feat(agglayer): process CLAIM notes from a rollup by mmagician · Pull Request #2573 · 0xMiden/protocol

mmagician · 2026-03-10T08:19:31Z

Summary

Support rollup deposits (mainnet_flag=0) in bridge-in verification, in addition to the existing mainnet-only path
Rollup deposits use two-level Merkle proof verification: first computing the local exit root from the leaf via calculate_root, then verifying it against the rollupExitRoot using smtProofRollupExitRoot
Add process_global_index_rollup MASM procedure and validate_rollup()/validate() Rust methods on GlobalIndex
Add Solidity test vector generator for rollup deposits and end-to-end integration test

Test plan

All 43 agglayer tests pass (including new case_3_rollup bridge-in integration test)
7 MASM global index unit tests pass (4 mainnet + 3 new rollup)
4 Rust GlobalIndex unit tests pass (1 existing + 3 new rollup)
make lint passes
Solidity test vector generation verified via forge test

🤖 Generated with Claude Code

Support rollup deposits (mainnet_flag=0) in bridge-in verification. Previously only mainnet deposits were supported and rollup deposits would panic. Rollup deposits use two-level Merkle proof verification: first computing the local exit root from the leaf, then verifying it against the rollup exit root. Closes #2394 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm

…pers Address review feedback: - Remove leading zeros assertion and mainnet flag validation from process_global_index_mainnet and process_global_index_rollup helpers. These are now done once in verify_leaf before branching. - The helpers now take [rollup_index_le, leaf_index_le] instead of the full 8-element global index. - In process_global_index_rollup, removed unnecessary byte-swap before asserting zero (zero is byte-order-independent). This is now moot since the mainnet flag is no longer checked in the helper. - Updated MASM unit tests to match the new helper signatures. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Copilot

Pull request overview

Adds support for processing CLAIM notes originating from rollups (mainnet_flag=0) by introducing rollup-aware global index handling and two-level Merkle proof verification, plus corresponding test vectors and integration coverage.

Changes:

Added rollup deposit test vectors and enabled bridge-in tests to run against them.
Introduced rollup global index processing in MASM and added Rust GlobalIndex::validate_rollup() / validate() helpers.
Added Solidity generator contract to produce rollup two-level Merkle proof vectors.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
crates/miden-testing/tests/agglayer/test_utils.rs	Adds rollup claim vector JSON + lazy parsed vector and exposes it via `ClaimDataSource`.
crates/miden-testing/tests/agglayer/global_index.rs	Extends MASM global index tests to cover both mainnet and rollup procedures.
crates/miden-testing/tests/agglayer/bridge_in.rs	Runs bridge-in integration test for rollup data source and creates destination account for rollup case.
crates/miden-agglayer/src/eth_types/global_index.rs	Adds rollup validation + shared leading-bits check + dispatching `validate()`, plus unit tests.
crates/miden-agglayer/solidity-compat/test/ClaimAssetTestVectorsRollupTx.t.sol	Adds Foundry test that generates rollup two-level Merkle proof vectors.
crates/miden-agglayer/solidity-compat/test-vectors/claim_asset_vectors_rollup_tx.json	Adds generated rollup test vector JSON consumed by Rust tests.
crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm	Adds `process_global_index_rollup` and rollup branch to `verify_leaf` using two-level verification.

Comments suppressed due to low confidence (1)

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm:1

movdn.n operates on the stack top, but at this point the stack top is leaf_index, not rollup_index. As written, movdn.9 will move leaf_index down (and the later movdn.9 movdn.9 will reshuffle again), which makes the comments about the resulting stack state inconsistent and can easily lead to calling calculate_root with mis-ordered arguments. This is especially risky because the current rollup test vector uses rollup_index=0 and leaf_index=0, so an ordering bug here can be masked. Rework these stack moves so that rollup_index (depth 1) is the value being moved/preserved (e.g., bring it to the top first with a swap/movup and only then move it down past the leaf value), and update the stack-shape comments to match the actual effects of the instructions.

use miden::agglayer::bridge::bridge_config

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

crates/miden-agglayer/solidity-compat/test/ClaimAssetTestVectorsRollupTx.t.sol

crates/miden-agglayer/solidity-compat/test-vectors/claim_asset_vectors_rollup_tx.json

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm

Include the auto-generated ERR_MAINNET_FLAG_INVALID constant in the committed errors file to fix the generated files CI check. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Use non-zero leafIndex (2) and indexRollup (5) in the rollup test vectors to exercise byte-ordering and stack manipulation paths. This exposed a bug in verify_leaf's rollup branch: the stack rearrangement after process_global_index_rollup had leaf_index and rollup_index in the wrong positions, which was masked when both were zero. Also restructure the rollup exit tree test helper to use an SMT-style approach (setLocalExitRootAt) matching the real PolygonRollupManager.getRollupExitRoot() construction. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Split ERR_BRIDGE_NOT_MAINNET into two errors: - ERR_BRIDGE_NOT_MAINNET: "mainnet flag must be 1 for a mainnet deposit" - ERR_BRIDGE_NOT_ROLLUP: "mainnet flag must be 0 for a rollup deposit" The previous error text "bridge not mainnet" read backwards when used in the rollup helper to reject a mainnet flag. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

- Fix stale comment in verify_leaf that incorrectly referenced "stack position 2" for the mainnet flag (it's at position 5) - Add explicit MASM test for the mainnet flag boolean validation (ERR_MAINNET_FLAG_INVALID) which rejects flag values >= 2 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Copilot

Pull request overview

Copilot reviewed 8 out of 8 changed files in this pull request and generated 3 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

crates/miden-agglayer/src/eth_types/global_index.rs

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm

crates/miden-testing/tests/agglayer/global_index.rs

- Fix comment referencing non-existent MAINNET_FLAG_STACK_POS constant - Remove test_mainnet_flag_rejects_invalid_value: it inlined the same logic as verify_leaf rather than testing verify_leaf itself, so it only proved u32lt.2 works on the value 3, not that verify_leaf actually performs the check Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Make validate_mainnet() and validate_rollup() private, exposing only validate() which checks the mainnet flag is a valid boolean (0 or 1) before dispatching. This fixes a bug where flag values >= 2 were incorrectly accepted by validate_rollup() since is_mainnet() returned false for any non-1 value. Add mainnet_flag() accessor and a test for invalid flag rejection. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Express SMT_PROOF_LOCAL_EXIT_ROOT_PTR and SMT_PROOF_ROLLUP_EXIT_ROOT_PTR relative to PROOF_DATA_PTR and each other, rather than hard-coding absolute values. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Reorganize constants into clear sections: proof data memory layout (with address map comment), leaf data, calculate_root locals, and data sizes. Define GLOBAL_INDEX_PTR relative to SMT_PROOF_ROLLUP_EXIT_ROOT_PTR for a fully chained layout. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm

Co-authored-by: Cursor Agent <cursoragent@cursor.com>

partylikeits1983

Looks great.

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm

partylikeits1983 · 2026-03-17T09:55:48Z

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm

+        exec.verify_merkle_proof
+        # => [verification_flag]
+
+        assert.err=ERR_SMT_ROOT_VERIFICATION_FAILED


Should we have different error messages for if the proof verification failed for rollup vs mainnet deposit?

Say, ERR_SMT_ROOT_VERIFICATION_FAILED_ROLLUP & ERR_SMT_ROOT_VERIFICATION_FAILED_MAINNET?

We could, but we only ever verify one of them per CLAIM note, and I think it should be clear for the submitted of the CLAIM note which path they're taking (by definition, they must know where this data came from, mainnet or rollup). So I think not strictly necessary, lmk if you disagree though (I'll merge as-is, this can be done in a follow up if needed)

partylikeits1983 · 2026-03-17T10:02:47Z

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm

@@ -189,17 +200,16 @@ end
 #!
 #! Invocation: exec
 pub proc process_global_index_mainnet


I think we should have a helper procedure which takes GLOBAL_INDEX[8] and returns leaf_index & sourceBridgeNetwork. Not for this PR, but for the nullifier tracking PR.

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm

Co-authored-by: Alexander John Lee <77119221+partylikeits1983@users.noreply.github.com>

mmagician added the agglayer PRs or issues related to AggLayer bridging integration label Mar 10, 2026

mmagician commented Mar 10, 2026

View reviewed changes

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm Outdated Show resolved Hide resolved

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm Outdated Show resolved Hide resolved

mmagician force-pushed the mmagician-claude/rollup-claims branch from 90d0479 to 54d7d66 Compare March 10, 2026 09:15

mmagician added the no changelog This PR does not require an entry in the `CHANGELOG.md` file label Mar 10, 2026

mmagician commented Mar 10, 2026

View reviewed changes

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm Outdated Show resolved Hide resolved

mmagician force-pushed the mmagician-claude/rollup-claims branch from 54d7d66 to cb4627e Compare March 10, 2026 09:33

mmagician added the pr-from-maintainers PRs that come from internal contributors or integration partners. They should be given priority label Mar 10, 2026

mmagician requested a review from Copilot March 10, 2026 09:34

Copilot started reviewing on behalf of mmagician March 10, 2026 09:35 View session

Copilot AI reviewed Mar 10, 2026

View reviewed changes

claude added 3 commits March 10, 2026 09:50

chore: regenerate agglayer error constants

39890f2

Include the auto-generated ERR_MAINNET_FLAG_INVALID constant in the committed errors file to fix the generated files CI check. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

mmagician marked this pull request as ready for review March 10, 2026 11:01

mmagician requested a review from Copilot March 10, 2026 11:01

Copilot started reviewing on behalf of mmagician March 10, 2026 11:01 View session

Copilot AI reviewed Mar 10, 2026

View reviewed changes

crates/miden-agglayer/src/eth_types/global_index.rs Outdated Show resolved Hide resolved

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm Outdated Show resolved Hide resolved

crates/miden-testing/tests/agglayer/global_index.rs Outdated Show resolved Hide resolved

claude added 4 commits March 10, 2026 11:13

mmagician commented Mar 10, 2026

View reviewed changes

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm Outdated Show resolved Hide resolved

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm Show resolved Hide resolved

Apply suggestions from code review

4fa6069

This comment was marked as resolved.

Sign in to view

Fix rollup deposit merkle proof stack setup (#2576)

8dfdc5c

Co-authored-by: Cursor Agent <cursoragent@cursor.com>

mmagician requested a review from partylikeits1983 March 10, 2026 17:18

mmagician mentioned this pull request Mar 17, 2026

feat: implement CLAIM note nullifier tracking #2610

Open

Merge branch 'agglayer' into mmagician-claude/rollup-claims

dc0d728

partylikeits1983 approved these changes Mar 17, 2026

View reviewed changes

partylikeits1983 reviewed Mar 17, 2026

View reviewed changes

crates/miden-agglayer/asm/agglayer/bridge/bridge_in.masm Show resolved Hide resolved

Apply suggestions from code review

152665a

Co-authored-by: Alexander John Lee <77119221+partylikeits1983@users.noreply.github.com>

mmagician enabled auto-merge March 17, 2026 10:13

mmagician disabled auto-merge March 17, 2026 11:58

mmagician merged commit c57ec7a into agglayer Mar 17, 2026
17 checks passed

mmagician deleted the mmagician-claude/rollup-claims branch March 17, 2026 11:58

mmagician mentioned this pull request Mar 17, 2026

AggLayer: process CLAIM notes from a rollup #2394

Closed

Conversation

mmagician commented Mar 10, 2026

Summary

Test plan

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

This comment was marked as resolved.

Uh oh!

partylikeits1983 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

partylikeits1983 Mar 17, 2026

Choose a reason for hiding this comment

Uh oh!

mmagician Mar 17, 2026

Choose a reason for hiding this comment

Uh oh!

partylikeits1983 Mar 17, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants