This pull request removes the `fsync` of index files in freezer.ModifyAncients function for
performance gain.
Originally, fsync is added after each freezer write operation to ensure
the written data is truly transferred into disk. Unfortunately, it turns
out `fsync` can be relatively slow, especially on
macOS (see https://github.com/ethereum/go-ethereum/issues/28754 for more
information).
In this pull request, fsync for index file is removed as it turns out
index file can be recovered even after a unclean shutdown. But fsync for data file is still kept, as
we have no meaningful way to validate the data correctness after unclean shutdown.
---
**But why do we need the `fsync` in the first place?**
As it's necessary for freezer to survive/recover after the machine crash
(e.g. power failure).
In linux, whenever the file write is performed, the file metadata update
and data update are
not necessarily performed at the same time. Typically, the metadata will
be flushed/journalled
ahead of the file data. Therefore, we make the pessimistic assumption
that the file is first
extended with invalid "garbage" data (normally zero bytes) and that
afterwards the correct
data replaces the garbage.
We have observed that the index file of the freezer often contain
garbage entry with zero value
(filenumber = 0, offset = 0) after a machine power failure. It proves
that the index file is extended
without the data being flushed. And this corruption can destroy the
whole freezer data eventually.
Performing fsync after each write operation can reduce the time window
for data to be transferred
to the disk and ensure the correctness of the data in the disk to the
greatest extent.
---
**How can we maintain this guarantee without relying on fsync?**
Because the items in the index file are strictly in order, we can
leverage this characteristic to
detect the corruption and truncate them when freezer is opened.
Specifically these validation
rules are performed for each index file:
For two consecutive index items:
- If their file numbers are the same, then the offset of the latter one
MUST not be less than that of the former.
- If the file number of the latter one is equal to that of the former
plus one, then the offset of the latter one MUST not be 0.
- If their file numbers are not equal, and the latter's file number is
not equal to the former plus 1, the latter one is valid
And also, for the first non-head item, it must refer to the earliest
data file, or the next file if the
earliest file is not sufficient to place the first item(very special
case, only theoretical possible
in tests)
With these validation rules, we can detect the invalid item in index
file with greatest possibility.
---
But unfortunately, these scenarios are not covered and could still lead
to a freezer corruption if it occurs:
**All items in index file are in zero value**
It's impossible to distinguish if they are truly zero (e.g. all the data
entries maintained in freezer
are zero size) or just the garbage left by OS. In this case, these index
items will be kept by truncating
the entire data file, namely the freezer is corrupted.
However, we can consider that the probability of this situation
occurring is quite low, and even
if it occurs, the freezer can be considered to be close to an empty
state. Rerun the state sync
should be acceptable.
**Index file is integral while relative data file is corrupted**
It might be possible the data file is corrupted whose file size is
extended correctly with garbage
filled (e.g. zero bytes). In this case, it's impossible to detect the
corruption by index validation.
We can either choose to `fsync` the data file, or blindly believe that
if index file is integral then
the data file could be integral with very high chance. In this pull
request, the first option is taken.
Reverts ethereum/go-ethereum#30495
You are free to create a proper Clear method if that's the best way. But
one that does a proper cleanup, not some hacky call to set gas which
screws up logs, metrics and everything along the way. Also doesn't work
for legacy pool local transactions.
The current code had a hack in the simulated code, now we have a hack in
live txpooling code. No, that's not acceptable. I want the live code to
be proper, meaningful API, meaningful comments, meaningful
implementation.
Here we move the method that drops all transactions by temporarily increasing the fee
into the TxPool itself. It's better to have it there because we can set it back to the
configured value afterwards. This resolves a TODO in the simulated backend.
Extends the opcontext interface to include accessor for code being executed in current context. While it is possible to get the code via `statedb.GetCode`, that approach doesn't work for initcode.
This pull request skips the state snapshot update if the base layer is
not existent, eliminating the numerous warning logs after an unclean
shutdown.
Specifically, Geth will rewind its chain head to a historical block
after unclean shutdown and state snapshot will be remained as unchanged
waiting for recovery. During this period of time, the snapshot is unusable
and all state updates should be ignored/skipped for state snapshot update.
This PR integrates witness-enabled block production, witness-creating
payload execution and stateless cross-validation into the `engine` API.
The purpose of the PR is to enable the following use-cases (for API
details, please see next section):
- Cross validating locally created blocks:
- Call `forkchoiceUpdatedWithWitness` instead of `forkchoiceUpdated` to
trigger witness creation too.
- Call `getPayload` as before to retrieve the new block and also the
above created witness.
- Call `executeStatelessPayload` against another client to
cross-validate the block.
- Cross validating locally processed blocks:
- Call `newPayloadWithWitness` instead of `newPayload` to trigger
witness creation too.
- Call `executeStatelessPayload` against another client to
cross-validate the block.
- Block production for stateless clients (local or MEV builders):
- Call `forkchoiceUpdatedWithWitness` instead of `forkchoiceUpdated` to
trigger witness creation too.
- Call `getPayload` as before to retrieve the new block and also the
above created witness.
- Propagate witnesses across the consensus libp2p network for stateless
Ethereum.
- Stateless validator validation:
- Call `executeStatelessPayload` with the propagated witness to
statelessly validate the block.
*Note, the various `WithWitness` methods could also *just be* an
additional boolean flag on the base methods, but this PR wanted to keep
the methods separate until a final consensus is reached on how to
integrate in production.*
---
The following `engine` API types are introduced:
```go
// StatelessPayloadStatusV1 is the result of a stateless payload execution.
type StatelessPayloadStatusV1 struct {
Status string `json:"status"`
StateRoot common.Hash `json:"stateRoot"`
ReceiptsRoot common.Hash `json:"receiptsRoot"`
ValidationError *string `json:"validationError"`
}
```
- Add `forkchoiceUpdatedWithWitnessV1,2,3` with same params and returns
as `forkchoiceUpdatedV1,2,3`, but triggering a stateless witness
building if block production is requested.
- Extend `getPayloadV2,3` to return `executionPayloadEnvelope` with an
additional `witness` field of type `bytes` iff created via
`forkchoiceUpdatedWithWitnessV2,3`.
- Add `newPayloadWithWitnessV1,2,3,4` with same params and returns as
`newPayloadV1,2,3,4`, but triggering a stateless witness creation during
payload execution to allow cross validating it.
- Extend `payloadStatusV1` with a `witness` field of type `bytes` if
returned by `newPayloadWithWitnessV1,2,3,4`.
- Add `executeStatelessPayloadV1,2,3,4` with same base params as
`newPayloadV1,2,3,4` and one more additional param (`witness`) of type
`bytes`. The method returns `statelessPayloadStatusV1`, which mirrors
`payloadStatusV1` but replaces `latestValidHash` with `stateRoot` and
`receiptRoot`.
After this PR, https://github.com/ethereum/go-ethereum/pull/28187, the
way to set the default logger is different. This PR only updates the way
to set logger in some test cases' comments that existed in the codebase
(since this commit
https://github.com/ethereum/go-ethereum/commit/b63e3c37a6). Although I
am not sure if it a good way to leave the code in the comment, it truly
makes me more efficiently to debug and fix the failing test cases.
Add changes from #30409 and #29338 to changelog.
---------
Co-authored-by: Martin HS <martin@swende.se>
Co-authored-by: Guillaume Ballet <3272758+gballet@users.noreply.github.com>
This change makes the code slightly easier for downstream-projects to extend with more signer-types, but if functionalily equivalent to the previous code.
Remove redundant address presence check in `makeGasSStoreFunc`.
This PR simplifies the `makeGasSStoreFunc` function by removing the
redundant check for address presence in the access list. The updated
code now only checks for slot presence, streamlining the logic and
eliminating unnecessary panic conditions.
This change removes the unnecessary address presence check, simplifying
the code and improving maintainability without affecting functionality.
The previous panic condition was intended as a canary during the testing
phases (i.e. _YOLOv2_) and is no longer needed.
#29995 has been reverted due to an unexpected flaw in the state snapshot
process.
Specifically, it attempts to stop the state snapshot generation, which
could potentially
cause the system to halt if the generation is not currently running.
This pull request ports the changes made in #29995 and fixes the flaw.
This is a successor PR to #25743. This PR is based on a new iteration of
the spec: https://github.com/ethereum/execution-apis/pull/484.
`eth_multicall` takes in a list of blocks, each optionally overriding
fields like number, timestamp, etc. of a base block. Each block can
include calls. At each block users can override the state. There are
extra features, such as:
- Include ether transfers as part of the logs
- Overriding precompile codes with evm bytecode
- Redirecting accounts to another address
## Breaking changes
This PR includes the following breaking changes:
- Block override fields of eth_call and debug_traceCall have had the
following fields renamed
- `coinbase` -> `feeRecipient`
- `random` -> `prevRandao`
- `baseFee` -> `baseFeePerGas`
---------
Co-authored-by: Gary Rong <garyrong0905@gmail.com>
Co-authored-by: Martin Holst Swende <martin@swende.se>
This pull request replaces the field pointer in journal entry with the
field itself, specifically the address of mutated account.
While it will introduce the extra allocation cost, but it's easier for
code reading. Let's measure the overhead overall to see if the change is
acceptable or not.
This pull request introduces a state.Reader interface for state
accessing.
The interface could be implemented in various ways. It can be pure trie
only reader, or the combination of trie and state snapshot. What's more,
this interface allows us to have more flexibility in the future, e.g.
the
archive reader (for accessing archive state).
Additionally, this pull request removes the following metrics
- `chain/snapshot/account/reads`
- `chain/snapshot/storage/reads`
This PR changes how sidechains are handled.
Before the merge, it was possible to import a chain with lower td and not set it as canonical. After the merge, we expect every chain that we get via InsertChain to be canonical. Non-canonical blocks can still be inserted
with InsertBlockWIthoutSetHead.
If during the InsertChain, the existing chain is not canonical anymore, we mark it as a sidechain and send the SideChainEvents normally.
This pull request fixes a flaw in prefetcher.
In verkle tree world, both accounts and storage slots are committed into
a single tree instance for state hashing. If the prefetcher is activated, we will
try to pull the trie for the prefetcher for performance speedup.
However, we had a special logic to skip pulling storage trie if the
storage root is empty. While it's true for merkle as we have nothing to
do with an empty storage trie, it's totally wrong for verkle. The consequences
for skipping pulling is the storage changes are committed into trie A, while the
account changes are committed into trie B (pulled from the prefetcher), boom.
This PR implements changes related to
[EIP-6800](https://eips.ethereum.org/EIPS/eip-6800) and
[EIP-4762](https://eips.ethereum.org/EIPS/eip-4762) spec updates.
A TL;DR of the changes is that `Version`, `Balance`, `Nonce` and
`CodeSize` are encoded in a single leaf named `BasicData`. For more
details, see the [_Header Values_ table in
EIP-6800](https://eips.ethereum.org/EIPS/eip-6800#header-values).
The motivation for this was simplifying access event patterns, reducing
code complexity, and, as a side effect, saving gas since fewer leaf
nodes must be accessed.
---------
Co-authored-by: Guillaume Ballet <3272758+gballet@users.noreply.github.com>
Co-authored-by: Felix Lange <fjl@twurst.com>
This PR adds the bulk verkle witness+proof production at the end of block
production. It reads all data from the tree in one swoop and produces
a verkle proof.
Co-authored-by: Felix Lange <fjl@twurst.com>
This is a follow-up to #29520, and a preparatory PR to a more thorough
change in the journalling system.
### API methods instead of `append` operations
This PR hides the journal-implementation details away, so that the
statedb invokes methods like `JournalCreate`, instead of explicitly
appending journal-events in a list. This means that it's up to the
journal whether to implement it as a sequence of events or
aggregate/merge events.
### Snapshot-management inside the journal
This PR also makes it so that management of valid snapshots is moved
inside the journal, exposed via the methods `Snapshot() int` and
`RevertToSnapshot(revid int, s *StateDB)`.
### SetCode
JournalSetCode journals the setting of code: it is implicit that the
previous values were "no code" and emptyCodeHash. Therefore, we can
simplify the setCode journal.
### Selfdestruct
The self-destruct journalling is a bit strange: we allow the
selfdestruct operation to be journalled several times. This makes it so
that we also are forced to store whether the account was already
destructed.
What we can do instead, is to only journal the first destruction, and
after that only journal balance-changes, but not journal the
selfdestruct itself.
This simplifies the journalling, so that internals about state
management does not leak into the journal-API.
### Preimages
Preimages were, for some reason, integrated into the journal management,
despite not being a consensus-critical data structure. This PR undoes
that.
---------
Co-authored-by: Gary Rong <garyrong0905@gmail.com>
This pull request adds a few more performance metrics, specifically:
- The average time cost of an account read
- The average time cost of a storage read
- The rate of account reads
- The rate of storage reads
To allow all error paths in `vm.EVM.create()` to consume the necessary
gas, there is currently a pattern of gating code on `if err == nil`
instead of returning as soon as the error occurs. The same behaviour can
be achieved by abstracting the gated code into a method that returns
immediately on error, improving readability and thus making it easier to
understand and maintain.
This PR refactors the genesis initialization a bit, s.th. we only
compute the blockhash once instead of twice as before (during hashAlloc
and flushAlloc)
This will significantly reduce the amount of memory allocated during
genesis init
---------
Co-authored-by: Gary Rong <garyrong0905@gmail.com>
Fixes a slight miscalculation in the downloader queue, which was not accurately taking block withdrawals into account when calculating the size of the items in the queue
Consistently use `uint64` for indices in `Memory` and drop lots of type
conversions from `uint64` to `int64`.
---------
Co-authored-by: lmittmann <lmittmann@users.noreply.github.com>
The struct-based tracing added in #29189 seems to have caused an issue
with the benchmark `BenchmarkTracerStepVsCallFrame`. On master we see
the following panic:
```console
BenchmarkTracerStepVsCallFrame
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x2 addr=0x40 pc=0x1019782f0]
goroutine 37 [running]:
github.com/ethereum/go-ethereum/eth/tracers/js.(*jsTracer).OnOpcode(0x140004c4000, 0x0, 0x10?, 0x989680, 0x1, {0x101ea2298, 0x1400000e258}, {0x1400000e258?, 0x14000155928?, 0x10173020c?}, ...)
/Users/matt/dev/go-ethereum/eth/tracers/js/goja.go:328 +0x140
github.com/ethereum/go-ethereum/core/vm.(*EVMInterpreter).Run(0x14000307da0, 0x140003cc0d0, {0x0, 0x0, 0x0}, 0x0)
...
FAIL github.com/ethereum/go-ethereum/core/vm/runtime 0.420s
FAIL
```
The issue seems to be that `OnOpcode` expects that `OnTxStart` has
already been called to initialize the `env` value in the tracer. The JS
tracer uses it in `OnOpcode` for the `GetRefund()` method.
This patch resolves the issue by reusing the `Call` method already
defined in `runtime_test.go` which correctly calls `OnTxStart`.
