core/state: track the block-level accessList

[rjl493456442]

This PR extends the journal to track the pre-transaction values of mutated balances, nonces, and code.

At the end of the transaction, these values are used to filter out no-op changes, such as balance transitions from a-> b->a. These changes are excluded from the block-level access list.

Additionally, there is a dedicated bal.ConstructionBlockAccessList objects for gathering the state reads and writes within the current transaction. These state writes will be keyed by the block accessList index.

[jwasinger]

It correctly catches the edge-case of a selfdestructing initcode in a prefunded account: a balance subtraction happens at the time the selfdestruct opcode is invoked.

[jwasinger]

Might be worth it to compare by hash here. I'm not sure how much performance difference it makes in practice.

[rjl493456442]

It's a good point.

In practice, the code change can be categorized as:

• nil -> contract code
• nil -> delegation
• delegation -> nil
• delegation a -> delegation b

for all these cases, the byte comparison is not expensive.

[jwasinger]

Did a first pass through this. tentatively LGTM but would be good to get more eyes on it.

[jwasinger]

Can you remind me what the plan was to track storage mutations?

[rjl493456442]

Can you remind me what the plan was to track storage mutations?

stateObject.dirtyStorage is exactly what we want.

		// Aggregate storage writes into the block-level access list.
		// All slots in the dirtyStorage set must have post-transaction
		// values that differ from their pre-transaction values.
		if s.db.stateAccessList != nil {
			s.db.stateAccessList.StorageWrite(s.db.blockAccessIndex, s.address, key, value)
		}