feat(flashblocks): Flashblocks engine sidecar #19479

0x00101010 · 2025-11-04T01:13:57Z

Towards: #18255, see bottom for follow up needed.

feat(op-reth): integrate flashblocks with consensus engine

Summary

This PR completes the integration of flashblocks with the consensus engine by:

Adding structured conversion infrastructure from flashblock sequences to execution payloads
Automatically spawning the consensus client when flashblocks are enabled
Improving error handling and observability with detailed logging

Integration

crates/optimism/rpc/src/eth/mod.rs

Added automatic spawning of FlashBlockConsensusClient when flashblocks are enabled
Wired up beacon_engine_handle to enable payload submission

crates/node/builder/src/rpc.rs

Exposed beacon_engine_handle in EthApiCtx for RPC builders to access

examples/custom-node/src/engine.rs

Added example implementation of From<&FlashBlockCompleteSequence> for custom execution data types
Shows how to derive custom extensions from flashblock sequences

Dependencies

crates/optimism/flashblocks/Cargo.toml

Added: op-alloy-rpc-types-engine (required for OpExecutionData)
Removed: ringbuffer (no longer needed with simplified FCU logic)

Testing

All existing tests pass
New unit tests for conversion functions
Clippy clean with all features enabled

Follow up needed

~~handle optional state root calculations scenario~~ Done
~~handle Jovian upgrade situation where blob_gas_used is not 0, need to update rollup-boost, op-rbuilder, and this repo for consistent types.~~ Done
~~ideally we move all the flashblocks relevant data structures into op-alloy-rpc-types-engine~~ Done
add functionality to work with different forks

0x00101010 · 2025-11-04T03:18:06Z

crates/optimism/flashblocks/src/convert.rs

+//! The goal is to have the conversion logic in one place to avoid missing steps.
+
+use crate::{
+    ExecutionPayloadBaseV1, ExecutionPayloadFlashblockDeltaV1, FlashBlockCompleteSequence,


@mattsse what's your thoughts on moving above data structures into op-alloy-rpc-types-engine?

Currently they're defined in rollup-boost and imported by op-rbuilder, etc.

If we're making reth to support flashblocks natively, it seems reasonable to move those into op-alloy-rpc-types-engine and update all other repos to refer to it.

This helps with

centralize commonly shared data structures

also we can move this convert.rs functionality directly into op-alloy-rpc-types-engine and sit alongside with OpExecutionData to share the same block / flashblock -> ExecutionData conversion logic

@mattsse what's your thoughts on moving above data structures into op-alloy-rpc-types-engine?

yeah that would be ideal, even if the format changes we can still make that work

mattsse

actually pretty simple, very nice

imo we should first offload all things flashblock types to op-alloy and then merge this

crates/optimism/flashblocks/src/consensus.rs

crates/optimism/flashblocks/src/convert.rs

crates/optimism/rpc/src/eth/mod.rs

crates/node/builder/src/rpc.rs

codspeed-hq · 2025-11-09T19:45:06Z

CodSpeed Performance Report

Merging #19479 will not alter performance

_{Comparing 0x00101010:fb-engine-sidecar (691d226) with main (95b8a85)}

Summary

✅ 81 untouched

klkvr · 2025-11-11T12:17:31Z

do you mind rebasing this on top of #19608 so that it's easier to review?

mattsse · 2025-11-11T12:58:20Z

@0x00101010 released new op-alloy patch with alloy-rs/op-alloy#606

0x00101010 · 2025-11-11T18:47:57Z

do you mind rebasing this on top of #19608 so that it's easier to review?

Done

klkvr · 2025-11-12T14:05:40Z

this needs rebase again

0x00101010 · 2025-11-12T16:15:34Z

this needs rebase again

@klkvr done

klkvr

this does make sense, left some questions

defer to @mattsse for final review

klkvr · 2025-11-13T09:49:24Z

crates/optimism/flashblocks/src/consensus.rs

+                safe_block_hash: B256::ZERO,
+                finalized_block_hash: B256::ZERO,


what's the motivation for setting those to zero? not sure how to properly set those if the only source of truth we have are flashblocks but maybe we could keep the head-32?

klkvr · 2025-11-13T09:50:51Z

crates/optimism/flashblocks/src/consensus.rs

+        // Replace payload's state_root with the calculated one. For flashblocks, there was an
+        // option to disable state root calculation for blocks, and in that case, the payload's
+        // state_root will be zero, and we'll need to locally calculate state_root before
+        // proceeding to call engine_newPayload.


I'm wondering if we should somehow remove the need to compute state root here entirely?

engine will anyway end up computing it when verifying the payload, so this would just mean we do the work twice? cc @mattsse

though I guess it's kind of tricky because engine relies on the newPayload to come with a computed block hash

klkvr · 2025-11-13T09:53:53Z

crates/optimism/rpc/src/eth/mod.rs

-
            ctx.components.task_executor().spawn(Box::pin(service.run(tx)));

+            info!(target: "reth::cli", "Launching FlashBlockConsensusClient");


should this be optional? e.g if you want to run op-node and a flashblocks engine client

github-project-automation bot added this to Reth Tracker Nov 4, 2025

github-project-automation bot moved this to Backlog in Reth Tracker Nov 4, 2025

0x00101010 force-pushed the fb-engine-sidecar branch 2 times, most recently from e28c5bc to 1f7443c Compare November 4, 2025 02:12

0x00101010 marked this pull request as ready for review November 4, 2025 03:11

0x00101010 requested review from Rjected, fgimenez, gakonst, klkvr and mattsse as code owners November 4, 2025 03:11

0x00101010 commented Nov 4, 2025

View reviewed changes

mattsse requested changes Nov 5, 2025

View reviewed changes

github-project-automation bot moved this from Backlog to In Progress in Reth Tracker Nov 5, 2025

0x00101010 force-pushed the fb-engine-sidecar branch from 1d7ba92 to a7d2396 Compare November 9, 2025 19:35

0x00101010 force-pushed the fb-engine-sidecar branch 2 times, most recently from 4ddf3ff to 1ae3b3b Compare November 9, 2025 19:57

0x00101010 requested a review from mattsse November 10, 2025 02:21

0x00101010 force-pushed the fb-engine-sidecar branch 3 times, most recently from 31a28cf to 919edf7 Compare November 11, 2025 18:47

0x00101010 force-pushed the fb-engine-sidecar branch from 919edf7 to 10c1564 Compare November 11, 2025 19:09

0x00101010 force-pushed the fb-engine-sidecar branch 2 times, most recently from 06db091 to 0bb1d9e Compare November 12, 2025 15:57

Implmeneted flashblocks engine sidecar

bde015c

0x00101010 force-pushed the fb-engine-sidecar branch from 0bb1d9e to bde015c Compare November 12, 2025 16:15

Make logging target consistent

691d226

klkvr reviewed Nov 13, 2025

View reviewed changes

		safe_block_hash: B256::ZERO,
		finalized_block_hash: B256::ZERO,


		ctx.components.task_executor().spawn(Box::pin(service.run(tx)));

		info!(target: "reth::cli", "Launching FlashBlockConsensusClient");

feat(flashblocks): Flashblocks engine sidecar #19479

Are you sure you want to change the base?

feat(flashblocks): Flashblocks engine sidecar #19479

Uh oh!

Conversation

0x00101010 commented Nov 4, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

feat(op-reth): integrate flashblocks with consensus engine

Summary

Integration

Dependencies

Testing

Follow up needed

Uh oh!

0x00101010 Nov 4, 2025

Choose a reason for hiding this comment

Uh oh!

mattsse Nov 5, 2025

Choose a reason for hiding this comment

Uh oh!

mattsse left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

codspeed-hq bot commented Nov 9, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

CodSpeed Performance Report

Merging #19479 will not alter performance

Summary

Uh oh!

klkvr commented Nov 11, 2025

Uh oh!

mattsse commented Nov 11, 2025

Uh oh!

0x00101010 commented Nov 11, 2025

Uh oh!

klkvr commented Nov 12, 2025

Uh oh!

0x00101010 commented Nov 12, 2025

Uh oh!

klkvr left a comment

Choose a reason for hiding this comment

Uh oh!

klkvr Nov 13, 2025

Choose a reason for hiding this comment

Uh oh!

klkvr Nov 13, 2025

Choose a reason for hiding this comment

Uh oh!

klkvr Nov 13, 2025

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

0x00101010 commented Nov 4, 2025 •

edited

Loading

codspeed-hq bot commented Nov 9, 2025 •

edited

Loading