feat: reduce the amount of key hashing required to fetch public keys

[nventuro]

Closes #7825

We now hash each set of public keys once instead of twice by having a smarter oracle function that always returns the correct value, instead of two that we need to call conditionally.

This reduces the gate count of a token transfer by ~3k, though this reduction is expected to be lower and closer to ~1k once #7954 is closed, since we'll then be performing fewer total key read operations.

[AztecBot]

Benchmark results

Metrics with a significant change:

• protocol_circuit_simulation_time_in_ms (private-kernel-tail-to-public): 776 (-41%)
• avm_simulation_time_ms (Token:mint_public): 43.4 (-39%)
• avm_simulation_time_ms (Token:assert_minter_and_mint): 67.8 (+70%)
• avm_simulation_time_ms (Token:transfer_public): 20.4 (-37%)
• avm_simulation_time_ms (Token:_increase_public_balance): 81.2 (+106%)

Detailed results

All benchmarks are run on txs on the Benchmarking contract on the repository. Each tx consists of a batch call to create_note and increment_balance, which guarantees that each tx has a private call, a nested private call, a public call, and a nested public call, as well as an emitted private note, an unencrypted log, and public storage read and write.

This benchmark source data is available in JSON format on S3 here.

Proof generation

Each column represents the number of threads used in proof generation.

Metric	1 threads	4 threads	16 threads	32 threads	64 threads
proof_construction_time_sha256_ms	5,783	1,559	706 (-1%)	754 (+1%)	764 (-1%)
proof_construction_time_sha256_30_ms	11,846	3,167 (-1%)	1,413	1,441 (-1%)	1,471 (-1%)
proof_construction_time_sha256_100_ms	45,423 (-1%)	12,039	5,412 (-2%)	5,386 (-2%)	5,362 (-1%)
proof_construction_time_poseidon_hash_ms	78.0	34.0	34.0 (-3%)	59.0 (+2%)	87.0 (-1%)
proof_construction_time_poseidon_hash_30_ms	1,533	423	202 (-1%)	228 (-2%)	267 (-2%)
proof_construction_time_poseidon_hash_100_ms	5,643	1,516	676	721 (+1%)	747 (-1%)

L2 block published to L1

Each column represents the number of txs on an L2 block published to L1.

Metric	4 txs	8 txs	16 txs
l1_rollup_calldata_size_in_bytes	4,324	7,844	14,852
l1_rollup_calldata_gas	49,648	92,496	177,500
l1_rollup_execution_gas	1,294,311	2,042,127	3,869,170
l2_block_processing_time_in_ms	253 (-1%)	446 (+2%)	788 (-2%)
l2_block_building_time_in_ms	9,263 (-2%)	18,192	36,040 (-1%)
l2_block_rollup_simulation_time_in_ms	9,262 (-2%)	18,192	36,040 (-1%)
l2_block_public_tx_process_time_in_ms	7,822 (-2%)	16,627	34,470 (-1%)

L2 chain processing

Each column represents the number of blocks on the L2 chain where each block has 8 txs.

Metric	3 blocks	5 blocks
node_history_sync_time_in_ms	2,983 (+1%)	3,842 (+3%)
node_database_size_in_bytes	12,669,008	16,711,760
pxe_database_size_in_bytes	16,254	26,813

Circuits stats

Stats on running time and I/O sizes collected for every kernel circuit run across all benchmarks.

Circuit	simulation_time_in_ms	witness_generation_time_in_ms	input_size_in_bytes	output_size_in_bytes	proving_time_in_ms
private-kernel-init	95.8 (+1%)	390 (-1%)	21,846	44,858	N/A
private-kernel-inner	173 (+1%)	694 (-1%)	72,545	45,005	N/A
private-kernel-reset-tiny	502 (+1%)	1,214	69,614	44,844	N/A
private-kernel-tail	270	193 (-2%)	50,760	52,256	N/A
base-parity	5.54 (-1%)	N/A	160	96.0	N/A
root-parity	33.1 (-1%)	N/A	69,084	96.0	N/A
base-rollup	2,854 (-1%)	N/A	187,817	664	N/A
root-rollup	38.5	N/A	54,525	716	N/A
public-kernel-setup	99.6	N/A	103,760	71,222	N/A
public-kernel-app-logic	106	N/A	103,599	71,222	N/A
public-kernel-tail	570	N/A	409,190	16,414	N/A
private-kernel-reset-small	496 (+3%)	N/A	66,533	45,629	N/A
private-kernel-tail-to-public	⚠️ 776 (-41%)	702 (-1%)	507,554	1,697	N/A
public-kernel-teardown	94.4	N/A	104,005	71,222	N/A
merge-rollup	19.2 (+1%)	N/A	35,742	664	N/A
undefined	N/A	N/A	N/A	N/A	67,795 (-2%)

Stats on running time collected for app circuits

Function	input_size_in_bytes	output_size_in_bytes	witness_generation_time_in_ms
ContractClassRegisterer:register	1,344	11,731	346
ContractInstanceDeployer:deploy	1,408	11,731	18.1
MultiCallEntrypoint:entrypoint	1,920	11,731	428
FeeJuice:deploy	1,376	11,731	400 (+2%)
SchnorrAccount:constructor	1,312	11,731	116 (+10%)
SchnorrAccount:entrypoint	2,304	11,731	442 (+3%)
Token:privately_mint_private_note	1,280	11,731	146 (+5%)
FPC:fee_entrypoint_public	1,344	11,731	21.5 (-23%)
Token:transfer	1,312	11,731	299 (+5%)
Benchmarking:create_note	1,344	11,731	112 (+12%)
SchnorrAccount:verify_private_authwit	1,280	11,731	27.7
Token:unshield	1,376	11,731	575 (+2%)
FPC:fee_entrypoint_private	1,376	11,731	764

AVM Simulation

Time to simulate various public functions in the AVM.

Function	time_ms	bytecode_size_in_bytes
FeeJuice:_increase_public_balance	54.2 (-1%)	8,139
FeeJuice:set_portal	9.48 (-10%)	2,362
Token:constructor	80.5 (-1%)	31,107
FPC:constructor	54.8 (+2%)	22,380
FeeJuice:mint_public	38.0 (+1%)	6,150
Token:mint_public	⚠️ 43.4 (-39%)	11,720
Token:assert_minter_and_mint	⚠️ 67.8 (+70%)	8,028
AuthRegistry:set_authorized	37.5 (-17%)	4,537
FPC:prepare_fee	243 (+6%)	8,812
Token:transfer_public	⚠️ 20.4 (-37%)	47,374
FPC:pay_refund	56.4 (-3%)	12,114
Benchmarking:increment_balance	972	7,450
Token:_increase_public_balance	⚠️ 81.2 (+106%)	8,960
FPC:pay_refund_with_shielded_rebate	68.1 (+3%)	12,663

Public DB Access

Time to access various public DBs.

Function	time_ms
get-nullifier-index	0.164 (-2%)

Tree insertion stats

The duration to insert a fixed batch of leaves into each tree type.

Metric	1 leaves	16 leaves	64 leaves	128 leaves	256 leaves	512 leaves	1024 leaves
batch_insert_into_append_only_tree_16_depth_ms	2.16	3.82 (+1%)	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_16_depth_hash_count	16.8	31.7	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_16_depth_hash_ms	0.112 (+1%)	0.108	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_32_depth_ms	N/A	N/A	11.6 (+3%)	18.0 (+1%)	31.1	59.6 (+2%)	111 (-3%)
batch_insert_into_append_only_tree_32_depth_hash_count	N/A	N/A	95.9	159	287	543	1,055
batch_insert_into_append_only_tree_32_depth_hash_ms	N/A	N/A	0.111 (+3%)	0.104	0.101	0.103 (+2%)	0.101 (-2%)
batch_insert_into_indexed_tree_20_depth_ms	N/A	N/A	14.5 (+1%)	25.8	44.1 (+1%)	83.1 (+1%)	160 (-1%)
batch_insert_into_indexed_tree_20_depth_hash_count	N/A	N/A	109	207	355	691	1,363
batch_insert_into_indexed_tree_20_depth_hash_ms	N/A	N/A	0.110	0.104	0.107 (+1%)	0.103 (+1%)	0.103
batch_insert_into_indexed_tree_40_depth_ms	N/A	N/A	16.4 (+1%)	N/A	N/A	N/A	N/A
batch_insert_into_indexed_tree_40_depth_hash_count	N/A	N/A	132	N/A	N/A	N/A	N/A
batch_insert_into_indexed_tree_40_depth_hash_ms	N/A	N/A	0.105 (+1%)	N/A	N/A	N/A	N/A

Miscellaneous

Transaction sizes based on how many contract classes are registered in the tx.

Metric	0 registered classes	1 registered classes
tx_size_in_bytes	64,779	668,997

Transaction size based on fee payment method

| Metric | |
| - | |

[benesjan]

Cute 🐣💖😊🌸✨

[nventuro]

This is failing because my new unconstrained helper creates an UnconstrainedContext, which requires calling some oracles (namely block number and timestamp), and some tests are running in a heavily mocked environment in which these are unavailable.

Fixing this is not terribly hard but also not trivial, so I'll leave this open since we're unsure about whether we'll keep key rotation at all.

[nventuro]

Closing this as we're changing how addresses and keys are related.