Ethereum’s weekly All Core Developer calls are a lot to keep up with, so this "Checkpoint" series aims for high-level updates roughly every 4-5 weeks, depending on what’s happening in core development. See the previous update here.
tl;dr:
Since the last Checkpoint, the Pectra upgrade shipped and core developers have maintained a heavy emphasis on keeping the next upgrade (Fusaka) lean to get PeerDAS, a scaling unlock, out the door. Fusaka testing is ongoing and developers are still optimistic about a 2025 release date. Proposals for the headlining feature of the upgrade following Fusaka (”Glamsterdam”, ~2026) are being solicited until June 20th.
Pectra
Pectra went without a hitch, delivering staking quality-of-life upgrades, a huge unlock for wallet UX, increased blobs, and more. Though Q1 testnet hiccups created a little more anxious anticipation than usual, the upgrade was smooth and all new features behaved as expected. It is now up to the app layer to take advantage of these feature unlocks for users - for staking providers to adopt maxEB, for wallet providers to support 7702, etc.
Fusaka
Fusaka’s headlining feature, PeerDAS, has been undergoing feature-specific testing since 2024 and recently concluded its seventh devnet. The first multi-feature Fusaka devnet launched last week and the next will go live during Berlin Blockchain Week.
The number of devnets required for thorough testing depends on a number of factors, like the number and complexity of the features and how easily they play together. This typically ranges from ~5-13 and each is usually very short-lived as each devnet adds or adjusts one or more features at a time. As devnets progress and features get implemented, you can watch these features move from the Considered status to the Scheduled status. Only when devnets have all features stable will the upgrade go live on a testnet.
Blob Parameter Only (BPO) fork testing is also ongoing! This will allow the number of blobs available to be adjusted outside of hard forks to enable faster scaling for L2s. As part of testing, blob counts are being adjusted up and down and the feature has been working as intended so far.
Gas limit testing is determining if a 60 million gas limit will be safe for Fusaka. Gas limit adjustments don’t require a hard fork and can be adjusted by validators independently, but commitments to testing their safety as part of an upgrade is helpful to getting higher defaults shipped in client software. Spoiler alert: 60m is looking good, even for home stakers.
Teams committed to upgrading the Sepolia testnet before the Hoodi testnet for Fusaka and going forward to preserve Hoodi as a critical app-testing environment prior to mainnet upgrades.
Glamsterdam
Glamsterdam conversation kicked off in ACDC #158 with two proposed headliners: ePBS & FOCIL. EIP champions made their case for these as the main features for the upgrade. Others have begun to make their cases as well. If you have an EIP that you think would be the best headliner for Glamsterdam, you can propose it using the proposal template in the EIPs category for discussion.
Timeline
- May 26 - Jun 20: Fork Focus Discussion & Headliner Proposals
- Jun 23 - July 17: Headliner Discussion & Finalization
- Jul 21 - Aug 21: Non-Headliner EIP Proposals
- Sep 4 - Sep 11: Non-Headliner EIP CFI Decisions
History Expiry
Besu has released history expiry on Sepolia with other clients soon to follow. History expiry is already experimentally available on mainnet with several clients but is user-configured instead of default. Once sufficiently tested by all clients on Sepolia, it will roll out in default software configurations. Progress can be followed in the #history-expiry channel in the Eth R&D Discord.
Process
A decision was made to move current upgrade testing conversation to Monday testing calls, now called All Core Devs Testing (ACDT), while Thursday calls should discuss future upgrade scoping. Upgrade process changes are ongoing and will likely see more reconfiguration during Berlin Blockchain Week sprints. All are invited to voice their opinions on Ethereum Magicians.
Core devs have remained truly intent on keeping Fusaka lean and have been aggressively rejecting anything that will slow its release down, which makes me hopeful for a 2025 Fusaka.
Old habits are hard to break and process changes are difficult to apply in decentralized governance - despite the decision to move Thursday calls to future fork scoping, we’ve had four calls since Pectra but only started discussing Glamsterdam in the last. To ship faster, we’ll need to stick well to the proposed Glamsterdam deadlines and figure out not just the processes that will make upgrades faster, but the process that people will actually use and stick to.
Fusaka development is on track though and if the goals are [ scale L1, scale blobs for L2, improve UX ] - we are certainly well on our way.
Relevant ACD calls
Now featuring ACDT(esting) calls as well!