Closes: #760
Solution:
- emit tx index to cosmos events
- rpc side try to use the events, but fallback to heavier approach when fails.
Update rpc/ethereum/namespaces/eth/api.go
changelog
fix lint
fix TxIndexFromEvents
fix
Update rpc/ethereum/backend/backend.go
Co-authored-by: Federico Kunze Küllmer <31522760+fedekunze@users.noreply.github.com>
This change fixes an insidious bug that unfortunately
tried to split values in multiples of 32, but unfortunately
due to the loop conditions, if the length of trace.Memory
was less than 32, nothing would be added; if the value wasn't
a multiple of 32, the ends wouldn't be added in.
Fixes#837
* Use effectiveGasPrice in ante handler for dynamic fee tx
Closes: #814
Solution:
- use effectiveGasPrice when check minimal-gas-prices, and deduct fee in ante handler
- implement an EthMempoolFeeDecorator
* add effectiveGasPrice to tx receipt
* changelog
* fix unit test
* fix comments
* add comments
* Apply suggestions from code review
Co-authored-by: Thomas Nguy <81727899+thomas-nguy@users.noreply.github.com>
* review suggestions
Co-authored-by: Thomas Nguy <81727899+thomas-nguy@users.noreply.github.com>
Co-authored-by: Federico Kunze Küllmer <31522760+fedekunze@users.noreply.github.com>
Uses the fastest slice making idiom of creating the well known
size of a slice using
make([]sdk.Attribute, len(response.Logs))
for i, log := range response.Logs {
txLogAttrs[i] = ...
}
instead of
make([]sdk.Attribute, 0)
for _, log := range response.Logs {
txLogAttrs = append(txLogAttrs, ...)
}
which had a few problems:
1. Using 0 for size then appending is quite slow yet we know the exact size
2. Using append instead of indexing is slower
If we examine the advisory at https://bencher.orijtech.com/perfclinic/sliceupdate/
and the verdict at https://bencher.orijtech.com/perfclinic/sliceupdate/#verdict
this new scheme shows a massive improvement in that call site.
Fixes#825
* x/evm/keeper: save 24B with Go in-condition variable idiom
The prior code doubly invoked (*ethereum/go-ethereum/core/types.Transaction).To()
which is quite expensive, and firstly copies 20 bytes each time, then
that gets rounded up to the proper size class/pointer alignment so on
64-bit machines 20B -> 24B.
Isolating a benchmark for this code per issue #826 shows this saves
quite a bit of bytes and some nano seconds which all count up towards
the transactions per seconds being processed:
```shell
$ benchstat before.txt after.txt
name old time/op new time/op delta
CopyAddr-8 38.4ns ± 3% 19.3ns ± 3% -49.66% (p=0.000 n=10+10)
name old alloc/op new alloc/op delta
CopyAddr-8 48.0B ± 0% 24.0B ± 0% -50.00% (p=0.000 n=10+10)
name old allocs/op new allocs/op delta
CopyAddr-8 2.00 ± 0% 1.00 ± 0% -50.00% (p=0.000 n=10+10)
```
Fixes#826
* changelog
* lint
* Revert stray changes that were used in testing
* rm log
Co-authored-by: Federico Kunze Küllmer <federico.kunze94@gmail.com>
* Problem: ForEachStorage sematic not compatible with go-ethereum
Solution:
- reversed the semantic of return value of the callback function.
* changelog
Co-authored-by: Federico Kunze Küllmer <31522760+fedekunze@users.noreply.github.com>
Closes#709
fix index
Apply suggestions from code review
Co-authored-by: Federico Kunze Küllmer <31522760+fedekunze@users.noreply.github.com>
fix lint
transaction decoding unit test
test BuildTx
fix lint
changelog
* Problem: state transition code is duplicated
Closes: #672
Solution:
- move gas refund out from ApplyMessage
- move check into ApplyMessage
- move evm construction into ApplyMessage
- ensure context stack is clean after ApplyMessage return
fix unit tests
undo rename
add underflow check
* improve performance
- don't duplicate params loading
- passing EVMConfig around as pointer
* Problem: No test on the transaction receipt api
Closes: #582
- add receipt rpc test for erc20 transfer logs
* lower gas fee
* build with go 1.17 in CI
* use go 1.17 in test-solidity
* fix merge
* test DynamicFeeTx against state_transition_benchmark_test
* add feemarketGenesis in the app setup param
* add dynamicTxFee flag to KeeperTestSuite
* add feemarketGenesis.BaseFee setup
* update TestAddLog
* fix gasFeeCap assignment in newMsgEthereumTx
* modify keeperTestSuite helper functions to support dynamicTxFee
* update test cases in grpc_query_test w/ dynamicTxFee
* update the evm keeper utils tests
* add dynamic tx fee in the ante tests
* remove duplicate type define
* fix error return type
* update changelog
Co-authored-by: Federico Kunze Küllmer <31522760+fedekunze@users.noreply.github.com>
* evm: fixed commented out simulations, pubsub, and handler tests
fixes#640
simulations are very basic: they can be built and executed,
but they don't generate any EVM-related transactions yet.
(It should be a matter of adding simulation-related code to the
modules + potentially extra helpers to the simulation.)
handler tests miss some extra assertions due to changes
in the return values snapshotting logic (ADR-001 and ADR-002).
Besides the test suites identified in the audit,
there's also "importer_test.go" which wasn't yet fixed.
(it'd require major rewriting + extra test resources)
* gofumpt
* evm: unit tests
* Add unit tests for DynamicFeeTx.Validate()
* Start get and set signature values tests
* get set values
* Add tests for GetTo()
* Add GetNonce test
* Add GetValue test
* Start copy test
* Add WIP newDynamicFeeTx test
* Add WIP legacy_tx_test
* pair programming session
* Add TestLegacyTxValidate
* Add TestLegacyTxSetSignatureValues & GetSignatureValues
* Add legacyTx tests
* Merge main, forgot to save one file
* Add AccessList tests
* Add chain Config (fork order)
* Add invalid genesis account test
* Add params tests
* Add WIP tracer test
* tracer tests
* Add FormatLogs tests
* Add NewNoOpTracer test
* Refactor to test suite
* Refactor Tx Test suits to only use TxDataTestSuite
* Update link to geth interpreter
* Update x/evm/types/params.go
* Refactor accessListTx Test suits to use TxDataTestSuite
Co-authored-by: Daniel Burckhardt <daniel.m.burckhardt@gmail.com>
* Problem: deep context stack efficienty is not benchmarked
Closes: #626
Solution:
- add a benchmark to demonstrate an extremely inefficiency in deep
context stack
* Update x/evm/keeper/benchmark_test.go
* prefix storage is irrelevant
* add comment to state_transition.go
Co-authored-by: Federico Kunze Küllmer <31522760+fedekunze@users.noreply.github.com>
