logos-storage/proof-aggregation
1
0
Fork 0
mirror of https://github.com/logos-storage/proof-aggregation.git synced 2026-01-02 13:53:13 +00:00
Code Issues Packages Projects Releases Wiki Activity

add utils description.

Browse Source
This commit is contained in:
M Alghazwi 2025-07-09 18:15:50 +02:00
parent 1723b182c7
commit 29a1a2fbb1
No known key found for this signature in database
GPG Key ID: 646E567CAD7DB607
1 changed files with 15 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
codex-plonky2-circuits/src/recursion/utils.rs
View File

@ -17,7 +17,8 @@ use crate::recursion::leaf::BUCKET_SIZE;
/// - `r` is in the range [0, bucket_size),
/// - `q` is in the range [0, num_buckets),
///
/// requires that the total range T = (bucket_size * num_buckets) is a power of 2.
/// requires that the total range T = (bucket_size * num_buckets) is a power of 2 (and so is bucket_size and num_buckets).
/// Assumes that `index` is in the range [0, T), range-checks `index` before calling this function.
pub fn split_index<
F: RichField + Extendable<D> + Poseidon2,
const D: usize,
@ -61,7 +62,9 @@ pub fn split_index<
Ok((q_val, r_val))
}
/// A helper that computes 2^r for a target r in [0, 32) using selection over 32 constants.
/// A helper that computes 2^r for a target r in [0, `BUCKET_SIZE`) using selection over `BUCKET_SIZE` constants.
/// assumes that r is in the range [0, `BUCKET_SIZE`), range-checks r before calling this function
/// if `r` is taken from `split_index` then it is already in the correct range
pub fn compute_power_of_two<
F: RichField + Extendable<D> + Poseidon2,
const D: usize,
@ -70,8 +73,6 @@ pub fn compute_power_of_two<
r: Target,
) -> crate::Result<Target>
{
// First range-check r so it is in [0, 32).
builder.range_check(r, BUCKET_SIZE);
let mut result = builder.zero();
for i in 0..BUCKET_SIZE {
let i_const = builder.constant(F::from_canonical_u64(i as u64));
@ -84,14 +85,18 @@ pub fn compute_power_of_two<
Ok(result)
}
/// Computes the flag buckets from a given index and flag.
/// Computes the flag buckets from a given index and flag (In-Circuit).
///
/// Given:
/// - `index` is a Target representing a number in T = [0, bucket_size * num_buckets),
/// - `flag` is a BoolTarget (true if the proof is real, false if dummy),
/// - `bucket_size` (e.g. 32 for Goldilocks) and `num_buckets` (e.g. 4 to fit 128 proofs),
/// this function returns a vector of Targets representing the computed flag buckets.
/// For bucket i, the value is:
/// - `bucket_size` is the number of flags per bucket (e.g. 32 for Goldilocks)
/// - `num_buckets` is the number of buckets (e.g. 4 to fit 128 proofs)
/// this function returns a vector of `num_buckets` Targets representing the computed flag buckets.
/// the flag buckets should contain zeroes everywhere except for the bucket that contains the flag.
///
/// The logic of this mini-circuit is as follows:
/// For bucket i in [0, num_buckets), the value is:
/// - flag * 2^(r) if i is the selected bucket (i.e. i == q), where (q, r) = split_index(index),
/// - 0 otherwise.
pub fn compute_flag_buckets<
@ -133,11 +138,12 @@ pub fn compute_flag_buckets<
/// Returns the number of buckets required to hold `t` flags,
/// where each bucket can hold up to BUCKET_SIZE flags.
/// bucket_count = ceil(t / BUCKET_SIZE)
pub fn bucket_count(t: usize) -> usize {
(t + BUCKET_SIZE -1) / BUCKET_SIZE
}
/// helper fn to generate hash of verifier data
/// helper fn to generate hash of verifier data (outside the circuit)
pub fn get_hash_of_verifier_data<
F: RichField + Extendable<D> + Poseidon2,
const D: usize,