Gap analysis, implement execution-spec-tests#10
Open
garyschulte wants to merge 39 commits intomainfrom
Open
Conversation
garyschulte
force-pushed
the
feat/gap-analysis
branch
3 times, most recently
from
7fe8bd0 to
6133bc2
Compare
Signed-off-by: garyschulte <garyschulte@gmail.com>
protocol schedule refactor (may remove in favor of namespace specific specId impls) initial implementation of host operations (storage ops) - they probably need work Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
Phase 1 — Caller state changes (enables first valid tx)
1. validateAgainstStateAndDeductCaller: create journal entries to actually deduct max_fee * gas_limit + value from caller (via journal.balanceTransfer or direct balance
change entry)
2. Pre-execution: increment caller nonce via journal
3. Pre-execution: warm access list addresses and storage slots, warm coinbase (EIP-3651)
4. Fix calculateInitialGas: zero bytes at 4 gas, non-zero at 16; access list: 2400/address + 1900/storage slot
Signed-off-by: garyschulte <garyschulte@gmail.com>
Phase 2 — Post-execution correctness (produces correct state root)
5. Implement gas refund accumulation from SSTORE (journal already tracks old values, need to compute refund from them)
6. reimburse_caller: effective_gas_price * (gas_limit - gas_used - refund) credited back to caller balance
7. reward_beneficiary: (effective_gas_price - base_fee) * gas_used credited to coinbase
Signed-off-by: garyschulte <garyschulte@gmail.com>
Phase 3 — State commitment
8. On success: call journal.commitTx(), then journal.finalize() to return State diff
9. On error/revert: call journal.discardTx()
10. Wire ExecuteCommitEvm.executeAndCommit() to apply the State diff to the underlying DB
Signed-off-by: garyschulte <garyschulte@gmail.com>
Phase 4 — Nested calls (needed for any non-trivial contract)
11. Implement frame loop in MainnetHandler.execute() with a call stack
12. Wire host.call() to push a child frame, execute it, and return result to parent's stack
Signed-off-by: garyschulte <garyschulte@gmail.com>
Phase 5 — EIP completeness
13. EIP-155 chain ID validation
14. EIP-3607 (reject tx from accounts with code)
15. EIP-4844 blob validation / excess blob gas
16. EIP-7702 authorization list
Signed-off-by: garyschulte <garyschulte@gmail.com>
Group 1 — Bytecode.default() fix (journal.zig:519)
- Bytecode.default() → Bytecode.new() (method didn't exist)
- Also fixed journal.zig:536: code.eip7702.address().isZero() → std.mem.eql(u8, &code.eip7702.address, &[_]u8{0} ** 20) (field access, not method; arrays have no
isZero())
Group 2 — Protocol schedule hardfork registration (protocol_schedule.zig)
- Removed RETURNDATASIZE/RETURNDATACOPY from Frontier → added to applyByzantiumChanges
- Removed SHL/SHR/SAR from applyByzantiumChanges → added to applyConstantinopleChanges
- Removed CREATE2 from Frontier → added to applyConstantinopleChanges
Group 3 — CREATE collision balance check (journal.zig:628)
- Added || balance != 0 to EIP-161 collision check
Group 4 — EXP gas spec-gating EIP-160 (gas_costs.zig, arithmetic.zig)
- Added G_EXPBYTE_FRONTIER = 10
- opExp now uses 10 pre-Spurious Dragon, 50 post
Group 5 — LOG topic memory leak (journal.zig)
- discardTx, commitTx, and deinit now free heap-allocated topic slices before clearing the log list
Group 6 — EIP-4844 blob validation (primitives/main.zig, validation.zig, mainnet_builder.zig)
- Added GAS_PER_BLOB, MAX_BLOB_NUMBER_PER_BLOCK, VERSIONED_HASH_VERSION_KZG constants
- calculateInitialGas adds blob intrinsic gas (131072 × blob_count)
- New validateBlobTx checks count, hash version, and max_fee_per_blob_gas ≥ blob_basefee
Group 7 — EIP-7702 authorization list (mainnet_builder.zig)
- preExecution now iterates authorization list (Prague+), validates chain_id and nonce, bumps authority nonce, and sets EIP-7702 delegation bytecode (or clears it for
zero-address authorizations)
Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
…aults and panics Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
… leak Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
…allout Signed-off-by: garyschulte <garyschulte@gmail.com>
…, balance and code checks Signed-off-by: garyschulte <garyschulte@gmail.com>
…rking Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
…or vec/static stack until we are zk proving to back up our cost expectations Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
…journal workarounds required by static stack allocation and recursion Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
…0.15 syntax
host.zig: touch precompile callee on successful call so it appears in
post-state for pre-EIP-161 forks (Frontier/Homestead). The callee was
loaded into evm_state by accountInfo() in the CALL opcode handler but
never marked touched, causing it to be excluded from the post-state.
For EIP-161+ forks the empty account is cleaned up by state-clear logic.
mainnet_builder.zig: detect EIP-7702 delegation designators stored as
legacy_analyzed bytecode by checking the raw 0xEF 0x01 0x00 prefix,
so pre-state fixtures with already-delegated accounts are re-delegatable.
validation.zig + primitives/main.zig: apply correct per-fork blob count
limits — Osaka uses per-tx limit of 6 (EIP-7594), Prague uses per-block
limit of 9 (EIP-7691), Cancun keeps 6.
mcl_wrapper.zig: fix array initializer syntax for Zig 0.15 compatibility
([1]T{.{}} ** N instead of .{.{}} ** N).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…ntirely, but that is something to consider since EOF is dead. Signed-off-by: garyschulte <garyschulte@gmail.com>
Signed-off-by: garyschulte <garyschulte@gmail.com>
garyschulte
force-pushed
the
feat/gap-analysis
branch
from
6133bc2 to
9639633
Compare
Signed-off-by: garyschulte <garyschulte@gmail.com>
garyschulte
pushed a commit
to garyschulte/zevm
that referenced
this pull request
Mar 24, 2026
* feat: EIP-7843 — SLOTNUM opcode (Amsterdam) Adds SLOTNUM (0x4B): pushes the beacon chain slot number from the block environment. Gas: G_BASE (2). Adds slot_number field to BlockEnv and BlockEnvBuilder (defaults to 0). Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix(eip-7843): make slot_number optional everywhere per review feedback - `BlockEnv.slot_number`: `u64` → `?u64`; default is `null` (absent on pre-Amsterdam / non-beacon chains) - `BlockEnvBuilder`: add `setSlotNumber(?u64)`; propagate `slot_number` through all existing setter methods (previously silently dropped) - `Host.slotNumber()`: returns `?u64` - `opSlotnum`: halt with `invalid_opcode` when slot number is absent - `spec_test/runner.zig`: initialize `slot_number` to `null` Addresses review comment from @garyschulte on PR 10d9e#10. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.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.
This PR is the result of gap-analysis necessary to have a mainnet-friendly evm interpreter
at a high level this PR:
Some things noted for future improvements but not included in this PR: