refactor(inbox): remove activation mechanism by dantaik · Pull Request #21475 · taikoxyz/taiko-mono

dantaik · 2026-03-18T02:19:55Z

Remove the explicit inbox activation flow that previously required a separate owner call before the Inbox could accept proposals. The activation is now assumed to have occurred in a prior contract version, with state properly initialized during contract upgrade.

This change simplifies the deployment flow and reduces gas consumption (~17 gas per proposal) by eliminating the activation check. The activationTimestamp state variable is retained but deprecated to maintain storage layout compatibility for proxy upgrades.

Remove the explicit inbox activation flow that required a separate owner call before the Inbox could accept proposals. The activation is now assumed to have occurred in a prior contract version, with state initialized during upgrade. - Remove activate() function and ActivationRequired guard check from Inbox.sol - Remove LibInboxSetup.activate() function and related error handling - Update deployment script to no longer call inbox.activate() - Update tests to simulate pre-activated state using vm.store cheatcodes - Keep activationTimestamp state variable (deprecated) for storage layout compatibility - Gas savings: ~17 gas per propose() call from removing the activation check

Rename to _activationTimestamp with internal visibility since it is deprecated and no longer part of the public API. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

File no longer contains activation tests after removing the activation mechanism. Rename to reflect its current contents. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: ba1548b07e

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

packages/protocol/contracts/layer1/core/impl/Inbox.sol

Instead of requiring a separate activation step, init() now sets up the core state (nextProposalId, lastProposalBlockId, genesis hash, etc.) directly. This removes the need for vm.store hacks in tests. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Move validateConfig, its constant, and errors from the external library into the Inbox contract as a private function, removing the DELEGATECALL overhead. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

packages/protocol/contracts/layer1/core/impl/Inbox.sol

…ation (#21493) Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

claude · 2026-03-21T08:54:59Z

Claude finished @dantaik's task in 1m 57s —— View job

PR Review: refactor(inbox): remove activation mechanism

Summary

Clean refactor that completes the activation removal. The initialization logic previously split across init() + activate() is now fully consolidated into init(). The overall approach is sound. A few things worth flagging:

Issues

1. Storage layout documentation mismatch (DevnetInbox_Layout.sol, MainnetInbox_Layout.sol)

The layout files (auto-generated) were updated to show _activationTimestamp (private naming), but the actual contract declares:

uint48 public activationTimestamp;  // Inbox.sol:127

The layout files are inconsistent with the live code. If someone reads these layout docs during an upgrade safety review, they'll see a variable name that doesn't exist. The layout script should be re-run against the current code (or the layout files were generated from an intermediate version where the variable was briefly renamed).

2. activator field in DeploymentConfig has stale semantics (DeployShastaContracts.s.sol)

The activator field is still validated non-zero and used as the initial inbox owner, but its original purpose (owning the inbox so it could call activate()) is gone. The field name now implies a concept that no longer exists. Not a bug, but confusing for operators setting up deployments. Consider renaming to initialOwner or adding a comment clarifying its current role.

3. _validateConfig inlining trades deployment size for simplicity

LibInboxSetup used public library functions specifically to reduce Inbox.sol deployment bytecode via external linking (this was explicitly documented in the old library comment). Now that _validateConfig is a private pure function in Inbox.sol, the constructor validation code is part of the deployment bytecode. Not a bug, but the gas report should show slightly larger deployment cost. Worth confirming the trade-off is acceptable.

Minor

The init() comment in Inbox.sol says _genesisBlockHash is "The block hash used as the genesis finalized block hash", but lastFinalizedBlockHash being set here isn't the L2 genesis block hash—it's the last Pacaya block hash used to anchor the chain. The comment is slightly misleading.
_coreState.lastProposalBlockId = 1 initialization means the first proposal can happen at block 2. This is correct, but a short inline comment explaining why (to avoid blockhash(0) == 0) would help future readers—the old LibInboxSetup.activate() had this explanation.

What's Good

Removing the ActivationRequired check from propose() is the right call—17 gas per proposal adds up at scale.
The Codex reviewer's concern about nextProposalId being 0 on fresh deployments is already handled: init() sets it to 1 (line 192) and seeds proposal 0's hash (line 198). dantaik's rebuttal is correct.
activationTimestamp being retained as public (per davidtaikocha's request) is correct for client compatibility.
Test cleanup is clean—InboxActivation.t.sol is properly replaced by InboxConfig.t.sol covering the inlined validation logic.
Deployment scripts correctly require L2_GENESIS_HASH env var and validate it non-zero.

dantaik · 2026-03-21T08:55:09Z

@davidtaikocha please take another look

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 635d7ad4a8

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-03-21T09:01:02Z

packages/protocol/contracts/layer1/core/impl/Inbox.sol

+        activationTimestamp = _activationTimestamp;
+        _coreState.nextProposalId = 1;
+        _coreState.lastProposalBlockId = 1;
+        _coreState.lastFinalizedTimestamp = uint48(block.timestamp);
+        _coreState.lastFinalizedBlockHash = _genesisBlockHash;


Keep a way to reseed the fork parent hash after deploy

Checked the mainnet/Hoodi deployment flow in packages/protocol/script/layer1/core/DeployShastaContracts.s.sol:125-128: we still predeploy Shasta and only switch SignalService at shastaForkTimestamp. This refactor now stores _coreState.lastFinalizedBlockHash once during init() and removes the only post-deployment path that used to update it. If the Pacaya head reorgs after deployment but before the fork, the first Shasta proof will hit ParentBlockHashMismatch in Inbox.prove() (packages/protocol/contracts/layer1/core/impl/Inbox.sol:283-286) and the fork is stuck until the proxy is redeployed.

Useful? React with 👍 / 👎.

chatgpt-codex-connector · 2026-03-21T09:01:02Z

packages/protocol/script/layer1/core/DeployShastaContracts.s.sol

                )
            ),
-            data: abi.encodeCall(Inbox.init, config.activator)
+            data: abi.encodeCall(Inbox.init, (config.activator, config.l2GenesisHash, uint48(block.timestamp)))


Initialize activationTimestamp from the configured fork time

In the mainnet/Hoodi path, this passes block.timestamp into Inbox.init() even though the same script already has the real fork time in config.shastaForkTimestamp. taiko-client treats activationTimestamp as the L1 block where Shasta started (packages/taiko-client/pkg/rpc/methods.go:1139-1155) and uses it to reset L1Current for the first Shasta proposal (packages/taiko-client/driver/state/l1_current.go:76-79). If these contracts are deployed ahead of the scheduled fork, proposal 1 will be anchored to the deployment block instead of the actual fork block, which skews client bootstrap and reorg recovery around genesis.

Useful? React with 👍 / 👎.

) Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

dantaik requested review from davidtaikocha and ggonzalez94 as code owners March 18, 2026 02:19

dantaik and others added 2 commits March 18, 2026 10:20

refactor(inbox): make deprecated activationTimestamp internal

ba1548b

Rename to _activationTimestamp with internal visibility since it is deprecated and no longer part of the public API. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

refactor(protocol): rename InboxActivation.t.sol to InboxConfig.t.sol

0d0c17b

File no longer contains activation tests after removing the activation mechanism. Rename to reflect its current contents. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

chatgpt-codex-connector bot reviewed Mar 18, 2026

View reviewed changes

packages/protocol/contracts/layer1/core/impl/Inbox.sol Show resolved Hide resolved

dantaik and others added 2 commits March 18, 2026 02:27

Update protocol generated artifacts

d39d2f2

dantaik added this to the Unzen Hard Fork (v6) milestone Mar 18, 2026

dantaik self-assigned this Mar 18, 2026

dantaik and others added 3 commits March 18, 2026 02:37

Update protocol generated artifacts

56940b3

refactor(inbox): inline LibInboxSetup into Inbox

247729b

Move validateConfig, its constant, and errors from the external library into the Inbox contract as a private function, removing the DELEGATECALL overhead. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Update protocol generated artifacts

a50a309

davidtaikocha reviewed Mar 18, 2026

View reviewed changes

packages/protocol/contracts/layer1/core/impl/Inbox.sol Outdated Show resolved Hide resolved

davidtaikocha added the area.protocol label Mar 18, 2026

dantaik marked this pull request as draft March 21, 2026 08:38

dantaik and others added 2 commits March 21, 2026 16:52

Merge branch 'main' into dantaik/inbox-rm-activation

02a0546

refactor(protocol): restore public activationTimestamp with initializ…

635d7ad

…ation (#21493) Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

dantaik marked this pull request as ready for review March 21, 2026 08:54

taikoxyz deleted a comment from claude bot Mar 21, 2026

dantaik enabled auto-merge March 21, 2026 08:55

dantaik requested a review from davidtaikocha March 21, 2026 08:55

Update protocol generated artifacts

ee4e40b

chatgpt-codex-connector bot reviewed Mar 21, 2026

View reviewed changes

dantaik and others added 2 commits March 21, 2026 18:41

refactor(inbox): clarify deployment semantics and initialization (#21494

4630009

) Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

Update protocol generated artifacts

9dbbdd8

davidtaikocha approved these changes Mar 21, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

refactor(inbox): remove activation mechanism#21475

refactor(inbox): remove activation mechanism#21475
dantaik wants to merge 13 commits intomainfrom
dantaik/inbox-rm-activation

dantaik commented Mar 18, 2026

Uh oh!

chatgpt-codex-connector bot left a comment

Uh oh!

Uh oh!

Uh oh!

claude bot commented Mar 21, 2026 •

edited

Loading

Uh oh!

dantaik commented Mar 21, 2026

Uh oh!

chatgpt-codex-connector bot left a comment

Uh oh!

chatgpt-codex-connector bot Mar 21, 2026

Uh oh!

chatgpt-codex-connector bot Mar 21, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

dantaik commented Mar 18, 2026

Uh oh!

chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

Uh oh!

claude bot commented Mar 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR Review: refactor(inbox): remove activation mechanism

Summary

Issues

Minor

What's Good

Uh oh!

dantaik commented Mar 21, 2026

Uh oh!

chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector bot Mar 21, 2026

Choose a reason for hiding this comment

Uh oh!

chatgpt-codex-connector bot Mar 21, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

claude bot commented Mar 21, 2026 •

edited

Loading