…UNT (bal-devnet-7)

EELS PR #2823 (commit eb80b438a) extends Policy A's top-level failure
refund so that a CREATE transaction that reverts or halts also returns
the intrinsic `STATE_BYTES_PER_NEW_ACCOUNT × cost_per_state_byte`
charge — paid via `add_intrinsic_gas` at tx start — back to the
reservoir. Mirrors the long-standing inner-CREATE failure rule.

Spec (fork.py::process_transaction):
  if tx_output.error is not None:
      tx_output.state_gas_left += tx_output.state_gas_used    # Policy A
      tx_output.state_gas_used = Uint(0)
      if isinstance(tx.to, Bytes0):
          new_account_refund = STATE_BYTES_PER_NEW_ACCOUNT * COST_PER_STATE_BYTE
          tx_output.state_gas_left += new_account_refund      # to reservoir
          tx_output.state_refund   += new_account_refund      # block-level

ethrex impl in `finalize_execution`, after the Policy A
execution-portion refund, when `is_create()`:
- `state_gas_reservoir += new_account_refund` (sender gets gas back)
- `state_gas_refund_absorbed += new_account_refund` (net block-level
  state_gas_used drops to 0 via the existing `net_state_gas_used`
  subtraction)
- `intrinsic_state_gas_charged -= new_account_refund` so
  `refund_sender`'s `regular_gas = raw_consumed - intrinsic_state - ...`
  formula doesn't double-deduct the now-refunded amount; the gas that
  was actually burned remains in the regular dimension where it sits.

Updates the existing `test_top_halt_phantom_drain_with_real_spill_under_policy_a`
test to assert the post-#2823 behavior: state_gas_used = 0 for a halted
CREATE tx (was state_new before).

…ate_gas_charged

Initial item-5 impl decremented `intrinsic_state_gas_charged` by
`new_account_refund` on top-level CREATE-tx failure, on the theory that
the refunded gas should not be subtracted from `raw_consumed` in the
regular-gas formula. That was wrong.

Per the EELS test_failed_create_tx_refunds_intrinsic_new_account
expectation:
  expected_gas_used = intrinsic_regular + regular_consumed
                    = gas_limit - create_state_gas
For an INVALID halt: expected = gas_limit - state_new.

`refund_sender`'s `regular_gas = raw_consumed - intrinsic_state - ...`
produces exactly that when `intrinsic_state_gas_charged` is left intact.
Clearing it overcounts regular by state_new (drops gas_used by state_new
in the wrong direction). Conceptually: the intrinsic burn is still in
`raw_consumed`, the refund returns it via the reservoir to the user,
and the regular-gas subtraction keeps it out of the regular dimension —
neither dimension counts it, which matches the spec.

Reservoir credit and `state_gas_refund_absorbed` bump (the two genuine
behavior changes from #2823) remain.

…-devnet-7)

Two fixes for the tx-level CREATE-collision path required by EELS
bal@v7.0.0:

1. Burn the forwarded gas in the regular dimension. EIP-684 + EELS
   `process_message_call` mark a CREATE-tx collision as
   `regular_gas_used = message.gas`. ethrex's collision branch in
   `handle_create_transaction` was returning with
   `current_call_frame.gas_remaining` still holding the
   post-intrinsic leftover, which leaked back to the sender as unused
   gas and never reached the regular dimension via
   `refund_sender`'s `raw_consumed - … + reclassified` formula. Now
   we zero `gas_remaining` so `raw_consumed = gas_limit` downstream.

2. Refund the intrinsic `STATE_BYTES_PER_NEW_ACCOUNT × CPSB` on
   collision. Previously the `!is_collision()` gate in
   `finalize_execution` excluded collision from BOTH Policy A and the
   item-5 NEW_ACCOUNT refund. Per EELS PR #2823 (commit eb80b438a)
   `process_transaction`, a tx-CREATE collision is just another
   failure variant that gets the intrinsic NEW_ACCOUNT refund:
       if tx_output.error is not None and isinstance(tx.to, Bytes0):
           tx_output.state_refund += create_state_gas
   Restructure the failure branch so Policy A still skips on
   collision (no execution state-gas to recover), but the
   `is_create()` intrinsic refund fires on every non-success.

After both: `regular_gas = gas_limit - intrinsic_state`,
`state_gas = 0`, header `gas_used = gas_limit - state_new` ✓
(matches EELS `test_create_tx_collision_refunds_intrinsic_new_account`).

Closes the 7-fixture CREATE-collision cluster in the bal-devnet-7
ef-test failures.

…t-7)

EELS PR #2823 (commit eb80b438a) simplifies `incorporate_child_on_error`
to:
    evm.state_gas_left += child_evm.state_gas_used + child_evm.state_gas_left

Removes the pre-#2823 subtraction of `child_evm.state_gas_refund` (the
"unwind on revert" of inline credits the child applied) on the rationale
that any inline credits the child applied are keyed to charges (its own
SSTORE or CREATE pre-charge) that are themselves rolled back, so the
matching `state_gas_left + state_gas_used` sum already reflects the
correct amount to return to the parent.

In ethrex's clamp-and-spill model the analogous tracking is the
`state_gas_refund_absorbed_snapshot` restore and the
`credit_against_drain_delta` subtraction in handle_return_call /
handle_return_create's Revert arms. Drop both.

Behavior change: an SSTORE 0→x→0 credit applied in a frame that later
reverts now leaks up to the parent's reservoir, matching the bal-7
test inversions:
- test_sstore_restoration_sub_frame_revert
- test_sstore_restoration_ancestor_revert
- test_subcall_set_clear_revert_pays_no_state_gas
and the broader stSStoreTest port group whose sub-contract calls
exercise this revert path.

Leaves `state_gas_used = state_gas_used_snapshot` (gross charges undone
on revert — same as before, matches EELS `state_gas_used` not flowing
to parent on error) and `state_gas_refund_pending = pending_snapshot`
(pending credits discarded with the child, per #2823 docstring).

The zkevm@v0.3.3 fixture bundle (the only bundle that ships
executionWitness, used by test-stateless-zkevm) is filled against an
older bal spec and disagrees with bal@v7.0.0 gas accounting:
storage_set/new_account/cpsb constants pre-recalibration plus pre-
EELS-#2815/#2816/#2823/#2827/#2828 refund-channel semantics.

Skips the 21 remaining gas mismatches in the eip8025_optional_proofs
filter (witness_codes_*, witness_state_*, validation_state_*),
analogous to the bal@v5.6.1 block already at the top of the list.
Re-enable once the zkevm bundle is regenerated against bal-7.

…UNT (bal-devnet-7)

EELS PR #2823 (commit eb80b438a) extends Policy A's top-level failure
refund so that a CREATE transaction that reverts or halts also returns
the intrinsic `STATE_BYTES_PER_NEW_ACCOUNT × cost_per_state_byte`
charge — paid via `add_intrinsic_gas` at tx start — back to the
reservoir. Mirrors the long-standing inner-CREATE failure rule.

Spec (fork.py::process_transaction):
  if tx_output.error is not None:
      tx_output.state_gas_left += tx_output.state_gas_used    # Policy A
      tx_output.state_gas_used = Uint(0)
      if isinstance(tx.to, Bytes0):
          new_account_refund = STATE_BYTES_PER_NEW_ACCOUNT * COST_PER_STATE_BYTE
          tx_output.state_gas_left += new_account_refund      # to reservoir
          tx_output.state_refund   += new_account_refund      # block-level

ethrex impl in `finalize_execution`, after the Policy A
execution-portion refund, when `is_create()`:
- `state_gas_reservoir += new_account_refund` (sender gets gas back)
- `state_gas_refund_absorbed += new_account_refund` (net block-level
  state_gas_used drops to 0 via the existing `net_state_gas_used`
  subtraction)
- `intrinsic_state_gas_charged -= new_account_refund` so
  `refund_sender`'s `regular_gas = raw_consumed - intrinsic_state - ...`
  formula doesn't double-deduct the now-refunded amount; the gas that
  was actually burned remains in the regular dimension where it sits.

Updates the existing `test_top_halt_phantom_drain_with_real_spill_under_policy_a`
test to assert the post-#2823 behavior: state_gas_used = 0 for a halted
CREATE tx (was state_new before).

…ate_gas_charged

Initial item-5 impl decremented `intrinsic_state_gas_charged` by
`new_account_refund` on top-level CREATE-tx failure, on the theory that
the refunded gas should not be subtracted from `raw_consumed` in the
regular-gas formula. That was wrong.

Per the EELS test_failed_create_tx_refunds_intrinsic_new_account
expectation:
  expected_gas_used = intrinsic_regular + regular_consumed
                    = gas_limit - create_state_gas
For an INVALID halt: expected = gas_limit - state_new.

`refund_sender`'s `regular_gas = raw_consumed - intrinsic_state - ...`
produces exactly that when `intrinsic_state_gas_charged` is left intact.
Clearing it overcounts regular by state_new (drops gas_used by state_new
in the wrong direction). Conceptually: the intrinsic burn is still in
`raw_consumed`, the refund returns it via the reservoir to the user,
and the regular-gas subtraction keeps it out of the regular dimension —
neither dimension counts it, which matches the spec.

Reservoir credit and `state_gas_refund_absorbed` bump (the two genuine
behavior changes from #2823) remain.

…-devnet-7)

Two fixes for the tx-level CREATE-collision path required by EELS
bal@v7.0.0:

1. Burn the forwarded gas in the regular dimension. EIP-684 + EELS
   `process_message_call` mark a CREATE-tx collision as
   `regular_gas_used = message.gas`. ethrex's collision branch in
   `handle_create_transaction` was returning with
   `current_call_frame.gas_remaining` still holding the
   post-intrinsic leftover, which leaked back to the sender as unused
   gas and never reached the regular dimension via
   `refund_sender`'s `raw_consumed - … + reclassified` formula. Now
   we zero `gas_remaining` so `raw_consumed = gas_limit` downstream.

2. Refund the intrinsic `STATE_BYTES_PER_NEW_ACCOUNT × CPSB` on
   collision. Previously the `!is_collision()` gate in
   `finalize_execution` excluded collision from BOTH Policy A and the
   item-5 NEW_ACCOUNT refund. Per EELS PR #2823 (commit eb80b438a)
   `process_transaction`, a tx-CREATE collision is just another
   failure variant that gets the intrinsic NEW_ACCOUNT refund:
       if tx_output.error is not None and isinstance(tx.to, Bytes0):
           tx_output.state_refund += create_state_gas
   Restructure the failure branch so Policy A still skips on
   collision (no execution state-gas to recover), but the
   `is_create()` intrinsic refund fires on every non-success.

After both: `regular_gas = gas_limit - intrinsic_state`,
`state_gas = 0`, header `gas_used = gas_limit - state_new` ✓
(matches EELS `test_create_tx_collision_refunds_intrinsic_new_account`).

Closes the 7-fixture CREATE-collision cluster in the bal-devnet-7
ef-test failures.

…t-7)

EELS PR #2823 (commit eb80b438a) simplifies `incorporate_child_on_error`
to:
    evm.state_gas_left += child_evm.state_gas_used + child_evm.state_gas_left

Removes the pre-#2823 subtraction of `child_evm.state_gas_refund` (the
"unwind on revert" of inline credits the child applied) on the rationale
that any inline credits the child applied are keyed to charges (its own
SSTORE or CREATE pre-charge) that are themselves rolled back, so the
matching `state_gas_left + state_gas_used` sum already reflects the
correct amount to return to the parent.

In ethrex's clamp-and-spill model the analogous tracking is the
`state_gas_refund_absorbed_snapshot` restore and the
`credit_against_drain_delta` subtraction in handle_return_call /
handle_return_create's Revert arms. Drop both.

Behavior change: an SSTORE 0→x→0 credit applied in a frame that later
reverts now leaks up to the parent's reservoir, matching the bal-7
test inversions:
- test_sstore_restoration_sub_frame_revert
- test_sstore_restoration_ancestor_revert
- test_subcall_set_clear_revert_pays_no_state_gas
and the broader stSStoreTest port group whose sub-contract calls
exercise this revert path.

Leaves `state_gas_used = state_gas_used_snapshot` (gross charges undone
on revert — same as before, matches EELS `state_gas_used` not flowing
to parent on error) and `state_gas_refund_pending = pending_snapshot`
(pending credits discarded with the child, per #2823 docstring).

The zkevm@v0.3.3 fixture bundle (the only bundle that ships
executionWitness, used by test-stateless-zkevm) is filled against an
older bal spec and disagrees with bal@v7.0.0 gas accounting:
storage_set/new_account/cpsb constants pre-recalibration plus pre-
EELS-#2815/#2816/#2823/#2827/#2828 refund-channel semantics.

Skips the 21 remaining gas mismatches in the eip8025_optional_proofs
filter (witness_codes_*, witness_state_*, validation_state_*),
analogous to the bal@v5.6.1 block already at the top of the list.
Re-enable once the zkevm bundle is regenerated against bal-7.

…UNT (bal-devnet-7)

EELS PR #2823 (commit eb80b438a) extends Policy A's top-level failure
refund so that a CREATE transaction that reverts or halts also returns
the intrinsic `STATE_BYTES_PER_NEW_ACCOUNT × cost_per_state_byte`
charge — paid via `add_intrinsic_gas` at tx start — back to the
reservoir. Mirrors the long-standing inner-CREATE failure rule.

Spec (fork.py::process_transaction):
  if tx_output.error is not None:
      tx_output.state_gas_left += tx_output.state_gas_used    # Policy A
      tx_output.state_gas_used = Uint(0)
      if isinstance(tx.to, Bytes0):
          new_account_refund = STATE_BYTES_PER_NEW_ACCOUNT * COST_PER_STATE_BYTE
          tx_output.state_gas_left += new_account_refund      # to reservoir
          tx_output.state_refund   += new_account_refund      # block-level

ethrex impl in `finalize_execution`, after the Policy A
execution-portion refund, when `is_create()`:
- `state_gas_reservoir += new_account_refund` (sender gets gas back)
- `state_gas_refund_absorbed += new_account_refund` (net block-level
  state_gas_used drops to 0 via the existing `net_state_gas_used`
  subtraction)
- `intrinsic_state_gas_charged -= new_account_refund` so
  `refund_sender`'s `regular_gas = raw_consumed - intrinsic_state - ...`
  formula doesn't double-deduct the now-refunded amount; the gas that
  was actually burned remains in the regular dimension where it sits.

Updates the existing `test_top_halt_phantom_drain_with_real_spill_under_policy_a`
test to assert the post-#2823 behavior: state_gas_used = 0 for a halted
CREATE tx (was state_new before).

…ate_gas_charged

Initial item-5 impl decremented `intrinsic_state_gas_charged` by
`new_account_refund` on top-level CREATE-tx failure, on the theory that
the refunded gas should not be subtracted from `raw_consumed` in the
regular-gas formula. That was wrong.

Per the EELS test_failed_create_tx_refunds_intrinsic_new_account
expectation:
  expected_gas_used = intrinsic_regular + regular_consumed
                    = gas_limit - create_state_gas
For an INVALID halt: expected = gas_limit - state_new.

`refund_sender`'s `regular_gas = raw_consumed - intrinsic_state - ...`
produces exactly that when `intrinsic_state_gas_charged` is left intact.
Clearing it overcounts regular by state_new (drops gas_used by state_new
in the wrong direction). Conceptually: the intrinsic burn is still in
`raw_consumed`, the refund returns it via the reservoir to the user,
and the regular-gas subtraction keeps it out of the regular dimension —
neither dimension counts it, which matches the spec.

Reservoir credit and `state_gas_refund_absorbed` bump (the two genuine
behavior changes from #2823) remain.

…-devnet-7)

Two fixes for the tx-level CREATE-collision path required by EELS
bal@v7.0.0:

1. Burn the forwarded gas in the regular dimension. EIP-684 + EELS
   `process_message_call` mark a CREATE-tx collision as
   `regular_gas_used = message.gas`. ethrex's collision branch in
   `handle_create_transaction` was returning with
   `current_call_frame.gas_remaining` still holding the
   post-intrinsic leftover, which leaked back to the sender as unused
   gas and never reached the regular dimension via
   `refund_sender`'s `raw_consumed - … + reclassified` formula. Now
   we zero `gas_remaining` so `raw_consumed = gas_limit` downstream.

2. Refund the intrinsic `STATE_BYTES_PER_NEW_ACCOUNT × CPSB` on
   collision. Previously the `!is_collision()` gate in
   `finalize_execution` excluded collision from BOTH Policy A and the
   item-5 NEW_ACCOUNT refund. Per EELS PR #2823 (commit eb80b438a)
   `process_transaction`, a tx-CREATE collision is just another
   failure variant that gets the intrinsic NEW_ACCOUNT refund:
       if tx_output.error is not None and isinstance(tx.to, Bytes0):
           tx_output.state_refund += create_state_gas
   Restructure the failure branch so Policy A still skips on
   collision (no execution state-gas to recover), but the
   `is_create()` intrinsic refund fires on every non-success.

After both: `regular_gas = gas_limit - intrinsic_state`,
`state_gas = 0`, header `gas_used = gas_limit - state_new` ✓
(matches EELS `test_create_tx_collision_refunds_intrinsic_new_account`).

Closes the 7-fixture CREATE-collision cluster in the bal-devnet-7
ef-test failures.

…t-7)

EELS PR #2823 (commit eb80b438a) simplifies `incorporate_child_on_error`
to:
    evm.state_gas_left += child_evm.state_gas_used + child_evm.state_gas_left

Removes the pre-#2823 subtraction of `child_evm.state_gas_refund` (the
"unwind on revert" of inline credits the child applied) on the rationale
that any inline credits the child applied are keyed to charges (its own
SSTORE or CREATE pre-charge) that are themselves rolled back, so the
matching `state_gas_left + state_gas_used` sum already reflects the
correct amount to return to the parent.

In ethrex's clamp-and-spill model the analogous tracking is the
`state_gas_refund_absorbed_snapshot` restore and the
`credit_against_drain_delta` subtraction in handle_return_call /
handle_return_create's Revert arms. Drop both.

Behavior change: an SSTORE 0→x→0 credit applied in a frame that later
reverts now leaks up to the parent's reservoir, matching the bal-7
test inversions:
- test_sstore_restoration_sub_frame_revert
- test_sstore_restoration_ancestor_revert
- test_subcall_set_clear_revert_pays_no_state_gas
and the broader stSStoreTest port group whose sub-contract calls
exercise this revert path.

Leaves `state_gas_used = state_gas_used_snapshot` (gross charges undone
on revert — same as before, matches EELS `state_gas_used` not flowing
to parent on error) and `state_gas_refund_pending = pending_snapshot`
(pending credits discarded with the child, per #2823 docstring).

The zkevm@v0.3.3 fixture bundle (the only bundle that ships
executionWitness, used by test-stateless-zkevm) is filled against an
older bal spec and disagrees with bal@v7.0.0 gas accounting:
storage_set/new_account/cpsb constants pre-recalibration plus pre-
EELS-#2815/#2816/#2823/#2827/#2828 refund-channel semantics.

Skips the 21 remaining gas mismatches in the eip8025_optional_proofs
filter (witness_codes_*, witness_state_*, validation_state_*),
analogous to the bal@v5.6.1 block already at the top of the list.
Re-enable once the zkevm bundle is regenerated against bal-7.

Two hand-rolled tests asserting specific `gas_used` values for
top-level CREATE-tx halt scenarios. Both encoded "Policy A" semantics
hard-coded to v7.0.0 numbers; the second one drifted after EELS
PR #2823's intrinsic NEW_ACCOUNT refund and again after the v7.1.x
spec series. ef-tests at tests-bal@v7.1.1 (2,145/2,145 green) cover
the same scenarios with fewer footguns, and the bespoke "Policy A"
shorthand wasn't carrying its weight.

Two hand-rolled tests asserting specific `gas_used` values for
top-level CREATE-tx halt scenarios. Both encoded "Policy A" semantics
hard-coded to v7.0.0 numbers; the second one drifted after EELS
PR #2823's intrinsic NEW_ACCOUNT refund and again after the v7.1.x
spec series. ef-tests at tests-bal@v7.1.1 (2,145/2,145 green) cover
the same scenarios with fewer footguns, and the bespoke "Policy A"
shorthand wasn't carrying its weight.

**Motivation**

Bring ethrex up to bal-devnet-7 (BAL fixtures `bal@v7.1.1`). Stacked
on top of #bal-devnet-6-pr (now in main).

**Description**

Aligns EIP-8037 state-gas accounting with bal-devnet-7 spec progression
(EELS PRs lambdaclass#2815 / lambdaclass#2816 / lambdaclass#2823 / lambdaclass#2827 / lambdaclass#2828 / lambdaclass#2836 / lambdaclass#2845 /
lambdaclass#2848),
bumps Amsterdam fixtures from `snobal-devnet-6@v1.1.0` to `bal@v7.1.1`,
and bumps the pinned hive version past the ethrex `--http.api` fix.

Main changes:

- EIP-8037 state-gas alignment with bal-devnet-7:
  - System-call state-gas reservoir.
  - Halt refunds spilled state gas (Policy A).
  - Tx-level CREATE failure refunds intrinsic `NEW_ACCOUNT`;
    `intrinsic_state_gas_charged` preserved across the failure path.
  - Tx-CREATE collision refund with regular-gas burn; billing matches
    EELS.
  - Cross-frame revert leaks inline credits.
  - Cross-frame revert reservoir formula fix.
  - Block-level `state_gas_used` subtracts `state_refund`.
- Remove same-tx SELFDESTRUCT state-gas refund (EELS PR lambdaclass#2845, v7.1.0).
- EIP-7702:
  - `set_delegation` refund via dedicated `state_refund` channel.
  - `set_delegation` refunds `AUTH_BASE` on existing delegation
    (EELS PR lambdaclass#2836).
  - `set_delegation` refunds `AUTH_BASE` on delegation clear
    (EELS PR lambdaclass#2848, v7.1.1).
- levm fixes pulled from main:
  - `revert` doesn't unmark the account as existing (lambdaclass#6592).
  - Account erroneously considered as existing after zero-value transfer
    (lambdaclass#6591).
- Tooling / tests:
  - Per-tx gas-dimension dump on block `gas_used` mismatch.
  - Bump Amsterdam fixtures to `bal@v7.1.1`.
  - Annotate BAL balance-mismatch errors with gas-equivalent diff and
    recognised state-gas constant multiples.
  - Unskip 74 bal-devnet-6 Amsterdam fixtures now passing.
  - Skip 21 stale EIP-8025 fixtures pinned at `bal@v5.7.0`
    (zkevm@v0.3.3 bundle, pre-bal-7).
  - Drop stale bal-devnet-6 known-issues entries from
    `docs/known_issues.md` and hive `KNOWN_EXCLUDED_TESTS`.
- CI:
  - Bump pinned hive version past the ethrex `--http.api` flag
    feature-detect fix (`c4d839b3`, hive lambdaclass#1485). Without this, hive
    starts ethrex with the default HTTP namespace allowlist
    (`eth,net,web3`) and tests touching `admin_*`/`debug_*`/`txpool_*`
    fail.

**Local test run**

`./run_test.sh` against `tests-bal@v7.1.1`: 2,145 / 2,145 pass.
`cargo test -p ethrex-test --tests`: 453 / 453 pass.

**Checklist**

- [ ] Updated `STORE_SCHEMA_VERSION` (crates/storage/lib.rs) if the PR
  includes breaking changes to the `Store` requiring a re-sync.

---------

Co-authored-by: Lucas Fiegl <iovoid@users.noreply.github.com>
Co-authored-by: Ivan Litteri <67517699+ilitteri@users.noreply.github.com>