## Why this should be merged
Avoids false positives from `goheader`.
## How this works
Checks out entire history to give the linter a reference point for new
issues only.
## How this was tested
With great frustration by waiting for CI.
## Why this should be merged
`goheader` linter apparently hasn't been running but now decided to and
is blocking another PR. It's not worth the time to investigatie why
(probably a version update).
## How this works
Update header comments as required.
## How this was tested
N/A
## Why this should be merged
Addition of recent, functionality constitutes a bump to the minor
version. Doing this on `main` first for consistency and will cut a
release candidate after.
## How this works
Magic
## How this was tested
Super magic
## Why this should be merged
To allow more thorough handling of duplicate state roots (or any other
info other users would like), additional information can be provided to
a call of `statedb.Commit`. This change allows arbitrary types to be
sent to `triedb` as well as the `SnapshotTree`. However, this is a
breaking change for those using the functionality already, since the
snapshot commit option is wrapped with another call.
## How this works
See the edited libevm test for usage.
## How this was tested
Edited test case to include `TrieDBUpdateOption` and ensures the payload
is sent.
## Why this should be merged
#185 was allowed to merge despite CI requiring the `lint` job.
## How this works
Fix linter issues.
## How this was tested
The first commit is an empty commit, expected to fail CI, to confirm
that the job is in fact required.
## Why this should be merged
Required for
[ACP-194](https://github.com/avalanche-foundation/ACPs/tree/main/ACPs/194-streaming-asynchronous-execution#gas-charged)
$\lambda$ bound on gas consumption.
## How this works
Hook into `core.StateTransition.TransitionDb()` as this is the bottom of
all execution paths (e.g. `core.ApplyTransaction()` as used in SAE,
`core.StateProcessor.Process(*Block,...)`, etc.). Once consumed gas is
no longer changing (i.e. after all spends and refunds), the transaction
limit is passed to the hook to determine the minimum consumption, which
is applied.
## How this was tested
Unit test via `core.ApplyTransaction()` as this is our entry point in
SAE.
---------
Signed-off-by: Arran Schlosberg <519948+ARR4N@users.noreply.github.com>
Co-authored-by: Stephen Buttolph <stephen@avalabs.org>
## Why this should be merged
Our old strategy of only cherry-picking on release branches made the
`main` branch incompatible with dependent repos.
## How this works
The goal of the old strategy was to avoid cherry-picking on the same
branch to which we would later merge the duplicated, upstream commit.
This can instead be achieved by tracking the cherry-picked commits and
reverting them as part of the geth sync. Since we already do this for
our own changes and a single cherry-pick (see #128), there's no need to
use the two approaches simultaneously.
This PR deletes the cherry-picking mechanism and removes the
release-branch test that enforced its proper usage. It will be followed
up by a series of PRs, one per cherry-pick that would have otherwise
been placed on release branches.
## How this was tested
n/a
## Why this should be merged
To override the standard trie behavior for custom databases in
`core/state/database.go`, custom methods may need to be defined on the
backend database.
## How this works
Exposes `backend` interface for the API
## How this was tested
No tests necessary - isn't used in `libevm`.
## Why this should be merged
Exposes API implementations for use by SAE.
## How this works
Type aliases and constructor propagation.
## How this was tested
N/A
## Why this should be merged
Importing some `avalanchego` test fixtures automatically imports
`coreth` packages that register extras, which affects testing of other
VMs that depend on `libevm`. All other registered extras can be cleared
with the pattern in this PR.
## How this works
Uses the `register.AtMostOnce` type already used in `params` and
`types`. I forgot to update the `vm` package when implementing the type.
## How this was tested
Existing unit tests.
## Why this should be merged
This enforces an assumption that is currently made by
`VerifyRangeProof`. If a key is a prefix of another key, the current
code can panic.
## How this works
Enforces all keys have the same length.
## How this was tested
Added a unit test.
## Why this should be merged
The release-branch CI job tests branch properties of the `git` DAG but
GitHub's `actions/checkout` by default creates a speculative merge
commit against which PR CI is run. Running against this hypothetical
situation makes the tests fail.
## How this works
[`actions/checkout` uses PR
tip](https://github.com/actions/checkout?tab=readme-ov-file#checkout-pull-request-head-commit-instead-of-merge-commit)
## How this was tested
Future run against a release branch.
---------
Signed-off-by: Quentin McGaw <quentin.mcgaw@gmail.com>
- Use grep `-E` flag instead of `-P` flag so it works on all platforms
- Add `-S` flag to `git cherry-pick` since it doesn't always sign all commits after resolving conflicts
## Why this should be merged
Debugging release-branch test failures in CI but not locally.
## How this works
Logs `HEAD` before testing branch properties.
## How this was tested
NA
## Why this should be merged
The `metrics.Enabled` global is misleadingly named and disables
_constructors_, not individual metrics when they're used. This resulted
in all `ava-labs/coreth` metrics being no-ops because it was only set to
`true` after they were all constructed. As the majority of metrics are
globals, they're created during package initialisation so simply setting
`metrics.Enabled = true` somewhere in `coreth` isn't a guarantee of
ordering. The old `coreth/metrics` package simply set the global to
`true` by default.
## How this works
`metrics.Enabled = true` at the point of declaration.
## How this was tested
Local inspection of metrics on a running binary.
## Why this should be merged
The CI job in #145 requires a local (to the runner) `main` branch and
since `actions/checkout` doesn't check it out, the workflow creates the
branch off `origin/main`. However, after merge of a PR to `main`, the
workflow is run on `main` itself and that step [fails because the branch
already
exists](https://github.com/ava-labs/libevm/actions/runs/13908218530/job/38916101218).
## How this works
Only call `git branch main origin/main` if `main` doesn't exist.
## How this was tested
We'll see once it's merged 🤷
> [!NOTE]
> This will be merged into `main` once the current target branch has
been merged.
## Why this should be merged
My original implementation was back to front and it's been bugging me.
## How this works
Originally `params.LibEVMVersion` was a `var` (because it needed to be
computed) and the test used a `const`. This change simply inverts the
two and moves some code around without any change in logic.
I bumped the minor version to 2 (a no-op when not on a release branch)
to bring it in line with the latest release candidate and to avoid
forgetting to do so when performing an actual release.
## How this was tested
Unit test of version-string constant.
## Why this should be merged
Safer release process by enforcing invariants of `release/*` branches as
automated in #137 to fulfil #25.
## How this works
New test for `go_tooling` CI job.
If the PR target branch is `main` then only the `params.ReleaseType` is
checked. If the target is neither `main` nor a release branch then the
test is skipped. The checks performed on `release/*` branches are
described in the test.
## How this was tested
Locally against a dummy release branch with deliberate problems created
by (a) including this PR's changes in the final commit and (b) not
updating the libevm version.
```
$ go test -v ./... --target_branch="release/v1.13.14-0.1.0.rc.3"
=== RUN TestCherryPicksFormat
--- PASS: TestCherryPicksFormat (0.39s)
=== RUN TestBranchProperties
=== RUN TestBranchProperties/branch_name
release_test.go:172:
Error Trace: .../ava-labs/libevm/libevm/tooling/release/release_test.go:172
Error: Not equal:
expected: "release/v1.13.14-0.1.0.beta"
actual : "release/v1.13.14-0.1.0.rc.3"
Diff:
--- Expected
+++ Actual
@@ -1 +1 @@
-release/v1.13.14-0.1.0.beta
+release/v1.13.14-0.1.0.rc.3
Test: TestBranchProperties/branch_name
release_test.go:175: On release branch; params.LibEVMReleaseType = "beta", which is unsuitable for release branches
=== RUN TestBranchProperties/commit_history
release_test.go:192: Forked from "main" at commit 4c6e50e7d1 (chore: module-renaming workflow inverts between `libevm` and `geth` (#152))
release_test.go:314: ### History since fork from default branch (8 commits):
release_test.go:316: internal/build, rpc: add missing HTTP response body Close() calls (#29223) by Shiming Zhang <wzshiming@hotmail.com>
release_test.go:316: core/state: fix bug in statedb.Copy and remove unnecessary preallocation (#29563) by Aaron Chen <aaronchen.lisp@gmail.com>
release_test.go:316: params: print time value instead of pointer in ConfigCompatError (#29514) by Nathan <Nathan.l@nodereal.io>
release_test.go:316: eth/gasprice: add query limit for FeeHistory to defend DDOS attack (#29644) by Nathan <Nathan.l@nodereal.io>
release_test.go:316: core/state/snapshot: add a missing lock (#30001) by maskpp <maskpp266@gmail.com>
release_test.go:316: crypto: add IsOnCurve check (#31100) by Felix Lange <fjl@twurst.com>
release_test.go:316: internal/ethapi: fix panic in debug methods (#31157) by Sina M <1591639+s1na@users.noreply.github.com>
release_test.go:316: x by Arran Schlosberg <me@arranschlosberg.com>
=== RUN TestBranchProperties/commit_history/cherry_picked_commits
release_test.go:314: ### Expected cherry-picks (7 commits):
release_test.go:316: internal/build, rpc: add missing HTTP response body Close() calls (#29223) by Shiming Zhang <wzshiming@hotmail.com>
release_test.go:316: core/state: fix bug in statedb.Copy and remove unnecessary preallocation (#29563) by Aaron Chen <aaronchen.lisp@gmail.com>
release_test.go:316: params: print time value instead of pointer in ConfigCompatError (#29514) by Nathan <Nathan.l@nodereal.io>
release_test.go:316: eth/gasprice: add query limit for FeeHistory to defend DDOS attack (#29644) by Nathan <Nathan.l@nodereal.io>
release_test.go:316: core/state/snapshot: add a missing lock (#30001) by maskpp <maskpp266@gmail.com>
release_test.go:316: crypto: add IsOnCurve check (#31100) by Felix Lange <fjl@twurst.com>
release_test.go:316: internal/ethapi: fix panic in debug methods (#31157) by Sina M <1591639+s1na@users.noreply.github.com>
=== RUN TestBranchProperties/commit_history/final_commit
release_test.go:365: Modified disallowed file "go.yml"
release_test.go:365: Modified disallowed file "go.mod"
release_test.go:365: Modified disallowed file "go.sum"
release_test.go:365: Modified disallowed file "release_test.go"
--- FAIL: TestBranchProperties (2.07s)
--- FAIL: TestBranchProperties/branch_name (0.00s)
--- FAIL: TestBranchProperties/commit_history (2.07s)
--- PASS: TestBranchProperties/commit_history/cherry_picked_commits (1.81s)
--- FAIL: TestBranchProperties/commit_history/final_commit (0.01s)
FAIL
FAIL github.com/ava-labs/libevm/libevm/tooling/release 2.712s
FAIL
```
## Why this should be merged
Allows `core/types/hashing.go` to be deleted from `coreth`.
## How this works
Exported function acts as a proxy for unexported upstream function.
## How this was tested
NA
## Why this should be merged
Originally I'd planned on doing an upstream sync by running the rename
workflow on the incoming commit and then merging it to `main` however
this resulted in hundreds of merge commits that were solely due to Go
imports. Renaming the module from `ava-labs/libevm` to
`ethereum/go-ethereum` removed 90% of conflicts (H/T @darioush). The
module will then need to be named `ava-labs/libevm` again, so the commit
history will probably[^1] look like this after an update:
```mermaid
---
config:
gitGraph:
mainBranchName: main
parallelCommits: true
---
gitGraph TB:
branch geth order:2
commit id:"geth@v1.15.2"
checkout main
commit id:"libevm@v1.13.14"
branch sync/v1.15.2
commit id:"[AUTO] rename to ethereum/go-ethereum"
merge geth
commit id:"[AUTO] rename to ava-labs/libevm"
checkout main
merge sync/v1.15.2 id:"libevm@v1.15.2"
```
[^1]: Specifics of and rationale behind the merge strategy are beyond
the scope of this PR.
## How this works
The current module name is determined with `go list -m` and the rename
from/to patterns are no longer hard-coded.
## How this was tested
Inspection of runs and resulting branch. Although these were the
bc8e5015694707ff682c444afff52a3801b35a53 workflow, the only other commit
in this PR is cosmetic (as seen in [this
run](https://github.com/ava-labs/libevm/actions/runs/13433886428) to
create f2cecaff9c01c86053644aafc9b7196229a84cbc).
1. Run on `main` @ d32c7e0090 changes from
`libevm` to `go-ethereum`:
a.
[Run](https://github.com/ava-labs/libevm/actions/runs/13432588947/job/37527588744)
b.
[Commit](6288df93f7)
c. Created branch `arr4n/auto/test-invertible-rename`
2. Re-run on `arr4n/auto/test-invertible-rename` changes back to
`libevm`:
a. [Run](https://github.com/ava-labs/libevm/actions/runs/13432810657)
b.
[Commit](6f4f94edd9)
c. [Auto-generated branch is identical to `main` at the time of
running](d32c7e0090...6f4f94edd9)
## Why this should be merged
To be able to generate RLP code using type aliases (defined e.g. in
`coreth`).
This can and should be PR-ed upstream as well I think.
## How this works
Call `types.Unalias()` at the start of `buildContext.makeOp()`. The
alternative of adding a `case` to the switch statement adds unnecessary
recursive calls while `types.Unalias()` is a no-op if not an alias.
## How this was tested
New `rlpgen` unit test with aliases of well-known types that have
special handling—these make their proper handling easy to spot when
inspecting generated code. A recursive alias in the same test also
demonstrates full alias resolution.
---------
Co-authored-by: Arran Schlosberg <me@arranschlosberg.com>
## Why this should be merged
We'll be building with Go 1.23 via `avalanchego` so should align
versions. Also unblocks #154.
## How this works
Bump `go.mod` version and run `go mod tidy`.
## How this was tested
CI on existing tests.
## Why this should be merged
`memsize` blocks usage of Go 1.23. We'll be building with 1.23 via
`avalanchego` anyway (which doesn't transitively import `memsize`) so we
want to run tests with the correct version. It will also unblock #154.
## How this works
✂️
We already cherry-pick the upstream commit
(e4675771ed) that removes `memsize` when
on release branches so we instead do it ourselves and no longer
cherry-pick. To avoid conflicts at the next `geth` sync, this PR should
be reverted before performing the merge. See #128.
## How this was tested
Existing tests.
## Why this should be merged
Adds logging of `libevm` modification of default behaviour.
## How this works
1. Introduces `log.Lazy` functions to allow expensive logging operations
to be computed i.f.f. required by the logging level.
2. Adds `Info` logging for registration of types and `Debug` logging for
all else.
3. Only paths that change behaviour in a potentially unpredictable
manner are logged; of note, RLP / JSON encoding is _not_ considered
unpredictable given that registered extras are logged.
4. The minimal viable package, `set`, was necessary because we don't
want to depend on `avalanchego` and the `hashicorp/go-set` latest
version requires a later version of Go. #153 tracks a swap to the latter
when possible.
The `eth/tracers/internal/tracetest` test flaked at least twice
(unrelated to these changes) so I've marked it as such since it's not
worth a separate PR.
## How this was tested
New unit test on `log.Lazy` + `set` methods. Existing CI for the rest as
it's a refactor.
## Why this should be merged
A separately cherry-picked commit, the fix for `StateDB.Copy()`,
requires Go 1.23 for use of the `maps` package. This, however, breaks
`memsize`, which was later removed later from `geth` for the same
reason.
## How this works
All release branches will now cherry-pick ethereum/go-ethereum#30253.
## How this was tested
Updated list of cherry-picks applied to #142, which [passed
tests](https://github.com/ava-labs/libevm/actions/runs/13371460014/job/37340683594).
- Pin golangci-lint action to v6 now that goheader doesn't run on the default branch pushes
- This addresses the goheader step misbehaving and linting all files on pushes
## Why this should be merged
Our initial release tags weren't well-formed for our needs. Although the
format was locked in in #81, I still stuffed up rc2. With the upcoming
rc3 release, we want to ensure that Go tooling picks it up as the
latest, which requires retracting rc2 as its format means it will take
higher precedence.
## How this works
`go.mod` `retract`
## How this was tested
n/a
## Why this should be merged
Branch-protection rules only require the Go linter (erroneously called
`lint`) by mistake. All linters are now dependencies of a single `lint`
job that remains as a PR gate. New linters will now be enforced by
default.
Closes#138
## How this works
GitHub Actions `need` configuration. The path limitation of running
`yamllint` is removed because it's such a cheap job so is ok to always
run.
## How this was tested
Inspection of CI run and PR's required jobs:
## Why this should be merged
Support for configurable `core/types.Block` with RLP encoding, including
interplay with `Body`.
## How this works
`Block` doesn't export most of its fields so relies on an internal type,
`extblock`, for RLP encoding. This type is modified to implement the
`rlp.Encoder` and `Decoder` methods as a point to inject hooks using
`rlp.Fields` (as in #120 for `Body`).
`Block` shares the same registered extra type as `Body`. Unlike
`Header`, which has its own field in a `Block`, the fields in `Body` are
promoted to be carried directly. This suggests that (at least for pure
data payloads) the modifications might be equivalent (and
`ava-labs/coreth` evidences this). Should different payloads be
absolutely required in the future, we can split the types—the
`RegisterExtras` signature is already too verbose though 😢.
## How this was tested
Explicit inclusion of a backwards-compatibility test for
`NOOPBlockBodyHooks` + implicit testing via the multiple upstream tests
in `block_test.go`. Re implicit testing: default behaviour is now to use
the noop hooks even when no registration is performed, but if we change
this then the tests in `block_test.go` can still be called as subtests
from a test that explicitly registers noops.
---------
Signed-off-by: Arran Schlosberg <519948+ARR4N@users.noreply.github.com>
Co-authored-by: Quentin McGaw <quentin.mcgaw@avalabs.org>
