core/txpool: move some validation to outside of mutex #27006

holiman · 2023-03-29T09:13:05Z

This PR is for discussion, I'm fine with not merging it if we choose to.

Currently, we do most of transaction validation while holding the txpool mutex: one exception being an early-on signature check. This PR changes that, so that we do all non-stateful checks before we enter the mutex arena. This means they can be performed in parallel, and to enable that, certain fields have been made atomic bools and uint64.

Right now, the PR does not remove those checks from the in-mutex-arena. This is because of some edge-cases that may happen around hardforks, and particularly when transactions are re-added due to a reorg, in which a transaction from a dropped block becomes invalid due to a change in the rules. I think maybe we don't need to cater for that particular edgecase, but I left it in for now.

I am not sure if this PR makes anything faster/better, but OTOH if we already today have one special "before mutex" check (signature check), this PR makes that a bit more structured.

core/txpool/txpool.go

holiman · 2023-03-29T14:51:09Z

core/txpool/txpool.go

+	if err := pool.validateTxBasics(tx, local); err != nil {
+		return err
+	}


We could skip this check, but that would require a bit of analysis that we don't hit some quirky edgecase due to the pool behavioural changes when a hardfork happens.

Like, we first validateBasics on x before shanghai, accept, then somehow the block with x is dropped, so x is re-added, but now we're past shanghai and the rule about initcode size must be enforced. So now we would add x to pending, but it cannot be added in a block.

If that's the "extent of the damage" then I guess we can drop the check here.

honestly I feel this check is unnecessary. I think the check in txpool doesn't need to be very strict.

Like the init-code-size check raised in shanghai, it's totally possible that the "large-code" transactions are accepted before the fork and stay in pool for long time because of low gas price. These transactions will be filtered out in miner.

So another question is: if these transactions are accepted into pool, how can they get evicted? Feel like they will stay until get evicted because of low price.

fjl · 2023-04-01T20:13:29Z

I think it's a good idea to apply these checks. Since the new method can work with or without the mutex held, can't we restructure things a bit so the main entry point for transactions from p2p doesn't perform the checks while holding the lock?

rjl493456442

I like this change and think we should accept it.

holiman · 2023-04-03T06:16:49Z

can't we restructure things a bit so the main entry point for transactions from p2p doesn't perform the checks while holding the lock?

That's already what this does

... the name of the PR is "core/txpool: move some validation to outside of mutex" :)

core/txpool/txpool.go

@gballet

* cmd/bootnode, cmd/utils: add 'pmp:<IP>' to --nat flag description (ethereum#26381) * cmd/evm: fix typo in README.md (ethereum#26500) * build: upgrade -dlgo version to Go 1.19.5 (ethereum#26472) * cmd/evm: typo fix in docs (ethereum#26506) * metrics/influxdb: fix time ticker leaks (ethereum#26507) * eth/filters: fix typo in comment (ethereum#26515) * les/fetcher : fix requestTimer leak (ethereum#26514) les/fetcher : fix requestTimer leak * core/txpool: check if initcode size is exceeded (ethereum#26504) * core/txpool: check if initcode size is exceeded * core/txpool: move check * all: remove Kiln testnet (ethereum#26522) Kiln was deprecated after the merge. * console, internal/jsre: fix autocomplete issues (ethereum#26518) Fixes ethereum#26505 where the console crashed when a property getter raised an exception during autocompletion. I also noticed while fixing this issue that autocomplete wasn't working for objects/fields with numbers in them (most importantly web3.<tab><tab>) which is also now fixed. * go.mod: upgrade to go-bip39 v1.1.0 (ethereum#26527) * go.sum: tidy (ethereum#26525) * .travis.yml: go1.19.5 for android builds (ethereum#26524) * params: define cancun and prague as timestamp based forks (ethereum#26481) * params: define cancun and osaka as timestamp based forks * core, params: change osaka to prague * params: fix * signer/core: don't require capital lettered reference types (ethereum#26462) * miner, cmd, eth: require explicit etherbase address (ethereum#26413) This change introduces a breaking change to miner.etherbase is configured. Previously, users did not need to explicitly set the etherbase address via flag, since 'first' local account was used as etherbase automatically. This change removes the "default first account" feature. In Proof-of-stake world, the fee recipient address is provided by CL, and not configured in Geth any more - meaning that miner.etherbase is mostly for legacy networks(pow, clique networks etc). * go.mod: upgrade to latest goja (ethereum#26523) Co-authored-by: Felix Lange <fjl@twurst.com> * go.sum: go mod tidy (ethereum#26536) * all: use http package to replace http method names (ethereum#26535) * cmd/utils: improve parsing of --miner.etherbase address (ethereum#26541) This fixes a regression where the flag did not accept values without the 0x prefix anymore. What's worse, if an invalid value was passed, the client would just log an INFO level message and continue. * all: make timestamp-based fork checks based on uint64 (ethereum#26474) This PR changes the API so that uint64 is used for fork timestamps. It's a good choice because types.Header also uses uint64 for time. Co-authored-by: Felix Lange <fjl@twurst.com> * all: implement withdrawals (EIP-4895) (ethereum#26484) This change implements withdrawals as specified in EIP-4895. Co-authored-by: lightclient@protonmail.com <lightclient@protonmail.com> Co-authored-by: marioevz <marioevz@gmail.com> Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: Felix Lange <fjl@twurst.com> * cmd/utils: fix error at geth startup in --dev mode (ethereum#26550) This fixes a regression in ethereum#26541 where we turned the miner address being missing into a startup error. The address was not configured in --dev mode. * eth/catalyst: error on nil withdrawals post-shanghai (ethereum#26549) This adds explicit checks for the presence of withdrawals in the engine API. Co-authored-by: Felix Lange <fjl@twurst.com> * eth/catalyst: implement exchangeCapabilities method (ethereum#26555) Spec: ethereum/execution-apis#364 * params: fix timestamp display in fork banner (ethereum#26553) * params: remove deprecated bootnodes (ethereum#26530) This change removes the Infura rinkeby bootnode as well as two deprecated ropsten bootnodes. Co-authored-by: Nicola Cocchiaro <ncocchiaro@users.noreply.github.com> Co-authored-by: Martin Holst Swende <martin@swende.se> * go.mod: gnark-crypto upgrade (ethereum#26545) * cmd/evm: add blocktest subcommand to evm (ethereum#26526) Adds blocktest subcommand to the evm command, which is very similar to statetest, but instead of loading a StateTest static test it loads a BlockchainTest from a json file and runs it. Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: lightclient <14004106+lightclient@users.noreply.github.com> * eth/catalyst: fix panic in TestWithdrawals (ethereum#26563) Fixes a regression introduced in ethereum#26549 * log: fix typo in comment (ethereum#26569) * cmd/devp2p: fix broken link in readme(ethereum#26576) fix broken link to DNS discovery tutorial * log: better sanitation (ethereum#26556) * ethclient/gethclient: fix typo (ethereum#26580) fix grammar typo * eth/tracer: rename to revertReason (ethereum#26574) * internal/ethapi: always return block withdrawals if present (ethereum#26565) The execution-apis specification says that the full list of withdrawals should always be returned when requesting a block over RPC: https://github.com/ethereum/execution-apis/blob/378c4304f75b5af2c9b5263c9c76b511e33e8984/src/schemas/block.yaml#L90-L94 This change adopts the expected behavior. * cmd/puppeth: remove puppeth * core: improve ambiguous block validation message (ethereum#26582) * eth/catalyst: return invalid params instead of invalid payload params (ethereum#26591) * cmd, eth, node: deprecate personal namespace (ethereum#26390) * eth: cmd: deprecate personal namespace * eth: cmd: move deprecation to node * node: disable toml of enablepersonal * node: disable personal on ipc as well * Update node/node.go Co-authored-by: Martin Holst Swende <martin@swende.se> * console: error -> warn * node: less roulette --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * accounts, build, mobile: remove Andriod and iOS support * core/vm: improve EVM instance reusability (ethereum#26341) This change improves reusability of the EVM struct. Two methods are added: - SetBlockContext(...) - SetTracer(...) Other attributes like the TransactionContext and the StateDB can already be updated. BlockContext and Tracer are partially not updateable right now. This change fixes it and opens the potential to reuse an EVM struct in more ways. Co-authored-by: Felix Lange <fjl@twurst.com> * eth/catalyst: implement engine_getPayloadBodiesByHash/Range methods (ethereum#26232) This change implements engine_getPayloadBodiesByHash and engine_getPayloadBodiesByRange, according to the specification at https://github.com/ethereum/execution-apis/blob/main/src/engine/shanghai.md#specification-4 . Co-authored-by: Martin Holst Swende <martin@swende.se> * readme: fix broken link (ethereum#26612) * rpc: fix off-by-one in ipc endpoint length check (ethereum#26614) This change fixes a minor flaw in the check for ipc endpoint length. The max_path_size is the max path that an ipc endpoint can have, which is 208. However, that size concerns the null-terminated pathname, so we need to account for an extra null-character too. * core/vm: add bn256ScalarMul testcase for zero scalar value (ethereum#26607) EIP-196 allows a zero value in the scalar argument to precompile `0x07`. This change adds a test for that case. * README: remove text about GPU mining (ethereum#26609) Co-authored-by: Felix Lange <fjl@twurst.com> * all: prepare for path-based trie storage (ethereum#26603) This PR moves some trie-related db accessor methods to a different file, and also removes the schema type. Instead of the schema type, a string is used to distinguish between hashbased/pathbased db accessors. This also moves some code from trie package to rawdb package. This PR is intended to be a no-functionality-change prep PR for ethereum#25963 . --------- Co-authored-by: Gary Rong <garyrong0905@gmail.com> * beacon/engine: move core/beacon to beacon/engine (ethereum#26616) This PR moves core/beacon to beacon/engine so that beacon-chain related code has its own top level package which also can house the the beacon lightclient-code. * build: upgrade to golangci-lint v1.51.1 (ethereum#26618) * tests: update (ethereum#26548) This updates the reference tests to the latest version. Co-authored-by: Felix Lange <fjl@twurst.com> * accounts/abi: fix integer encoding/decoding (ethereum#26568) This PR fixes this abi encoder/decoder to be more stringent. * rpc: remove DecimalOrHex type (ethereum#26629) It's the same as math.HexOrDecimal64, which has more uses across the codebase. * eth/catalyst,miner: include withdrawals in payload id calculation (ethereum#26554) According to the spec the payloadID needs to be random or dependent on all arguments, to prevent two payloads from clashing. This change adds withdrawals into the payload derivation. --------- Co-authored-by: lightclient@protonmail.com <lightclient@protonmail.com> Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: Felix Lange <fjl@twurst.com> * light: fix receiver name from Python style to Go (ethereum#26631) Co-authored-by: Halil Yildirim <halilgolang@gmail.com> * log: allow tabs in log messages (ethereum#26630) * log: allow tabs in log messages This fixes a regression where panic reports in RPC handlers were quoted because they contain tab characters. * Update format.go * all: remove database commit callback, rework noderesolver (ethereum#26637) This change ports some changes from the main PBSS PR: - get rid of callback function in `trie.Database.Commit` which is not required anymore - rework the `nodeResolver` in `trie.Iterator` to make it compatible with multiple state scheme - some other shallow changes in tests and typo-fixes * eth/catalyst: make getPayloadBodiesByRange take hex inputs (ethereum#26624) Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: Marius van der Wijden <m.vanderwijden@live.de> * all: remove support for Ropsten (ethereum#26644) * ethdb: pebble backend (64bit platforms only) (ethereum#26517) * ethdb: use pebble Co-authored-by: Gary Rong <garyrong0905@gmail.com> foo update * apply suggested changes * flags: go format node: fix ddir lookup mistake accounts/abi/bind: fix go.mod replacement for generated binding deps: update pebble + with fix 32-bit build * ethdb/pebble: respect max memtable size * core/rawdb, ethdb: enable pebble on non-32bit platforms only * core/rawdb: fix build tags, fix some review concerns * core/rawdb: refactor methods for database opening * core/rawdb: remove erroneous build tag * cmd/geth: fix the flag default handling + testcase * cmd/geth: improve testing regarding custom backends * ethdb/pebble, deps: update pebble dependency * core/rawdb: replace method with Open * ethdb/pebble: several updates for pebble (ethereum#49) * ethdb/pebble: fix size count in batch * ethdb/pebble: disable seek compaction * ethdb/pebble: more fixes * ethdb, core, cmd: polish and fixes (ethereum#50) * cmd/utils, core/rawdb, ethdb/pebble: address some review concerns * Update flags.go * ethdb/pebble: minor refactors * ethdb/pebble: avoid copy on batch replay * ethdb: fix compilation flaw * cmd: fix test fail due to mismatching error message * cmd/geth, node: rename backingdb to db.engine --------- Co-authored-by: Jared Wasinger <j-wasinger@hotmail.com> Co-authored-by: rjl493456442 <garyrong0905@gmail.com> Co-authored-by: Péter Szilágyi <peterke@gmail.com> * eth/catalyst: disallow forkchoiceupdate v1 post-shanghai (ethereum#26645) * common/prque: generic priority queue (ethereum#26290) * common, core, eth, les, trie: make prque generic * les/vflux/server: fixed issues in priorityPool * common, core, eth, les, trie: make priority also generic in prque * les/flowcontrol: add test case for priority accumulator overflow * les/flowcontrol: avoid priority value overflow * common/prque: use int priority in some tests No need to convert to int64 when we can just change the type used by the queue. * common/prque: remove comment about int64 range --------- Co-authored-by: Zsolt Felfoldi <zsfelfoldi@gmail.com> Co-authored-by: Felix Lange <fjl@twurst.com> * ethdb/pebble: fix nil callbacks (ethereum#26650) * eth/downloader: fix timeout resurrection panic (ethereum#26652) * common/prque, eth/downloader: fix timeout resurrection panic * common/prque: revert -1 hack for les, temporaryly! * core/state, trie: remove unused error-return from trie Commit operation (ethereum#26641) * go.mod: update pebble to latest master (ethereum#26654) * core/vm: set tracer-observable `value` of a delegatecall to match parent `value` (ethereum#26632) This is a breaking change in the tracing hooks API as well as semantics of the callTracer: - CaptureEnter hook provided a nil value argument in case of DELEGATECALL. However to stay consistent with how delegate calls behave in EVM this hook is changed to pass in the value of the parent call. - callTracer will return parent call's value for DELEGATECALL frames. --------- Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * ethdb: add benchmark test suite (ethereum#26659) * params: schedule shanghai fork on sepolia (ethereum#26662) * params: schedule shanghai fork on sepolia * params: u64 -> newUint64 * eth/filters: avoid block body retrieval when no matching logs (ethereum#25199) Logs stored on disk have minimal information. Contextual information such as block number, index of log in block, index of transaction in block are filled in upon request. We can fill in all these fields only having the block header and list of receipts. But determining the transaction hash of a log requires the block body. The goal of this PR is postponing this retrieval until we are sure we the transaction hash. It happens often that the header bloom filter signals there might be matches in a block, but after actually checking them reveals the logs do not match. We want to avoid fetching the body in this case. Note that this changes the semantics of Backend.GetLogs. Downstream callers of GetLogs now assume log context fields have not been derived, and need to call DeriveFields on the logs if necessary. * eth/tracers: more fork overrides in traceBlockToFile (ethereum#26655) This change allows all post-Berlin forks to be specified as overrides for futureForkBlock in the config parameter for traceBlockToFile. * tests/fuzzers: supply gnark multiexp config, fixes ethereum#26669 (ethereum#26670) This change fixes a fuzzer which broke when we updated the gnark dependency earlier. * cmd/devp2p: reduce output of node crawler (ethereum#26674) Our discovery crawler spits out a huge amount of logs, most of which is pretty non-interesting. This change moves the very verbose output to Debug, and adds a 8-second status log message giving the general idea about what's going on. * params: update mainnet + rinkeby CHT (ethereum#26677) This change updates the CHT entries for mainnet and rinkeby * eth/filters: replace atomic pointer with value (ethereum#26689) * eth/filters: replace atomic.Pointer * fix * improve Co-authored-by: Martin Holst Swende <martin@swende.se> --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * p2p/dnsdisc: fix tests with Go 1.20 (ethereum#26690) * eth/catalyst: return error if withdrawals are nil post-shanghai (ethereum#26691) Spec: https://github.com/ethereum/execution-apis/blob/main/src/engine/shanghai.md#request * ethdb/pebble: Fix `MemTableStopWritesThreshold` (ethereum#26692) MemTableStopWritesThreshold was set to the max size of all memtables before blocking writing but should be set to the max number of memtables. This is documented [here](https://github.com/cockroachdb/pebble/blob/master/options.go#L738-L742). * eth/downloader: handle missing withdrawals if empty list is expected (ethereum#26675) This PR relaxes the block body ingress handling a bit: if block body withdrawals are missing (but expected to be empty), the body withdrawals are set to 'empty list' before being passed to upper layers. This fixes an issue where a block passed from EthereumJS to geth was deemed invalid. * params: go-ethereum v1.11.0 stable * params: begin v1.11.1 release cycle * travis, build: update Go to 1.20.1 (ethereum#26653) travis, build: update Go to 1.20 * core: check genesis state presence by disk read (ethereum#26703) * core, eth/downloader: make body validation more strict (ethereum#26704) * eth/downloader: fix empty-body case in queue fetchresult (ethereum#26707) * eth/downloader: fix typo (ethereum#26716) * all: remove deprecated uses of math.rand (ethereum#26710) This PR is a (superior) alternative to ethereum#26708, it handles deprecation, primarily two specific cases. `rand.Seed` is typically used in two ways - `rand.Seed(time.Now().UnixNano())` -- we seed it, just to be sure to get some random, and not always get the same thing on every run. This is not needed, with global seeding, so those are just removed. - `rand.Seed(1)` this is typically done to ensure we have a stable test. If we rely on this, we need to fix up the tests to use a deterministic prng-source. A few occurrences like this has been replaced with a proper custom source. `rand.Read` has been replaced by `crypto/rand`.`Read` in this PR. * params: go-ethereum v1.11.1 stable * params: begin v1.11.2 release cycle * eth/catalyst: send INVALID instead of INVALID_BLOCK_HASH (ethereum#26696) This change will break one hive test, but pass another and it will be the better way going forward * ci: disable coverage reporting in appveyor and travis * eth/catalyst: request too large error (ethereum#26722) The method `GetPayloadBodiesByRangeV1` now returns "-38004: Too large request" error if the requested range is too large, according to spec Co-authored-by: Martin Holst Swende <martin@swende.se> * core/trie: remove trie tracer (ethereum#26665) This PR contains a small portion of the full pbss PR, namely Remove the tracer from trie (and comitter), and instead using an accessList. Related changes to the Nodeset. --------- Co-authored-by: Gary Rong <garyrong0905@gmail.com> * rpc: fix unmarshaling of null result in CallContext (ethereum#26723) The change fixes unmarshaling of JSON null results into json.RawMessage. --------- Co-authored-by: Jason Yuan <jason.yuan@curvegrid.com> Co-authored-by: Jason Yuan <jason.yuan869@gmail.com> * build: ship bootstrapper Go along with builder for PPA (ethereum#26731) * build: fix setting env var, temp early exit * build: fix gobootstrap path for the PPA * build: add some PPA debug logs, sigh * internal/build: revert raising the chunk size for PPA * build: yet another weird PPA fix * build: fix (finaly?) the PPA env vars for Go bootstrapping * build: fix Go 1.19.0 bootstrapper issues on 386 PPA * build: enable Lunar Lobster PPA builds * Revert "core/trie: remove trie tracer (ethereum#26665)" (ethereum#26732) This reverts commit 7c749c9. * cmd/geth: clarify dumpconfig options (ethereum#26729) Clarifies the documentation around dumpconfi Signed-off-by: Sungwoo Kim <git@sung-woo.kim> * core, eth: merge snap-sync chain download progress logs (ethereum#26676) * core: fix accessor mismatch for genesis state (ethereum#26747) * core/rawdb: expose chain freezer constructor without internals (ethereum#26748) * all: use unified emptyRootHash and emptyCodeHash (ethereum#26718) The EmptyRootHash and EmptyCodeHash are defined everywhere in the codebase, this PR replaces all of them with unified one defined in core/types package, and also defines constants for TxRoot, WithdrawalsRoot and UncleRoot * eth/filters: fix a breaking change and return rpctransaction (ethereum#26757) * eth/filters: fix a breaking change and return rpctransaction * eth/filters: fix test cases --------- Co-authored-by: Catror <me@catror.com> * common/math: allow HexOrDecimal to accept unquoted decimals too (ethereum#26758) * params: release Geth v1.11.2 * params: begin v.1.11.3 release cycle * log: improve documentation (ethereum#26753) Add usage examples * core/rawdb, node: use standalone flock dependency (ethereum#26633) * eth: use the last announced finalized block as the sync ancient limit (ethereum#26685) * cmd/devp2p: faster crawling + less verbose dns updates (ethereum#26697) This improves the speed of DHT crawling by using concurrent requests. It also removes logging of individual DNS updates. * eth/tracers: add native flatCallTracer (aka parity style tracer) (ethereum#26377) Adds support for a native call tracer with the Parity format, which outputs call frames in a flat array. This tracer accepts the following options: - `convertParityErrors: true` will convert error messages to match those of Parity - `includePrecompiles: true` will report all calls to precompiles. The default matches Parity's behavior where CALL and STATICCALLs to precompiles are excluded Incompatibilities with Parity include: - Parity removes the result object in case of failure. This behavior is maintained with the exception of reverts. Revert output usually contains useful information, i.e. Solidity revert reason. - The `gasUsed` field accounts for intrinsic gas (e.g. 21000 for simple transfers) and refunds unlike Parity - Block rewards are not reported Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * core: improve withdrawal index assignment in GenerateChain (ethereum#26756) This fixes an issue where the withdrawal index was not calculated correctly for multiple withdrawals in a single block. Co-authored-by: Gary Rong <garyrong0905@gmail.com> Co-authored-by: Felix Lange <fjl@twurst.com> * ethdb/pebble: fix range compaction (ethereum#26771) * ethdb/pebble: fix range compaction * ethdb/pebble: add comment * ethdb/pebble: fix max memorytable size (ethereum#26776) * ethclient: include withdrawals in ethclient block responses (ethereum#26778) * include withdrawals in ethclient responses * omit empty withdrawals array in json serialization * all: change chain head markers from block to header (ethereum#26777) * core/rawdb, ethdb/pebble: disable pebble on openbsd (ethereum#26801) * core: fix a merge fault (ethereum#26802) * README, go.mod, event, internal/version: bump min Go to 1.19 (ethereum#26803) * travi: remove strange leftover Go version * core, params: schedule Shanghai on goerli (ethereum#26795) * core: params: schedule Shanghai on goerli * core/forkid: fix comment * eth: remove admin.peers[i].eth.head and difficulty (ethereum#26804) * core/types: add EffectiveGasPrice in Receipt (ethereum#26713) This change adds a struct field EffectiveGasPrice in types.Receipt. The field is present in RPC responses, but not in the Go struct, and thus can't easily be accessed via ethclient. Co-authored-by: PulsarAI <dev@pulsar-systems.fi> * core, eth/catalyst: fix race conditions in tests (ethereum#26790) Fixes a race in TestNewPayloadOnInvalidTerminalBlock where setting the TTD raced with the miner. Solution: set the TTD on the blockchain config not the genesis config. Also fixes a race in CopyHeader which resulted in race reports all over the place. * metrics: improve accuracy of CPU gauges (ethereum#26793) This PR changes metrics collection to actually measure the time interval between collections, rather than assume 3 seconds. I did some ad hoc profiling, and on slower hardware (eg, my Raspberry Pi 4) I routinely saw intervals between 3.3 - 3.5 seconds, with some being as high as 4.5 seconds. This will generally cause the CPU gauge readings to be too high, and in some cases can cause impossibly large values for the CPU load metrics (eg. greater than 400 for a 4 core CPU). --------- Co-authored-by: Felix Lange <fjl@twurst.com> * ethclient: fix panic when requesting missing blocks (ethereum#26817) This fixes a regression introduced by ethereum#26723. Fixes ethereum#26816. * core, miner: revert block gas counter in case of invalid transaction (ethereum#26799) This change fixes a flaw where, in certain scenarios, the block sealer did not accurately reset the remaining gas after failing to include an invalid transaction. Fixes ethereum#26791 * internal/ethapi: add tests for transaction types JSON marshal/unmarshal (ethereum#26667) Checks that Transaction.MarshalJSON and newRPCTransaction JSON output can be parsed by Transaction.UnmarshalJSON --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * cmd/evm: correct `alloc` for `t8n` testdata (ethereum#26822) Fixes a minor error in the testdata * eth/tracers/native: set created address to nil in case of failure (ethereum#26779) Fixes ethereum#26073 * accounts/usbwallet: mitigate ledger app chunking issue (ethereum#26773) This PR mitigates an issue with Ledger's on-device RLP deserialization, see LedgerHQ/app-ethereum#409 Ledger's RLP deserialization code does not validate the length of the RLP list received, and it may prematurely enter the signing flow when a APDU chunk boundary falls immediately before the EIP-155 chain_id when deserializing a transaction. Since the chain_id is uninitialized, it is 0 during this signing flow. This may cause the user to accidentally sign the transaction with chain_id = 0. That signature would be returned from the device 1 packet earlier than expected by the communication loop. The device blocks the second-to-last packet waiting for the signer flow, and then errors on the successive packet (which contains the chain_id, zeroed r, and zeroed s) Since the signature's early arrival causes successive errors during the communication process, geth does not parse the improper signature produced by the device, and therefore no improperly-signed transaction can be created. User funds are not at risk. We mitigate by selecting the highest chunk size that leaves at least 4 bytes in the final chunk. * beacon/engine: don't omit empty withdrawals in ExecutionPayloadBodies (ethereum#26698) This ensures the "withdrawals" field will always be present in responses to getPayloadBodiesByRangeV1 and getPayloadBodiesByHashV1. --------- Co-authored-by: Felix Lange <fjl@twurst.com> * build: update to go 1.20.2 (ethereum#26824) * params: go-ethereum v1.11.3 stable * params: begin v1.11.4 release cycle * core/rawdb: find smallest block stored in key-value store when chain gapped (ethereum#26719) This change prints out more information about the problem, in the case where geth detects a gap between leveldb and ancients, so we can determine more exactly where the gap is (what the first missing is). Also prints out more metadata. --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * signer/core: accept all solidity primitive types for EIP-712 signing (ethereum#26770) Accept all primitive types in Solidity for EIP-712 from intN, uintN, intN[], uintN[] for N as 0 to 256 in multiples of 8 --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * params: remove EF azure bootnodes (ethereum#26828) * core/vm: use golang native big.Int (ethereum#26834) reverts ethereum#26021, to use the upstream bigint instead. * core/vm: fix typo in comment (ethereum#26838) fixes eip 220 -> 2200 * core/forkid: fix issue in validation test (ethereum#26544) This changes the test to match the comment description. Using timestampedConfig in this test case is incorrect, the comment says 'local is at Gray Glacier' and isn't aware of more forks. * cmd/evm: update readmes for the tests (ethereum#26841) * core, core/types: plain Message struct (ethereum#25977) Here, the core.Message interface turns into a plain struct and types.Message gets removed. This is a breaking change to packages core and core/types. While we do not promise API stability for package core, we do for core/types. An exception can be made for types.Message, since it doesn't have any purpose apart from invoking the state transition in package core. types.Message was also marked deprecated by the same commit it got added in, 4dca5d4 (November 2016). The core.Message interface was added in December 2014, in commit db49417, for the purpose of 'testing' state transitions. It's the same change that made transaction struct fields private. Before that, the state transition used *types.Transaction directly. Over time, multiple implementations of the interface accrued across different packages, since constructing a Message is required whenever one wants to invoke the state transition. These implementations all looked very similar, a struct with private fields exposing the fields as accessor methods. By changing Message into a struct with public fields we can remove all these useless interface implementations. It will also hopefully simplify future changes to the type with less updates to apply across all of go-ethereum when a field is added to Message. --------- Co-authored-by: Felix Lange <fjl@twurst.com> * travis: only build PPAs nightly, not on every push, too heavy (ethereum#26846) * p2p: small comment typo (ethereum#26850) Update server.go * core: add Timestamp method in BlockGen (ethereum#26844) Since forks are now scheduled by block time, it can be necessary to check the timestamp of a block while generating transactions. * core/txpool: implement additional DoS defenses (ethereum#26648) This adds two new rules to the transaction pool: - A future transaction can not evict a pending transaction. - A transaction can not overspend available funds of a sender. --- Co-authored-by: dwn1998 <42262393+dwn1998@users.noreply.github.com> Co-authored-by: Martin Holst Swende <martin@swende.se> * params: go-ethereum v1.11.4 stable * params: begin v1.11.5 release cycle * tests: define `MuirGlacier` fork (ethereum#26856) add muir glacier to t8n * code/vm: fix comment typo (ethereum#26865) it should be constantinople rather than contantinople * core: minor code refactor (ethereum#26852) * core: refactor code * core: drop it from this anonymous goroutine func * core/txpool: use priceList.Put instead of heap.Push (ethereum#26863) Minor refactor to use the 'intended' accessor * eth: return error if 'safe' or 'finalized' tag used pre-merge (ethereum#26862) Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: Felix Lange <fjl@twurst.com> * .travis.yml: reenable PPA build on tag push (ethereum#26873) * core/state, trie: port changes from PBSS (ethereum#26763) * p2p/discover: pass invalid discv5 packets to Unhandled channel (ethereum#26699) This makes it possible to run another protocol alongside discv5, by reading unhandled packets from the channel. * all: update links in documentation (ethereum#26882) Co-authored-by: Stephen Flynn <stephen.flynn@gapac.com> * Increase websocket frame size (from erigon rpc client) (ethereum#26883) This increases the maximum allowed message size to 32MB. Originally submitted at erigontech/erigon#2739 example block failure: https://etherscan.io/tx/0x1317d973a55cedf9b0f2df6ea48e8077dd176f5444a3423368a46d6e4db89982#internal * cmd/devp2p, cmd/geth: add version in --help output (ethereum#26895) Not sure why this was removed, it's pretty useful to see the version also in --help. * core: show db error-info in case of mismatched hash root (ethereum#26870) When a database failure occurs, bubble it up a into statedb, and report it in suitable places, such as during a 'bad block' report. * consensus: improve consensus engine definition (ethereum#26871) Makes clear the distinction between Finalize and FinalizedAndAssemble: - In Finalize function, a series of state operations are applied according to consensus rules. The statedb is mutated and the root hash can be checked and compared afterwards. This function should be used in block processing(receive afrom network and apply it locally) but not block generation. - In FinalizeAndAssemble function, after applying state mutations, the block is also to be assembled with the latest state root computed, updating the header. This function should be used in block generation only. * eth/catalyst: increase update consensus timeout (ethereum#26840) Increases the time between consensus updates that we give the CL before we start warning the user. * internal/ethapi: avoid int overflow in GetTransactionReceipt (ethereum#26911) * trie, accounts/abi: add error-checks (ethereum#26914) * rlp: support for uint256 (ethereum#26898) This adds built-in support in package rlp for encoding, decoding and generating code dealing with uint256.Int. --------- Co-authored-by: Felix Lange <fjl@twurst.com> * eth: fix output file permissions in admin_exportChain (ethereum#26912) * api: Use 0700 file permissions for ExportChain * change perm to 0644 * Update api.go --------- Co-authored-by: Felix Lange <fjl@twurst.com> * trie: reduce unit test time (ethereum#26918) * core/txpool: use atomic int added in go1.19 (ethereum#26913) Makes use of atomic.Uint64 instead of atomic by pointer * params: schedule shanghai fork on mainnet (ethereum#26908) Schedules the shanghai hardfork on timestamp 1681338455 as discussed on ACDE 157: ethereum/execution-specs#727 * core/txpool: allow future local transactions (ethereum#26930) Local transactions should not be subject to the "future shouldn't churn pending txs" rule * params: go-ethereum v1.11.5 stable * params: begin v1.11.6 release cycle * build: allow building nightly archives via cron jobs (ethereum#26938) * log: add special casing of uint256 into the logger (ethereum#26936) * core/rawdb: use atomic int added in go1.19 (ethereum#26935) * core/vm: expose jumptable constructors (ethereum#26880) When interacting with geth as a library to e.g. produce state tests, it is desirable to obtain the consensus-correct jumptable definition for a given fork. This changes adds accessors so the instructionset can be obtained and characteristics about opcodes can be inspected. * eth/catalyst: fix races (ethereum#26950) * core/rawdb: update freezertable read meter (ethereum#26946) The meter for "for measuring the effective amount of data read" within the freezertable was never updated. This change remedies that. --------- Signed-off-by: jsvisa <delweng@gmail.com> * cmd/evm, tests: record preimages if dump is expected (ethereum#26955) With ethereum#25287 we made it so that preimages were not recorded by default. This had the side effect that the evm command is no longer able to dump state since it does a preimage lookup to determine the address represented by a key. This change enables the recording of preimages when the dump command is given. * core/state: add account address to Trie slot accessors (ethereum#26934) This changes the Trie interface to add the plain account address as a parameter to all storage-related methods. After the introduction of the TryAccount* functions, TryGet, TryUpdate and TryDelete are now only meant to read an account's storage. In their current form, they assume that an account storage is stored in a separate trie, and that the hashing of the slot is independent of its account's address. The proposed structure for a stateless storage breaks these two assumptions: the hashing of a slot key requires the address and all slots and accounts are stored in a single trie. This PR therefore adds an address parameter to the interface. It is ignored in the MPT version, so this change has no functional impact, however it will reduce the diff size when merging verkle trees. * metrics: add cpu counters (ethereum#26796) This PR adds counter metrics for the CPU system and the Geth process. Currently the only metrics available for these items are gauges. Gauges are fine when the consumer scrapes metrics data at the same interval as Geth produces new values (every 3 seconds), but it is likely that most consumers will not scrape that often. Intervals of 10, 15, or maybe even 30 seconds are probably more common. So the problem is, how does the consumer estimate what the CPU was doing in between scrapes. With a counter, it's easy ... you just subtract two successive values and divide by the time to get a nice, accurate average. But with a gauge, you can't do that. A gauge reading is an instantaneous picture of what was happening at that moment, but it gives you no idea about what was going on between scrapes. Taking an average of values is meaningless. * metrics/influxdb: use smaller dependency and reuse code between v1 and v2 reporters (ethereum#26963) This change switches to use the smaller influxdata/influxdb1-client package instead of depending on the whole infuxdb package. The new smaller client is very similar to the influxdb-v2 client, which made it possible to refactor the two reporters to reuse code a lot more. * eth/gasprice: change feehistory input type from int to uint64 (ethereum#26922) Change input param type from int to uint64 * go.mod: update golang.org/x/tools (ethereum#26960) * rlp/rlpgen: print want/expect output string if mismatch (ethereum#26932) Co-authored-by: Marius van der Wijden <m.vanderwijden@live.de> * ethclient: ensure returned subscription is nil on error (ethereum#26976) * core/state, trie: remove Try prefix in Trie accessors (ethereum#26975) This change renames StateTrie methods to remove the Try* prefix. We added the Trie methods with prefix 'Try' a long time ago, working around the problem that most existing methods of Trie did not return the database error. This weird naming convention has persisted until now. Co-authored-by: Gary Rong <garyrong0905@gmail.com> * metrics/librato: ensure resp.body closed (ethereum#26969) This change ensures that we call Close on a http response body, in various places in the source code (mostly tests) * core/vm: use atomic.Bool (ethereum#26951) Make use of new atomic types --------- Co-authored-by: Felix Lange <fjl@twurst.com> Co-authored-by: Martin Holst Swende <martin@swende.se> * core/bloombits: use atomic type (ethereum#26993) * core/state: use atomic.Bool (ethereum#26992) * graphql: fix data races (ethereum#26965) Fixes multiple data races caused by the fact that resolving fields are done concurrently by the graphql library. It also enforces caching at the stateobject level for account fields. * eth/tracers/native: prevent panic for LOG edge-cases (ethereum#26848) This PR fixes OOM panic in the callTracer as well as panicing on opcode validation errors (e.g. stack underflow) in callTracer and prestateTracer. Co-authored-by: Martin Holst Swende <martin@swende.se> * internal/debug: add log.logfmt flag to set logging to use logfmt (ethereum#26970) * docs: update outdated DeriveSha docs comment (ethereum#26968) * remove @gballet as a GraphQL codeowner (ethereum#27012) * core: use atomic type (ethereum#27011) * graphql: revert storage access regression (ethereum#27007) * cmd/geth: Add `--log.format` cli param (ethereum#27001) Removes the new --log.logfmt directive and hides --log.json, replacing both with log.format=(json|logfmt|terminal). The hidden log.json option is still respected if log.format is not specified for backwards compatibility. Co-authored-by: Martin Holst Swende <martin@swende.se> * ethdb/pebble: use atomic type (ethereum#27014) * common: fix json marshaller MixedcaseAddress (ethereum#26998) Fix the json marshaller of MixedcaseAddress * eth/catalyst: improve consensus heartbeat (ethereum#26896) improve the heartbeat function that is no longer suitable in the current situation Co-authored-by: “openex27” <“openexkevin@gmail.com”> * miner: use atomic type (ethereum#27013) Use the new typed atomics in the miner package * accounts/abi/bind: handle UnpackLog with zero topics (ethereum#26920) Adds error handling for the case that UnpackLog or UnpackLogIntoMap is called with a log that has zero topics. --------- Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * cmd/evm: use correct parent number for t8n base fee calculation (ethereum#27032) Currently the t8n tool uses the same block number for the current block and its parent while calculating the base fee. This causes incorrect base fee calculation for the london fork block. This commit sets the parent block number to be one less than the current block number * go.mod : update snappy (ethereum#27027) * common: delete MakeName (ethereum#27023) common,p2p: remove unused function MakeName * cmd/geth: enable log rotation (ethereum#26843) This change enables log rotation, which can be activated using the flag --log.rotate. Additional parameters that can be given are: - log.maxsize to set maximum size before files are rotated, - log.maxbackups to set how many files are retailed, - log.maxage to configure max age of rotated files, - log.compress whether to compress rotated files The way to configure location of the logfile(s) is left unchanged, via the `log.logfile` parameter. --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * cmd, miner, signer: avoid panic if keystore is not available (ethereum#27039) * cmd, miner, singer: avoid panic if keystore is not available * cmd/geth: print warning instead of panic * test/fuzzers: fuzz rlp handling of big.Int and uint256.Int (ethereum#26917) test/fuzzers: fuzz rlp handling of big.Lnt and uint256.Int * core/txpool: move some validation to outside of mutex (ethereum#27006) Currently, most of transaction validation while holding the txpool mutex: one exception being an early-on signature check. This PR changes that, so that we do all non-stateful checks before we entering the mutex area. This means they can be performed in parallel, and to enable that, certain fields have been made atomic bools and uint64. * eth/downloader: use atomic types (ethereum#27030) * eth/downloader: use atomic type * Update eth/downloader/downloader_test.go Co-authored-by: Martin Holst Swende <martin@swende.se> * Update eth/downloader/downloader_test.go Co-authored-by: Martin Holst Swende <martin@swende.se> --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * core/vm: clarify comment (ethereum#27045) * consensus, core/typer: add 4844 excessDataGas to header, tie it to Cancun (ethereum#27046) * consensus/misc, params: add EIP-4844 blobfee conversions (ethereum#27041) * consensus/misc, params: add EIP-4844 blobfee conversions * consensus/misc: pull in fakeExponential test cases * consensus/misc: reuse bigints * consensus/misc: nit renames, additional larger testcase --------- Co-authored-by: Roberto Bayardo <bayardo@alum.mit.edu> Co-authored-by: Martin Holst Swende <martin@swende.se> * eth/tracers: report correct gasLimit in call tracers (ethereum#27029) This includes a semantic change to the `callTracer` as well as `flatCallTracer`. The value of field `gas` in the **first** call frame will change as follows: - It previously contained gas available after initial deductions (i.e. tx costs) - It will now contain the full tx gasLimit value Signed-off-by: jsvisa <delweng@gmail.com> * all: remove debug-field from vm config (ethereum#27048) This PR removes the Debug field from vmconfig, making it so that if a tracer is set, debug=true is implied. --------- Co-authored-by: 0xTylerHolmes <tyler@ethereum.org> Co-authored-by: Sina Mahmoodi <1591639+s1na@users.noreply.github.com> * metrics: make gauge_float64 and counter_float64 lock free (ethereum#27025) Makes the float-gauges lock-free name old time/op new time/op delta CounterFloat64Parallel-8 1.45µs ±10% 0.85µs ± 6% -41.65% (p=0.008 n=5+5) --------- Co-authored-by: Exca-DK <dev@DESKTOP-RI45P4J.localdomain> Co-authored-by: Martin Holst Swende <martin@swende.se> * eth/tracers: use atomic type (ethereum#27031) Use the new atomic types in package eth/tracers --------- Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * build: upgrade -dlgo version to Go 1.20.3 * core/txpool: disallow future churn by remote txs (ethereum#26907) Prior to this change, it was possible that transactions are erroneously deemed as 'future' although they are in fact 'pending', causing them to be dropped due to 'future' not being allowed to replace 'pending'. This change fixes that, by doing a more in-depth inspection of the queue. * core, miner: drop transactions from the same sender when error occurs (ethereum#27038) This PR unifies the error handling in miner. Whenever an error occur while applying a transaction, the transaction should be regarded as invalid and all following transactions from the same sender not executable because of the nonce restriction. The only exception is the `nonceTooLow` error which is handled separately. * params: new sepolia bootnodes (ethereum#27099) New sepolia bootnodes managed by EF devops * cmd/devp2p: fix erroneous log output in crawler (ethereum#27089) cmd/devp2p: fix log of ignored recent nodes counter * signer/core: rename testdata files (ethereum#27063) Sets a meaningful name on test-files * core: fix comment to reflect function name (ethereum#27070) * params: remove `EIP150Hash` from chainconfig (ethereum#27087) 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> * p2p: access embedded fields of Server directly (ethereum#27078) * consensus/ethash: use atomic type (ethereum#27068) * cmd/devp2p: make crawler-route53-updater less verbose (ethereum#27116) Follow-up to ethereum#26697, makes the crawler less verbose on route53-based scenarios. It also changes the loglevel from debug to info on Updates, which are typically the root, and can be interesting to see. * cmd/geth: rename --vmodule to --log.vmodule (ethereum#27071) renames `--vmodule` to `--log.vmodule`, and prints a warning if the old form is used. * core/vm: order opcodes properly (ethereum#27113) * metrics: use atomic type (ethereum#27121) * all: refactor trie API (ethereum#26995) 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. * params: go-ethereum v1.11.6 stable * fix merge conflicts * add market gas price to simulate duplicate nonce behavior (#7) --------- Signed-off-by: Sungwoo Kim <git@sung-woo.kim> Signed-off-by: jsvisa <delweng@gmail.com> Co-authored-by: Seungbae Yu <dbadoy4874@gmail.com> Co-authored-by: Ikko Eltociear Ashimine <eltociear@gmail.com> Co-authored-by: Marius Kjærstad <sandakersmann@users.noreply.github.com> Co-authored-by: ucwong <ucwong@126.com> Co-authored-by: Paul <41552663+molecula451@users.noreply.github.com> Co-authored-by: Marius van der Wijden <m.vanderwijden@live.de> Co-authored-by: Zachinquarantine <Zachinquarantine@protonmail.com> Co-authored-by: Sina Mahmoodi <1591639+s1na@users.noreply.github.com> Co-authored-by: meehow <michal@ert.pl> Co-authored-by: rjl493456442 <garyrong0905@gmail.com> Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: Felix Lange <fjl@twurst.com> Co-authored-by: Shude Li <islishude@gmail.com> Co-authored-by: lightclient@protonmail.com <lightclient@protonmail.com> Co-authored-by: marioevz <marioevz@gmail.com> Co-authored-by: Nicola Cocchiaro <3538109+ncocchiaro@users.noreply.github.com> Co-authored-by: Nicola Cocchiaro <ncocchiaro@users.noreply.github.com> Co-authored-by: lightclient <14004106+lightclient@users.noreply.github.com> Co-authored-by: Pascal Marco Caversaccio <pcaversaccio@users.noreply.github.com> Co-authored-by: Mio <mshimmeris@gmail.com> Co-authored-by: Péter Szilágyi <peterke@gmail.com> Co-authored-by: lmittmann <3458786+lmittmann@users.noreply.github.com> Co-authored-by: delihiros <delihiros@gmail.com> Co-authored-by: raulk <raul.kripalani@gmail.com> Co-authored-by: Felföldi Zsolt <zsfelfoldi@gmail.com> Co-authored-by: zhiqiangxu <652732310@qq.com> Co-authored-by: halilylm <65048618+halilylm@users.noreply.github.com> Co-authored-by: Halil Yildirim <halilgolang@gmail.com> Co-authored-by: jwasinger <j-wasinger@hotmail.com> Co-authored-by: Chris Ziogas <ziogaschr@gmail.com> Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> Co-authored-by: Darioush Jalali <darioush.jalali@avalabs.org> Co-authored-by: Patrick O'Grady <prohb125@gmail.com> Co-authored-by: Roman Krasiuk <rokrassyuk@gmail.com> Co-authored-by: Jason Yuan <jason.yuan@curvegrid.com> Co-authored-by: Jason Yuan <jason.yuan869@gmail.com> Co-authored-by: Sungwoo Kim <git@sung-woo.kim> Co-authored-by: Yier <90763233+yierx@users.noreply.github.com> Co-authored-by: Catror <me@catror.com> Co-authored-by: Nate Armstrong <naterarmstrong@gmail.com> Co-authored-by: Dan Cline <6798349+Rjected@users.noreply.github.com> Co-authored-by: Peter (bitfly) <1674920+peterbitfly@users.noreply.github.com> Co-authored-by: PulsarAI <dev@pulsar-systems.fi> Co-authored-by: turboboost55 <7891737+turboboost55@users.noreply.github.com> Co-authored-by: Adrian Sutton <adrian@symphonious.net> Co-authored-by: Guruprasad Kamath <48196632+gurukamath@users.noreply.github.com> Co-authored-by: James Prestwich <10149425+prestwich@users.noreply.github.com> Co-authored-by: Daniel Fernandes <711733+daferna@users.noreply.github.com> Co-authored-by: Rafael Matias <rafael@skyle.net> Co-authored-by: xiyang <90125263+JBossBC@users.noreply.github.com> Co-authored-by: Roberto Bayardo <bayardo@alum.mit.edu> Co-authored-by: panicalways <113693386+panicalways@users.noreply.github.com> Co-authored-by: dwn1998 <42262393+dwn1998@users.noreply.github.com> Co-authored-by: s7v7nislands <s7v7nislands@gmail.com> Co-authored-by: Stephen Flynn <ssflynn@gmail.com> Co-authored-by: Stephen Flynn <stephen.flynn@gapac.com> Co-authored-by: Jonathan Otto <jonathan.otto@gmail.com> Co-authored-by: Delweng <delweng@gmail.com> Co-authored-by: Guillaume Ballet <3272758+gballet@users.noreply.github.com> Co-authored-by: aaronbuchwald <aaron.buchwald56@gmail.com> Co-authored-by: ucwong <ethereum2k@gmail.com> Co-authored-by: norwnd <112318969+norwnd@users.noreply.github.com> Co-authored-by: Adrian Sutton <adrian@oplabs.co> Co-authored-by: David Murdoch <187813+davidmurdoch@users.noreply.github.com> Co-authored-by: openex <openexkevin@gmail.com> Co-authored-by: “openex27” <“openexkevin@gmail.com”> Co-authored-by: sudeep <sudeepdino008@gmail.com> Co-authored-by: joohhnnn <68833933+joohhnnn@users.noreply.github.com> Co-authored-by: 0xTylerHolmes <tyler@ethereum.org> Co-authored-by: Exca-DK <85954505+Exca-DK@users.noreply.github.com> Co-authored-by: Exca-DK <dev@DESKTOP-RI45P4J.localdomain> Co-authored-by: Marius Kjærstad <mkjaerstad@protonmail.com> Co-authored-by: Parithosh Jayanthi <parithosh@indenwolken.xyz> Co-authored-by: noel <72006780+0x00Duke@users.noreply.github.com> Co-authored-by: Taeguk Kwon <xornrbboy@gmail.com> Co-authored-by: Anusha <63559942+anusha-ctrl@users.noreply.github.com> Co-authored-by: Alex Beregszaszi <alex@rtfs.hu> Co-authored-by: susannapaxos <86749396+susannapaxos@users.noreply.github.com>

Currently, most of transaction validation while holding the txpool mutex: one exception being an early-on signature check. This PR changes that, so that we do all non-stateful checks before we entering the mutex area. This means they can be performed in parallel, and to enable that, certain fields have been made atomic bools and uint64.

@gballet

* ethdb/pebble: fix nil callbacks (ethereum#26650) * eth/downloader: fix timeout resurrection panic (ethereum#26652) * common/prque, eth/downloader: fix timeout resurrection panic * common/prque: revert -1 hack for les, temporaryly! * core/state, trie: remove unused error-return from trie Commit operation (ethereum#26641) * go.mod: update pebble to latest master (ethereum#26654) * core/vm: set tracer-observable `value` of a delegatecall to match parent `value` (ethereum#26632) This is a breaking change in the tracing hooks API as well as semantics of the callTracer: - CaptureEnter hook provided a nil value argument in case of DELEGATECALL. However to stay consistent with how delegate calls behave in EVM this hook is changed to pass in the value of the parent call. - callTracer will return parent call's value for DELEGATECALL frames. --------- Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * ethdb: add benchmark test suite (ethereum#26659) * params: schedule shanghai fork on sepolia (ethereum#26662) * params: schedule shanghai fork on sepolia * params: u64 -> newUint64 * eth/filters: avoid block body retrieval when no matching logs (ethereum#25199) Logs stored on disk have minimal information. Contextual information such as block number, index of log in block, index of transaction in block are filled in upon request. We can fill in all these fields only having the block header and list of receipts. But determining the transaction hash of a log requires the block body. The goal of this PR is postponing this retrieval until we are sure we the transaction hash. It happens often that the header bloom filter signals there might be matches in a block, but after actually checking them reveals the logs do not match. We want to avoid fetching the body in this case. Note that this changes the semantics of Backend.GetLogs. Downstream callers of GetLogs now assume log context fields have not been derived, and need to call DeriveFields on the logs if necessary. * eth/tracers: more fork overrides in traceBlockToFile (ethereum#26655) This change allows all post-Berlin forks to be specified as overrides for futureForkBlock in the config parameter for traceBlockToFile. * tests/fuzzers: supply gnark multiexp config, fixes ethereum#26669 (ethereum#26670) This change fixes a fuzzer which broke when we updated the gnark dependency earlier. * cmd/devp2p: reduce output of node crawler (ethereum#26674) Our discovery crawler spits out a huge amount of logs, most of which is pretty non-interesting. This change moves the very verbose output to Debug, and adds a 8-second status log message giving the general idea about what's going on. * params: update mainnet + rinkeby CHT (ethereum#26677) This change updates the CHT entries for mainnet and rinkeby * eth/filters: replace atomic pointer with value (ethereum#26689) * eth/filters: replace atomic.Pointer * fix * improve Co-authored-by: Martin Holst Swende <martin@swende.se> --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * p2p/dnsdisc: fix tests with Go 1.20 (ethereum#26690) * eth/catalyst: return error if withdrawals are nil post-shanghai (ethereum#26691) Spec: https://github.com/ethereum/execution-apis/blob/main/src/engine/shanghai.md#request * ethdb/pebble: Fix `MemTableStopWritesThreshold` (ethereum#26692) MemTableStopWritesThreshold was set to the max size of all memtables before blocking writing but should be set to the max number of memtables. This is documented [here](https://github.com/cockroachdb/pebble/blob/master/options.go#L738-L742). * eth/downloader: handle missing withdrawals if empty list is expected (ethereum#26675) This PR relaxes the block body ingress handling a bit: if block body withdrawals are missing (but expected to be empty), the body withdrawals are set to 'empty list' before being passed to upper layers. This fixes an issue where a block passed from EthereumJS to geth was deemed invalid. * params: go-ethereum v1.11.0 stable * params: begin v1.11.1 release cycle * travis, build: update Go to 1.20.1 (ethereum#26653) travis, build: update Go to 1.20 * core: check genesis state presence by disk read (ethereum#26703) * core, eth/downloader: make body validation more strict (ethereum#26704) * eth/downloader: fix empty-body case in queue fetchresult (ethereum#26707) * eth/downloader: fix typo (ethereum#26716) * all: remove deprecated uses of math.rand (ethereum#26710) This PR is a (superior) alternative to ethereum#26708, it handles deprecation, primarily two specific cases. `rand.Seed` is typically used in two ways - `rand.Seed(time.Now().UnixNano())` -- we seed it, just to be sure to get some random, and not always get the same thing on every run. This is not needed, with global seeding, so those are just removed. - `rand.Seed(1)` this is typically done to ensure we have a stable test. If we rely on this, we need to fix up the tests to use a deterministic prng-source. A few occurrences like this has been replaced with a proper custom source. `rand.Read` has been replaced by `crypto/rand`.`Read` in this PR. * params: go-ethereum v1.11.1 stable * params: begin v1.11.2 release cycle * eth/catalyst: send INVALID instead of INVALID_BLOCK_HASH (ethereum#26696) This change will break one hive test, but pass another and it will be the better way going forward * ci: disable coverage reporting in appveyor and travis * eth/catalyst: request too large error (ethereum#26722) The method `GetPayloadBodiesByRangeV1` now returns "-38004: Too large request" error if the requested range is too large, according to spec Co-authored-by: Martin Holst Swende <martin@swende.se> * core/trie: remove trie tracer (ethereum#26665) This PR contains a small portion of the full pbss PR, namely Remove the tracer from trie (and comitter), and instead using an accessList. Related changes to the Nodeset. --------- Co-authored-by: Gary Rong <garyrong0905@gmail.com> * rpc: fix unmarshaling of null result in CallContext (ethereum#26723) The change fixes unmarshaling of JSON null results into json.RawMessage. --------- Co-authored-by: Jason Yuan <jason.yuan@curvegrid.com> Co-authored-by: Jason Yuan <jason.yuan869@gmail.com> * build: ship bootstrapper Go along with builder for PPA (ethereum#26731) * build: fix setting env var, temp early exit * build: fix gobootstrap path for the PPA * build: add some PPA debug logs, sigh * internal/build: revert raising the chunk size for PPA * build: yet another weird PPA fix * build: fix (finaly?) the PPA env vars for Go bootstrapping * build: fix Go 1.19.0 bootstrapper issues on 386 PPA * build: enable Lunar Lobster PPA builds * Revert "core/trie: remove trie tracer (ethereum#26665)" (ethereum#26732) This reverts commit 7c749c9. * cmd/geth: clarify dumpconfig options (ethereum#26729) Clarifies the documentation around dumpconfi Signed-off-by: Sungwoo Kim <git@sung-woo.kim> * core, eth: merge snap-sync chain download progress logs (ethereum#26676) * core: fix accessor mismatch for genesis state (ethereum#26747) * core/rawdb: expose chain freezer constructor without internals (ethereum#26748) * all: use unified emptyRootHash and emptyCodeHash (ethereum#26718) The EmptyRootHash and EmptyCodeHash are defined everywhere in the codebase, this PR replaces all of them with unified one defined in core/types package, and also defines constants for TxRoot, WithdrawalsRoot and UncleRoot * eth/filters: fix a breaking change and return rpctransaction (ethereum#26757) * eth/filters: fix a breaking change and return rpctransaction * eth/filters: fix test cases --------- Co-authored-by: Catror <me@catror.com> * common/math: allow HexOrDecimal to accept unquoted decimals too (ethereum#26758) * params: release Geth v1.11.2 * params: begin v.1.11.3 release cycle * log: improve documentation (ethereum#26753) Add usage examples * core/rawdb, node: use standalone flock dependency (ethereum#26633) * eth: use the last announced finalized block as the sync ancient limit (ethereum#26685) * cmd/devp2p: faster crawling + less verbose dns updates (ethereum#26697) This improves the speed of DHT crawling by using concurrent requests. It also removes logging of individual DNS updates. * eth/tracers: add native flatCallTracer (aka parity style tracer) (ethereum#26377) Adds support for a native call tracer with the Parity format, which outputs call frames in a flat array. This tracer accepts the following options: - `convertParityErrors: true` will convert error messages to match those of Parity - `includePrecompiles: true` will report all calls to precompiles. The default matches Parity's behavior where CALL and STATICCALLs to precompiles are excluded Incompatibilities with Parity include: - Parity removes the result object in case of failure. This behavior is maintained with the exception of reverts. Revert output usually contains useful information, i.e. Solidity revert reason. - The `gasUsed` field accounts for intrinsic gas (e.g. 21000 for simple transfers) and refunds unlike Parity - Block rewards are not reported Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * core: improve withdrawal index assignment in GenerateChain (ethereum#26756) This fixes an issue where the withdrawal index was not calculated correctly for multiple withdrawals in a single block. Co-authored-by: Gary Rong <garyrong0905@gmail.com> Co-authored-by: Felix Lange <fjl@twurst.com> * ethdb/pebble: fix range compaction (ethereum#26771) * ethdb/pebble: fix range compaction * ethdb/pebble: add comment * ethdb/pebble: fix max memorytable size (ethereum#26776) * ethclient: include withdrawals in ethclient block responses (ethereum#26778) * include withdrawals in ethclient responses * omit empty withdrawals array in json serialization * all: change chain head markers from block to header (ethereum#26777) * core/rawdb, ethdb/pebble: disable pebble on openbsd (ethereum#26801) * core: fix a merge fault (ethereum#26802) * README, go.mod, event, internal/version: bump min Go to 1.19 (ethereum#26803) * travi: remove strange leftover Go version * core, params: schedule Shanghai on goerli (ethereum#26795) * core: params: schedule Shanghai on goerli * core/forkid: fix comment * eth: remove admin.peers[i].eth.head and difficulty (ethereum#26804) * core/types: add EffectiveGasPrice in Receipt (ethereum#26713) This change adds a struct field EffectiveGasPrice in types.Receipt. The field is present in RPC responses, but not in the Go struct, and thus can't easily be accessed via ethclient. Co-authored-by: PulsarAI <dev@pulsar-systems.fi> * core, eth/catalyst: fix race conditions in tests (ethereum#26790) Fixes a race in TestNewPayloadOnInvalidTerminalBlock where setting the TTD raced with the miner. Solution: set the TTD on the blockchain config not the genesis config. Also fixes a race in CopyHeader which resulted in race reports all over the place. * metrics: improve accuracy of CPU gauges (ethereum#26793) This PR changes metrics collection to actually measure the time interval between collections, rather than assume 3 seconds. I did some ad hoc profiling, and on slower hardware (eg, my Raspberry Pi 4) I routinely saw intervals between 3.3 - 3.5 seconds, with some being as high as 4.5 seconds. This will generally cause the CPU gauge readings to be too high, and in some cases can cause impossibly large values for the CPU load metrics (eg. greater than 400 for a 4 core CPU). --------- Co-authored-by: Felix Lange <fjl@twurst.com> * ethclient: fix panic when requesting missing blocks (ethereum#26817) This fixes a regression introduced by ethereum#26723. Fixes ethereum#26816. * core, miner: revert block gas counter in case of invalid transaction (ethereum#26799) This change fixes a flaw where, in certain scenarios, the block sealer did not accurately reset the remaining gas after failing to include an invalid transaction. Fixes ethereum#26791 * internal/ethapi: add tests for transaction types JSON marshal/unmarshal (ethereum#26667) Checks that Transaction.MarshalJSON and newRPCTransaction JSON output can be parsed by Transaction.UnmarshalJSON --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * cmd/evm: correct `alloc` for `t8n` testdata (ethereum#26822) Fixes a minor error in the testdata * eth/tracers/native: set created address to nil in case of failure (ethereum#26779) Fixes ethereum#26073 * accounts/usbwallet: mitigate ledger app chunking issue (ethereum#26773) This PR mitigates an issue with Ledger's on-device RLP deserialization, see LedgerHQ/app-ethereum#409 Ledger's RLP deserialization code does not validate the length of the RLP list received, and it may prematurely enter the signing flow when a APDU chunk boundary falls immediately before the EIP-155 chain_id when deserializing a transaction. Since the chain_id is uninitialized, it is 0 during this signing flow. This may cause the user to accidentally sign the transaction with chain_id = 0. That signature would be returned from the device 1 packet earlier than expected by the communication loop. The device blocks the second-to-last packet waiting for the signer flow, and then errors on the successive packet (which contains the chain_id, zeroed r, and zeroed s) Since the signature's early arrival causes successive errors during the communication process, geth does not parse the improper signature produced by the device, and therefore no improperly-signed transaction can be created. User funds are not at risk. We mitigate by selecting the highest chunk size that leaves at least 4 bytes in the final chunk. * beacon/engine: don't omit empty withdrawals in ExecutionPayloadBodies (ethereum#26698) This ensures the "withdrawals" field will always be present in responses to getPayloadBodiesByRangeV1 and getPayloadBodiesByHashV1. --------- Co-authored-by: Felix Lange <fjl@twurst.com> * build: update to go 1.20.2 (ethereum#26824) * params: go-ethereum v1.11.3 stable * params: begin v1.11.4 release cycle * core/rawdb: find smallest block stored in key-value store when chain gapped (ethereum#26719) This change prints out more information about the problem, in the case where geth detects a gap between leveldb and ancients, so we can determine more exactly where the gap is (what the first missing is). Also prints out more metadata. --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * signer/core: accept all solidity primitive types for EIP-712 signing (ethereum#26770) Accept all primitive types in Solidity for EIP-712 from intN, uintN, intN[], uintN[] for N as 0 to 256 in multiples of 8 --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * params: remove EF azure bootnodes (ethereum#26828) * core/vm: use golang native big.Int (ethereum#26834) reverts ethereum#26021, to use the upstream bigint instead. * core/vm: fix typo in comment (ethereum#26838) fixes eip 220 -> 2200 * core/forkid: fix issue in validation test (ethereum#26544) This changes the test to match the comment description. Using timestampedConfig in this test case is incorrect, the comment says 'local is at Gray Glacier' and isn't aware of more forks. * cmd/evm: update readmes for the tests (ethereum#26841) * core, core/types: plain Message struct (ethereum#25977) Here, the core.Message interface turns into a plain struct and types.Message gets removed. This is a breaking change to packages core and core/types. While we do not promise API stability for package core, we do for core/types. An exception can be made for types.Message, since it doesn't have any purpose apart from invoking the state transition in package core. types.Message was also marked deprecated by the same commit it got added in, 4dca5d4 (November 2016). The core.Message interface was added in December 2014, in commit db49417, for the purpose of 'testing' state transitions. It's the same change that made transaction struct fields private. Before that, the state transition used *types.Transaction directly. Over time, multiple implementations of the interface accrued across different packages, since constructing a Message is required whenever one wants to invoke the state transition. These implementations all looked very similar, a struct with private fields exposing the fields as accessor methods. By changing Message into a struct with public fields we can remove all these useless interface implementations. It will also hopefully simplify future changes to the type with less updates to apply across all of go-ethereum when a field is added to Message. --------- Co-authored-by: Felix Lange <fjl@twurst.com> * travis: only build PPAs nightly, not on every push, too heavy (ethereum#26846) * p2p: small comment typo (ethereum#26850) Update server.go * core: add Timestamp method in BlockGen (ethereum#26844) Since forks are now scheduled by block time, it can be necessary to check the timestamp of a block while generating transactions. * core/txpool: implement additional DoS defenses (ethereum#26648) This adds two new rules to the transaction pool: - A future transaction can not evict a pending transaction. - A transaction can not overspend available funds of a sender. --- Co-authored-by: dwn1998 <42262393+dwn1998@users.noreply.github.com> Co-authored-by: Martin Holst Swende <martin@swende.se> * params: go-ethereum v1.11.4 stable * params: begin v1.11.5 release cycle * tests: define `MuirGlacier` fork (ethereum#26856) add muir glacier to t8n * code/vm: fix comment typo (ethereum#26865) it should be constantinople rather than contantinople * core: minor code refactor (ethereum#26852) * core: refactor code * core: drop it from this anonymous goroutine func * core/txpool: use priceList.Put instead of heap.Push (ethereum#26863) Minor refactor to use the 'intended' accessor * eth: return error if 'safe' or 'finalized' tag used pre-merge (ethereum#26862) Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: Felix Lange <fjl@twurst.com> * .travis.yml: reenable PPA build on tag push (ethereum#26873) * core/state, trie: port changes from PBSS (ethereum#26763) * p2p/discover: pass invalid discv5 packets to Unhandled channel (ethereum#26699) This makes it possible to run another protocol alongside discv5, by reading unhandled packets from the channel. * all: update links in documentation (ethereum#26882) Co-authored-by: Stephen Flynn <stephen.flynn@gapac.com> * Increase websocket frame size (from erigon rpc client) (ethereum#26883) This increases the maximum allowed message size to 32MB. Originally submitted at erigontech/erigon#2739 example block failure: https://etherscan.io/tx/0x1317d973a55cedf9b0f2df6ea48e8077dd176f5444a3423368a46d6e4db89982#internal * cmd/devp2p, cmd/geth: add version in --help output (ethereum#26895) Not sure why this was removed, it's pretty useful to see the version also in --help. * core: show db error-info in case of mismatched hash root (ethereum#26870) When a database failure occurs, bubble it up a into statedb, and report it in suitable places, such as during a 'bad block' report. * consensus: improve consensus engine definition (ethereum#26871) Makes clear the distinction between Finalize and FinalizedAndAssemble: - In Finalize function, a series of state operations are applied according to consensus rules. The statedb is mutated and the root hash can be checked and compared afterwards. This function should be used in block processing(receive afrom network and apply it locally) but not block generation. - In FinalizeAndAssemble function, after applying state mutations, the block is also to be assembled with the latest state root computed, updating the header. This function should be used in block generation only. * eth/catalyst: increase update consensus timeout (ethereum#26840) Increases the time between consensus updates that we give the CL before we start warning the user. * internal/ethapi: avoid int overflow in GetTransactionReceipt (ethereum#26911) * trie, accounts/abi: add error-checks (ethereum#26914) * rlp: support for uint256 (ethereum#26898) This adds built-in support in package rlp for encoding, decoding and generating code dealing with uint256.Int. --------- Co-authored-by: Felix Lange <fjl@twurst.com> * eth: fix output file permissions in admin_exportChain (ethereum#26912) * api: Use 0700 file permissions for ExportChain * change perm to 0644 * Update api.go --------- Co-authored-by: Felix Lange <fjl@twurst.com> * trie: reduce unit test time (ethereum#26918) * core/txpool: use atomic int added in go1.19 (ethereum#26913) Makes use of atomic.Uint64 instead of atomic by pointer * params: schedule shanghai fork on mainnet (ethereum#26908) Schedules the shanghai hardfork on timestamp 1681338455 as discussed on ACDE 157: ethereum/execution-specs#727 * core/txpool: allow future local transactions (ethereum#26930) Local transactions should not be subject to the "future shouldn't churn pending txs" rule * params: go-ethereum v1.11.5 stable * params: begin v1.11.6 release cycle * build: allow building nightly archives via cron jobs (ethereum#26938) * log: add special casing of uint256 into the logger (ethereum#26936) * core/rawdb: use atomic int added in go1.19 (ethereum#26935) * core/vm: expose jumptable constructors (ethereum#26880) When interacting with geth as a library to e.g. produce state tests, it is desirable to obtain the consensus-correct jumptable definition for a given fork. This changes adds accessors so the instructionset can be obtained and characteristics about opcodes can be inspected. * eth/catalyst: fix races (ethereum#26950) * core/rawdb: update freezertable read meter (ethereum#26946) The meter for "for measuring the effective amount of data read" within the freezertable was never updated. This change remedies that. --------- Signed-off-by: jsvisa <delweng@gmail.com> * cmd/evm, tests: record preimages if dump is expected (ethereum#26955) With ethereum#25287 we made it so that preimages were not recorded by default. This had the side effect that the evm command is no longer able to dump state since it does a preimage lookup to determine the address represented by a key. This change enables the recording of preimages when the dump command is given. * core/state: add account address to Trie slot accessors (ethereum#26934) This changes the Trie interface to add the plain account address as a parameter to all storage-related methods. After the introduction of the TryAccount* functions, TryGet, TryUpdate and TryDelete are now only meant to read an account's storage. In their current form, they assume that an account storage is stored in a separate trie, and that the hashing of the slot is independent of its account's address. The proposed structure for a stateless storage breaks these two assumptions: the hashing of a slot key requires the address and all slots and accounts are stored in a single trie. This PR therefore adds an address parameter to the interface. It is ignored in the MPT version, so this change has no functional impact, however it will reduce the diff size when merging verkle trees. * metrics: add cpu counters (ethereum#26796) This PR adds counter metrics for the CPU system and the Geth process. Currently the only metrics available for these items are gauges. Gauges are fine when the consumer scrapes metrics data at the same interval as Geth produces new values (every 3 seconds), but it is likely that most consumers will not scrape that often. Intervals of 10, 15, or maybe even 30 seconds are probably more common. So the problem is, how does the consumer estimate what the CPU was doing in between scrapes. With a counter, it's easy ... you just subtract two successive values and divide by the time to get a nice, accurate average. But with a gauge, you can't do that. A gauge reading is an instantaneous picture of what was happening at that moment, but it gives you no idea about what was going on between scrapes. Taking an average of values is meaningless. * metrics/influxdb: use smaller dependency and reuse code between v1 and v2 reporters (ethereum#26963) This change switches to use the smaller influxdata/influxdb1-client package instead of depending on the whole infuxdb package. The new smaller client is very similar to the influxdb-v2 client, which made it possible to refactor the two reporters to reuse code a lot more. * eth/gasprice: change feehistory input type from int to uint64 (ethereum#26922) Change input param type from int to uint64 * go.mod: update golang.org/x/tools (ethereum#26960) * rlp/rlpgen: print want/expect output string if mismatch (ethereum#26932) Co-authored-by: Marius van der Wijden <m.vanderwijden@live.de> * ethclient: ensure returned subscription is nil on error (ethereum#26976) * core/state, trie: remove Try prefix in Trie accessors (ethereum#26975) This change renames StateTrie methods to remove the Try* prefix. We added the Trie methods with prefix 'Try' a long time ago, working around the problem that most existing methods of Trie did not return the database error. This weird naming convention has persisted until now. Co-authored-by: Gary Rong <garyrong0905@gmail.com> * metrics/librato: ensure resp.body closed (ethereum#26969) This change ensures that we call Close on a http response body, in various places in the source code (mostly tests) * core/vm: use atomic.Bool (ethereum#26951) Make use of new atomic types --------- Co-authored-by: Felix Lange <fjl@twurst.com> Co-authored-by: Martin Holst Swende <martin@swende.se> * core/bloombits: use atomic type (ethereum#26993) * core/state: use atomic.Bool (ethereum#26992) * graphql: fix data races (ethereum#26965) Fixes multiple data races caused by the fact that resolving fields are done concurrently by the graphql library. It also enforces caching at the stateobject level for account fields. * eth/tracers/native: prevent panic for LOG edge-cases (ethereum#26848) This PR fixes OOM panic in the callTracer as well as panicing on opcode validation errors (e.g. stack underflow) in callTracer and prestateTracer. Co-authored-by: Martin Holst Swende <martin@swende.se> * internal/debug: add log.logfmt flag to set logging to use logfmt (ethereum#26970) * docs: update outdated DeriveSha docs comment (ethereum#26968) * remove @gballet as a GraphQL codeowner (ethereum#27012) * core: use atomic type (ethereum#27011) * graphql: revert storage access regression (ethereum#27007) * cmd/geth: Add `--log.format` cli param (ethereum#27001) Removes the new --log.logfmt directive and hides --log.json, replacing both with log.format=(json|logfmt|terminal). The hidden log.json option is still respected if log.format is not specified for backwards compatibility. Co-authored-by: Martin Holst Swende <martin@swende.se> * ethdb/pebble: use atomic type (ethereum#27014) * common: fix json marshaller MixedcaseAddress (ethereum#26998) Fix the json marshaller of MixedcaseAddress * eth/catalyst: improve consensus heartbeat (ethereum#26896) improve the heartbeat function that is no longer suitable in the current situation Co-authored-by: “openex27” <“openexkevin@gmail.com”> * miner: use atomic type (ethereum#27013) Use the new typed atomics in the miner package * accounts/abi/bind: handle UnpackLog with zero topics (ethereum#26920) Adds error handling for the case that UnpackLog or UnpackLogIntoMap is called with a log that has zero topics. --------- Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * cmd/evm: use correct parent number for t8n base fee calculation (ethereum#27032) Currently the t8n tool uses the same block number for the current block and its parent while calculating the base fee. This causes incorrect base fee calculation for the london fork block. This commit sets the parent block number to be one less than the current block number * go.mod : update snappy (ethereum#27027) * common: delete MakeName (ethereum#27023) common,p2p: remove unused function MakeName * cmd/geth: enable log rotation (ethereum#26843) This change enables log rotation, which can be activated using the flag --log.rotate. Additional parameters that can be given are: - log.maxsize to set maximum size before files are rotated, - log.maxbackups to set how many files are retailed, - log.maxage to configure max age of rotated files, - log.compress whether to compress rotated files The way to configure location of the logfile(s) is left unchanged, via the `log.logfile` parameter. --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * cmd, miner, signer: avoid panic if keystore is not available (ethereum#27039) * cmd, miner, singer: avoid panic if keystore is not available * cmd/geth: print warning instead of panic * test/fuzzers: fuzz rlp handling of big.Int and uint256.Int (ethereum#26917) test/fuzzers: fuzz rlp handling of big.Lnt and uint256.Int * core/txpool: move some validation to outside of mutex (ethereum#27006) Currently, most of transaction validation while holding the txpool mutex: one exception being an early-on signature check. This PR changes that, so that we do all non-stateful checks before we entering the mutex area. This means they can be performed in parallel, and to enable that, certain fields have been made atomic bools and uint64. * eth/downloader: use atomic types (ethereum#27030) * eth/downloader: use atomic type * Update eth/downloader/downloader_test.go Co-authored-by: Martin Holst Swende <martin@swende.se> * Update eth/downloader/downloader_test.go Co-authored-by: Martin Holst Swende <martin@swende.se> --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * core/vm: clarify comment (ethereum#27045) * consensus, core/typer: add 4844 excessDataGas to header, tie it to Cancun (ethereum#27046) * consensus/misc, params: add EIP-4844 blobfee conversions (ethereum#27041) * consensus/misc, params: add EIP-4844 blobfee conversions * consensus/misc: pull in fakeExponential test cases * consensus/misc: reuse bigints * consensus/misc: nit renames, additional larger testcase --------- Co-authored-by: Roberto Bayardo <bayardo@alum.mit.edu> Co-authored-by: Martin Holst Swende <martin@swende.se> * eth/tracers: report correct gasLimit in call tracers (ethereum#27029) This includes a semantic change to the `callTracer` as well as `flatCallTracer`. The value of field `gas` in the **first** call frame will change as follows: - It previously contained gas available after initial deductions (i.e. tx costs) - It will now contain the full tx gasLimit value Signed-off-by: jsvisa <delweng@gmail.com> * all: remove debug-field from vm config (ethereum#27048) This PR removes the Debug field from vmconfig, making it so that if a tracer is set, debug=true is implied. --------- Co-authored-by: 0xTylerHolmes <tyler@ethereum.org> Co-authored-by: Sina Mahmoodi <1591639+s1na@users.noreply.github.com> * metrics: make gauge_float64 and counter_float64 lock free (ethereum#27025) Makes the float-gauges lock-free name old time/op new time/op delta CounterFloat64Parallel-8 1.45µs ±10% 0.85µs ± 6% -41.65% (p=0.008 n=5+5) --------- Co-authored-by: Exca-DK <dev@DESKTOP-RI45P4J.localdomain> Co-authored-by: Martin Holst Swende <martin@swende.se> * eth/tracers: use atomic type (ethereum#27031) Use the new atomic types in package eth/tracers --------- Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * build: upgrade -dlgo version to Go 1.20.3 * core/txpool: disallow future churn by remote txs (ethereum#26907) Prior to this change, it was possible that transactions are erroneously deemed as 'future' although they are in fact 'pending', causing them to be dropped due to 'future' not being allowed to replace 'pending'. This change fixes that, by doing a more in-depth inspection of the queue. * core, miner: drop transactions from the same sender when error occurs (ethereum#27038) This PR unifies the error handling in miner. Whenever an error occur while applying a transaction, the transaction should be regarded as invalid and all following transactions from the same sender not executable because of the nonce restriction. The only exception is the `nonceTooLow` error which is handled separately. * params: new sepolia bootnodes (ethereum#27099) New sepolia bootnodes managed by EF devops * cmd/devp2p: fix erroneous log output in crawler (ethereum#27089) cmd/devp2p: fix log of ignored recent nodes counter * signer/core: rename testdata files (ethereum#27063) Sets a meaningful name on test-files * core: fix comment to reflect function name (ethereum#27070) * params: remove `EIP150Hash` from chainconfig (ethereum#27087) 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> * p2p: access embedded fields of Server directly (ethereum#27078) * consensus/ethash: use atomic type (ethereum#27068) * cmd/devp2p: make crawler-route53-updater less verbose (ethereum#27116) Follow-up to ethereum#26697, makes the crawler less verbose on route53-based scenarios. It also changes the loglevel from debug to info on Updates, which are typically the root, and can be interesting to see. * cmd/geth: rename --vmodule to --log.vmodule (ethereum#27071) renames `--vmodule` to `--log.vmodule`, and prints a warning if the old form is used. * core/vm: order opcodes properly (ethereum#27113) * metrics: use atomic type (ethereum#27121) * all: refactor trie API (ethereum#26995) 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. * params: go-ethereum v1.11.6 stable * dev: chg: regression changes for bor after merge * dev: chg: more regression changes for bor after merge * dev: chg: txpool_test regression changes after merge * dev: chg: gomock re-generate mocks for backend interface * dev: chg: regression changes after develop is merged into upstream-merge * dev: chg: further fixes merging develop into upstream-merge * dev: chg: apply changes to NewParallelBlockChain * dev: chg: solve some TODOs * dev: fix: CreateConsensusEngine for new ethereum objects * dev: fix: NewParallelBlockChain using Genesis * dev: fix: build ci.go * dev: fix: thelper and tparallel lint * dev: fix: http related and nilnil lint errors * dev: fix: ineffassign lint errors * dev: chg: comment position fix * dev: fix: govet lint errors * dev: fix: error related lint issues * dev: fix: bodyclose lint issues * dev: fix: some wsl lint issues * dev: fix: more wsl lint issues * dev: fix: errorcheck lint issues * dev: fix: solve more lint issues * dev: fix: more wsl lint issues * dev: fix: more errcheck lint issues * dev: fix: most of wsl lint issues * dev: fix: all remaining lint issues * dev: fix: t.Parallel called multiple times * dev: fix: tests failing due to t.Parallel * fix : runtime testcases * fix : testcase : config, addTxWithChain, burnAmount * fix : ethhash bor burn contract * fix : genspec config * fix : freezer, TestStateProcessorErrors * core,eth,miner: fix initial test cases (ethereum#922) * fix : TestTransactionIndices, testBeaconSync * fix : TestBeaconSync66 * core,eth: fix TestDeduplication, TestSyncAccountPerformance, TestTraceTransaction * fix : TestTxIndexer * rm : multiple coinbase balance * fix : testCommitInterruptExperimentBor * adding balance send to burntcontract back * Fix header encoding test * core,eth/tracers: fix TestPrestateWithDiffModeTracer * consensus/bor,eth/filters,miner,params,tests: fix mocks * fix : TestGraphQLConcurrentResolvers * fix : TestBuildPayload * common,core,miner: fix goleaks,duplicate init and inconsistent mutex (un)locks * fix : lint * fix : lint * fix : test-integration * fix : test-integration * core: restore AddFeeTransferLog post miner tipping * fix : TestGraphQLConcurrentResolvers * fix : deadlocks * rm: t.parallel from testQueueTimeLimiting tests * Merge branch 'develop' into mardizzone/upstream-merge * fix : lint * fix : test-integration * fix : TxDependency * add rpc.enabledeprecatedpersonal flag --------- Signed-off-by: Sungwoo Kim <git@sung-woo.kim> Signed-off-by: jsvisa <delweng@gmail.com> Co-authored-by: Péter Szilágyi <peterke@gmail.com> Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: Chris Ziogas <ziogaschr@gmail.com> Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> Co-authored-by: rjl493456442 <garyrong0905@gmail.com> Co-authored-by: Marius van der Wijden <m.vanderwijden@live.de> Co-authored-by: Sina Mahmoodi <1591639+s1na@users.noreply.github.com> Co-authored-by: Darioush Jalali <darioush.jalali@avalabs.org> Co-authored-by: Felix Lange <fjl@twurst.com> Co-authored-by: Patrick O'Grady <prohb125@gmail.com> Co-authored-by: ucwong <ucwong@126.com> Co-authored-by: Roman Krasiuk <rokrassyuk@gmail.com> Co-authored-by: Jason Yuan <jason.yuan@curvegrid.com> Co-authored-by: Jason Yuan <jason.yuan869@gmail.com> Co-authored-by: Sungwoo Kim <git@sung-woo.kim> Co-authored-by: Yier <90763233+yierx@users.noreply.github.com> Co-authored-by: Catror <me@catror.com> Co-authored-by: Nate Armstrong <naterarmstrong@gmail.com> Co-authored-by: Dan Cline <6798349+Rjected@users.noreply.github.com> Co-authored-by: Peter (bitfly) <1674920+peterbitfly@users.noreply.github.com> Co-authored-by: PulsarAI <dev@pulsar-systems.fi> Co-authored-by: turboboost55 <7891737+turboboost55@users.noreply.github.com> Co-authored-by: Adrian Sutton <adrian@symphonious.net> Co-authored-by: Guruprasad Kamath <48196632+gurukamath@users.noreply.github.com> Co-authored-by: James Prestwich <10149425+prestwich@users.noreply.github.com> Co-authored-by: Daniel Fernandes <711733+daferna@users.noreply.github.com> Co-authored-by: Rafael Matias <rafael@skyle.net> Co-authored-by: xiyang <90125263+JBossBC@users.noreply.github.com> Co-authored-by: Roberto Bayardo <bayardo@alum.mit.edu> Co-authored-by: panicalways <113693386+panicalways@users.noreply.github.com> Co-authored-by: dwn1998 <42262393+dwn1998@users.noreply.github.com> Co-authored-by: s7v7nislands <s7v7nislands@gmail.com> Co-authored-by: lightclient <14004106+lightclient@users.noreply.github.com> Co-authored-by: Stephen Flynn <ssflynn@gmail.com> Co-authored-by: Stephen Flynn <stephen.flynn@gapac.com> Co-authored-by: Jonathan Otto <jonathan.otto@gmail.com> Co-authored-by: Delweng <delweng@gmail.com> Co-authored-by: Guillaume Ballet <3272758+gballet@users.noreply.github.com> Co-authored-by: aaronbuchwald <aaron.buchwald56@gmail.com> Co-authored-by: ucwong <ethereum2k@gmail.com> Co-authored-by: norwnd <112318969+norwnd@users.noreply.github.com> Co-authored-by: jwasinger <j-wasinger@hotmail.com> Co-authored-by: Adrian Sutton <adrian@oplabs.co> Co-authored-by: David Murdoch <187813+davidmurdoch@users.noreply.github.com> Co-authored-by: openex <openexkevin@gmail.com> Co-authored-by: “openex27” <“openexkevin@gmail.com”> Co-authored-by: sudeep <sudeepdino008@gmail.com> Co-authored-by: joohhnnn <68833933+joohhnnn@users.noreply.github.com> Co-authored-by: 0xTylerHolmes <tyler@ethereum.org> Co-authored-by: Exca-DK <85954505+Exca-DK@users.noreply.github.com> Co-authored-by: Exca-DK <dev@DESKTOP-RI45P4J.localdomain> Co-authored-by: Marius Kjærstad <mkjaerstad@protonmail.com> Co-authored-by: Parithosh Jayanthi <parithosh@indenwolken.xyz> Co-authored-by: noel <72006780+0x00Duke@users.noreply.github.com> Co-authored-by: Taeguk Kwon <xornrbboy@gmail.com> Co-authored-by: Anusha <63559942+anusha-ctrl@users.noreply.github.com> Co-authored-by: Alex Beregszaszi <alex@rtfs.hu> Co-authored-by: Shivam Sharma <shivam691999@gmail.com> Co-authored-by: Raneet Debnath <35629432+Raneet10@users.noreply.github.com> Co-authored-by: Raneet Debnath <raneetdebnath10@gmail.com> Co-authored-by: Jerry <jerrycgh@gmail.com> Co-authored-by: Manav Darji <manavdarji.india@gmail.com>

* build: yet another weird PPA fix * build: fix (finaly?) the PPA env vars for Go bootstrapping * build: fix Go 1.19.0 bootstrapper issues on 386 PPA * build: enable Lunar Lobster PPA builds * Revert "core/trie: remove trie tracer (#26665)" (#26732) This reverts commit 7c749c947a9d5181f5f2c1b3fdb5ea6b0e401e8e. * cmd/geth: clarify dumpconfig options (#26729) Clarifies the documentation around dumpconfi Signed-off-by: Sungwoo Kim <git@sung-woo.kim> * core, eth: merge snap-sync chain download progress logs (#26676) * core: fix accessor mismatch for genesis state (#26747) * core/rawdb: expose chain freezer constructor without internals (#26748) * all: use unified emptyRootHash and emptyCodeHash (#26718) The EmptyRootHash and EmptyCodeHash are defined everywhere in the codebase, this PR replaces all of them with unified one defined in core/types package, and also defines constants for TxRoot, WithdrawalsRoot and UncleRoot * eth/filters: fix a breaking change and return rpctransaction (#26757) * eth/filters: fix a breaking change and return rpctransaction * eth/filters: fix test cases --------- Co-authored-by: Catror <me@catror.com> * common/math: allow HexOrDecimal to accept unquoted decimals too (#26758) * params: release Geth v1.11.2 * params: begin v.1.11.3 release cycle * log: improve documentation (#26753) Add usage examples * core/rawdb, node: use standalone flock dependency (#26633) * eth: use the last announced finalized block as the sync ancient limit (#26685) * cmd/devp2p: faster crawling + less verbose dns updates (#26697) This improves the speed of DHT crawling by using concurrent requests. It also removes logging of individual DNS updates. * eth/tracers: add native flatCallTracer (aka parity style tracer) (#26377) Adds support for a native call tracer with the Parity format, which outputs call frames in a flat array. This tracer accepts the following options: - `convertParityErrors: true` will convert error messages to match those of Parity - `includePrecompiles: true` will report all calls to precompiles. The default matches Parity's behavior where CALL and STATICCALLs to precompiles are excluded Incompatibilities with Parity include: - Parity removes the result object in case of failure. This behavior is maintained with the exception of reverts. Revert output usually contains useful information, i.e. Solidity revert reason. - The `gasUsed` field accounts for intrinsic gas (e.g. 21000 for simple transfers) and refunds unlike Parity - Block rewards are not reported Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * core: improve withdrawal index assignment in GenerateChain (#26756) This fixes an issue where the withdrawal index was not calculated correctly for multiple withdrawals in a single block. Co-authored-by: Gary Rong <garyrong0905@gmail.com> Co-authored-by: Felix Lange <fjl@twurst.com> * ethdb/pebble: fix range compaction (#26771) * ethdb/pebble: fix range compaction * ethdb/pebble: add comment * ethdb/pebble: fix max memorytable size (#26776) * ethclient: include withdrawals in ethclient block responses (#26778) * include withdrawals in ethclient responses * omit empty withdrawals array in json serialization * all: change chain head markers from block to header (#26777) * core/rawdb, ethdb/pebble: disable pebble on openbsd (#26801) * core: fix a merge fault (#26802) * README, go.mod, event, internal/version: bump min Go to 1.19 (#26803) * travi: remove strange leftover Go version * core, params: schedule Shanghai on goerli (#26795) * core: params: schedule Shanghai on goerli * core/forkid: fix comment * eth: remove admin.peers[i].eth.head and difficulty (#26804) * core/types: add EffectiveGasPrice in Receipt (#26713) This change adds a struct field EffectiveGasPrice in types.Receipt. The field is present in RPC responses, but not in the Go struct, and thus can't easily be accessed via ethclient. Co-authored-by: PulsarAI <dev@pulsar-systems.fi> * core, eth/catalyst: fix race conditions in tests (#26790) Fixes a race in TestNewPayloadOnInvalidTerminalBlock where setting the TTD raced with the miner. Solution: set the TTD on the blockchain config not the genesis config. Also fixes a race in CopyHeader which resulted in race reports all over the place. * metrics: improve accuracy of CPU gauges (#26793) This PR changes metrics collection to actually measure the time interval between collections, rather than assume 3 seconds. I did some ad hoc profiling, and on slower hardware (eg, my Raspberry Pi 4) I routinely saw intervals between 3.3 - 3.5 seconds, with some being as high as 4.5 seconds. This will generally cause the CPU gauge readings to be too high, and in some cases can cause impossibly large values for the CPU load metrics (eg. greater than 400 for a 4 core CPU). --------- Co-authored-by: Felix Lange <fjl@twurst.com> * ethclient: fix panic when requesting missing blocks (#26817) This fixes a regression introduced by #26723. Fixes #26816. * core, miner: revert block gas counter in case of invalid transaction (#26799) This change fixes a flaw where, in certain scenarios, the block sealer did not accurately reset the remaining gas after failing to include an invalid transaction. Fixes #26791 * internal/ethapi: add tests for transaction types JSON marshal/unmarshal (#26667) Checks that Transaction.MarshalJSON and newRPCTransaction JSON output can be parsed by Transaction.UnmarshalJSON --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * cmd/evm: correct `alloc` for `t8n` testdata (#26822) Fixes a minor error in the testdata * eth/tracers/native: set created address to nil in case of failure (#26779) Fixes #26073 * accounts/usbwallet: mitigate ledger app chunking issue (#26773) This PR mitigates an issue with Ledger's on-device RLP deserialization, see https://github.com/LedgerHQ/app-ethereum/issues/409 Ledger's RLP deserialization code does not validate the length of the RLP list received, and it may prematurely enter the signing flow when a APDU chunk boundary falls immediately before the EIP-155 chain_id when deserializing a transaction. Since the chain_id is uninitialized, it is 0 during this signing flow. This may cause the user to accidentally sign the transaction with chain_id = 0. That signature would be returned from the device 1 packet earlier than expected by the communication loop. The device blocks the second-to-last packet waiting for the signer flow, and then errors on the successive packet (which contains the chain_id, zeroed r, and zeroed s) Since the signature's early arrival causes successive errors during the communication process, geth does not parse the improper signature produced by the device, and therefore no improperly-signed transaction can be created. User funds are not at risk. We mitigate by selecting the highest chunk size that leaves at least 4 bytes in the final chunk. * beacon/engine: don't omit empty withdrawals in ExecutionPayloadBodies (#26698) This ensures the "withdrawals" field will always be present in responses to getPayloadBodiesByRangeV1 and getPayloadBodiesByHashV1. --------- Co-authored-by: Felix Lange <fjl@twurst.com> * build: update to go 1.20.2 (#26824) * params: go-ethereum v1.11.3 stable * params: begin v1.11.4 release cycle * core/rawdb: find smallest block stored in key-value store when chain gapped (#26719) This change prints out more information about the problem, in the case where geth detects a gap between leveldb and ancients, so we can determine more exactly where the gap is (what the first missing is). Also prints out more metadata. --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * signer/core: accept all solidity primitive types for EIP-712 signing (#26770) Accept all primitive types in Solidity for EIP-712 from intN, uintN, intN[], uintN[] for N as 0 to 256 in multiples of 8 --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * params: remove EF azure bootnodes (#26828) * core/vm: use golang native big.Int (#26834) reverts #26021, to use the upstream bigint instead. * core/vm: fix typo in comment (#26838) fixes eip 220 -> 2200 * core/forkid: fix issue in validation test (#26544) This changes the test to match the comment description. Using timestampedConfig in this test case is incorrect, the comment says 'local is at Gray Glacier' and isn't aware of more forks. * cmd/evm: update readmes for the tests (#26841) * core, core/types: plain Message struct (#25977) Here, the core.Message interface turns into a plain struct and types.Message gets removed. This is a breaking change to packages core and core/types. While we do not promise API stability for package core, we do for core/types. An exception can be made for types.Message, since it doesn't have any purpose apart from invoking the state transition in package core. types.Message was also marked deprecated by the same commit it got added in, 4dca5d4db7 (November 2016). The core.Message interface was added in December 2014, in commit db494170dc, for the purpose of 'testing' state transitions. It's the same change that made transaction struct fields private. Before that, the state transition used *types.Transaction directly. Over time, multiple implementations of the interface accrued across different packages, since constructing a Message is required whenever one wants to invoke the state transition. These implementations all looked very similar, a struct with private fields exposing the fields as accessor methods. By changing Message into a struct with public fields we can remove all these useless interface implementations. It will also hopefully simplify future changes to the type with less updates to apply across all of go-ethereum when a field is added to Message. --------- Co-authored-by: Felix Lange <fjl@twurst.com> * travis: only build PPAs nightly, not on every push, too heavy (#26846) * p2p: small comment typo (#26850) Update server.go * core: add Timestamp method in BlockGen (#26844) Since forks are now scheduled by block time, it can be necessary to check the timestamp of a block while generating transactions. * core/txpool: implement additional DoS defenses (#26648) This adds two new rules to the transaction pool: - A future transaction can not evict a pending transaction. - A transaction can not overspend available funds of a sender. --- Co-authored-by: dwn1998 <42262393+dwn1998@users.noreply.github.com> Co-authored-by: Martin Holst Swende <martin@swende.se> * params: go-ethereum v1.11.4 stable * params: begin v1.11.5 release cycle * tests: define `MuirGlacier` fork (#26856) add muir glacier to t8n * code/vm: fix comment typo (#26865) it should be constantinople rather than contantinople * core: minor code refactor (#26852) * core: refactor code * core: drop it from this anonymous goroutine func * core/txpool: use priceList.Put instead of heap.Push (#26863) Minor refactor to use the 'intended' accessor * eth: return error if 'safe' or 'finalized' tag used pre-merge (#26862) Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: Felix Lange <fjl@twurst.com> * .travis.yml: reenable PPA build on tag push (#26873) * core/state, trie: port changes from PBSS (#26763) * p2p/discover: pass invalid discv5 packets to Unhandled channel (#26699) This makes it possible to run another protocol alongside discv5, by reading unhandled packets from the channel. * all: update links in documentation (#26882) Co-authored-by: Stephen Flynn <stephen.flynn@gapac.com> * Increase websocket frame size (from erigon rpc client) (#26883) This increases the maximum allowed message size to 32MB. Originally submitted at https://github.com/ledgerwatch/erigon/pull/2739 example block failure: https://etherscan.io/tx/0x1317d973a55cedf9b0f2df6ea48e8077dd176f5444a3423368a46d6e4db89982#internal * cmd/devp2p, cmd/geth: add version in --help output (#26895) Not sure why this was removed, it's pretty useful to see the version also in --help. * core: show db error-info in case of mismatched hash root (#26870) When a database failure occurs, bubble it up a into statedb, and report it in suitable places, such as during a 'bad block' report. * consensus: improve consensus engine definition (#26871) Makes clear the distinction between Finalize and FinalizedAndAssemble: - In Finalize function, a series of state operations are applied according to consensus rules. The statedb is mutated and the root hash can be checked and compared afterwards. This function should be used in block processing(receive afrom network and apply it locally) but not block generation. - In FinalizeAndAssemble function, after applying state mutations, the block is also to be assembled with the latest state root computed, updating the header. This function should be used in block generation only. * eth/catalyst: increase update consensus timeout (#26840) Increases the time between consensus updates that we give the CL before we start warning the user. * internal/ethapi: avoid int overflow in GetTransactionReceipt (#26911) * trie, accounts/abi: add error-checks (#26914) * rlp: support for uint256 (#26898) This adds built-in support in package rlp for encoding, decoding and generating code dealing with uint256.Int. --------- Co-authored-by: Felix Lange <fjl@twurst.com> * eth: fix output file permissions in admin_exportChain (#26912) * api: Use 0700 file permissions for ExportChain * change perm to 0644 * Update api.go --------- Co-authored-by: Felix Lange <fjl@twurst.com> * trie: reduce unit test time (#26918) * core/txpool: use atomic int added in go1.19 (#26913) Makes use of atomic.Uint64 instead of atomic by pointer * params: schedule shanghai fork on mainnet (#26908) Schedules the shanghai hardfork on timestamp 1681338455 as discussed on ACDE 157: https://github.com/ethereum/execution-specs/pull/727 * core/txpool: allow future local transactions (#26930) Local transactions should not be subject to the "future shouldn't churn pending txs" rule * params: go-ethereum v1.11.5 stable * params: begin v1.11.6 release cycle * build: allow building nightly archives via cron jobs (#26938) * log: add special casing of uint256 into the logger (#26936) * core/rawdb: use atomic int added in go1.19 (#26935) * core/vm: expose jumptable constructors (#26880) When interacting with geth as a library to e.g. produce state tests, it is desirable to obtain the consensus-correct jumptable definition for a given fork. This changes adds accessors so the instructionset can be obtained and characteristics about opcodes can be inspected. * eth/catalyst: fix races (#26950) * core/rawdb: update freezertable read meter (#26946) The meter for "for measuring the effective amount of data read" within the freezertable was never updated. This change remedies that. --------- Signed-off-by: jsvisa <delweng@gmail.com> * cmd/evm, tests: record preimages if dump is expected (#26955) With #25287 we made it so that preimages were not recorded by default. This had the side effect that the evm command is no longer able to dump state since it does a preimage lookup to determine the address represented by a key. This change enables the recording of preimages when the dump command is given. * core/state: add account address to Trie slot accessors (#26934) This changes the Trie interface to add the plain account address as a parameter to all storage-related methods. After the introduction of the TryAccount* functions, TryGet, TryUpdate and TryDelete are now only meant to read an account's storage. In their current form, they assume that an account storage is stored in a separate trie, and that the hashing of the slot is independent of its account's address. The proposed structure for a stateless storage breaks these two assumptions: the hashing of a slot key requires the address and all slots and accounts are stored in a single trie. This PR therefore adds an address parameter to the interface. It is ignored in the MPT version, so this change has no functional impact, however it will reduce the diff size when merging verkle trees. * metrics: add cpu counters (#26796) This PR adds counter metrics for the CPU system and the Geth process. Currently the only metrics available for these items are gauges. Gauges are fine when the consumer scrapes metrics data at the same interval as Geth produces new values (every 3 seconds), but it is likely that most consumers will not scrape that often. Intervals of 10, 15, or maybe even 30 seconds are probably more common. So the problem is, how does the consumer estimate what the CPU was doing in between scrapes. With a counter, it's easy ... you just subtract two successive values and divide by the time to get a nice, accurate average. But with a gauge, you can't do that. A gauge reading is an instantaneous picture of what was happening at that moment, but it gives you no idea about what was going on between scrapes. Taking an average of values is meaningless. * metrics/influxdb: use smaller dependency and reuse code between v1 and v2 reporters (#26963) This change switches to use the smaller influxdata/influxdb1-client package instead of depending on the whole infuxdb package. The new smaller client is very similar to the influxdb-v2 client, which made it possible to refactor the two reporters to reuse code a lot more. * eth/gasprice: change feehistory input type from int to uint64 (#26922) Change input param type from int to uint64 * go.mod: update golang.org/x/tools (#26960) * rlp/rlpgen: print want/expect output string if mismatch (#26932) Co-authored-by: Marius van der Wijden <m.vanderwijden@live.de> * ethclient: ensure returned subscription is nil on error (#26976) * core/state, trie: remove Try prefix in Trie accessors (#26975) This change renames StateTrie methods to remove the Try* prefix. We added the Trie methods with prefix 'Try' a long time ago, working around the problem that most existing methods of Trie did not return the database error. This weird naming convention has persisted until now. Co-authored-by: Gary Rong <garyrong0905@gmail.com> * metrics/librato: ensure resp.body closed (#26969) This change ensures that we call Close on a http response body, in various places in the source code (mostly tests) * core/vm: use atomic.Bool (#26951) Make use of new atomic types --------- Co-authored-by: Felix Lange <fjl@twurst.com> Co-authored-by: Martin Holst Swende <martin@swende.se> * core/bloombits: use atomic type (#26993) * core/state: use atomic.Bool (#26992) * graphql: fix data races (#26965) Fixes multiple data races caused by the fact that resolving fields are done concurrently by the graphql library. It also enforces caching at the stateobject level for account fields. * eth/tracers/native: prevent panic for LOG edge-cases (#26848) This PR fixes OOM panic in the callTracer as well as panicing on opcode validation errors (e.g. stack underflow) in callTracer and prestateTracer. Co-authored-by: Martin Holst Swende <martin@swende.se> * internal/debug: add log.logfmt flag to set logging to use logfmt (#26970) * docs: update outdated DeriveSha docs comment (#26968) * remove @gballet as a GraphQL codeowner (#27012) * core: use atomic type (#27011) * graphql: revert storage access regression (#27007) * cmd/geth: Add `--log.format` cli param (#27001) Removes the new --log.logfmt directive and hides --log.json, replacing both with log.format=(json|logfmt|terminal). The hidden log.json option is still respected if log.format is not specified for backwards compatibility. Co-authored-by: Martin Holst Swende <martin@swende.se> * ethdb/pebble: use atomic type (#27014) * common: fix json marshaller MixedcaseAddress (#26998) Fix the json marshaller of MixedcaseAddress * eth/catalyst: improve consensus heartbeat (#26896) improve the heartbeat function that is no longer suitable in the current situation Co-authored-by: “openex27” <“openexkevin@gmail.com”> * miner: use atomic type (#27013) Use the new typed atomics in the miner package * accounts/abi/bind: handle UnpackLog with zero topics (#26920) Adds error handling for the case that UnpackLog or UnpackLogIntoMap is called with a log that has zero topics. --------- Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * cmd/evm: use correct parent number for t8n base fee calculation (#27032) Currently the t8n tool uses the same block number for the current block and its parent while calculating the base fee. This causes incorrect base fee calculation for the london fork block. This commit sets the parent block number to be one less than the current block number * go.mod : update snappy (#27027) * common: delete MakeName (#27023) common,p2p: remove unused function MakeName * cmd/geth: enable log rotation (#26843) This change enables log rotation, which can be activated using the flag --log.rotate. Additional parameters that can be given are: - log.maxsize to set maximum size before files are rotated, - log.maxbackups to set how many files are retailed, - log.maxage to configure max age of rotated files, - log.compress whether to compress rotated files The way to configure location of the logfile(s) is left unchanged, via the `log.logfile` parameter. --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * cmd, miner, signer: avoid panic if keystore is not available (#27039) * cmd, miner, singer: avoid panic if keystore is not available * cmd/geth: print warning instead of panic * test/fuzzers: fuzz rlp handling of big.Int and uint256.Int (#26917) test/fuzzers: fuzz rlp handling of big.Lnt and uint256.Int * core/txpool: move some validation to outside of mutex (#27006) Currently, most of transaction validation while holding the txpool mutex: one exception being an early-on signature check. This PR changes that, so that we do all non-stateful checks before we entering the mutex area. This means they can be performed in parallel, and to enable that, certain fields have been made atomic bools and uint64. * eth/downloader: use atomic types (#27030) * eth/downloader: use atomic type * Update eth/downloader/downloader_test.go Co-authored-by: Martin Holst Swende <martin@swende.se> * Update eth/downloader/downloader_test.go Co-authored-by: Martin Holst Swende <martin@swende.se> --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * core/vm: clarify comment (#27045) * consensus, core/typer: add 4844 excessDataGas to header, tie it to Cancun (#27046) * consensus/misc, params: add EIP-4844 blobfee conversions (#27041) * consensus/misc, params: add EIP-4844 blobfee conversions * consensus/misc: pull in fakeExponential test cases * consensus/misc: reuse bigints * consensus/misc: nit renames, additional larger testcase --------- Co-authored-by: Roberto Bayardo <bayardo@alum.mit.edu> Co-authored-by: Martin Holst Swende <martin@swende.se> * eth/tracers: report correct gasLimit in call tracers (#27029) This includes a semantic change to the `callTracer` as well as `flatCallTracer`. The value of field `gas` in the **first** call frame will change as follows: - It previously contained gas available after initial deductions (i.e. tx costs) - It will now contain the full tx gasLimit value Signed-off-by: jsvisa <delweng@gmail.com> * all: remove debug-field from vm config (#27048) This PR removes the Debug field from vmconfig, making it so that if a tracer is set, debug=true is implied. --------- Co-authored-by: 0xTylerHolmes <tyler@ethereum.org> Co-authored-by: Sina Mahmoodi <1591639+s1na@users.noreply.github.com> * metrics: make gauge_float64 and counter_float64 lock free (#27025) Makes the float-gauges lock-free name old time/op new time/op delta CounterFloat64Parallel-8 1.45µs ±10% 0.85µs ± 6% -41.65% (p=0.008 n=5+5) --------- Co-authored-by: Exca-DK <dev@DESKTOP-RI45P4J.localdomain> Co-authored-by: Martin Holst Swende <martin@swende.se> * eth/tracers: use atomic type (#27031) Use the new atomic types in package eth/tracers --------- Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * build: upgrade -dlgo version to Go 1.20.3 * core/txpool: disallow future churn by remote txs (#26907) Prior to this change, it was possible that transactions are erroneously deemed as 'future' although they are in fact 'pending', causing them to be dropped due to 'future' not being allowed to replace 'pending'. This change fixes that, by doing a more in-depth inspection of the queue. * core, miner: drop transactions from the same sender when error occurs (#27038) This PR unifies the error handling in miner. Whenever an error occur while applying a transaction, the transaction should be regarded as invalid and all following transactions from the same sender not executable because of the nonce restriction. The only exception is the `nonceTooLow` error which is handled separately. * params: new sepolia bootnodes (#27099) New sepolia bootnodes managed by EF devops * cmd/devp2p: fix erroneous log output in crawler (#27089) cmd/devp2p: fix log of ignored recent nodes counter * signer/core: rename testdata files (#27063) Sets a meaningful name on test-files * core: fix comment to reflect function name (#27070) * params: remove `EIP150Hash` from chainconfig (#27087) 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> * p2p: access embedded fields of Server directly (#27078) * consensus/ethash: use atomic type (#27068) * cmd/devp2p: make crawler-route53-updater less verbose (#27116) Follow-up to #26697, makes the crawler less verbose on route53-based scenarios. It also changes the loglevel from debug to info on Updates, which are typically the root, and can be interesting to see. * cmd/geth: rename --vmodule to --log.vmodule (#27071) renames `--vmodule` to `--log.vmodule`, and prints a warning if the old form is used. * core/vm: order opcodes properly (#27113) * metrics: use atomic type (#27121) * all: refactor trie API (#26995) 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. * params: go-ethereum v1.11.6 stable * dev: chg: regression changes for bor after merge * dev: chg: more regression changes for bor after merge * Use `big.Int#Set` instead of `big.Int#SetBytes` when possible (#896) There is no reason to make the bytes transformation and back to big int to assign a big int to another one. The `Set` method should be used instead. * dev: chg: txpool_test regression changes after merge * dev: chg: gomock re-generate mocks for backend interface * dev: chg: regression changes after develop is merged into upstream-merge * dev: chg: further fixes merging develop into upstream-merge * dev: chg: apply changes to NewParallelBlockChain * dev: chg: solve some TODOs * dev: fix: CreateConsensusEngine for new ethereum objects * dev: fix: NewParallelBlockChain using Genesis * dev: fix: build ci.go * dev: fix: thelper and tparallel lint * dev: fix: http related and nilnil lint errors * dev: fix: ineffassign lint errors * dev: chg: comment position fix * dev: fix: govet lint errors * dev: fix: error related lint issues * dev: fix: bodyclose lint issues * dev: fix: some wsl lint issues * dev: fix: more wsl lint issues * dev: fix: errorcheck lint issues * dev: fix: solve more lint issues * dev: fix: more wsl lint issues * dev: fix: more errcheck lint issues * dev: fix: most of wsl lint issues * dev: fix: all remaining lint issues * fixed MVHashMap being enabled even when paralellUniverse is set to 0 (#900) * dev: fix: t.Parallel called multiple times * dev: fix: tests failing due to t.Parallel * fix : runtime testcases * fix : testcase : config, addTxWithChain, burnAmount * fix : ethhash bor burn contract * fix : genspec config * dev: chg: pos-1465: remove snyk and sonarqube (#908) * fix : freezer, TestStateProcessorErrors * core,eth,miner: fix initial test cases (#922) * fix : TestTransactionIndices, testBeaconSync * fix : TestBeaconSync66 * core,eth: fix TestDeduplication, TestSyncAccountPerformance, TestTraceTransaction * fix : TestTxIndexer * rm : multiple coinbase balance * fix : testCommitInterruptExperimentBor * adding balance send to burntcontract back * Fix header encoding test * Fix worker tracing `nil` pointer exception (#899) The `baseFee` can be `nil` in certain situations, the tracing must not use `baseFee` unless it was checked otherwise if such case happen, Go is going to panic. * core,eth/tracers: fix TestPrestateWithDiffModeTracer * fixed go files * consensus/bor,eth/filters,miner,params,tests: fix mocks * fix : TestGraphQLConcurrentResolvers * fix : TestBuildPayload * common,core,miner: fix goleaks,duplicate init and inconsistent mutex (un)locks * fix : lint * fix : lint * Move NumSpeculativeProcs from module variable to function parameter (#931) This will prevent data races when more than one parallel execution are running at the same time. * fix : test-integration * fix : test-integration * Calculate tx dependency only when mining is enabled (#935) * core: restore AddFeeTransferLog post miner tipping * fix : TestGraphQLConcurrentResolvers * mardizzone/POS-1605: go and deps upgrade (#929) * dev: chg: pos-1605: fix govulns by bumping golang version and replacing mongodb version * dev: chg: pos-1605: update some dependencies to solve dependabot vulns * dev: chg: update linter for compatibility with upgraded go version * dev: chg: update go version in golangci lintere * dev: chg: update go version in ci workflow, docker and travis * dev: chg: change govulncheck version to a stable fixed one * dev: chg: fix govulncheck version * dev: chg: remove vulncheck version * dev: chg: adapt tests based on go-ethereum for go version upgrade * dev: chg: use explicit go version for templum action * dev: chg: fix lint * dev: chg: fix lint * dev: chg: 1.20.x explicitly in packager * dev: chg: ignore some additional leak to be fixed * dev: chg: ignore some additional leak to be fixed * dev: chg: fix lint * rpc: add execution pool metrics (#919) * rpc: add execution pool metrics * rpc: stop execution pool and report metrics using ticker * fix lint * update go.mod, update metric report interval * handle empty workerpool case to fix tests * fix lint * refactor ep metrics collection based on each service * remove log * rpc: convert processed metric to histogram * lint * Moved TxDependency Metadata to ExtraData in Block Header (#930) * moved TxDependency Metadata to ExtraData in Block Header * updated unit tests * lints * small bug fix * small improvement * minor bug fix * addressed comments * bug fix, only using the BlockExtraData structure after the Parallel Universe HF * addressed comments * fixed an error * bug fix * encoding nil ValidatorBytes and TxDependency in prepare if the current block is not the last block of the sprint * bug fix * fix : deadlocks * rm: t.parallel from testQueueTimeLimiting tests * Merge branch 'develop' into mardizzone/upstream-merge * fix : lint * fix : test-integration * fix : TxDependency * Update discord link in README.md (#938) * add rpc.enabledeprecatedpersonal flag * Upstream merge from go-ethereum/v1.11.6 (#901) * ethdb/pebble: fix nil callbacks (#26650) * eth/downloader: fix timeout resurrection panic (#26652) * common/prque, eth/downloader: fix timeout resurrection panic * common/prque: revert -1 hack for les, temporaryly! * core/state, trie: remove unused error-return from trie Commit operation (#26641) * go.mod: update pebble to latest master (#26654) * core/vm: set tracer-observable `value` of a delegatecall to match parent `value` (#26632) This is a breaking change in the tracing hooks API as well as semantics of the callTracer: - CaptureEnter hook provided a nil value argument in case of DELEGATECALL. However to stay consistent with how delegate calls behave in EVM this hook is changed to pass in the value of the parent call. - callTracer will return parent call's value for DELEGATECALL frames. --------- Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * ethdb: add benchmark test suite (#26659) * params: schedule shanghai fork on sepolia (#26662) * params: schedule shanghai fork on sepolia * params: u64 -> newUint64 * eth/filters: avoid block body retrieval when no matching logs (#25199) Logs stored on disk have minimal information. Contextual information such as block number, index of log in block, index of transaction in block are filled in upon request. We can fill in all these fields only having the block header and list of receipts. But determining the transaction hash of a log requires the block body. The goal of this PR is postponing this retrieval until we are sure we the transaction hash. It happens often that the header bloom filter signals there might be matches in a block, but after actually checking them reveals the logs do not match. We want to avoid fetching the body in this case. Note that this changes the semantics of Backend.GetLogs. Downstream callers of GetLogs now assume log context fields have not been derived, and need to call DeriveFields on the logs if necessary. * eth/tracers: more fork overrides in traceBlockToFile (#26655) This change allows all post-Berlin forks to be specified as overrides for futureForkBlock in the config parameter for traceBlockToFile. * tests/fuzzers: supply gnark multiexp config, fixes #26669 (#26670) This change fixes a fuzzer which broke when we updated the gnark dependency earlier. * cmd/devp2p: reduce output of node crawler (#26674) Our discovery crawler spits out a huge amount of logs, most of which is pretty non-interesting. This change moves the very verbose output to Debug, and adds a 8-second status log message giving the general idea about what's going on. * params: update mainnet + rinkeby CHT (#26677) This change updates the CHT entries for mainnet and rinkeby * eth/filters: replace atomic pointer with value (#26689) * eth/filters: replace atomic.Pointer * fix * improve Co-authored-by: Martin Holst Swende <martin@swende.se> --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * p2p/dnsdisc: fix tests with Go 1.20 (#26690) * eth/catalyst: return error if withdrawals are nil post-shanghai (#26691) Spec: https://github.com/ethereum/execution-apis/blob/main/src/engine/shanghai.md#request * ethdb/pebble: Fix `MemTableStopWritesThreshold` (#26692) MemTableStopWritesThreshold was set to the max size of all memtables before blocking writing but should be set to the max number of memtables. This is documented [here](https://github.com/cockroachdb/pebble/blob/master/options.go#L738-L742). * eth/downloader: handle missing withdrawals if empty list is expected (#26675) This PR relaxes the block body ingress handling a bit: if block body withdrawals are missing (but expected to be empty), the body withdrawals are set to 'empty list' before being passed to upper layers. This fixes an issue where a block passed from EthereumJS to geth was deemed invalid. * params: go-ethereum v1.11.0 stable * params: begin v1.11.1 release cycle * travis, build: update Go to 1.20.1 (#26653) travis, build: update Go to 1.20 * core: check genesis state presence by disk read (#26703) * core, eth/downloader: make body validation more strict (#26704) * eth/downloader: fix empty-body case in queue fetchresult (#26707) * eth/downloader: fix typo (#26716) * all: remove deprecated uses of math.rand (#26710) This PR is a (superior) alternative to https://github.com/ethereum/go-ethereum/pull/26708, it handles deprecation, primarily two specific cases. `rand.Seed` is typically used in two ways - `rand.Seed(time.Now().UnixNano())` -- we seed it, just to be sure to get some random, and not always get the same thing on every run. This is not needed, with global seeding, so those are just removed. - `rand.Seed(1)` this is typically done to ensure we have a stable test. If we rely on this, we need to fix up the tests to use a deterministic prng-source. A few occurrences like this has been replaced with a proper custom source. `rand.Read` has been replaced by `crypto/rand`.`Read` in this PR. * params: go-ethereum v1.11.1 stable * params: begin v1.11.2 release cycle * eth/catalyst: send INVALID instead of INVALID_BLOCK_HASH (#26696) This change will break one hive test, but pass another and it will be the better way going forward * ci: disable coverage reporting in appveyor and travis * eth/catalyst: request too large error (#26722) The method `GetPayloadBodiesByRangeV1` now returns "-38004: Too large request" error if the requested range is too large, according to spec Co-authored-by: Martin Holst Swende <martin@swende.se> * core/trie: remove trie tracer (#26665) This PR contains a small portion of the full pbss PR, namely Remove the tracer from trie (and comitter), and instead using an accessList. Related changes to the Nodeset. --------- Co-authored-by: Gary Rong <garyrong0905@gmail.com> * rpc: fix unmarshaling of null result in CallContext (#26723) The change fixes unmarshaling of JSON null results into json.RawMessage. --------- Co-authored-by: Jason Yuan <jason.yuan@curvegrid.com> Co-authored-by: Jason Yuan <jason.yuan869@gmail.com> * build: ship bootstrapper Go along with builder for PPA (#26731) * build: fix setting env var, temp early exit * build: fix gobootstrap path for the PPA * build: add some PPA debug logs, sigh * internal/build: revert raising the chunk size for PPA * build: yet another weird PPA fix * build: fix (finaly?) the PPA env vars for Go bootstrapping * build: fix Go 1.19.0 bootstrapper issues on 386 PPA * build: enable Lunar Lobster PPA builds * Revert "core/trie: remove trie tracer (#26665)" (#26732) This reverts commit 7c749c947a9d5181f5f2c1b3fdb5ea6b0e401e8e. * cmd/geth: clarify dumpconfig options (#26729) Clarifies the documentation around dumpconfi Signed-off-by: Sungwoo Kim <git@sung-woo.kim> * core, eth: merge snap-sync chain download progress logs (#26676) * core: fix accessor mismatch for genesis state (#26747) * core/rawdb: expose chain freezer constructor without internals (#26748) * all: use unified emptyRootHash and emptyCodeHash (#26718) The EmptyRootHash and EmptyCodeHash are defined everywhere in the codebase, this PR replaces all of them with unified one defined in core/types package, and also defines constants for TxRoot, WithdrawalsRoot and UncleRoot * eth/filters: fix a breaking change and return rpctransaction (#26757) * eth/filters: fix a breaking change and return rpctransaction * eth/filters: fix test cases --------- Co-authored-by: Catror <me@catror.com> * common/math: allow HexOrDecimal to accept unquoted decimals too (#26758) * params: release Geth v1.11.2 * params: begin v.1.11.3 release cycle * log: improve documentation (#26753) Add usage examples * core/rawdb, node: use standalone flock dependency (#26633) * eth: use the last announced finalized block as the sync ancient limit (#26685) * cmd/devp2p: faster crawling + less verbose dns updates (#26697) This improves the speed of DHT crawling by using concurrent requests. It also removes logging of individual DNS updates. * eth/tracers: add native flatCallTracer (aka parity style tracer) (#26377) Adds support for a native call tracer with the Parity format, which outputs call frames in a flat array. This tracer accepts the following options: - `convertParityErrors: true` will convert error messages to match those of Parity - `includePrecompiles: true` will report all calls to precompiles. The default matches Parity's behavior where CALL and STATICCALLs to precompiles are excluded Incompatibilities with Parity include: - Parity removes the result object in case of failure. This behavior is maintained with the exception of reverts. Revert output usually contains useful information, i.e. Solidity revert reason. - The `gasUsed` field accounts for intrinsic gas (e.g. 21000 for simple transfers) and refunds unlike Parity - Block rewards are not reported Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * core: improve withdrawal index assignment in GenerateChain (#26756) This fixes an issue where the withdrawal index was not calculated correctly for multiple withdrawals in a single block. Co-authored-by: Gary Rong <garyrong0905@gmail.com> Co-authored-by: Felix Lange <fjl@twurst.com> * ethdb/pebble: fix range compaction (#26771) * ethdb/pebble: fix range compaction * ethdb/pebble: add comment * ethdb/pebble: fix max memorytable size (#26776) * ethclient: include withdrawals in ethclient block responses (#26778) * include withdrawals in ethclient responses * omit empty withdrawals array in json serialization * all: change chain head markers from block to header (#26777) * core/rawdb, ethdb/pebble: disable pebble on openbsd (#26801) * core: fix a merge fault (#26802) * README, go.mod, event, internal/version: bump min Go to 1.19 (#26803) * travi: remove strange leftover Go version * core, params: schedule Shanghai on goerli (#26795) * core: params: schedule Shanghai on goerli * core/forkid: fix comment * eth: remove admin.peers[i].eth.head and difficulty (#26804) * core/types: add EffectiveGasPrice in Receipt (#26713) This change adds a struct field EffectiveGasPrice in types.Receipt. The field is present in RPC responses, but not in the Go struct, and thus can't easily be accessed via ethclient. Co-authored-by: PulsarAI <dev@pulsar-systems.fi> * core, eth/catalyst: fix race conditions in tests (#26790) Fixes a race in TestNewPayloadOnInvalidTerminalBlock where setting the TTD raced with the miner. Solution: set the TTD on the blockchain config not the genesis config. Also fixes a race in CopyHeader which resulted in race reports all over the place. * metrics: improve accuracy of CPU gauges (#26793) This PR changes metrics collection to actually measure the time interval between collections, rather than assume 3 seconds. I did some ad hoc profiling, and on slower hardware (eg, my Raspberry Pi 4) I routinely saw intervals between 3.3 - 3.5 seconds, with some being as high as 4.5 seconds. This will generally cause the CPU gauge readings to be too high, and in some cases can cause impossibly large values for the CPU load metrics (eg. greater than 400 for a 4 core CPU). --------- Co-authored-by: Felix Lange <fjl@twurst.com> * ethclient: fix panic when requesting missing blocks (#26817) This fixes a regression introduced by #26723. Fixes #26816. * core, miner: revert block gas counter in case of invalid transaction (#26799) This change fixes a flaw where, in certain scenarios, the block sealer did not accurately reset the remaining gas after failing to include an invalid transaction. Fixes #26791 * internal/ethapi: add tests for transaction types JSON marshal/unmarshal (#26667) Checks that Transaction.MarshalJSON and newRPCTransaction JSON output can be parsed by Transaction.UnmarshalJSON --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * cmd/evm: correct `alloc` for `t8n` testdata (#26822) Fixes a minor error in the testdata * eth/tracers/native: set created address to nil in case of failure (#26779) Fixes #26073 * accounts/usbwallet: mitigate ledger app chunking issue (#26773) This PR mitigates an issue with Ledger's on-device RLP deserialization, see https://github.com/LedgerHQ/app-ethereum/issues/409 Ledger's RLP deserialization code does not validate the length of the RLP list received, and it may prematurely enter the signing flow when a APDU chunk boundary falls immediately before the EIP-155 chain_id when deserializing a transaction. Since the chain_id is uninitialized, it is 0 during this signing flow. This may cause the user to accidentally sign the transaction with chain_id = 0. That signature would be returned from the device 1 packet earlier than expected by the communication loop. The device blocks the second-to-last packet waiting for the signer flow, and then errors on the successive packet (which contains the chain_id, zeroed r, and zeroed s) Since the signature's early arrival causes successive errors during the communication process, geth does not parse the improper signature produced by the device, and therefore no improperly-signed transaction can be created. User funds are not at risk. We mitigate by selecting the highest chunk size that leaves at least 4 bytes in the final chunk. * beacon/engine: don't omit empty withdrawals in ExecutionPayloadBodies (#26698) This ensures the "withdrawals" field will always be present in responses to getPayloadBodiesByRangeV1 and getPayloadBodiesByHashV1. --------- Co-authored-by: Felix Lange <fjl@twurst.com> * build: update to go 1.20.2 (#26824) * params: go-ethereum v1.11.3 stable * params: begin v1.11.4 release cycle * core/rawdb: find smallest block stored in key-value store when chain gapped (#26719) This change prints out more information about the problem, in the case where geth detects a gap between leveldb and ancients, so we can determine more exactly where the gap is (what the first missing is). Also prints out more metadata. --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * signer/core: accept all solidity primitive types for EIP-712 signing (#26770) Accept all primitive types in Solidity for EIP-712 from intN, uintN, intN[], uintN[] for N as 0 to 256 in multiples of 8 --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * params: remove EF azure bootnodes (#26828) * core/vm: use golang native big.Int (#26834) reverts #26021, to use the upstream bigint instead. * core/vm: fix typo in comment (#26838) fixes eip 220 -> 2200 * core/forkid: fix issue in validation test (#26544) This changes the test to match the comment description. Using timestampedConfig in this test case is incorrect, the comment says 'local is at Gray Glacier' and isn't aware of more forks. * cmd/evm: update readmes for the tests (#26841) * core, core/types: plain Message struct (#25977) Here, the core.Message interface turns into a plain struct and types.Message gets removed. This is a breaking change to packages core and core/types. While we do not promise API stability for package core, we do for core/types. An exception can be made for types.Message, since it doesn't have any purpose apart from invoking the state transition in package core. types.Message was also marked deprecated by the same commit it got added in, 4dca5d4db7 (November 2016). The core.Message interface was added in December 2014, in commit db494170dc, for the purpose of 'testing' state transitions. It's the same change that made transaction struct fields private. Before that, the state transition used *types.Transaction directly. Over time, multiple implementations of the interface accrued across different packages, since constructing a Message is required whenever one wants to invoke the state transition. These implementations all looked very similar, a struct with private fields exposing the fields as accessor methods. By changing Message into a struct with public fields we can remove all these useless interface implementations. It will also hopefully simplify future changes to the type with less updates to apply across all of go-ethereum when a field is added to Message. --------- Co-authored-by: Felix Lange <fjl@twurst.com> * travis: only build PPAs nightly, not on every push, too heavy (#26846) * p2p: small comment typo (#26850) Update server.go * core: add Timestamp method in BlockGen (#26844) Since forks are now scheduled by block time, it can be necessary to check the timestamp of a block while generating transactions. * core/txpool: implement additional DoS defenses (#26648) This adds two new rules to the transaction pool: - A future transaction can not evict a pending transaction. - A transaction can not overspend available funds of a sender. --- Co-authored-by: dwn1998 <42262393+dwn1998@users.noreply.github.com> Co-authored-by: Martin Holst Swende <martin@swende.se> * params: go-ethereum v1.11.4 stable * params: begin v1.11.5 release cycle * tests: define `MuirGlacier` fork (#26856) add muir glacier to t8n * code/vm: fix comment typo (#26865) it should be constantinople rather than contantinople * core: minor code refactor (#26852) * core: refactor code * core: drop it from this anonymous goroutine func * core/txpool: use priceList.Put instead of heap.Push (#26863) Minor refactor to use the 'intended' accessor * eth: return error if 'safe' or 'finalized' tag used pre-merge (#26862) Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: Felix Lange <fjl@twurst.com> * .travis.yml: reenable PPA build on tag push (#26873) * core/state, trie: port changes from PBSS (#26763) * p2p/discover: pass invalid discv5 packets to Unhandled channel (#26699) This makes it possible to run another protocol alongside discv5, by reading unhandled packets from the channel. * all: update links in documentation (#26882) Co-authored-by: Stephen Flynn <stephen.flynn@gapac.com> * Increase websocket frame size (from erigon rpc client) (#26883) This increases the maximum allowed message size to 32MB. Originally submitted at https://github.com/ledgerwatch/erigon/pull/2739 example block failure: https://etherscan.io/tx/0x1317d973a55cedf9b0f2df6ea48e8077dd176f5444a3423368a46d6e4db89982#internal * cmd/devp2p, cmd/geth: add version in --help output (#26895) Not sure why this was removed, it's pretty useful to see the version also in --help. * core: show db error-info in case of mismatched hash root (#26870) When a database failure occurs, bubble it up a into statedb, and report it in suitable places, such as during a 'bad block' report. * consensus: improve consensus engine definition (#26871) Makes clear the distinction between Finalize and FinalizedAndAssemble: - In Finalize function, a series of state operations are applied according to consensus rules. The statedb is mutated and the root hash can be checked and compared afterwards. This function should be used in block processing(receive afrom network and apply it locally) but not block generation. - In FinalizeAndAssemble function, after applying state mutations, the block is also to be assembled with the latest state root computed, updating the header. This function should be used in block generation only. * eth/catalyst: increase update consensus timeout (#26840) Increases the time between consensus updates that we give the CL before we start warning the user. * internal/ethapi: avoid int overflow in GetTransactionReceipt (#26911) * trie, accounts/abi: add error-checks (#26914) * rlp: support for uint256 (#26898) This adds built-in support in package rlp for encoding, decoding and generating code dealing with uint256.Int. --------- Co-authored-by: Felix Lange <fjl@twurst.com> * eth: fix output file permissions in admin_exportChain (#26912) * api: Use 0700 file permissions for ExportChain * change perm to 0644 * Update api.go --------- Co-authored-by: Felix Lange <fjl@twurst.com> * trie: reduce unit test time (#26918) * core/txpool: use atomic int added in go1.19 (#26913) Makes use of atomic.Uint64 instead of atomic by pointer * params: schedule shanghai fork on mainnet (#26908) Schedules the shanghai hardfork on timestamp 1681338455 as discussed on ACDE 157: https://github.com/ethereum/execution-specs/pull/727 * core/txpool: allow future local transactions (#26930) Local transactions should not be subject to the "future shouldn't churn pending txs" rule * params: go-ethereum v1.11.5 stable * params: begin v1.11.6 release cycle * build: allow building nightly archives via cron jobs (#26938) * log: add special casing of uint256 into the logger (#26936) * core/rawdb: use atomic int added in go1.19 (#26935) * core/vm: expose jumptable constructors (#26880) When interacting with geth as a library to e.g. produce state tests, it is desirable to obtain the consensus-correct jumptable definition for a given fork. This changes adds accessors so the instructionset can be obtained and characteristics about opcodes can be inspected. * eth/catalyst: fix races (#26950) * core/rawdb: update freezertable read meter (#26946) The meter for "for measuring the effective amount of data read" within the freezertable was never updated. This change remedies that. --------- Signed-off-by: jsvisa <delweng@gmail.com> * cmd/evm, tests: record preimages if dump is expected (#26955) With #25287 we made it so that preimages were not recorded by default. This had the side effect that the evm command is no longer able to dump state since it does a preimage lookup to determine the address represented by a key. This change enables the recording of preimages when the dump command is given. * core/state: add account address to Trie slot accessors (#26934) This changes the Trie interface to add the plain account address as a parameter to all storage-related methods. After the introduction of the TryAccount* functions, TryGet, TryUpdate and TryDelete are now only meant to read an account's storage. In their current form, they assume that an account storage is stored in a separate trie, and that the hashing of the slot is independent of its account's address. The proposed structure for a stateless storage breaks these two assumptions: the hashing of a slot key requires the address and all slots and accounts are stored in a single trie. This PR therefore adds an address parameter to the interface. It is ignored in the MPT version, so this change has no functional impact, however it will reduce the diff size when merging verkle trees. * metrics: add cpu counters (#26796) This PR adds counter metrics for the CPU system and the Geth process. Currently the only metrics available for these items are gauges. Gauges are fine when the consumer scrapes metrics data at the same interval as Geth produces new values (every 3 seconds), but it is likely that most consumers will not scrape that often. Intervals of 10, 15, or maybe even 30 seconds are probably more common. So the problem is, how does the consumer estimate what the CPU was doing in between scrapes. With a counter, it's easy ... you just subtract two successive values and divide by the time to get a nice, accurate average. But with a gauge, you can't do that. A gauge reading is an instantaneous picture of what was happening at that moment, but it gives you no idea about what was going on between scrapes. Taking an average of values is meaningless. * metrics/influxdb: use smaller dependency and reuse code between v1 and v2 reporters (#26963) This change switches to use the smaller influxdata/influxdb1-client package instead of depending on the whole infuxdb package. The new smaller client is very similar to the influxdb-v2 client, which made it possible to refactor the two reporters to reuse code a lot more. * eth/gasprice: change feehistory input type from int to uint64 (#26922) Change input param type from int to uint64 * go.mod: update golang.org/x/tools (#26960) * rlp/rlpgen: print want/expect output string if mismatch (#26932) Co-authored-by: Marius van der Wijden <m.vanderwijden@live.de> * ethclient: ensure returned subscription is nil on error (#26976) * core/state, trie: remove Try prefix in Trie accessors (#26975) This change renames StateTrie methods to remove the Try* prefix. We added the Trie methods with prefix 'Try' a long time ago, working around the problem that most existing methods of Trie did not return the database error. This weird naming convention has persisted until now. Co-authored-by: Gary Rong <garyrong0905@gmail.com> * metrics/librato: ensure resp.body closed (#26969) This change ensures that we call Close on a http response body, in various places in the source code (mostly tests) * core/vm: use atomic.Bool (#26951) Make use of new atomic types --------- Co-authored-by: Felix Lange <fjl@twurst.com> Co-authored-by: Martin Holst Swende <martin@swende.se> * core/bloombits: use atomic type (#26993) * core/state: use atomic.Bool (#26992) * graphql: fix data races (#26965) Fixes multiple data races caused by the fact that resolving fields are done concurrently by the graphql library. It also enforces caching at the stateobject level for account fields. * eth/tracers/native: prevent panic for LOG edge-cases (#26848) This PR fixes OOM panic in the callTracer as well as panicing on opcode validation errors (e.g. stack underflow) in callTracer and prestateTracer. Co-authored-by: Martin Holst Swende <martin@swende.se> * internal/debug: add log.logfmt flag to set logging to use logfmt (#26970) * docs: update outdated DeriveSha docs comment (#26968) * remove @gballet as a GraphQL codeowner (#27012) * core: use atomic type (#27011) * graphql: revert storage access regression (#27007) * cmd/geth: Add `--log.format` cli param (#27001) Removes the new --log.logfmt directive and hides --log.json, replacing both with log.format=(json|logfmt|terminal). The hidden log.json option is still respected if log.format is not specified for backwards compatibility. Co-authored-by: Martin Holst Swende <martin@swende.se> * ethdb/pebble: use atomic type (#27014) * common: fix json marshaller MixedcaseAddress (#26998) Fix the json marshaller of MixedcaseAddress * eth/catalyst: improve consensus heartbeat (#26896) improve the heartbeat function that is no longer suitable in the current situation Co-authored-by: “openex27” <“openexkevin@gmail.com”> * miner: use atomic type (#27013) Use the new typed atomics in the miner package * accounts/abi/bind: handle UnpackLog with zero topics (#26920) Adds error handling for the case that UnpackLog or UnpackLogIntoMap is called with a log that has zero topics. --------- Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * cmd/evm: use correct parent number for t8n base fee calculation (#27032) Currently the t8n tool uses the same block number for the current block and its parent while calculating the base fee. This causes incorrect base fee calculation for the london fork block. This commit sets the parent block number to be one less than the current block number * go.mod : update snappy (#27027) * common: delete MakeName (#27023) common,p2p: remove unused function MakeName * cmd/geth: enable log rotation (#26843) This change enables log rotation, which can be activated using the flag --log.rotate. Additional parameters that can be given are: - log.maxsize to set maximum size before files are rotated, - log.maxbackups to set how many files are retailed, - log.maxage to configure max age of rotated files, - log.compress whether to compress rotated files The way to configure location of the logfile(s) is left unchanged, via the `log.logfile` parameter. --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * cmd, miner, signer: avoid panic if keystore is not available (#27039) * cmd, miner, singer: avoid panic if keystore is not available * cmd/geth: print warning instead of panic * test/fuzzers: fuzz rlp handling of big.Int and uint256.Int (#26917) test/fuzzers: fuzz rlp handling of big.Lnt and uint256.Int * core/txpool: move some validation to outside of mutex (#27006) Currently, most of transaction validation while holding the txpool mutex: one exception being an early-on signature check. This PR changes that, so that we do all non-stateful checks before we entering the mutex area. This means they can be performed in parallel, and to enable that, certain fields have been made atomic bools and uint64. * eth/downloader: use atomic types (#27030) * eth/downloader: use atomic type * Update eth/downloader/downloader_test.go Co-authored-by: Martin Holst Swende <martin@swende.se> * Update eth/downloader/downloader_test.go Co-authored-by: Martin Holst Swende <martin@swende.se> --------- Co-authored-by: Martin Holst Swende <martin@swende.se> * core/vm: clarify comment (#27045) * consensus, core/typer: add 4844 excessDataGas to header, tie it to Cancun (#27046) * consensus/misc, params: add EIP-4844 blobfee conversions (#27041) * consensus/misc, params: add EIP-4844 blobfee conversions * consensus/misc: pull in fakeExponential test cases * consensus/misc: reuse bigints * consensus/misc: nit renames, additional larger testcase --------- Co-authored-by: Roberto Bayardo <bayardo@alum.mit.edu> Co-authored-by: Martin Holst Swende <martin@swende.se> * eth/tracers: report correct gasLimit in call tracers (#27029) This includes a semantic change to the `callTracer` as well as `flatCallTracer`. The value of field `gas` in the **first** call frame will change as follows: - It previously contained gas available after initial deductions (i.e. tx costs) - It will now contain the full tx gasLimit value Signed-off-by: jsvisa <delweng@gmail.com> * all: remove debug-field from vm config (#27048) This PR removes the Debug field from vmconfig, making it so that if a tracer is set, debug=true is implied. --------- Co-authored-by: 0xTylerHolmes <tyler@ethereum.org> Co-authored-by: Sina Mahmoodi <1591639+s1na@users.noreply.github.com> * metrics: make gauge_float64 and counter_float64 lock free (#27025) Makes the float-gauges lock-free name old time/op new time/op delta CounterFloat64Parallel-8 1.45µs ±10% 0.85µs ± 6% -41.65% (p=0.008 n=5+5) --------- Co-authored-by: Exca-DK <dev@DESKTOP-RI45P4J.localdomain> Co-authored-by: Martin Holst Swende <martin@swende.se> * eth/tracers: use atomic type (#27031) Use the new atomic types in package eth/tracers --------- Co-authored-by: Martin Holst Swende <martin@swende.se> Co-authored-by: Sina Mahmoodi <itz.s1na@gmail.com> * build: upgrade -dlgo version to Go 1.20.3 * core/txpool: disallow future churn by remote txs (#26907) Prior to this change, it was possible that transactions are erroneously deemed as 'future' although they are in fact 'pending', causing them to be dropped due to 'future' not being allowed to replace 'pending'. This change fixes that, by doing a more in-depth inspection of the queue. * core, miner: drop transactions from the same sender when error occurs (#27038) This PR unifies the error handling in miner. Whenever an error occur while applying a transaction, the transaction should be regarded as invalid and all following transactions from the same sender not executable because of the nonce restriction. The only exception is the `nonceTooLow` error which is handled separately. * params: new sepolia bootnodes (#27099) New sepolia bootnodes managed by EF devops * cmd/devp2p: fix erroneous log output in crawler (#27089) cmd/devp2p: fix log of ignored recent nodes counter * signer/core: rename testdata files (#27063) Sets a meaningful na…

Currently, most of transaction validation while holding the txpool mutex: one exception being an early-on signature check. This PR changes that, so that we do all non-stateful checks before we entering the mutex area. This means they can be performed in parallel, and to enable that, certain fields have been made atomic bools and uint64.

core/txpool: move some validation to outside of mutex

300ef9d

holiman requested review from karalabe and rjl493456442 as code owners March 29, 2023 09:13

holiman commented Mar 29, 2023

View reviewed changes

core/txpool/txpool.go Outdated Show resolved Hide resolved

holiman added 2 commits March 29, 2023 07:13

Update core/txpool/txpool.go

0defec2

Update txpool.go

acbc02a

holiman commented Mar 29, 2023

View reviewed changes

core/txpool: fix test regression due to changed order of checks

e3ce6ca

rjl493456442 reviewed Apr 3, 2023

View reviewed changes

holiman commented Apr 3, 2023

View reviewed changes

core/txpool/txpool.go Outdated Show resolved Hide resolved

core/txpool: drop basics-validation internally

0ceb08f

rjl493456442 approved these changes Apr 3, 2023

View reviewed changes

holiman merged commit beda6c4 into ethereum:master Apr 3, 2023

holiman added this to the 1.11.6 milestone Apr 3, 2023

holiman deleted the txpool_validation branch October 11, 2023 07:28

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

core/txpool: move some validation to outside of mutex #27006

core/txpool: move some validation to outside of mutex #27006

holiman commented Mar 29, 2023

holiman Mar 29, 2023

rjl493456442 Apr 3, 2023 •

edited

Loading

fjl commented Apr 1, 2023

rjl493456442 left a comment

holiman commented Apr 3, 2023 •

edited

Loading

core/txpool: move some validation to outside of mutex #27006

core/txpool: move some validation to outside of mutex #27006

Conversation

holiman commented Mar 29, 2023

holiman Mar 29, 2023

Choose a reason for hiding this comment

rjl493456442 Apr 3, 2023 • edited Loading

Choose a reason for hiding this comment

fjl commented Apr 1, 2023

rjl493456442 left a comment

Choose a reason for hiding this comment

holiman commented Apr 3, 2023 • edited Loading

rjl493456442 Apr 3, 2023 •

edited

Loading

holiman commented Apr 3, 2023 •

edited

Loading