feat(avm): gzip avm bytecode

[fcarreiro]

We can use this to reduce bytecode size (just as Brillig does) until we do bytecode validation.

AvmAcvmInteropTest::constant_field_avm: compressed 139 to 95 bytes
AvmAcvmInteropTest::call_acvm_from_avm: compressed 6343 to 1486 bytes
AvmAcvmInteropTest::avm_to_acvm_call: compressed 2145 to 552 bytes
AvmAcvmInteropTest::new_nullifier: compressed 91 to 74 bytes
AvmAcvmInteropTest::test_authwit_send_money: compressed 9090 to 1977 bytes
AvmInitializerTest::read_storage_immutable: compressed 2078 to 560 bytes
AvmInitializerTest::constructor: compressed 73769 to 6638 bytes
AvmNestedCallsTest::set_storage_single: compressed 719 to 201 bytes
AvmNestedCallsTest::nested_call_to_add_with_gas: compressed 48796 to 3976 bytes
AvmNestedCallsTest::assert_same: compressed 1758 to 452 bytes
AvmNestedCallsTest::new_nullifier: compressed 91 to 74 bytes
AvmNestedCallsTest::nested_static_call_to_add: compressed 48643 to 3871 bytes
AvmNestedCallsTest::nested_static_call_to_set_storage: compressed 47167 to 3573 bytes
AvmNestedCallsTest::add_args_return: compressed 140 to 92 bytes
AvmNestedCallsTest::nested_call_to_add: compressed 48258 to 3843 bytes
AvmNestedCallsTest::create_same_nullifier_in_nested_call: compressed 46760 to 3527 bytes
AvmNestedCallsTest::create_different_nullifier_in_nested_call: compressed 46809 to 3540 bytes
AvmTest::get_chain_id: compressed 111 to 81 bytes
AvmTest::to_radix_le: compressed 41037 to 2164 bytes
AvmTest::read_storage_list: compressed 1186 to 345 bytes
AvmTest::set_storage_map: compressed 5335 to 930 bytes
AvmTest::new_nullifier: compressed 91 to 74 bytes
AvmTest::set_opcode_u32: compressed 116 to 83 bytes
AvmTest::assertion_failure: compressed 2107 to 495 bytes
AvmTest::set_opcode_small_field: compressed 139 to 106 bytes
AvmTest::read_storage_map: compressed 4987 to 890 bytes
AvmTest::get_address: compressed 111 to 81 bytes
AvmTest::pedersen_hash_with_index: compressed 199 to 114 bytes
AvmTest::get_sender: compressed 111 to 81 bytes
AvmTest::add_storage_map: compressed 10417 to 1742 bytes
AvmTest::set_storage_list: compressed 864 to 219 bytes
AvmTest::get_fee_per_da_gas: compressed 111 to 81 bytes
AvmTest::u128_addition_overflow: compressed 46973 to 2808 bytes
AvmTest::read_storage_single: compressed 728 to 227 bytes
AvmTest::test_get_contract_instance_raw: compressed 1576 to 442 bytes
AvmTest::test_get_contract_instance: compressed 44797 to 2953 bytes
AvmTest::get_version: compressed 111 to 80 bytes
AvmTest::check_selector: compressed 44567 to 3039 bytes
AvmTest::get_transaction_fee: compressed 111 to 81 bytes
AvmTest::add_u128: compressed 2390 to 593 bytes
AvmTest::nullifier_collision: compressed 97 to 76 bytes
AvmTest::pedersen_hash: compressed 199 to 113 bytes
AvmTest::modulo2: compressed 177 to 104 bytes
AvmTest::debug_logging: compressed 8916 to 1718 bytes
AvmTest::assert_nullifier_exists: compressed 1964 to 502 bytes
AvmTest::l1_to_l2_msg_exists: compressed 162 to 101 bytes
AvmTest::note_hash_exists: compressed 162 to 102 bytes
AvmTest::emit_unencrypted_log: compressed 47663 to 3297 bytes
AvmTest::send_l2_to_l1_msg: compressed 105 to 79 bytes
AvmTest::get_args_hash: compressed 176 to 105 bytes
AvmTest::get_block_number: compressed 111 to 80 bytes
AvmTest::set_read_storage_single: compressed 1372 to 350 bytes
AvmTest::get_fee_per_l2_gas: compressed 111 to 81 bytes
AvmTest::keccak_hash: compressed 559 to 220 bytes
AvmTest::sha256_hash: compressed 537 to 207 bytes
AvmTest::nullifier_exists: compressed 158 to 99 bytes
AvmTest::new_note_hash: compressed 91 to 74 bytes
AvmTest::set_opcode_u8: compressed 113 to 82 bytes
AvmTest::add_args_return: compressed 140 to 92 bytes
AvmTest::get_storage_address: compressed 111 to 81 bytes
AvmTest::emit_nullifier_and_check: compressed 4100 to 682 bytes
AvmTest::u128_from_integer_overflow: compressed 43568 to 2048 bytes
AvmTest::poseidon2_hash: compressed 55082 to 4986 bytes
AvmTest::set_opcode_u64: compressed 120 to 86 bytes
AvmTest::get_timestamp: compressed 111 to 80 bytes
AvmTest::set_storage_single: compressed 719 to 201 bytes
AvmTest::set_opcode_big_field: compressed 263 to 137 bytes

[fcarreiro]

• feat(avm): gzip avm bytecode #6475 👈
• master

This stack of pull requests is managed by Graphite. Learn more about stacking.

Join @fcarreiro and the rest of your teammates on Graphite

[AztecBot]

Benchmark results

Metrics with a significant change:

• tx_size_in_bytes (1): 668,029 (-37%)
• tx_sequencer_processing_time_ms (1): 1,567 (+40%)
• app_circuit_proving_time_in_ms (Test:emit_nullifier): 2,362 (+40%)
• app_circuit_proving_time_in_ms (FPC:fee_entrypoint_public): 13,340 (+59%)
• protocol_circuit_witness_generation_time_in_ms (base-parity): 206 (+96%)
• protocol_circuit_witness_generation_time_in_ms (merge-rollup): 183 (+112%)
• protocol_circuit_proving_time_in_ms (base-parity): 3,461 (+25%)
• protocol_circuit_proving_time_in_ms (public-kernel-app-logic): 640 (-16%)
• l2_block_public_tx_process_time_in_ms (32): 51,226 (+16%)
• l2_block_public_tx_process_time_in_ms (64): 102,913 (+18%)

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.

L2 block published to L1

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

Metric	8 txs	32 txs	64 txs
l1_rollup_calldata_size_in_bytes	740	740	740
l1_rollup_calldata_gas	6,680	6,680	6,680
l1_rollup_execution_gas	586,865	586,865	586,865
l2_block_processing_time_in_ms	1,617 (+1%)	6,080 (+1%)	12,047 (+1%)
l2_block_building_time_in_ms	27,001	105,929 (-1%)	210,437 (-1%)
l2_block_rollup_simulation_time_in_ms	26,815	105,272 (-1%)	209,178 (-1%)
l2_block_public_tx_process_time_in_ms	12,033 (+8%)	⚠️ 51,226 (+16%)	⚠️ 102,913 (+18%)

L2 chain processing

Each column represents the number of blocks on the L2 chain where each block has 16 txs.

Metric	5 blocks	10 blocks
node_history_sync_time_in_ms	18,030 (-2%)	33,697 (-1%)
node_database_size_in_bytes	21,373,008 (-6%)	38,019,152 (-4%)
pxe_database_size_in_bytes	29,868	59,425

Circuits stats

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

Circuit	protocol_circuit_simulation_time_in_ms	protocol_circuit_witness_generation_time_in_ms	protocol_circuit_proving_time_in_ms	protocol_circuit_input_size_in_bytes	protocol_circuit_output_size_in_bytes	protocol_circuit_proof_size_in_bytes	protocol_circuit_num_public_inputs	protocol_circuit_size_in_gates
private-kernel-init	140 (-2%)	2,616	20,596 (-1%)	19,396	46,118	62,400	1,883	1,048,576
private-kernel-inner	417	3,677 (+2%)	38,524 (-1%)	72,583	46,118	62,400	1,883	2,097,152
private-kernel-reset	674 (-1%)	4,988 (+14%)	45,901 (+2%)	219,992	46,118	62,400	1,883	2,097,152
private-kernel-tail	441 (-1%)	4,747 (-2%)	37,190 (+1%)	70,968	64,320	10,656	266	2,097,152
base-parity	8.25 (+1%)	⚠️ 206 (+96%)	⚠️ 3,461 (+25%)	128	64.0	2,208	2.00	131,072
root-parity	51.7 (+1%)	132 (+5%)	40,595 (+1%)	27,064	64.0	2,720	18.0	2,097,152
base-rollup	833 (+12%)	5,239 (+4%)	43,369 (+1%)	111,126	925	3,104	30.0	2,097,152
root-rollup	72.8	171 (-1%)	8,315 (+2%)	11,454	789	3,424	40.0	524,288
public-kernel-app-logic	155	255 (+1%)	⚠️ 640 (-16%)	78,387	68,775	90,432	2,759	4,096
public-kernel-tail	534 (-1%)	638 (-2%)	834 (+7%)	184,419	7,723	10,144	250	512
public-kernel-setup	132 (+2%)	295	838 (+2%)	110,522	68,775	90,432	2,759	4,096
public-kernel-teardown	139 (+6%)	345 (-12%)	908 (-8%)	114,417	68,775	90,432	2,759	4,096
merge-rollup	6.88 (+2%)	⚠️ 183 (+112%)	1,533	2,696	925	3,104	30.0	65,536
private-kernel-tail-to-public	N/A	9,090 (-5%)	43,590 (+3%)	N/A	N/A	90,944	2,775	2,097,152

Stats on running time collected for app circuits

Function	app_circuit_proof_size_in_bytes	app_circuit_proving_time_in_ms	app_circuit_size_in_gates	app_circuit_num_public_inputs
SchnorrAccount:entrypoint	13,600	47,654 (+3%)	2,097,152	358
Test:emit_nullifier	13,600	⚠️ 2,362 (+40%)	65,536	358
FPC:fee_entrypoint_public	13,600	⚠️ 13,340 (+59%)	524,288	358
Token:transfer	13,600	44,726 (+9%)	2,097,152	358
FPC:fee_entrypoint_private	13,600	11,371 (+15%)	524,288	358
Token:unshield	13,600	34,051 (+1%)	2,097,152	358
SchnorrAccount:spend_private_authwit	13,600	2,602 (+1%)	131,072	358

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	512 leaves	1024 leaves	2048 leaves	4096 leaves	32 leaves
batch_insert_into_append_only_tree_16_depth_ms	13.3 (+1%)	21.4	N/A	N/A	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_16_depth_hash_count	16.7	31.8	N/A	N/A	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_16_depth_hash_ms	0.775 (+1%)	0.660	N/A	N/A	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_32_depth_ms	N/A	N/A	61.8	97.3 (-2%)	316	611	1,195	2,361	N/A
batch_insert_into_append_only_tree_32_depth_hash_count	N/A	N/A	95.9	159	543	1,055	2,079	4,127	N/A
batch_insert_into_append_only_tree_32_depth_hash_ms	N/A	N/A	0.634	0.601 (-2%)	0.574	0.572	0.567	0.566	N/A
batch_insert_into_indexed_tree_20_depth_ms	N/A	N/A	74.2 (+1%)	144	457	896	1,790 (+1%)	3,549	N/A
batch_insert_into_indexed_tree_20_depth_hash_count	N/A	N/A	106	208	692	1,363	2,707	5,395	N/A
batch_insert_into_indexed_tree_20_depth_hash_ms	N/A	N/A	0.647	0.646	0.619	0.617	0.616	0.617	N/A
batch_insert_into_indexed_tree_40_depth_ms	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A	80.9
batch_insert_into_indexed_tree_40_depth_hash_count	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A	108
batch_insert_into_indexed_tree_40_depth_hash_ms	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A	0.711

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	68,448	⚠️ 668,029 (-37%)

Transaction size based on fee payment method

| Metric | |
| - | |

Transaction processing duration by data writes.

Metric	0 new note hashes	1 new note hashes	2 new note hashes
tx_pxe_processing_time_ms	28,098	3,073	91,839 (+2%)

Metric	0 public data writes	1 public data writes	2 public data writes	4 public data writes	8 public data writes
tx_sequencer_processing_time_ms	827 (-11%)	⚠️ 1,567 (+40%)	1,326 (+5%)	2,148 (+7%)	1,341 (-1%)