refactor(bb): more namespaces under bb (#4348)

Co-authored-by: ludamad <adam@aztecprotocol.com>

barretenberg/cpp/src/barretenberg/benchmark/decrypt_bench/main.cpp

@@ -1,6 +1,8 @@
 #include "barretenberg/ecc/curves/grumpkin/grumpkin.hpp"
 #include <chrono>
 
+using namespace bb;
+
 int main(int, char**)
 {
 

barretenberg/cpp/src/barretenberg/benchmark/poseidon2_bench/poseidon2.bench.cpp

@@ -3,6 +3,7 @@
 #include <benchmark/benchmark.h>
 
 using namespace benchmark;
+using namespace bb;
 
 grumpkin::fq poseidon_function(const size_t count)
 {

barretenberg/cpp/src/barretenberg/common/parallel_for_atomic_pool.cpp

@@ -100,6 +100,7 @@ void ThreadPool::worker_loop(size_t /*unused*/)
 }
 } // namespace
 
+namespace bb {
 /**
  * A thread pooled strategy that uses atomics to prevent needing constantly lock on a queue.
  * The main thread acts as a worker also, and when it completes, it spins until thread workers are done.
@@ -112,3 +113,4 @@ void parallel_for_atomic_pool(size_t num_iterations, const std::function<void(si
     pool.start_tasks(num_iterations, func);
     // info("done");
 }
+} // namespace bb

barretenberg/cpp/src/barretenberg/common/parallel_for_moody.cpp

@@ -10,6 +10,7 @@
 #include <thread>
 #include <vector>
 
+namespace {
 class ThreadPool {
   public:
     ThreadPool(size_t num_threads)
@@ -85,7 +86,9 @@ class ThreadPool {
         }
     }
 };
+} // namespace
 
+namespace bb {
 /**
  * A Thread pooled strategy that uses a popular lock-free multiple-producer multiple-consume queue library by
  * "moodycamel" as the underlying mechanism to distribute work and join on completion.
@@ -97,3 +100,4 @@ void parallel_for_moody(size_t num_iterations, const std::function<void(size_t)>
 
     pool.start_tasks(func, num_iterations);
 }
+} // namespace bb

barretenberg/cpp/src/barretenberg/common/parallel_for_mutex_pool.cpp

@@ -115,6 +115,7 @@ void ThreadPool::worker_loop(size_t /*unused*/)
 }
 } // namespace
 
+namespace bb {
 /**
  * A thread pooled strategy that uses std::mutex for protection. Each worker increments the "iteration" and processes.
  * The main thread acts as a worker also, and when it completes, it spins until thread workers are done.
@@ -127,3 +128,4 @@ void parallel_for_mutex_pool(size_t num_iterations, const std::function<void(siz
     pool.start_tasks(num_iterations, func);
     // info("done");
 }
+} // namespace bb

barretenberg/cpp/src/barretenberg/common/parallel_for_omp.cpp

@@ -1,6 +1,7 @@
 #include <cstddef>
 #include <functional>
 
+namespace bb {
 void parallel_for_omp(size_t num_iterations, const std::function<void(size_t)>& func)
 {
 #ifndef NO_OMP_MULTITHREADING
@@ -10,3 +11,4 @@ void parallel_for_omp(size_t num_iterations, const std::function<void(size_t)>&
         func(i);
     }
 }
+} // namespace bb

barretenberg/cpp/src/barretenberg/common/parallel_for_queued.cpp

@@ -8,6 +8,7 @@
 #include <thread>
 #include <vector>
 
+namespace {
 class ThreadPool {
   public:
     ThreadPool(size_t num_threads);
@@ -100,7 +101,9 @@ void ThreadPool::worker_loop(size_t /*unused*/)
     }
     // info("worker exit ", worker_num);
 }
+} // namespace
 
+namespace bb {
 /**
  * A thread pooled strategey that assumed that thread pools would be more efficient than spawning threads.
  * Every iteration becomes a task in a queue. That's probably not very efficient.
@@ -120,3 +123,4 @@ void parallel_for_queued(size_t num_iterations, const std::function<void(size_t)
     pool.wait();
     // info("pool finished work");
 }
+} // namespace bb

barretenberg/cpp/src/barretenberg/common/parallel_for_spawning.cpp

@@ -1,6 +1,6 @@
 #include "thread.hpp"
-// #include "log.hpp"
 
+namespace bb {
 /**
  * A very simple strategy. Spawn a worker thread for every iteration (but no more than num cores).
  * Worker threads tight-loop incrementing an atomic variable from 0-num_iterations, until num_iterations reached.
@@ -40,3 +40,4 @@ void parallel_for_spawning(size_t num_iterations, const std::function<void(size_
     }
     // info("joined!\n\n");
 }
+} // namespace bb

barretenberg/cpp/src/barretenberg/common/thread.cpp

@@ -36,6 +36,7 @@
  * Haven't done deeper analysis. Defaulting to mutex_pool.
  */
 
+namespace bb {
 // 64 core aws r5.
 // pippenger run: pippenger_bench/1048576
 // coset_fft run: coset_fft_bench_parallel/4194304
@@ -223,4 +224,41 @@ void run_loop_in_parallel_if_effective_internal(size_t num_points,
 template void run_loop_in_parallel_if_effective_internal(
     size_t, const std::function<void(size_t, size_t)>&, size_t, size_t, size_t, size_t, size_t, size_t, size_t);
 template void run_loop_in_parallel_if_effective_internal(
-    size_t, const std::function<void(size_t, size_t, size_t)>&, size_t, size_t, size_t, size_t, size_t, size_t, size_t);
+    size_t, const std::function<void(size_t, size_t, size_t)>&, size_t, size_t, size_t, size_t, size_t, size_t, size_t);
+
+/**
+ * @brief calculates number of threads to create based on minimum iterations per thread
+ * @details Finds the number of cpus with get_num_cpus(), and calculates `desired_num_threads`
+ * Returns the min of `desired_num_threads` and `max_num_threads`.
+ * Note that it will not calculate a power of 2 necessarily, use `calculate_num_threads_pow2` instead
+ *
+ * @param num_iterations
+ * @param min_iterations_per_thread
+ * @return size_t
+ */
+size_t calculate_num_threads(size_t num_iterations, size_t min_iterations_per_thread)
+{
+    size_t max_num_threads = get_num_cpus(); // number of available threads
+    size_t desired_num_threads = num_iterations / min_iterations_per_thread;
+    size_t num_threads = std::min(desired_num_threads, max_num_threads); // fewer than max if justified
+    num_threads = num_threads > 0 ? num_threads : 1;                     // ensure num_threads is at least 1
+    return num_threads;
+}
+
+/**
+ * @brief calculates number of threads to create based on minimum iterations per thread, guaranteed power of 2
+ * @details Same functionality as `calculate_num_threads` but guaranteed power of 2
+ * @param num_iterations
+ * @param min_iterations_per_thread
+ * @return size_t
+ */
+size_t calculate_num_threads_pow2(size_t num_iterations, size_t min_iterations_per_thread)
+{
+    size_t max_num_threads = get_num_cpus_pow2(); // number of available threads (power of 2)
+    size_t desired_num_threads = num_iterations / min_iterations_per_thread;
+    desired_num_threads = static_cast<size_t>(1ULL << numeric::get_msb(desired_num_threads));
+    size_t num_threads = std::min(desired_num_threads, max_num_threads); // fewer than max if justified
+    num_threads = num_threads > 0 ? num_threads : 1;                     // ensure num_threads is at least 1
+    return num_threads;
+}
+} // namespace bb

barretenberg/cpp/src/barretenberg/common/thread.hpp

@@ -7,6 +7,8 @@
 #include <thread>
 #include <vector>
 
+namespace bb {
+
 inline size_t get_num_cpus()
 {
 #ifdef NO_MULTITHREADING
@@ -19,7 +21,7 @@ inline size_t get_num_cpus()
 // For algorithms that need to be divided amongst power of 2 threads.
 inline size_t get_num_cpus_pow2()
 {
-    return static_cast<size_t>(1ULL << bb::numeric::get_msb(get_num_cpus()));
+    return static_cast<size_t>(1ULL << numeric::get_msb(get_num_cpus()));
 }
 
 void parallel_for(size_t num_iterations, const std::function<void(size_t)>& func);
@@ -89,4 +91,30 @@ inline void run_loop_in_parallel_if_effective_with_index(size_t num_points,
                                                group_element_doublings_per_iteration,
                                                scalar_multiplications_per_iteration,
                                                sequential_copy_ops_per_iteration);
-}
+}
+
+const size_t DEFAULT_MIN_ITERS_PER_THREAD = 1 << 4;
+
+/**
+ * @brief calculates number of threads to create based on minimum iterations per thread
+ * @details Finds the number of cpus with get_num_cpus(), and calculates `desired_num_threads`
+ * Returns the min of `desired_num_threads` and `max_num_theads`.
+ * Note that it will not calculate a power of 2 necessarily, use `calculate_num_threads_pow2` instead
+ *
+ * @param num_iterations
+ * @param min_iterations_per_thread
+ * @return size_t
+ */
+size_t calculate_num_threads(size_t num_iterations, size_t min_iterations_per_thread = DEFAULT_MIN_ITERS_PER_THREAD);
+
+/**
+ * @brief calculates number of threads to create based on minimum iterations per thread, guaranteed power of 2
+ * @details Same functionality as `calculate_num_threads` but guaranteed power of 2
+ * @param num_iterations
+ * @param min_iterations_per_thread
+ * @return size_t
+ */
+size_t calculate_num_threads_pow2(size_t num_iterations,
+                                  size_t min_iterations_per_thread = DEFAULT_MIN_ITERS_PER_THREAD);
+
+} // namespace bb

barretenberg/cpp/src/barretenberg/crypto/blake2s/c_bind.cpp

@@ -4,8 +4,6 @@
 
 using namespace bb;
 
-extern "C" {
-
 WASM_EXPORT void blake2s(uint8_t const* data, out_buf32 out)
 {
     std::vector<uint8_t> inputv;
@@ -31,4 +29,3 @@ WASM_EXPORT void blake2s_to_field_(uint8_t const* data, fr::out_buf r)
     auto result = bb::fr::serialize_from_buffer(output.data());
     bb::fr::serialize_to_buffer(result, r);
 }
-}

barretenberg/cpp/src/barretenberg/crypto/blake2s/c_bind.hpp

@@ -3,11 +3,5 @@
 #include <cstddef>
 #include <cstdint>
 
-extern "C" {
-
-using namespace bb;
-
 WASM_EXPORT void blake2s(uint8_t const* data, out_buf32 r);
-
 WASM_EXPORT void blake2s_to_field_(uint8_t const* data, fr::out_buf r);
-}

barretenberg/cpp/src/barretenberg/crypto/ecdsa/c_bind.cpp

@@ -1,6 +1,9 @@
 #include "ecdsa.hpp"
 #include <barretenberg/ecc/curves/secp256k1/secp256k1.hpp>
 
+using namespace bb;
+using namespace bb::crypto;
+
 WASM_EXPORT void ecdsa__compute_public_key(uint8_t const* private_key, uint8_t* public_key_buf)
 {
     auto priv_key = from_buffer<secp256k1::fr>(private_key);
@@ -18,9 +21,9 @@ WASM_EXPORT void ecdsa__construct_signature(uint8_t const* message,
     using serialize::write;
     auto priv_key = from_buffer<secp256k1::fr>(private_key);
     secp256k1::g1::affine_element pub_key = secp256k1::g1::one * priv_key;
-    bb::crypto::ecdsa_key_pair<secp256k1::fr, secp256k1::g1> key_pair = { priv_key, pub_key };
+    ecdsa_key_pair<secp256k1::fr, secp256k1::g1> key_pair = { priv_key, pub_key };
 
-    auto sig = bb::crypto::ecdsa_construct_signature<Sha256Hasher, secp256k1::fq, secp256k1::fr, secp256k1::g1>(
+    auto sig = ecdsa_construct_signature<Sha256Hasher, secp256k1::fq, secp256k1::fr, secp256k1::g1>(
         std::string((char*)message, msg_len), key_pair);
     write(output_sig_r, sig.r);
     write(output_sig_s, sig.s);
@@ -39,10 +42,9 @@ WASM_EXPORT void ecdsa__recover_public_key_from_signature(uint8_t const* message
     std::copy(sig_s, sig_s + 32, s.begin());
     const uint8_t v = *sig_v;
 
-    bb::crypto::ecdsa_signature sig = { r, s, v };
-    auto recovered_pub_key =
-        bb::crypto::ecdsa_recover_public_key<Sha256Hasher, secp256k1::fq, secp256k1::fr, secp256k1::g1>(
-            std::string((char*)message, msg_len), sig);
+    ecdsa_signature sig = { r, s, v };
+    auto recovered_pub_key = ecdsa_recover_public_key<Sha256Hasher, secp256k1::fq, secp256k1::fr, secp256k1::g1>(
+        std::string((char*)message, msg_len), sig);
     serialize::write(output_pub_key, recovered_pub_key);
 }
 
@@ -59,7 +61,7 @@ WASM_EXPORT bool ecdsa__verify_signature(uint8_t const* message,
     std::copy(sig_s, sig_s + 32, s.begin());
     const uint8_t v = *sig_v;
 
-    bb::crypto::ecdsa_signature sig = { r, s, v };
-    return bb::crypto::ecdsa_verify_signature<Sha256Hasher, secp256k1::fq, secp256k1::fr, secp256k1::g1>(
+    ecdsa_signature sig = { r, s, v };
+    return ecdsa_verify_signature<Sha256Hasher, secp256k1::fq, secp256k1::fr, secp256k1::g1>(
         std::string((char*)message, msg_len), pubk, sig);
 }