refactor(avm): separate alu finalization

[fcarreiro]

PIL

• Moved ALU lookups to alu.pil. These are mostly range checks and I expect them to be different once @IlyasRidhuan finishes setting up the gadget changes.

CPP

• Made ALU entry smaller and encapsulated finalize.

[fcarreiro]

• refactor(avm): separate alu finalization #8069 👈
• 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:

• avm_simulation_time_ms (Token:mint_public): 68.8 (+60%)
• avm_simulation_time_ms (Token:assert_minter_and_mint): 43.1 (-35%)

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	5,790 (+1%)	1,556	707	759	770 (-1%)
proof_construction_time_sha256_30_ms	11,840	3,166 (-1%)	1,412 (-1%)	1,437	1,467 (-1%)
proof_construction_time_sha256_100_ms	45,435	11,880 (-1%)	5,425 (-2%)	5,417 (-1%)	5,360 (-2%)
proof_construction_time_poseidon_hash_ms	78.0	34.0	34.0	56.0 (-5%)	88.0 (+1%)
proof_construction_time_poseidon_hash_30_ms	1,537	421	202	233	265 (-1%)
proof_construction_time_poseidon_hash_100_ms	5,676 (+1%)	1,516	682	737 (+2%)	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	49,696	92,436	177,680
l1_rollup_execution_gas	1,383,190	2,130,898	3,958,181
l2_block_processing_time_in_ms	259 (+5%)	434 (-1%)	821
l2_block_building_time_in_ms	9,005 (+1%)	17,372 (-1%)	35,066 (+2%)
l2_block_rollup_simulation_time_in_ms	9,005 (+1%)	17,372 (-1%)	35,065 (+2%)
l2_block_public_tx_process_time_in_ms	7,604 (+1%)	15,867 (-1%)	33,532 (+2%)

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	2,827	3,690 (-4%)
node_database_size_in_bytes	12,607,568	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	82.8 (-7%)	400 (+1%)	21,502	44,858	N/A
private-kernel-inner	158 (+2%)	728 (+3%)	72,180	45,005	N/A
private-kernel-reset-tiny	471	891 (+2%)	65,502	44,844	N/A
private-kernel-tail	200 (+1%)	161 (-1%)	50,606	52,256	N/A
base-parity	5.59 (+1%)	N/A	160	96.0	N/A
root-parity	33.2	N/A	69,084	96.0	N/A
base-rollup	2,749	N/A	187,817	664	N/A
root-rollup	39.0	N/A	54,525	716	N/A
public-kernel-setup	83.0 (-1%)	N/A	103,760	71,222	N/A
public-kernel-app-logic	95.1 (-1%)	N/A	103,599	71,222	N/A
public-kernel-tail	555	N/A	409,190	16,414	N/A
private-kernel-reset-small	466 (-1%)	N/A	66,085	45,629	N/A
private-kernel-tail-to-public	990 (-4%)	649 (-1%)	473,529	1,697	N/A
public-kernel-teardown	82.4	N/A	104,005	71,222	N/A
merge-rollup	19.1 (-1%)	N/A	35,742	664	N/A
undefined	N/A	N/A	N/A	N/A	66,591 (+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	343
ContractInstanceDeployer:deploy	1,408	11,731	18.3
MultiCallEntrypoint:entrypoint	1,920	11,731	421
FeeJuice:deploy	1,376	11,731	390
SchnorrAccount:constructor	1,312	11,731	104 (-1%)
SchnorrAccount:entrypoint	2,304	11,731	429
Token:privately_mint_private_note	1,280	11,731	138 (-1%)
FPC:fee_entrypoint_public	1,344	11,731	26.9
Token:transfer	1,312	11,731	274 (+1%)
Benchmarking:create_note	1,344	11,731	100 (-3%)
SchnorrAccount:verify_private_authwit	1,280	11,731	27.4 (-1%)
Token:unshield	1,376	11,731	553 (-1%)
FPC:fee_entrypoint_private	1,376	11,731	749 (+1%)

AVM Simulation

Time to simulate various public functions in the AVM.

Function	time_ms	bytecode_size_in_bytes
FeeJuice:_increase_public_balance	54.0 (+2%)	7,739
FeeJuice:set_portal	11.8 (-1%)	2,354
Token:constructor	79.8	26,525
FPC:constructor	51.4 (-2%)	18,383
FeeJuice:mint_public	38.6 (+2%)	5,877
Token:mint_public	⚠️ 68.8 (+60%)	10,917
Token:assert_minter_and_mint	⚠️ 43.1 (-35%)	7,512
AuthRegistry:set_authorized	46.4 (+2%)	4,391
FPC:prepare_fee	237 (+2%)	7,712
Token:transfer_public	34.4 (+27%)	39,946
FPC:pay_refund	55.6 (-4%)	10,811
Benchmarking:increment_balance	929 (-1%)	6,929
Token:_increase_public_balance	39.4 (-1%)	8,433
FPC:pay_refund_with_shielded_rebate	64.2 (-1%)	11,360

Public DB Access

Time to access various public DBs.

Function	time_ms
get-nullifier-index	0.155 (-6%)

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	2.22 (+1%)	3.93 (+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	0.116 (+1%)	0.112 (+1%)	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_32_depth_ms	N/A	N/A	11.7	17.9 (+1%)	31.7 (+4%)	58.4 (-1%)	115 (+1%)
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	0.113	0.104 (+1%)	0.103 (+3%)	0.101 (-1%)	0.102
batch_insert_into_indexed_tree_20_depth_ms	N/A	N/A	14.5	25.8 (+1%)	44.5 (+3%)	81.8 (-1%)	162
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	0.111	0.104 (+1%)	0.108 (+3%)	0.102 (-1%)	0.103
batch_insert_into_indexed_tree_40_depth_ms	N/A	N/A	16.5	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	0.106	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 | |
| - | |

[jeanmon]

Thx @fcarreiro. This is really a nice clean up.

[jeanmon]

Suggested change

	sel_shift_which {t_sub_s_bits ,two_pow_t_sub_s} in main.sel_rng_8 {main.clk, powers.power_of_2};
	sel_shift_which {t_sub_s_bits, two_pow_t_sub_s} in main.sel_rng_8 {main.clk, powers.power_of_2};