Function init_backend_with_config

pub fn init_backend_with_config(config: BackendConfig)

Responsible for initializing the C++ commitment backend and producing a specified number of generators.

This function initializes the backend so that computations can proceed either in the CPU or in the GPU. It is only called once. The backend type must be specified during build time using the following flags:

cargo build --features pip-cpu # only the Pippenger commitment CPU is used
cargo build --features naive-gpu # only the naive commitment GPU is used
cargo build --features naive-cpu # only the naive commitment CPU is used

Once the backend is initialized, it is not possible to change to another one. Therefore, if the GPU feature is specified during build time, then it is not possible to use the CPU to do the computations. In the case, no feature is specified during build time, the backend will initialize with the pip-cpu as default.

During this initialization process, the user can also specify a num_precomputed_generators value, which is used to pre-generate some generators. Those are later used in the commitment computation, preventing the generators from being created over and over again.

Any compute function will call this init_backend_with_precomputation securing that the backend is always in a proper state.

To guarantee that the code inside this function is not initialized multiple times, we use the std::sync::Once scheme.

Arguments

• num_precomputed_generators - The total number of generators to be precomputed. Those are used later during the commitment computation. Pre-computing may be beneficial, as it can save computational time.

Panics

If the backend initialization fails.

Example - Initializing the Backend with provided Configuration values

Backends need to be initialized before the commitment computation is called. You may want to call this function at the beginning of your program to prevent later initialization overhead. Specifying a config.num_precomputed_generators > 0 forces the config.num_precomputed_generators generators to be computed and stored at the CPU memory. Later, those are used with the commitment computation.

// Copyright 2023-present Space and Time Labs, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

use blitzar::compute::*;
use curve25519_dalek::ristretto::CompressedRistretto;

fn main() {
    /////////////////////////////////////////////
    // Initialize the backend with a custom configuration
    /////////////////////////////////////////////
    let num_precomputed_generators: u64 = 7;
    init_backend_with_config(BackendConfig {
        num_precomputed_generators,
    });

    /////////////////////////////////////////////
    // Define the data vectors that will be used in the computation
    // and do the actual commitment computation
    /////////////////////////////////////////////
    let data: &[u64] = &[40, 32, 21, 10, 20, 35, 444];
    let mut commitments = vec![CompressedRistretto::default(); 1];
    compute_curve25519_commitments(&mut commitments, &[data.into()], 0_u64);
    assert_ne!(commitments[0], CompressedRistretto::default())
}