feat(avm-transpiler): implement tags for SET and others

[fcarreiro]

Generates tags for

• BinaryFieldOp
• BinaryIntOp
• Const (SET on the AVM)
• CAST on the AVM

Gotcha (1): Brillig seems to only support fields of <= 126 bits. We now support CONST of a field (Brillig) and transpile it to a SET<u128> + CAST<field>.

It's very difficult to test this e2e in the avm_test contract since Noir does a lot of optimizations. Moreover, Noir currently generates code that compares different bit sizes and uses unsupported bit sizes.

As an example

#[aztec(public-vm)]
fn setOpcodeUint64() -> pub u64 {
    1 << 60 as u64
}

produces (using nargo compile --package avm_test_contract --show-ssa --print-acir)

Compiled ACIR for AvmTest::avm_setOpcodeUint64 (unoptimized):
current witness index : 0
public parameters indices : []
return value indices : [0]
BRILLIG: inputs: []
outputs: [Simple(Witness(0))]
[Const { destination: MemoryAddress(0), bit_size: 32, value: Value { inner: 1025 } }, CalldataCopy { destination_address: MemoryAddress(1024), size: 0, offset: 0 }, Call { location: 5 }, Mov { destination: MemoryAddress(1024), source: MemoryAddress(2) }, Stop { return_data_offset: 1024, return_data_size: 1 }, Const { destination: MemoryAddress(2), bit_size: 64, value: Value { inner: 2⁶⁰ } }, Mov { destination: MemoryAddress(3), source: MemoryAddress(2) }, Mov { destination: MemoryAddress(2), source: MemoryAddress(3) }, Return]

Initial SSA:
brillig fn avm_setOpcodeUint64 f0 {
  b0():
    call f1()
    return u64 2⁶⁰
}

...
After Dead Instruction Elimination:
brillig fn avm_setOpcodeUint64 f0 {
  b0():
    return u64 2⁶⁰
}

Closes #4268.

[fcarreiro]

Current dependencies on/for this PR:

• master
  • PR feat(avm-transpiler): implement tags for SET and others #4545 👈

This stack of pull requests is managed by Graphite.

[AztecBot]

Benchmark results

Metrics with a significant change:

• note_history_trial_decrypting_time_in_ms (5): 305 (+260%)

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 da2f5ed3 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	45,700 (-1%)	180,580 (-1%)	720,100 (-1%)
l1_rollup_calldata_gas	222,796 (-2%)	871,036 (-2%)	3,465,700 (-2%)
l1_rollup_execution_gas	315,987 (-1%)	989,484 (-2%)	3,686,078 (-2%)
l2_block_processing_time_in_ms	1,160 (+1%)	4,397	17,388 (+1%)
note_successful_decrypting_time_in_ms	323 (+4%)	1,024 (+2%)	3,864 (+2%)
note_trial_decrypting_time_in_ms	36.7 (+66%)	44.0 (-5%)	154 (+6%)
l2_block_building_time_in_ms	18,381	72,879	292,767
l2_block_rollup_simulation_time_in_ms	13,479	53,478	215,392
l2_block_public_tx_process_time_in_ms	4,875	19,312	77,102

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	15,131	28,489
note_history_successful_decrypting_time_in_ms	2,429 (-5%)	4,653
note_history_trial_decrypting_time_in_ms	⚠️ 305 (+260%)	217 (+23%)
node_database_size_in_bytes	17,363,024	33,288,272
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	307	44,528	25,825
private-kernel-ordering	197	44,145	16,193
base-rollup	1,457	130,286	933
root-rollup	83.6 (+1%)	4,192	729
private-kernel-inner	436	71,331	25,825
public-kernel-private-input	249	34,687	27,457
public-kernel-non-first-iteration	248	34,729	27,457
merge-rollup	7.85	2,712	933

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	128 leaves	64 leaves	512 leaves	1024 leaves	2048 leaves	8192 leaves
batch_insert_into_append_only_tree_16_depth_ms	10.3	10.7 (-4%)	12.7 (+1%)	16.8 (-1%)	23.4 (+2%)	64.9 (+1%)	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_16_depth_hash_count	16.9	17.5	23.0	31.6	47.0	143	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_16_depth_hash_ms	0.596	0.595 (-4%)	0.542 (+1%)	0.520 (-1%)	0.486 (+1%)	0.445 (+1%)	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_32_depth_ms	N/A	N/A	N/A	N/A	N/A	75.0 (-5%)	47.9	239	461	907	3,585 (+1%)
batch_insert_into_append_only_tree_32_depth_hash_count	N/A	N/A	N/A	N/A	N/A	159	96.0	543	1,055	2,079	8,223
batch_insert_into_append_only_tree_32_depth_hash_ms	N/A	N/A	N/A	N/A	N/A	0.463 (-5%)	0.491	0.436	0.432	0.430	0.430 (+1%)
batch_insert_into_indexed_tree_20_depth_ms	N/A	N/A	N/A	N/A	N/A	105	56.9	349	689	1,380 (+1%)	5,450
batch_insert_into_indexed_tree_20_depth_hash_count	N/A	N/A	N/A	N/A	N/A	197	104	691	1,363	2,707	10,771
batch_insert_into_indexed_tree_20_depth_hash_ms	N/A	N/A	N/A	N/A	N/A	0.496 (-1%)	0.498	0.475	0.473	0.477 (+1%)	0.473
batch_insert_into_indexed_tree_40_depth_ms	N/A	N/A	N/A	56.7	N/A	N/A	N/A	N/A	N/A	N/A	N/A
batch_insert_into_indexed_tree_40_depth_hash_count	N/A	N/A	N/A	94.1	N/A	N/A	N/A	N/A	N/A	N/A	N/A
batch_insert_into_indexed_tree_40_depth_hash_ms	N/A	N/A	N/A	0.579	N/A	N/A	N/A	N/A	N/A	N/A	N/A

Miscellaneous

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

Metric	0 deployed contracts	1 deployed contracts
tx_size_in_bytes	16,827	40,014

Transaction processing duration by data writes.

Metric	0 new commitments	1 new commitments
tx_pxe_processing_time_ms	555	1,314

Metric	0 public data writes	1 public data writes
tx_sequencer_processing_time_ms	0.529 (+3%)	597

[dbanks12]

Looks mostly good. Pretty much just requesting a few comments!

[github-actions]

Changes to circuit sizes

Generated at commit: 81f8ca9142aafe6d9466ed291abfbbc9bf6472f0, compared to commit: d10774605f7c39175034e664b9dfcca8f7327106

🧾 Summary (100% most significant diffs)

Program	ACIR opcodes (+/-)	%	Circuit size (+/-)	%
private_kernel_inner	-16 ✅	-0.02%	-32 ✅	-0.01%
private_kernel_init	-16 ✅	-0.04%	-32 ✅	-0.01%
public_kernel_app_logic_simulated	-16 ✅	-1.84%	-32 ✅	-1.84%
public_kernel_setup_simulated	-16 ✅	-1.84%	-32 ✅	-1.84%
private_kernel_inner_simulated	-16 ✅	-1.96%	-32 ✅	-1.95%
private_kernel_init_simulated	-16 ✅	-1.96%	-32 ✅	-1.95%
public_kernel_setup	-2,379 ✅	-5.02%	-10,253 ✅	-4.05%
public_kernel_app_logic	-2,379 ✅	-5.02%	-10,253 ✅	-4.05%

---

Full diff report 👇

Program	ACIR opcodes (+/-)	%	Circuit size (+/-)	%
private_kernel_inner	93,263 (-16)	-0.02%	522,659 (-32)	-0.01%
private_kernel_init	36,330 (-16)	-0.04%	331,113 (-32)	-0.01%
public_kernel_app_logic_simulated	852 (-16)	-1.84%	1,707 (-32)	-1.84%
public_kernel_setup_simulated	852 (-16)	-1.84%	1,707 (-32)	-1.84%
private_kernel_inner_simulated	801 (-16)	-1.96%	1,605 (-32)	-1.95%
private_kernel_init_simulated	801 (-16)	-1.96%	1,605 (-32)	-1.95%
public_kernel_setup	44,982 (-2,379)	-5.02%	242,825 (-10,253)	-4.05%
public_kernel_app_logic	44,974 (-2,379)	-5.02%	242,817 (-10,253)	-4.05%