and instead using a method.
This change does not modify any behaviour.
* 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
This is a spin-out from the EIP-4844 devnet branch, containing just the Engine API modifications
and nothing else. The newPayloadV3 endpoint won't really work in this version, but we need the
data structures for testing so I'd like to get this in early.
Co-authored-by: Marius van der Wijden <m.vanderwijden@live.de>
This change removes PoW header syncing related code from LES and also deletes
duplicated packages les/catalyst, les/downloader and les/fetcher. These package copies
were created because people wanted to make changes in their eth/ counterparts, but weren't
able to adapt LES code to the API changes.
The struct logger (or opcode tracer) was missing the return data field even
if this was explicitly enabled by user via `"enableReturnData": true` in the config.
This PR fixes this issue.
Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com>
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.
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.
