refactor: bye bye shared ptrs for ultra/goblin ultra proving_keys :) #5407

lucasxia01 · 2024-03-22T20:59:55Z

Partially addresses #4885.

Removes shared_ptr usage for ultra/goblin_ultra proving key.

Shouldn't affect the benchmarks much, but for some reason it seems to have decreased slightly.

--------------------------------------------------------------------------------
Benchmark                      Time             CPU   Iterations UserCounters...
--------------------------------------------------------------------------------
ClientIVCBench/Full/6      22099 ms        17172 ms            1 Decider::construct_proof=1 Decider::construct_proof(t)=754.682M ECCVMComposer::compute_commitment_key=1 ECCVMComposer::compute_commitment_key(t)=3.65834M ECCVMComposer::compute_witness=1 ECCVMComposer::compute_witness(t)=130.21M ECCVMComposer::create_prover=1 ECCVMComposer::create_prover(t)=149.963M ECCVMComposer::create_proving_key=1 ECCVMComposer::create_proving_key(t)=15.8747M ECCVMProver::construct_proof=1 ECCVMProver::construct_proof(t)=1.75623G Goblin::merge=11 Goblin::merge(t)=142.793M GoblinTranslatorCircuitBuilder::constructor=1 GoblinTranslatorCircuitBuilder::constructor(t)=58.4782M GoblinTranslatorProver=1 GoblinTranslatorProver(t)=144.954M GoblinTranslatorProver::construct_proof=1 GoblinTranslatorProver::construct_proof(t)=962.221M ProtoGalaxyProver_::accumulator_update_round=10 ProtoGalaxyProver_::accumulator_update_round(t)=294.093M ProtoGalaxyProver_::combiner_quotient_round=10 ProtoGalaxyProver_::combiner_quotient_round(t)=5.93984G ProtoGalaxyProver_::perturbator_round=10 ProtoGalaxyProver_::perturbator_round(t)=1.29776G ProtoGalaxyProver_::preparation_round=10 ProtoGalaxyProver_::preparation_round(t)=4.1465G ProtogalaxyProver::fold_instances=10 ProtogalaxyProver::fold_instances(t)=11.6782G ProverInstance(Circuit&)=11 ProverInstance(Circuit&)(t)=1.97873G batch_mul_with_endomorphism=30 batch_mul_with_endomorphism(t)=565.226M commit=426 commit(t)=4.01591G compute_combiner=10 compute_combiner(t)=5.93777G compute_perturbator=9 compute_perturbator(t)=1.29745G compute_univariate=48 compute_univariate(t)=1.42388G construct_circuits=6 construct_circuits(t)=4.44634G
Benchmarking lock deleted.
client_ivc_bench.json                                                            100% 3999   115.8KB/s   00:00    
function                                        ms     % sum
construct_circuits(t)                         4446    20.33%
ProverInstance(Circuit&)(t)                   1979     9.05%
ProtogalaxyProver::fold_instances(t)         11678    53.40%
Decider::construct_proof(t)                    755     3.45%
ECCVMComposer::create_prover(t)                150     0.69%
ECCVMProver::construct_proof(t)               1756     8.03%
GoblinTranslatorProver::construct_proof(t)     962     4.40%
Goblin::merge(t)                               143     0.65%

Total time accounted for: 21869ms/22099ms = 98.96%

Major contributors:
function                                        ms    % sum
commit(t)                                     4016   18.36%
compute_combiner(t)                           5938   27.15%
compute_perturbator(t)                        1297    5.93%
compute_univariate(t)                         1424    6.51%

Breakdown of ProtogalaxyProver::fold_instances:
ProtoGalaxyProver_::preparation_round(t)           4147    35.51%
ProtoGalaxyProver_::perturbator_round(t)           1298    11.11%
ProtoGalaxyProver_::combiner_quotient_round(t)     5940    50.86%
ProtoGalaxyProver_::accumulator_update_round(t)     294     2.52%

Please read contributing guidelines and remove this line.

…tion_parameters

…instance-2

…ded 900ms) >:(

…packages into lx/clean-shared-ptr

AztecBot · 2024-03-22T21:37:52Z

Benchmark results

Metrics with a significant change:

circuit_simulation_time_in_ms (private-kernel-inner): 412 (+79%)
circuit_simulation_time_in_ms (public-kernel-app-logic): 246 (+154%)
tx_pxe_processing_time_ms (0): 2,166 (+17%)
tx_pxe_processing_time_ms (1): 1,416 (+16%)
tx_sequencer_processing_time_ms (1): 858 (+46%)
l2_block_building_time_in_ms (8): 14,803 (+20%)
l2_block_building_time_in_ms (32): 54,884 (+21%)
l2_block_building_time_in_ms (64): 109,433 (+22%)
l2_block_public_tx_process_time_in_ms (8): 6,947 (+44%)
l2_block_public_tx_process_time_in_ms (32): 27,406 (+44%)
l2_block_public_tx_process_time_in_ms (64): 55,772 (+46%)
note_trial_decrypting_time_in_ms (64): 108 (+194%)

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 f9be2f2f 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	64 txs
l1_rollup_calldata_size_in_bytes	676	676	676
l1_rollup_calldata_gas	6,412	6,418	6,412
l1_rollup_execution_gas	585,745	585,751	585,745
l2_block_processing_time_in_ms	1,351 (+2%)	5,007 (+5%)	9,056 (-3%)
note_successful_decrypting_time_in_ms	192 (-16%)	553 (-15%)	978 (-3%)
note_trial_decrypting_time_in_ms	45.1 (+30%)	71.1 (+22%)	⚠️ 108 (+194%)
l2_block_building_time_in_ms	⚠️ 14,803 (+20%)	⚠️ 54,884 (+21%)	⚠️ 109,433 (+22%)
l2_block_rollup_simulation_time_in_ms	7,836 (+4%)	27,426 (+5%)	53,565 (+4%)
l2_block_public_tx_process_time_in_ms	⚠️ 6,947 (+44%)	⚠️ 27,406 (+44%)	⚠️ 55,772 (+46%)

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	14,232 (+1%)	26,184 (-2%)
note_history_successful_decrypting_time_in_ms	1,261 (-1%)	2,338 (-5%)
note_history_trial_decrypting_time_in_ms	93.0 (+4%)	120 (-19%)
node_database_size_in_bytes	18,653,264	35,027,024
pxe_database_size_in_bytes	29,859	59,414

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 (+15%)	44,378	28,214
private-kernel-ordering	190 (+11%)	52,880	14,296
base-parity	3,874 (+10%)	128	311
root-parity	1,366 (+8%)	1,244	311
base-rollup	16,417 (+9%)	165,760	861
root-rollup	49.9 (+1%)	4,359	725
private-kernel-inner	⚠️ 412 (+79%)	73,794	28,214
public-kernel-app-logic	⚠️ 246 (+154%)	35,251	28,217
public-kernel-tail	155 (+11%)	40,928	28,217
merge-rollup	9.22 (+3%)	2,568	861
public-kernel-teardown	92.5 (-1%)	35,252	28,217
public-kernel-setup	91.7	35,252	28,217

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	512 leaves	1024 leaves	2048 leaves	4096 leaves	32 leaves
batch_insert_into_append_only_tree_16_depth_ms	10.0 (-1%)	16.0 (-2%)	N/A	N/A	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_16_depth_hash_count	16.8	31.6	N/A	N/A	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_16_depth_hash_ms	0.583	0.494 (-2%)	N/A	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	46.2 (-2%)	71.4 (-1%)	233	440 (-2%)	998 (+14%)	1,713 (-1%)	N/A
batch_insert_into_append_only_tree_32_depth_hash_count	N/A	N/A	96.0	159	543	1,055	2,079	4,127	N/A
batch_insert_into_append_only_tree_32_depth_hash_ms	N/A	N/A	0.474 (-2%)	0.441 (-1%)	0.424	0.412 (-1%)	0.475 (+14%)	0.410 (-1%)	N/A
batch_insert_into_indexed_tree_20_depth_ms	N/A	N/A	54.0 (-1%)	105 (-2%)	335 (-1%)	651 (-2%)	1,329 (+1%)	2,602 (-1%)	N/A
batch_insert_into_indexed_tree_20_depth_hash_count	N/A	N/A	104 (-1%)	207	691	1,363	2,707	5,395	N/A
batch_insert_into_indexed_tree_20_depth_hash_ms	N/A	N/A	0.477 (-1%)	0.476 (-1%)	0.458	0.448 (-2%)	0.460 (+1%)	0.453 (-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	N/A	61.1
batch_insert_into_indexed_tree_40_depth_hash_count	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A	109
batch_insert_into_indexed_tree_40_depth_hash_ms	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A	0.534

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	14,967	495,185

Transaction size based on fee payment method

Metric	none fee payment method	native fee payment method	fpc_public fee payment method	fpc_private fee payment method
tx_with_fee_size_in_bytes	773	905	1,161	1,377

Transaction processing duration by data writes.

Metric	0 new note hashes	1 new note hashes	2 new note hashes
tx_pxe_processing_time_ms	⚠️ 2,166 (+17%)	⚠️ 1,416 (+16%)	5,749 (+1%)

Metric	0 public data writes	1 public data writes	2 public data writes	3 public data writes	4 public data writes	5 public data writes	8 public data writes
tx_sequencer_processing_time_ms	11.5 (+13%)	⚠️ 858 (+46%)	442 (-2%)	1,173 (-1%)	579 (-4%)	2,031 (-1%)	563 (-1%)

lucasxia01 · 2024-03-22T21:43:23Z

barretenberg/cpp/src/barretenberg/flavor/flavor.hpp

@@ -165,19 +165,6 @@ class VerificationKey_ : public PrecomputedCommitments {
        this->log_circuit_size = numeric::get_msb(circuit_size);
        this->num_public_inputs = num_public_inputs;
    };
-
-    template <typename ProvingKeyPtr> VerificationKey_(const ProvingKeyPtr& proving_key)


this function with a provingkeyptr template is pretty bad

Instead, I put them in the specific flavor classes

lucasxia01 · 2024-03-22T21:45:34Z

barretenberg/cpp/src/barretenberg/flavor/goblin_ultra.hpp

@@ -452,6 +471,7 @@ class GoblinUltraFlavor {
     */
    class ProverPolynomials : public AllEntities<Polynomial> {
      public:
+        // TODO(https://github.com/AztecProtocol/barretenberg/issues/925), proving_key could be const ref


extra todo added from prev PR

codygunton

Approved conditional on responding to the comments, in particular to the one relating to reference type data member.

codygunton · 2024-03-25T18:24:11Z

barretenberg/cpp/src/barretenberg/execution_trace/execution_trace.hpp

@@ -54,7 +54,7 @@ template <class Flavor> class ExecutionTrace_ {
     */
    static void add_wires_and_selectors_to_proving_key(TraceData& trace_data,
                                                       Builder& builder,
-                                                       const std::shared_ptr<typename Flavor::ProvingKey>& proving_key);
+                                                       typename Flavor::ProvingKey& proving_key);


You can use the alias here rather than this heavier notation.

codygunton · 2024-03-25T18:25:53Z

barretenberg/cpp/src/barretenberg/plonk/composer/standard_composer.cpp

@@ -41,7 +41,7 @@ std::shared_ptr<plonk::proving_key> StandardComposer::compute_proving_key(Circui
        subgroup_size, circuit_constructor.public_inputs.size(), crs, CircuitType::STANDARD);

    // Construct and add to proving key the wire, selector and copy constraint polynomials
-    Trace::populate(circuit_constructor, circuit_proving_key);
+    Trace::populate(circuit_constructor, *circuit_proving_key);


I guess it would have spiraled here to handle not through a pointer here? Seems reasonable to leave as a pointer in that case since Plonk is on its way out.

codygunton · 2024-03-25T18:49:52Z

barretenberg/cpp/src/barretenberg/ultra_honk/oink_prover.hpp

@@ -43,7 +43,7 @@ template <IsUltraFlavor Flavor> class OinkProver {
    using FF = typename Flavor::FF;

  public:
-    std::shared_ptr<ProvingKey> proving_key;
+    ProvingKey& proving_key;


Reference memmbers are the devil! Was this intentional?

Please change or try to convince me this is the way.

you're right that this is not great. I think changing to unique_ptr might be better? I used a reference here because OinkProver's lifetime is basically in the call to prove().

lucasxia01 · 2024-03-26T18:22:55Z

barretenberg/cpp/src/barretenberg/ultra_honk/oink_prover.cpp

@@ -26,6 +26,7 @@ template <IsUltraFlavor Flavor> OinkProverOutput<Flavor> OinkProver<Flavor>::pro
    execute_grand_product_computation_round();

    return OinkProverOutput<Flavor>{
+        .proving_key = std::move(proving_key),


not sure if this one is necessary too. Do we need to do 2 std::moves() in order to avoid copying when returning?

Hm, my sense is it's better to have these though perhaps the compiler would already move? So yeah, also not sure.

…s temporarily

🤖 I have created a release *beep* *boop* --- <details><summary>aztec-package: 0.32.1</summary> ## [0.32.1](aztec-package-v0.32.0...aztec-package-v0.32.1) (2024-04-02) ### Miscellaneous * Explicit type imports ([#5519](#5519)) ([2a217de](2a217de)) </details> <details><summary>barretenberg.js: 0.32.1</summary> ## [0.32.1](barretenberg.js-v0.32.0...barretenberg.js-v0.32.1) (2024-04-02) ### Features * Earthly split runners, structure reverts ([#5524](#5524)) ([fcb8787](fcb8787)) ### Bug Fixes * **ci:** Turn on earthly for everyone ([#5423](#5423)) ([bea3fcb](bea3fcb)) </details> <details><summary>aztec-cli: 0.32.1</summary> ## [0.32.1](aztec-cli-v0.32.0...aztec-cli-v0.32.1) (2024-04-02) ### Miscellaneous * Explicit type imports ([#5519](#5519)) ([2a217de](2a217de)) </details> <details><summary>aztec-packages: 0.32.1</summary> ## [0.32.1](aztec-packages-v0.32.0...aztec-packages-v0.32.1) (2024-04-02) ### Features * **acir_gen:** Fold attribute at compile-time and initial non inlined ACIR ([#5341](#5341)) ([a979150](a979150)) * **acvm:** Execute multiple circuits ([#5380](#5380)) ([bb71920](bb71920)) * Dont double check num bits in brillig vm ([#5489](#5489)) ([a18288d](a18288d)) * Earthly split runners, structure reverts ([#5524](#5524)) ([fcb8787](fcb8787)) * Parallel gtest ([#5498](#5498)) ([349ea59](349ea59)) ### Bug Fixes * **ci:** Turn on earthly for everyone ([#5423](#5423)) ([bea3fcb](bea3fcb)) * Cpp cache and add other e2e ([#5512](#5512)) ([4118bcd](4118bcd)) * Require noir-packages-test to finish ([#5505](#5505)) ([191f0df](191f0df)) * Univariate evals not set in ECCVM prover ([#5529](#5529)) ([f9a2b7c](f9a2b7c)) ### Miscellaneous * Add goblin ops in add_gates_to_ensure_all_polys_are_non_zero ([#5468](#5468)) ([b9041e4](b9041e4)) * **avm:** Add 15 additional 16-bit registers in ALU trace of AVM circuit ([#5503](#5503)) ([8725c39](8725c39)) * **avm:** Migrate memory data structure in AVM circuit to unordered map ([#5506](#5506)) ([ccd09aa](ccd09aa)) * Build contracts and protocol circuits sequentially if not enough ram ([#5499](#5499)) ([ea072b6](ea072b6)) * Bye bye shared ptrs for ultra/goblin ultra proving_keys :) ([#5407](#5407)) ([b94d0db](b94d0db)) * Clean up compute_next_accumulator ([#5516](#5516)) ([f9be2f2](f9be2f2)) * Explicit type imports ([#5519](#5519)) ([2a217de](2a217de)) * Improve caching in noir Earthfile ([#5513](#5513)) ([5d1fb44](5d1fb44)) * Inject fetcher instead of using global ([#5502](#5502)) ([a066544](a066544)) * Make get notes return all notes at beginning of array [#4991](#4991) ([#5321](#5321)) ([5c5b627](5c5b627)) * Move alphas generation to oink ([#5515](#5515)) ([3b964f3](3b964f3)) * Replace relative paths to noir-protocol-circuits ([a689e4e](a689e4e)) * Replace relative paths to noir-protocol-circuits ([db1bab5](db1bab5)) * Replace relative paths to noir-protocol-circuits ([b2ab64b](b2ab64b)) * Replace relative paths to noir-protocol-circuits ([1f468db](1f468db)) * Run nargo format for noir-projects ([#5483](#5483)) ([277168f](277168f)) </details> <details><summary>barretenberg: 0.32.1</summary> ## [0.32.1](barretenberg-v0.32.0...barretenberg-v0.32.1) (2024-04-02) ### Features * **acvm:** Execute multiple circuits ([#5380](#5380)) ([bb71920](bb71920)) * Earthly split runners, structure reverts ([#5524](#5524)) ([fcb8787](fcb8787)) * Parallel gtest ([#5498](#5498)) ([349ea59](349ea59)) ### Bug Fixes * **ci:** Turn on earthly for everyone ([#5423](#5423)) ([bea3fcb](bea3fcb)) * Cpp cache and add other e2e ([#5512](#5512)) ([4118bcd](4118bcd)) * Univariate evals not set in ECCVM prover ([#5529](#5529)) ([f9a2b7c](f9a2b7c)) ### Miscellaneous * Add goblin ops in add_gates_to_ensure_all_polys_are_non_zero ([#5468](#5468)) ([b9041e4](b9041e4)) * **avm:** Add 15 additional 16-bit registers in ALU trace of AVM circuit ([#5503](#5503)) ([8725c39](8725c39)) * **avm:** Migrate memory data structure in AVM circuit to unordered map ([#5506](#5506)) ([ccd09aa](ccd09aa)) * Bye bye shared ptrs for ultra/goblin ultra proving_keys :) ([#5407](#5407)) ([b94d0db](b94d0db)) * Clean up compute_next_accumulator ([#5516](#5516)) ([f9be2f2](f9be2f2)) * Move alphas generation to oink ([#5515](#5515)) ([3b964f3](3b964f3)) </details> --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please).

🤖 I have created a release *beep* *boop* --- <details><summary>aztec-package: 0.32.1</summary> ## [0.32.1](AztecProtocol/aztec-packages@aztec-package-v0.32.0...aztec-package-v0.32.1) (2024-04-02) ### Miscellaneous * Explicit type imports ([#5519](AztecProtocol/aztec-packages#5519)) ([2a217de](AztecProtocol/aztec-packages@2a217de)) </details> <details><summary>barretenberg.js: 0.32.1</summary> ## [0.32.1](AztecProtocol/aztec-packages@barretenberg.js-v0.32.0...barretenberg.js-v0.32.1) (2024-04-02) ### Features * Earthly split runners, structure reverts ([#5524](AztecProtocol/aztec-packages#5524)) ([fcb8787](AztecProtocol/aztec-packages@fcb8787)) ### Bug Fixes * **ci:** Turn on earthly for everyone ([#5423](AztecProtocol/aztec-packages#5423)) ([bea3fcb](AztecProtocol/aztec-packages@bea3fcb)) </details> <details><summary>aztec-cli: 0.32.1</summary> ## [0.32.1](AztecProtocol/aztec-packages@aztec-cli-v0.32.0...aztec-cli-v0.32.1) (2024-04-02) ### Miscellaneous * Explicit type imports ([#5519](AztecProtocol/aztec-packages#5519)) ([2a217de](AztecProtocol/aztec-packages@2a217de)) </details> <details><summary>aztec-packages: 0.32.1</summary> ## [0.32.1](AztecProtocol/aztec-packages@aztec-packages-v0.32.0...aztec-packages-v0.32.1) (2024-04-02) ### Features * **acir_gen:** Fold attribute at compile-time and initial non inlined ACIR ([#5341](AztecProtocol/aztec-packages#5341)) ([a979150](AztecProtocol/aztec-packages@a979150)) * **acvm:** Execute multiple circuits ([#5380](AztecProtocol/aztec-packages#5380)) ([bb71920](AztecProtocol/aztec-packages@bb71920)) * Dont double check num bits in brillig vm ([#5489](AztecProtocol/aztec-packages#5489)) ([a18288d](AztecProtocol/aztec-packages@a18288d)) * Earthly split runners, structure reverts ([#5524](AztecProtocol/aztec-packages#5524)) ([fcb8787](AztecProtocol/aztec-packages@fcb8787)) * Parallel gtest ([#5498](AztecProtocol/aztec-packages#5498)) ([349ea59](AztecProtocol/aztec-packages@349ea59)) ### Bug Fixes * **ci:** Turn on earthly for everyone ([#5423](AztecProtocol/aztec-packages#5423)) ([bea3fcb](AztecProtocol/aztec-packages@bea3fcb)) * Cpp cache and add other e2e ([#5512](AztecProtocol/aztec-packages#5512)) ([4118bcd](AztecProtocol/aztec-packages@4118bcd)) * Require noir-packages-test to finish ([#5505](AztecProtocol/aztec-packages#5505)) ([191f0df](AztecProtocol/aztec-packages@191f0df)) * Univariate evals not set in ECCVM prover ([#5529](AztecProtocol/aztec-packages#5529)) ([f9a2b7c](AztecProtocol/aztec-packages@f9a2b7c)) ### Miscellaneous * Add goblin ops in add_gates_to_ensure_all_polys_are_non_zero ([#5468](AztecProtocol/aztec-packages#5468)) ([b9041e4](AztecProtocol/aztec-packages@b9041e4)) * **avm:** Add 15 additional 16-bit registers in ALU trace of AVM circuit ([#5503](AztecProtocol/aztec-packages#5503)) ([8725c39](AztecProtocol/aztec-packages@8725c39)) * **avm:** Migrate memory data structure in AVM circuit to unordered map ([#5506](AztecProtocol/aztec-packages#5506)) ([ccd09aa](AztecProtocol/aztec-packages@ccd09aa)) * Build contracts and protocol circuits sequentially if not enough ram ([#5499](AztecProtocol/aztec-packages#5499)) ([ea072b6](AztecProtocol/aztec-packages@ea072b6)) * Bye bye shared ptrs for ultra/goblin ultra proving_keys :) ([#5407](AztecProtocol/aztec-packages#5407)) ([b94d0db](AztecProtocol/aztec-packages@b94d0db)) * Clean up compute_next_accumulator ([#5516](AztecProtocol/aztec-packages#5516)) ([f9be2f2](AztecProtocol/aztec-packages@f9be2f2)) * Explicit type imports ([#5519](AztecProtocol/aztec-packages#5519)) ([2a217de](AztecProtocol/aztec-packages@2a217de)) * Improve caching in noir Earthfile ([#5513](AztecProtocol/aztec-packages#5513)) ([5d1fb44](AztecProtocol/aztec-packages@5d1fb44)) * Inject fetcher instead of using global ([#5502](AztecProtocol/aztec-packages#5502)) ([a066544](AztecProtocol/aztec-packages@a066544)) * Make get notes return all notes at beginning of array [#4991](AztecProtocol/aztec-packages#4991) ([#5321](AztecProtocol/aztec-packages#5321)) ([5c5b627](AztecProtocol/aztec-packages@5c5b627)) * Move alphas generation to oink ([#5515](AztecProtocol/aztec-packages#5515)) ([3b964f3](AztecProtocol/aztec-packages@3b964f3)) * Replace relative paths to noir-protocol-circuits ([a689e4e](AztecProtocol/aztec-packages@a689e4e)) * Replace relative paths to noir-protocol-circuits ([db1bab5](AztecProtocol/aztec-packages@db1bab5)) * Replace relative paths to noir-protocol-circuits ([b2ab64b](AztecProtocol/aztec-packages@b2ab64b)) * Replace relative paths to noir-protocol-circuits ([1f468db](AztecProtocol/aztec-packages@1f468db)) * Run nargo format for noir-projects ([#5483](AztecProtocol/aztec-packages#5483)) ([277168f](AztecProtocol/aztec-packages@277168f)) </details> <details><summary>barretenberg: 0.32.1</summary> ## [0.32.1](AztecProtocol/aztec-packages@barretenberg-v0.32.0...barretenberg-v0.32.1) (2024-04-02) ### Features * **acvm:** Execute multiple circuits ([#5380](AztecProtocol/aztec-packages#5380)) ([bb71920](AztecProtocol/aztec-packages@bb71920)) * Earthly split runners, structure reverts ([#5524](AztecProtocol/aztec-packages#5524)) ([fcb8787](AztecProtocol/aztec-packages@fcb8787)) * Parallel gtest ([#5498](AztecProtocol/aztec-packages#5498)) ([349ea59](AztecProtocol/aztec-packages@349ea59)) ### Bug Fixes * **ci:** Turn on earthly for everyone ([#5423](AztecProtocol/aztec-packages#5423)) ([bea3fcb](AztecProtocol/aztec-packages@bea3fcb)) * Cpp cache and add other e2e ([#5512](AztecProtocol/aztec-packages#5512)) ([4118bcd](AztecProtocol/aztec-packages@4118bcd)) * Univariate evals not set in ECCVM prover ([#5529](AztecProtocol/aztec-packages#5529)) ([f9a2b7c](AztecProtocol/aztec-packages@f9a2b7c)) ### Miscellaneous * Add goblin ops in add_gates_to_ensure_all_polys_are_non_zero ([#5468](AztecProtocol/aztec-packages#5468)) ([b9041e4](AztecProtocol/aztec-packages@b9041e4)) * **avm:** Add 15 additional 16-bit registers in ALU trace of AVM circuit ([#5503](AztecProtocol/aztec-packages#5503)) ([8725c39](AztecProtocol/aztec-packages@8725c39)) * **avm:** Migrate memory data structure in AVM circuit to unordered map ([#5506](AztecProtocol/aztec-packages#5506)) ([ccd09aa](AztecProtocol/aztec-packages@ccd09aa)) * Bye bye shared ptrs for ultra/goblin ultra proving_keys :) ([#5407](AztecProtocol/aztec-packages#5407)) ([b94d0db](AztecProtocol/aztec-packages@b94d0db)) * Clean up compute_next_accumulator ([#5516](AztecProtocol/aztec-packages#5516)) ([f9be2f2](AztecProtocol/aztec-packages@f9be2f2)) * Move alphas generation to oink ([#5515](AztecProtocol/aztec-packages#5515)) ([3b964f3](AztecProtocol/aztec-packages@3b964f3)) </details> --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please).

lucasxia01 added 21 commits March 19, 2024 15:46

moved sorted_polynomials to proving key

d132a10

make prover polys constructor from proving key

eec17c6

eliminated initialize_prover_polynomials

45c5ee2

moved compute_sorted_list_accumulator to proving_key

810d369

test fix

e76c972

moved compute_sorted_accumulator_polynomials and made updates to rela…

6078832

…tion_parameters

test fix (prover polys wasn't being updated correctly)

b0855d5

add OinkProverOutput and make one prove() function

f4aa979

proverpolynomials now is just a temporary object during oink

1f4e307

Merge remote-tracking branch 'origin/master' into lx/refactor-prover-…

3fe93cc

…instance-2

moved compute_logderivative_inverse() from instance to proving_key

2798eaf

moved compute_grand_product_polynomials from instance to proving key

4030744

removed the last instance reference that is non instance->proving_key!

e11bb7e

oink says bye bye to instance 👋

b4c4354

small fixes to tests

2b67229

removed stupid shared_ptr calling copy constructor of proving_key (ad…

abd0bd7

…ded 900ms) >:(

deleting copy constructor of proving key to prevent copying

8923c8c

updated based on comments

c425e3d

bye bye shared ptrs for ultra/goblin ultra proving_keys :)

2b6aae9

bye bye shared ptrs for ultra/goblin ultra proving_keys :)

2d019cd

Merge branch 'lx/clean-shared-ptr' of github.com:AztecProtocol/aztec-…

994c767

…packages into lx/clean-shared-ptr

lucasxia01 self-assigned this Mar 22, 2024

lucasxia01 marked this pull request as draft March 22, 2024 21:05

fixed pg prover (shouldn't move too early)

de50495

remove comments

c7192bc

lucasxia01 commented Mar 22, 2024

View reviewed changes

lucasxia01 added 2 commits March 22, 2024 22:09

proving key gets copied instead of shared

7e27f9c

Merge remote-tracking branch 'origin/master' into lx/clean-shared-ptr

70ff641

lucasxia01 marked this pull request as ready for review March 25, 2024 15:01

Merge remote-tracking branch 'origin/master' into lx/clean-shared-ptr

6397673

codygunton self-requested a review March 25, 2024 18:22

codygunton approved these changes Mar 25, 2024

View reviewed changes

lucasxia01 added 2 commits March 26, 2024 18:14

Merge remote-tracking branch 'origin/master' into lx/clean-shared-ptr

d1a4965

get rid of reference member var and use std::move()s instead

03238e8

lucasxia01 commented Mar 26, 2024

View reviewed changes

lucasxia01 added 4 commits March 26, 2024 19:50

fix for pg tests

35ae332

removing oink prover as a member variable of ultra_prover to fix test…

7b4a20e

…s temporarily

Merge remote-tracking branch 'origin/master' into lx/clean-shared-ptr

7f79def

Merge remote-tracking branch 'origin/master' into lx/clean-shared-ptr

bf9a7f7

lucasxia01 merged commit b94d0db into master Apr 1, 2024
148 of 150 checks passed

lucasxia01 deleted the lx/clean-shared-ptr branch April 1, 2024 19:05

AztecBot mentioned this pull request Apr 1, 2024

chore(master): Release 0.32.1 #5484

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

refactor: bye bye shared ptrs for ultra/goblin ultra proving_keys :) #5407

refactor: bye bye shared ptrs for ultra/goblin ultra proving_keys :) #5407

lucasxia01 commented Mar 22, 2024 •

edited

Loading

AztecBot commented Mar 22, 2024 •

edited

Loading

L2 block published to L1

L2 chain processing

Circuits stats

Tree insertion stats

Miscellaneous

lucasxia01 Mar 22, 2024

lucasxia01 Mar 22, 2024

lucasxia01 Mar 22, 2024

codygunton left a comment

codygunton Mar 25, 2024

codygunton Mar 25, 2024

lucasxia01 Mar 26, 2024

codygunton Mar 25, 2024

codygunton Mar 25, 2024

lucasxia01 Mar 25, 2024

lucasxia01 Mar 26, 2024

codygunton Apr 1, 2024

refactor: bye bye shared ptrs for ultra/goblin ultra proving_keys :) #5407

refactor: bye bye shared ptrs for ultra/goblin ultra proving_keys :) #5407

Conversation

lucasxia01 commented Mar 22, 2024 • edited Loading

AztecBot commented Mar 22, 2024 • edited Loading

Benchmark results

L2 block published to L1

L2 chain processing

Circuits stats

Tree insertion stats

Miscellaneous

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

codygunton left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

lucasxia01 commented Mar 22, 2024 •

edited

Loading

AztecBot commented Mar 22, 2024 •

edited

Loading