Repository navigation
chore: port-to-v6 rolling ports - #25597
Merged
Merged
Conversation
AztecBot
added
ci-draft
spalladino
approved these changes
Oct 8, 2026
spalladino
marked this pull request as ready for review
October 8, 2026 01:50
spalladino
requested review from
iAmMichaelConnor,
just-mitch and
koenmtb1
as code owners
October 8, 2026 01:50
IlyasRidhuan
enabled auto-merge (squash)
October 8, 2026 05:18
IlyasRidhuan
disabled auto-merge
October 8, 2026 05:18
Removes `@ledwards2225` from every entry in `CODEOWNERS`, as requested by Tom French. The 12 affected entries cover `/barretenberg/`, the `noir-protocol-circuits` paths (kernels, rollup, parity, blob), the circuit JSON files, and `noir-contracts`. Each one still has at least two owners afterward, so no path is left without an owner. --- *Created by [claudebox](https://claudebox.work/v2/sessions/c853494dbde67212/jobs/1) · group: `slackbot` · requested by Tom (@TomAFrench) · [Slack thread](https://aztecfoundation.slack.com/archives/D0B586H14KG/p1791215548634289?thread_ts=1791215548.634289&cid=D0B586H14KG)* (cherry picked from commit bb004df)
## What changed `FeeLib.sol` prices the L1 work a sequencer and a prover must recover with two constants. They were placeholders taken from a forge benchmark run under `--evm-version amsterdam`. This PR replaces them with values measured on Sepolia after Glamsterdam: | Constant | Old | New | |---|---|---| | `L1_GAS_PER_CHECKPOINT_PROPOSED` | 500,000 | 520,000 | | `L1_GAS_PER_EPOCH_VERIFIED` | 4,000,000 | 2,500,000 | Sepolia measurements: - **Propose** (through Multicall3 `aggregate3`): median 487k; ~514k per checkpoint on average once the `setupEpoch` folded into each epoch's first checkpoint is counted; the most expensive regular propose was 520k. - **Epoch proof** (first proof submitted for an epoch): median 2.36M, max 2.51M. The governance proposal is at AztecProtocol/governance#68. ## Mirrors updated - `l1-contracts/test/fixtures/fee_data_points.json` is recomputed for the new constants. Only the sequencer, prover and protocol fee components change. The congestion multiplier and the L1 fee inputs stay the same. The recompute script reproduces the current fixture exactly with the old constants (0 of 672 points differ), so the only thing that moves is the constants. - `labs-patches/0011` makes the same change to the TypeScript port (`yarn-project/stdlib/src/gas/fee_math.ts`), updates the constants table in the stdlib gas README, and updates the epoch constant hardcoded in the single-node `fees_test.ts` e2e. The patch stays in the series until the aztec-node PR lands. Nothing else in the tree mirrors these values: they are not in the constants codegen, and no docs state them. ## Testing - `forge test` in `l1-contracts` (excluding the verifier-dependent scripts, `shouting.t.sol` and `UniswapPortalTest`): 1370 passed, 0 failed. The fee suites under `test/fees/` (43 tests) pass both before and after the change. - In aztec-node, `stdlib/src/gas` and `stdlib/src/tx`, the sequencer-client `global_variable_builder` tests and the aztec-node `next_block` tests all pass. `fee_predictor.test.ts` deploys the rollup to anvil and checks the TypeScript predictions against `getManaMinFeeAt`. It fails when only the TypeScript side is updated (`625010100000 != 598135100000`). It passes once the L1 contracts carry the new constants too. (cherry picked from commit 828486f)
… GSE cap (#25600) Measures BN254 proof-of-possession verification gas against the GSE cap, and publishes a calibration fixture and a conservative cost model for tooling that screens validator keys before registration. ## What Under `test/shared/bn254-pop-gas/`: - `PopGasBase.sol`: a counting copy of `hashToPoint` (checked against `BN254Lib.g1ToDigestPoint` on every run). It measures in-wrapper gas and binary-searches the minimum `{gas: g}` stipend. - `BN254G2TestLib.sol`: test-only G2 scalar multiplication to build `pk2`. Reproduces all 50 `pk2` values in `bn254_constants.json`. - `PopGasVectors.t.sol` (runs in CI): every vector is a valid PoP whose attempts, sqrt calls, rejections, root choice and digest match the library; the corpus covers all branches; the recorded gas matches the active EVM version; the model bounds every vector. - `PopGasScan.t.sol` and `PopGasGenerate.t.sol`: opt-in (env-gated) scanning and regeneration. - `scripts/bn254_pop_gas_model.py`: runs forge per EVM version, fits the model, and writes `test/fixtures/bn254_pop_gas_vectors.json`. The output is deterministic. - `README.md`: method, tables, commands and fixture format. ## Results On this path, Osaka and the next fork (`amsterdam` in Foundry 1.8.4) price identically. Modexp is 4,016 gas per sqrt call, versus 1,338 on Prague. Model (minimum stipend, fitted on 2,152 keys, max residual 8 gas): `gas ≤ 132,694 + 515·attempts + 4,580·sqrtCalls + 0.096·attempts²` | Cap | Honest keys over the cap | |---|---| | 250k | ~1 in 28,559 | | 300k | ~1 in 2.48M | | 500k | 9.6e-15 | | 1M | 5.1e-33 | `hashToPoint` is unbounded, so no finite worst case is claimed. ## Commands ``` cd l1-contracts python3 scripts/bn254_pop_gas_model.py # regenerate (Foundry 1.8.4 on PATH) forge test --match-contract PopGasVectorsTest --evm-version {amsterdam,osaka,prague} ``` (cherry picked from commit 5727158)
…fixes (#25619) Four L1 fixes, one commit each. ## Changes - **fix(l1): bind rollup registration payload to its predecessor.** `RegisterNewRollupVersionPayload` records the Registry's canonical rollup at deployment as an immutable `PREDECESSOR`, and its first action, `assertPredecessorIsCanonical()`, reverts unless that rollup is still canonical. A registration payload therefore only executes as a transition from the rollup that was canonical when it was deployed. The constructor signature is unchanged. The scenario tests now import the production payload instead of a test-only copy, which is removed. - **fix(l1): do not designate a fully exited candidate as escape hatch proposer.** `EscapeHatch.selectCandidates()` draws from the candidate-set snapshot, but the drawn candidate may have since left the set entirely. In that case the hatch is now left prepared without a designated proposer, the same outcome as an empty snapshot. - **fix(l1): sample escape hatch RANDAO after the candidate set freeze.** The escape hatch now accounts for the rollup's own `lagInEpochsForRandao` when it queries the sample seed, so the RANDAO used for the draw is the one checkpointed in the seed epoch, after the freeze, for any rollup lag. This adds `getRollupLagInEpochsForRandao()` to `EscapeHatch`/`IEscapeHatch`, sets `AZTEC_LAG_IN_EPOCHS_FOR_RANDAO` to 1 in `network-defaults.json`, and corrects the `ValidatorSelectionLib` docs on the RANDAO lag. - **fix(staking): require a per-deposit gas floor in flushEntryQueue.** Before each `GSE.deposit`, `flushEntryQueue` requires enough gas for the deposit (including the live proof-of-possession gas limit) plus the refund path, and otherwise reverts with `Staking__InsufficientFlushGas(required, available)`. The reserves are sized for the Glamsterdam gas schedule. `DeployRollupLib` no longer flushes inside the deploy broadcast. Instead, `forge_broadcast.js` runs the new `FlushEntryQueue.s.sol` as a separate broadcast. ## Testing - `forge fmt --check`, `forge build` and `scripts/check_contract_sizes.sh` pass in `l1-contracts`. - New and updated tests: `AddRollup.t.sol`, `escape-hatch/unit/selectCandidates.t.sol` (run with rollup RANDAO lags of 2 and 1), `escape-hatch/unit/getters.t.sol`, `staking/flushGasFloor.t.sol`, `staking/tmnt221.t.sol`, `staking/tmnt333.t.sol` and `governance/gse/gse/tmnt395.t.sol`. - Full `forge test` passes. `FlushGasFloorMainnetFork.t.sol` skips without `MAINNET_FORK_RPC`, and the measurement suites under `test/staking/flush-gas-floor/` skip unless `FLUSH_GAS_FLOOR_MEASURE` is set. (cherry picked from commit 9f6ddc4)
AztecBot
force-pushed
the
cb/port-to-v6
branch
from
October 8, 2026 05:23
f421519 to
8ac3065
Compare
IlyasRidhuan
enabled auto-merge (squash)
October 8, 2026 05:36
Collaborator
Author
Flakey Tests🤖 says: This CI run detected 2 tests that failed, but were tolerated due to a .test_patterns.yml entry. |
IlyasRidhuan
pushed a commit
that referenced
this pull request
Oct 8, 2026
…t to v6] (#25624) Port of [aztec-packages#25521](#25521) to `v6`, on its own branch so that the rolling port PR [aztec-packages#25597](#25597) can merge without it (see *Testnet compatibility pins*). The two PRs touch different files and apply in either order. Together, the labs series is #25597's `0011` followed by `0012`–`0018` from this PR, and that combination applies cleanly to v6's labs pin `ee1312cf`. Three commits: 1. **Raw pick** of `a36ddcc1c4c`, with git's conflict markers, in the three world-state and genesis paths that sit in different places on v6 and next. 2. **`fix(port):`**, which resolves those conflicts and regenerates the labs patches `0012`–`0014` against v6's labs pin. 3. **`fix(port):`**, which adds `0018` to pin labs' testnet compatibility test to the V3 roots. ## Foundation side - `genesis_protocol_nullifiers.hpp` keeps v6's `bb::world_state` layout and takes the new seed set. The only seed the V3 separator moves is ContractInstanceRegistry's registration nullifier, `0x2c1eb0…` → `0x16f917…`. - `world_state.test.cpp` keeps asserting the `GENESIS_*` macros. On v6 those come from `aztec_constants.hpp`, generated from the in-tree `constants.nr`. - `regenerate_genesis_constants.sh` stays as v6 has it. The source change's `genesis_*_root()` functions (and the `--check` hunk for them) are there for next's wsdb package, which reads its constants from a pinned foundation release. - Everything else (AVM pil and C++, codegen selections, `constants.nr`, `Prover.toml`s, L1 checkpoint fixtures) picks clean. ## Labs side (`labs-patches`) v6's labs pin `ee1312cf` is 101 commits ahead of next's `54e3383c` and has its own standard-contract pin, so next's `0012`–`0014` did not fit it. Those three patches were regenerated on a full build of this tree: - `0012`: the separator rename and the address vectors from next, applied unchanged, plus the address tables the standard-contracts generator derives from v6's pinned artifacts under V3. - `0013`: the migration note, with v6's V2 → V3 table. - `0014`: `noir-contracts/bootstrap.sh pin-standard-build` with V3, plus the regenerated tables. The re-pin is intentional: HandshakeRegistry compiles the separator in through `get_public_keys`, and V3 moves these addresses regardless. - `0015`–`0017` are byte-identical to next's. - `0018`: the testnet compatibility pins (below). | Contract | V2 (v6 today) | V3 (this PR) | | --- | --- | --- | | AuthRegistry | `0x2e05e8fb…f7c9` | `0x2ba87146…3a9c` | | MultiCallEntrypoint | `0x09b93e8e…5137` | `0x265552a4…fcc2` | | PublicChecks | `0x0468c918…dbbe` | `0x2cfebe97…d21a` | | HandshakeRegistry | `0x24d2effa…0fc7` | `0x198aa5f2…6040` | PublicChecks' V3 address matches next's exactly, because its V2 address was the same on both lines. The other three differ from next because v6's pinned artifacts differ. The re-pin also changes AuthRegistry's and MultiCallEntrypoint's bytecode: aztec-nr has moved on since v6's pin was cut, while nargo is the same build. PublicChecks' bytecode is unchanged. The pinned artifacts do not compile in the `STANDARD_*_ADDRESS` stamps (checked by recompiling with both stamp sets), so the pin and the address tables agree. ## Testnet compatibility pins `labs/yarn-project/aztec/src/testnet_compatibility.test.ts` pins the VK tree root, the protocol-contracts hash and the genesis archive root, and V3 moves all three. `0018` sets them to the values the V3 tree produces on v6, as reported by the test in CI3 run [37721289208](https://github.com/AztecProtocol/aztec-packages/actions/runs/37721289208): | Pin | V2 | V3 | | --- | --- | --- | | VK tree root | `0x031139ab…f75d` | `0x13eeb174…9e02` | | Protocol-contracts hash | `0x0030cdae…b378` | `0x0f54271c…54d7` | | Genesis archive root | `0x2ef904bb…bfb6` | `0x29eb2c52…e5c4` | Because of this, the v6 line is no longer compatible with the current testnet deployment; that is the intended effect of shipping V3 on v6. ## Checks run locally - Built with `make labs-yarn-project`, plus the labs `yarn-project` build: no generator drift after regeneration. - `labs-patches/bootstrap.sh check` passes. The series also applies after #25597's `0011`. - Jest: - stdlib `derivation`, `contract_address` and `complete_address` (14/14) - `noir_test_gen` (6/6) - `standard_contract_data` drift (5/5) - `contract_instance_published_event` (1/1) - `key_store` (4/4) - PXE storage backwards compatibility (16/16) - Noir: - aztec-nr `domain_separators_are_valid` - HandshakeRegistry `recipient_signature` (8/8) - protocol `types` crate: `constants_tests` (4/4) and `address::aztec_address` (9/9) - C++: `world_state_tests` `GetInitialTreeInfoForAllTrees`. - `regenerate_genesis_constants.sh --check`: up to date. Measured from v6's own artifacts, the roots equal the `constants.nr` values. Everything else is covered by CI3, which passes on `1e629c1`. The next step after this lands on v6 and a V3 rc is published: upstream `0012`–`0018` to aztec-node `main`, then bump and drain the queue as in [aztec-packages#25594](#25594). Refs #25521 --- *Created by [claudebox](https://claudebox.work/v2/sessions/9666bbcd1ae659f5/jobs/8) · group: `slackbot` · requested by Santiago Palladino · [Slack thread](https://aztecprotocol.slack.com/archives/C0AGN2WT3CP/p1791216786568079?thread_ts=1791216786.568079&cid=C0AGN2WT3CP)* --------- Co-authored-by: Michael Connor <mike@aztecprotocol.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Rolling integration PR carrying merged
port-to-v6PRs fromnextontov6. One commit per source PR (plus afix(port):commit after any conflicted pick).Ported
labs-patches/0011-…to v6's otherwise empty patch queue. The patch's pre-image blobs match v6's labs pinee1312cf, so it applies there unchanged.BN254Lib.sol,Bn254LibWrapper.sol,BN254Fixtures.t.sol,bn254_constants.json) are identical on v6 and next.Created by claudebox · group:
slackbot· Slack thread