blockifier: update patricia resource calc according to new OS impl

[Yoni-Starkware]

No description provided.

[reviewable-StarkWare]

This change is 

[Yoni-Starkware]

Reviewable status: 0 of 2 files reviewed, 1 unresolved discussion (waiting on @dorimedini-starkware and @noaov1)

---

crates/blockifier/src/bouncer_test.rs line 457 at r1 (raw file):

                527194,
            ),
        }

We have two acesses, one of the is write. So we saved one traversal:

In [2]: 702922 - (1 * 5722 + 100 * 16) * 24
Out[2]: 527194

In [3]: 542410 - (1 * 4050 + 100 * 16) * 24
Out[3]: 406810

Code quote:

    expect![
        r#"
        BouncerWeights {
            l1_gas: 10,
            message_segment_length: 10,
            n_events: 10,
            state_diff_size: 14,
            sierra_gas: GasAmount(
                406810,
            ),
            n_txs: 20,
            proving_gas: GasAmount(
                527194,
            ),
        }

[github-actions]

Artifacts upload workflows:

• Started at 2025-11-11T07:58:29Z
• Started at 2025-11-11T08:18:19Z
• Started at 2025-11-11T08:52:02Z
• Started at 2025-11-11T08:53:50Z
• Started at 2025-11-11T08:54:39Z

[dorimedini-starkware]

@dorimedini-starkware reviewed 2 of 2 files at r1, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @noaov1)

---

crates/blockifier/src/bouncer.rs line 852 at r1 (raw file):

        n_visited_storage_entries,
        // TODO(Yoni) consider counting here the global contract tree and the aliases as well.
        state_changes_keys.storage_keys.len(),

this includes aliases? we can ignore the block hash update I guess but we can't ignore aliases

Code quote:

state_changes_keys.storage_keys.len(),

[Yoni-Starkware]

Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @noaov1)

---

crates/blockifier/src/bouncer.rs line 852 at r1 (raw file):

Previously, dorimedini-starkware wrote…

this includes aliases? we can ignore the block hash update I guess but we can't ignore aliases

No, it's not. I don't want to include it yet since it will double the estimation (alias access per every storage access).
We are already overshooting badly here. The alias is a special case, maybe I could estimate a multi-update there properly

[noaov1]

@noaov1 reviewed 1 of 2 files at r1, 1 of 1 files at r2, all commit messages.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @Yoni-Starkware)

---

crates/blockifier/src/bouncer.rs line 955 at r1 (raw file):

///   1. n_visited_storage_entries includes both read and write accesses, and may overlap with
///      n_modified_storage_entries (if the first access to a cell was write) and my not (if a
///      cell was read by a previous transaction and is now modified).

What about a write overlap? If a previous transaction has already written to that cell?

Suggestion:

///   1. n_visited_storage_entries includes both read and write accesses, and may overlap with
///      n_modified_storage_entries (if the first access to a cell was write) and may not (if a
///      cell was read by a previous transaction and is now modified).

[noaov1]

@noaov1 reviewed 1 of 2 files at r1.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @Yoni-Starkware)

[Yoni-Starkware]

Reviewable status: 1 of 2 files reviewed, 1 unresolved discussion (waiting on @noaov1)

---

crates/blockifier/src/bouncer.rs line 955 at r1 (raw file):

Previously, noaov1 (Noa Oved) wrote…

What about a write overlap? If a previous transaction has already written to that cell?

I wouldn't appear in state_changes_keys. Only the first change is counted. We do not remove cancelled writes from it.

Update names and docs; hopefully, we won't fall into this hole in the future

[dorimedini-starkware]

@dorimedini-starkware reviewed 1 of 1 files at r5, all commit messages.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @noaov1)

[noaov1]

@noaov1 reviewed 1 of 1 files at r5, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @Yoni-Starkware)