chore(avm): better log_derivative_inverse_round

[fcarreiro]

Moved it to the prover cpp and used metaprogramming to cycle through relations.

I'm expecting the compiler to be smart enough to remove recursion. I think this is quite common in metaprogramming.

[fcarreiro]

• chore(avm): avoid including flavor where possible #7361
• chore(avm): better log_derivative_inverse_round #7360 👈
• chore(avm): smaller prover #7359
• chore(avm): smaller transcript #7357
• 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:

• proof_construction_time_sha256_30_ms (32): 1,781 (+25%)
• protocol_circuit_witness_generation_time_in_ms (private-kernel-init): 370 (-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.

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	5,724	1,546	707	748 (+1%)	772
proof_construction_time_sha256_30_ms	11,741	3,138	1,412 (+1%)	⚠️ 1,781 (+25%)	1,462
proof_construction_time_sha256_100_ms	43,986 (+1%)	11,769	5,438	5,384 (-1%)	5,378 (+1%)
proof_construction_time_poseidon_hash_ms	78.0	34.0	34.0	57.0 (-2%)	88.0
proof_construction_time_poseidon_hash_30_ms	1,514	416	201 (+1%)	219 (-3%)	264 (-1%)
proof_construction_time_poseidon_hash_100_ms	5,733	1,563	726 (+1%)	801 (+2%)	791

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	1,412	1,412	1,412
l1_rollup_calldata_gas	9,464	9,464	9,476
l1_rollup_execution_gas	611,203	611,354	611,517
l2_block_processing_time_in_ms	758	1,424 (+1%)	2,785 (+3%)
l2_block_building_time_in_ms	20,889	41,983 (+1%)	81,975
l2_block_rollup_simulation_time_in_ms	20,889	41,982 (+1%)	81,975
l2_block_public_tx_process_time_in_ms	17,867	38,722 (+1%)	78,811

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	7,102	9,951 (+2%)
node_database_size_in_bytes	12,443,728	16,212,048
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	proving_time_in_ms	input_size_in_bytes	output_size_in_bytes	proof_size_in_bytes	num_public_inputs	size_in_gates
private-kernel-init	100	⚠️ 370 (-18%)	12,946	19,418	53,878	71,872	2,179	524,288
private-kernel-inner	303	729 (-1%)	49,534 (+1%)	80,374	53,878	71,872	2,179	2,097,152
private-kernel-tail	1,045 (+1%)	2,629 (+3%)	45,923	61,201	62,057	14,912	399	2,097,152
base-parity	6.11 (-1%)	1,481	2,631	128	64.0	2,208	2.00	131,072
root-parity	48.6	65.6 (-5%)	42,370 (+1%)	27,100	64.0	2,720	18.0	2,097,152
base-rollup	6,609 (+1%)	4,917 (+2%)	91,092 (+1%)	170,330	728	3,648	47.0	4,194,304
root-rollup	109	79.3 (-3%)	24,384 (+3%)	25,253	620	3,456	41.0	1,048,576
public-kernel-setup	531 (-1%)	2,382	41,945 (-1%)	102,121	80,278	106,912	3,274	2,097,152
public-kernel-app-logic	504 (+1%)	3,384 (+2%)	42,219 (-4%)	102,121	80,278	106,912	3,274	2,097,152
public-kernel-tail	1,143	26,521 (+9%)	184,025	399,014	10,014	14,912	399	8,388,608
private-kernel-reset-small	286	1,075 (-1%)	26,309 (+1%)	78,065	53,878	71,872	2,179	1,048,576
public-kernel-teardown	484 (-1%)	3,384 (+2%)	44,760 (+4%)	102,121	80,278	106,912	3,274	2,097,152
merge-rollup	29.4	N/A	N/A	16,486	728	N/A	N/A	N/A
private-kernel-tail-to-public	N/A	8,111 (-2%)	50,229 (+1%)	N/A	N/A	106,912	3,274	2,097,152

Stats on running time collected for app circuits

Function	input_size_in_bytes	output_size_in_bytes	witness_generation_time_in_ms	proof_size_in_bytes	proving_time_in_ms	size_in_gates	num_public_inputs
ContractClassRegisterer:register	1,344	8,792	390 (-1%)	N/A	N/A	N/A	N/A
ContractInstanceDeployer:deploy	1,408	8,792	23.9	N/A	N/A	N/A	N/A
MultiCallEntrypoint:entrypoint	1,920	8,792	1,149	N/A	N/A	N/A	N/A
GasToken:deploy	1,376	8,792	810	N/A	N/A	N/A	N/A
SchnorrAccount:constructor	1,312	8,792	473	N/A	N/A	N/A	N/A
SchnorrAccount:entrypoint	2,304	8,792	1,590	14,720	54,787	2,097,152	393
Token:privately_mint_private_note	1,280	8,792	597 (-3%)	N/A	N/A	N/A	N/A
FPC:fee_entrypoint_public	1,344	8,792	234 (+1%)	14,720	12,203 (+7%)	524,288	393
Token:transfer	1,312	8,792	1,750 (-1%)	14,720	12,671 (+2%)	524,288	393
AuthRegistry:set_authorized (avm)	19,226	N/A	N/A	91,264	1,359 (+1%)	N/A	N/A
FPC:prepare_fee (avm)	26,668	N/A	N/A	91,328	2,846 (-2%)	N/A	N/A
Token:transfer_public (avm)	42,918	N/A	N/A	91,328	3,937 (+4%)	N/A	N/A
AuthRegistry:consume (avm)	33,104	N/A	N/A	91,264	2,783	N/A	N/A
FPC:pay_refund (avm)	36,833	N/A	N/A	91,296	24,843 (+9%)	N/A	N/A
Benchmarking:create_note	1,344	8,792	467	N/A	N/A	N/A	N/A
SchnorrAccount:verify_private_authwit	1,280	8,792	40.5	N/A	N/A	N/A	N/A
Token:unshield	1,376	8,792	1,448 (-2%)	N/A	N/A	N/A	N/A
FPC:fee_entrypoint_private	1,376	8,792	1,989 (-1%)	N/A	N/A	N/A	N/A

AVM Simulation

Time to simulate various public functions in the AVM.

Function	time_ms	bytecode_size_in_bytes
GasToken:_increase_public_balance	69.6 (+1%)	13,790
GasToken:set_portal	16.5 (-3%)	3,339
Token:constructor	97.4 (+7%)	23,692
FPC:constructor	62.2	13,592
GasToken:mint_public	50.2 (+1%)	10,158
Token:mint_public	493 (+5%)	19,034
Token:assert_minter_and_mint	55.3 (+6%)	12,925
AuthRegistry:set_authorized	29.1 (+1%)	7,812
FPC:prepare_fee	107 (+1%)	15,062
Token:transfer_public	41.3 (-5%)	31,218
FPC:pay_refund	129 (+4%)	25,260
Benchmarking:increment_balance	2,201	15,267
Token:_increase_public_balance	54.5	15,006
FPC:pay_refund_with_shielded_rebate	114 (-3%)	26,347

Public DB Access

Time to access various public DBs.

Function	time_ms
get-nullifier-index	0.158 (+3%)

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	10.5 (+1%)	16.8 (+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	0.604 (+1%)	0.516 (+1%)	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_32_depth_ms	N/A	N/A	48.3	75.7 (-1%)	131 (-1%)	246	472 (+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	0.493	0.465 (-1%)	0.451	0.446	0.441 (+1%)
batch_insert_into_indexed_tree_20_depth_ms	N/A	N/A	59.4 (+1%)	112 (+1%)	183 (+1%)	355 (+1%)	696 (+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	0.501	0.503 (+1%)	0.485 (+1%)	0.480 (+1%)	0.477 (+1%)
batch_insert_into_indexed_tree_40_depth_ms	N/A	N/A	72.9 (+1%)	N/A	N/A	N/A	N/A
batch_insert_into_indexed_tree_40_depth_hash_count	N/A	N/A	133	N/A	N/A	N/A	N/A
batch_insert_into_indexed_tree_40_depth_hash_ms	N/A	N/A	0.519 (+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	74,057	667,841

Transaction size based on fee payment method

| Metric | |
| - | |

[fcarreiro]

Merge activity

• Jul 8, 7:28 AM EDT: @fcarreiro started a stack merge that includes this pull request via Graphite.
• Jul 8, 7:31 AM EDT: @fcarreiro merged this pull request with Graphite.