msm-webgpu 0.1.0

Multi-Scalar Multiplication (MSM) implementation for WebGPU.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39

const MAX_NUM_INVOCATIONS = 2048u;

struct MsmLen {
    val: u32,
}

@group(0) @binding(0)
var<storage, read_write> points: array<JacobianPoint>;
@group(0) @binding(1)
var<storage, read_write> scalars: array<ScalarField>;

@group(0) @binding(2)
var<storage, read_write> buckets: array<JacobianPoint, MAX_NUM_INVOCATIONS * TotalBuckets>;
@group(0) @binding(3)
var<storage, read_write> windows: array<JacobianPoint, MAX_NUM_INVOCATIONS * NumWindows>;

@group(0) @binding(4)
var<uniform> msm_len: MsmLen;

@compute @workgroup_size(1)
fn test_bucket_accumulation(
    @builtin(global_invocation_id) global_id: vec3<u32>,
    @builtin(local_invocation_id) local_id: vec3<u32>
) {
    let gidx = global_id.x;
    let lidx = local_id.x;

    bucket_accumulation_phase(gidx);
}

@compute @workgroup_size(1)
fn test_bucket_reduction(
    @builtin(global_invocation_id) global_id: vec3<u32>,
    @builtin(local_invocation_id) local_id: vec3<u32>
) {
    let gidx = global_id.x;
    let lidx = local_id.x;
    bucket_reduction_phase(gidx);
}