feat: pil lookups w/ xor table example

[Maddiaa0]

depends on : https://github.com/AztecProtocol/powdr/pull/37/files

Lookups allow us to assert that values in a set of columns equal some values in another set of columns.

This approach uses the log derivative method to perform lookups, where the lookup relation requires four extra columns ( as well as the read and write columns ) to function.

• One selector column to initiate adding a value into the READ side of the lookup table ( lhs in pil )
• One selector column to initiate adding a value into the WRITE side of the lookup table ( rhs in pil ) - { adding to the table }
• One selector column to compute inverses - ( helper column )
• One column to store the number of times a WRITE term is read - ( helper column )

The selector columns will need to be defined manually - as in the ToyAvm example.
The inverse column and the counts columns are automatically included by the code gen. ( this means there are two extra columns per lookup ( which is a little bit inefficient; but will do for now ).

The name of the inverse column is decided by the attribute tag #[attribute] above the lookup - which is required.
The name of the counts column is ${attribute}_counts.

Note: that in the circuit builder you are currently required to set the counts column value manually.

syntax:

  #[lookup_xor] // attribute
    q_xor { xor_a, xor_b, xor_c } in q_xor_table { table_xor_a, table_xor_b, table_xor_c };
    // lhs selector, lhs columns        // rhs selector, rhs columns

closes: #2995

Co-authored-by: Rumata888 [email protected]

[AztecBot]

Benchmark results

Metrics with a significant change:

• note_successful_decrypting_time_in_ms (8): 358 (-25%)

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.

Values are compared against data from master at commit a2b267bf and shown if the difference exceeds 1%.

L2 block published to L1

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

Metric	8 txs	32 txs	128 txs
l1_rollup_calldata_size_in_bytes	45,444	179,588	716,132
l1_rollup_calldata_gas	223,056	867,884	3,449,492
l1_rollup_execution_gas	867,165	3,745,937	23,306,257
l2_block_processing_time_in_ms	1,341 (-1%)	5,096	21,191 (+1%)
note_successful_decrypting_time_in_ms	⚠️ 358 (-25%)	1,137 (+3%)	4,134 (+1%)
note_trial_decrypting_time_in_ms	67.8 (-24%)	36.0 (-9%)	136 (-5%)
l2_block_building_time_in_ms	15,016 (+1%)	59,595	238,775
l2_block_rollup_simulation_time_in_ms	10,667 (+1%)	42,365 (+1%)	170,251
l2_block_public_tx_process_time_in_ms	4,317	17,131	68,261 (+1%)

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	15,204 (+1%)	29,079
note_history_successful_decrypting_time_in_ms	2,908 (+8%)	5,268 (-5%)
note_history_trial_decrypting_time_in_ms	118 (+7%)	168 (+14%)
node_database_size_in_bytes	3,618,127	3,726,813
pxe_database_size_in_bytes	29,923	59,478

Circuits stats

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

Circuit	circuit_simulation_time_in_ms	circuit_input_size_in_bytes	circuit_output_size_in_bytes
private-kernel-init	211	44,709	25,561
private-kernel-ordering	137	30,953	14,809
base-rollup	1,078	128,374	881
root-rollup	82.5 (-3%)	4,088	889
private-kernel-inner	288	71,236	25,561
public-kernel-private-input	203	31,891	25,561
public-kernel-non-first-iteration	201 (+1%)	31,933	25,561
merge-rollup	9.70	2,608	881

Tree insertion stats

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

Metric	1 leaves	2 leaves	8 leaves	16 leaves	32 leaves	64 leaves	128 leaves	512 leaves	1024 leaves	2048 leaves	8192 leaves
batch_insert_into_append_only_tree_ms	11.0 (-1%)	88.1	13.6 (-2%)	17.7 (-2%)	23.7 (+2%)	49.4 (-1%)	88.4	242 (-2%)	469	909	3,563
batch_insert_into_indexed_tree_ms	N/A	N/A	N/A	64.5	N/A	59.4	110 (-1%)	358 (-2%)	731	1,401	5,737 (+3%)

Miscellaneous

Transaction sizes based on how many contracts are deployed in the tx.

Metric	0 deployed contracts	1 deployed contracts
tx_size_in_bytes	15,487	31,877

Transaction processing duration by data writes.

Metric	0 new commitments	1 new commitments
tx_pxe_processing_time_ms	369 (+7%)	970

Metric	1 public data writes
tx_sequencer_processing_time_ms	530

[jeanmon]

LGTM. Please address/look at my comment before merging.