feat(bb): stack traces for check_circuit

[ludamad]

Overview:
If you are in a scenario where you have a failing call to check_circuit and wish to get more information out of it than just the gate index, you can use this feature to get a stack trace, see example below.

Usage instructions:

• On ubuntu (or our mainframe accounts) use sudo apt-get install libdw-dev to support trace printing
• Use cmake --preset clang16-dbg-fast-circuit-check-traces and cmake --build --preset clang16-dbg-fast-circuit-check-traces to enable the backward-cpp dependency through the CHECK_CIRCUIT_STACKTRACES CMake variable.
• Run any case where you have a failing check_circuit call, you will now have a stack trace illuminating where this constraint was added in code.

Caveats:

• This works best for code that is not overly generic, i.e. where just the sequence of function calls carries a lot of information. It is possible to tag extra data along with the stack trace, this can be done as a followup, please leave feedback if desired.
• There are certain functions like assert_equals that can cause gates that occur before them to fail. If this would be useful to automatically report, please leave feedback.

Example:

[ RUN      ] standard_circuit_constructor.test_check_circuit_broken
Stack trace (most recent call last):
#4    Source "_deps/gtest-src/googletest/src/gtest.cc", line 2845, in Run
       2842:   if (!Test::HasFatalFailure() && !Test::IsSkipped()) {
       2843:     // This doesn't throw as all user code that can throw are wrapped into
       2844:     // exception handling code.
      >2845:     test->Run();
       2846:   }
       2847: 
       2848:   if (test != nullptr) {
#3    Source "_deps/gtest-src/googletest/src/gtest.cc", line 2696, in Run
       2693:   // GTEST_SKIP().
       2694:   if (!HasFatalFailure() && !IsSkipped()) {
       2695:     impl->os_stack_trace_getter()->UponLeavingGTest();
      >2696:     internal::HandleExceptionsInMethodIfSupported(this, &Test::TestBody,
       2697:                                                   "the test body");
       2698:   }
#2  | Source "_deps/gtest-src/googletest/src/gtest.cc", line 2657, in HandleSehExceptionsInMethodIfSupported<testing::Test, void>
    |  2655: #if GTEST_HAS_EXCEPTIONS
    |  2656:     try {
    | >2657:       return HandleSehExceptionsInMethodIfSupported(object, method, location);
    |  2658:     } catch (const AssertionException&) {  // NOLINT
    |  2659:       // This failure was reported already.
      Source "_deps/gtest-src/googletest/src/gtest.cc", line 2621, in HandleExceptionsInMethodIfSupported<testing::Test, void>
       2618:   }
       2619: #else
       2620:   (void)location;
      >2621:   return (object->*method)();
       2622: #endif  // GTEST_HAS_SEH
       2623: }
#1    Source "/mnt/user-data/adam/aztec-packages/barretenberg/cpp/src/barretenberg/circuit_checker/standard_circuit_builder.test.cpp", line 464, in TestBody
        461:     uint32_t d_idx = circuit_constructor.add_variable(d);
        462:     circuit_constructor.create_add_gate({ a_idx, b_idx, c_idx, fr::one(), fr::one(), fr::neg_one(), fr::zero() });
        463: 
      > 464:     circuit_constructor.create_add_gate({ d_idx, c_idx, a_idx, fr::one(), fr::neg_one(), fr::neg_one(), fr::zero() });
        465: 
        466:     bool result = CircuitChecker::check(circuit_constructor);
        467:     EXPECT_EQ(result, false);
#0    Source "/mnt/user-data/adam/aztec-packages/barretenberg/cpp/src/barretenberg/stdlib_circuit_builders/standard_circuit_builder.cpp", line 22, in create_add_gate
         19: {
         20:     this->assert_valid_variables({ in.a, in.b, in.c });
         21: 
      >  22:     blocks.arithmetic.populate_wires(in.a, in.b, in.c);
         23:     blocks.arithmetic.q_m().emplace_back(FF::zero());
         24:     blocks.arithmetic.q_1().emplace_back(in.a_scaling);
         25:     blocks.arithmetic.q_2().emplace_back(in.b_scaling);
gate number4

[ludamad]

I just couldn't stand looking at bool == false :)

[Rumata888]

Snob :)

[AztecBot]

Benchmark results

Metrics with a significant change:

• proof_construction_time_poseidon_hash_100_ms (16): 934 (+29%)
• proof_construction_time_poseidon_hash_30_ms (16): 284 (+41%)
• proof_construction_time_sha256_ms (16): 923 (+31%)

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	5,711	1,554	⚠️ 923 (+31%)	855 (+13%)	817 (+6%)
proof_construction_time_sha256_30_ms	11,867 (+1%)	3,156	1,409	1,424	1,467
proof_construction_time_sha256_100_ms	44,682 (-1%)	11,742	5,427	5,431 (+1%)	5,367 (+1%)
proof_construction_time_poseidon_hash_ms	78.0	34.0	44.0 (+29%)	65.0 (+14%)	89.0 (+2%)
proof_construction_time_poseidon_hash_30_ms	1,514	419	⚠️ 284 (+41%)	263 (+14%)	289 (+7%)
proof_construction_time_poseidon_hash_100_ms	5,724	1,561	⚠️ 934 (+29%)	876 (+13%)	842 (+6%)

L2 block published to L1

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

Metric	4 txs	8 txs
l1_rollup_calldata_size_in_bytes	1,412	1,412
l1_rollup_calldata_gas	9,476	9,460
l1_rollup_execution_gas	608,012	607,996
l2_block_processing_time_in_ms	697 (+1%)	1,286 (-1%)
l2_block_building_time_in_ms	22,492 (+1%)	45,645
l2_block_rollup_simulation_time_in_ms	22,366 (+1%)	45,297
l2_block_public_tx_process_time_in_ms	19,214 (+1%)	41,889

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	5,916 (-1%)	8,455 (-1%)
node_database_size_in_bytes	9,510,992	12,812,368
pxe_database_size_in_bytes	7,774	15,127

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	136	498	12,243 (+2%)	20,634	64,614	89,536	2,731	524,288
private-kernel-inner	407	1,101 (+9%)	45,270	92,326	64,614	89,536	2,731	2,097,152
private-kernel-tail	577 (-1%)	2,819	46,720 (-2%)	96,545	68,261	11,648	297	2,097,152
base-parity	6.52 (-1%)	1,759	2,568 (-1%)	128	64.0	2,208	2.00	131,072
root-parity	51.0 (-4%)	49.4 (-1%)	34,323	27,100	64.0	2,720	18.0	2,097,152
base-rollup	6,003	2,334 (+1%)	65,509	119,738	756	3,648	47.0	4,194,304
root-rollup	111 (-1%)	63.4 (+4%)	19,579	25,309	620	3,456	41.0	1,048,576
public-kernel-app-logic	567	3,525 (+1%)	38,891 (+1%)	108,073	86,550	116,768	3,582	2,097,152
public-kernel-tail	1,137	23,509 (-2%)	154,660	403,238	7,646	11,648	297	8,388,608
private-kernel-reset-small	587	2,068 (-3%)	40,893 (+1%)	120,737	64,614	89,536	2,731	2,097,152
public-kernel-setup	661 (-1%)	2,682 (+1%)	37,622	108,073	86,550	116,768	3,582	2,097,152
public-kernel-teardown	565 (-1%)	3,535	39,445 (+1%)	108,073	86,550	116,768	3,582	2,097,152
merge-rollup	31.9 (+6%)	N/A	N/A	16,542	756	N/A	N/A	N/A
private-kernel-tail-to-public	N/A	8,689	86,322 (-1%)	N/A	N/A	116,768	3,582	4,194,304

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	9,944	462 (-2%)	N/A	N/A	N/A	N/A
ContractInstanceDeployer:deploy	1,408	9,944	42.5 (+1%)	N/A	N/A	N/A	N/A
MultiCallEntrypoint:entrypoint	1,920	9,944	1,808 (+1%)	N/A	N/A	N/A	N/A
SchnorrAccount:constructor	1,312	9,944	1,454	N/A	N/A	N/A	N/A
SchnorrAccount:entrypoint	2,304	9,944	2,828	16,768	50,421	2,097,152	457
Token:privately_mint_private_note	1,280	9,944	1,660 (+1%)	N/A	N/A	N/A	N/A
FPC:fee_entrypoint_public	1,344	9,944	1,133	16,768	10,075	524,288	457
Token:transfer	1,376	9,944	5,406	16,768	47,937 (-1%)	2,097,152	457
Benchmarking:create_note	1,344	9,944	1,392	N/A	N/A	N/A	N/A
SchnorrAccount:spend_private_authwit	1,280	9,944	77.1	N/A	N/A	N/A	N/A
Token:unshield	1,376	9,944	3,896 (-1%)	N/A	N/A	N/A	N/A
FPC:fee_entrypoint_private	1,376	9,944	4,801 (-1%)	N/A	N/A	N/A	N/A

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	32 leaves
batch_insert_into_append_only_tree_16_depth_ms	10.5	17.0	N/A	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_16_depth_hash_count	16.7	31.7	N/A	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_16_depth_hash_ms	0.610	0.523	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	48.7 (-1%)	76.0	133	246	474	N/A
batch_insert_into_append_only_tree_32_depth_hash_count	N/A	N/A	95.9	159	288	543	1,055	N/A
batch_insert_into_append_only_tree_32_depth_hash_ms	N/A	N/A	0.498	0.468	0.455	0.447	0.444	N/A
batch_insert_into_indexed_tree_20_depth_ms	N/A	N/A	58.6	112	183 (-1%)	351	629	N/A
batch_insert_into_indexed_tree_20_depth_hash_count	N/A	N/A	107	208	355	686	1,263	N/A
batch_insert_into_indexed_tree_20_depth_hash_ms	N/A	N/A	0.505	0.503	0.484 (-1%)	0.480	0.472 (+1%)	N/A
batch_insert_into_indexed_tree_40_depth_ms	N/A	N/A	N/A	N/A	N/A	N/A	N/A	58.9
batch_insert_into_indexed_tree_40_depth_hash_count	N/A	N/A	N/A	N/A	N/A	N/A	N/A	101
batch_insert_into_indexed_tree_40_depth_hash_ms	N/A	N/A	N/A	N/A	N/A	N/A	N/A	0.551

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	79,470	665,267

Transaction size based on fee payment method

| Metric | |
| - | |

[Rumata888]

Awesome