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
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
// SPDX-License-Identifier: AGPL-3.0-or-later
//! Edge Convolution for Graph Neural Networks
//!
//! **Pure WGSL**: Single implementation via WebGPU shader
//! Learns edge features by aggregating neighbor information using CSR-format edges.
//!
//! Reference: "Dynamic Graph CNN for Learning on Point Clouds" by Wang et al. (2019)
use crate::device::{DeviceCapabilities, WorkloadType};
use crate::error::{BarracudaError, Result};
use crate::tensor::Tensor;
#[repr(C)]
#[derive(Copy, Clone, Debug, bytemuck::Pod, bytemuck::Zeroable)]
struct EdgeConvParams {
num_nodes: u32,
feature_dim: u32,
output_dim: u32,
num_edges: u32,
}
/// Edge convolution for graph neural networks (Dynamic Graph CNN).
pub struct EdgeConv {
node_features: Tensor,
/// CSR row offsets: [`num_nodes` + 1] entries
edge_offsets: Tensor,
/// CSR column indices: [`num_edges`] neighbor node indices
edge_targets: Tensor,
mlp_weight: Tensor,
mlp_bias: Tensor,
num_edges: u32,
}
impl EdgeConv {
/// Create `EdgeConv` operation with CSR-format edge storage
/// # Arguments
/// * `node_features` - Node features [`num_nodes`, `feature_dim`]
/// * `edge_offsets` - CSR row offsets [`num_nodes` + 1] (stored as f32, cast to u32 in shader)
/// * `edge_targets` - CSR column indices [`num_edges`] (stored as f32, cast to u32 in shader)
/// * `mlp_weight` - MLP weight matrix [`output_dim`, 2 * `feature_dim`]
/// * `mlp_bias` - MLP bias vector [`output_dim`]
///
/// # Errors
///
/// Returns [`Err`] if buffer allocation, GPU dispatch, or buffer
/// readback fails (e.g. device lost or out of memory).
pub fn new(
node_features: Tensor,
edge_offsets: Tensor,
edge_targets: Tensor,
mlp_weight: Tensor,
mlp_bias: Tensor,
) -> Result<Self> {
let num_edges = edge_targets.len() as u32;
Ok(Self {
node_features,
edge_offsets,
edge_targets,
mlp_weight,
mlp_bias,
num_edges,
})
}
/// WGSL shader source (embedded at compile time)
fn wgsl_shader() -> &'static str {
{
static SHADER: std::sync::LazyLock<String> = std::sync::LazyLock::new(|| {
include_str!("../shaders/gnn/edge_conv_f64.wgsl").to_string()
});
SHADER.as_str()
}
}
/// Execute `EdgeConv` on tensor
/// # Errors
/// Returns [`Err`] if buffer allocation, GPU dispatch, or buffer
/// readback fails (e.g. device lost or out of memory).
pub fn execute(self) -> Result<Tensor> {
let device = self.node_features.device();
let node_shape = self.node_features.shape();
if node_shape.len() != 2 {
return Err(BarracudaError::invalid_op(
"EdgeConv",
format!(
"node_features must be 2D [num_nodes, feature_dim], got shape {node_shape:?}"
),
));
}
let num_nodes = node_shape[0];
let feature_dim = node_shape[1];
let output_dim = self.mlp_bias.len();
// Create output buffer: [num_nodes, output_dim]
let output_size = num_nodes * output_dim;
let output_buffer = device.create_buffer_f32(output_size)?;
let params = EdgeConvParams {
num_nodes: num_nodes as u32,
feature_dim: feature_dim as u32,
output_dim: output_dim as u32,
num_edges: self.num_edges,
};
let params_buffer = device
.device
.create_buffer_init(&wgpu::util::BufferInitDescriptor {
label: Some("EdgeConv Params"),
contents: bytemuck::cast_slice(&[params]),
usage: wgpu::BufferUsages::UNIFORM,
});
// Create bind group layout (7 bindings to match evolved WGSL)
let bind_group_layout =
device
.device
.create_bind_group_layout(&wgpu::BindGroupLayoutDescriptor {
label: Some("EdgeConv Bind Group Layout"),
entries: &[
// binding 0: node_features (storage read)
wgpu::BindGroupLayoutEntry {
binding: 0,
visibility: wgpu::ShaderStages::COMPUTE,
ty: wgpu::BindingType::Buffer {
ty: wgpu::BufferBindingType::Storage { read_only: true },
has_dynamic_offset: false,
min_binding_size: None,
},
count: None,
},
// binding 1: edge_offsets (storage read)
wgpu::BindGroupLayoutEntry {
binding: 1,
visibility: wgpu::ShaderStages::COMPUTE,
ty: wgpu::BindingType::Buffer {
ty: wgpu::BufferBindingType::Storage { read_only: true },
has_dynamic_offset: false,
min_binding_size: None,
},
count: None,
},
// binding 2: edge_targets (storage read)
wgpu::BindGroupLayoutEntry {
binding: 2,
visibility: wgpu::ShaderStages::COMPUTE,
ty: wgpu::BindingType::Buffer {
ty: wgpu::BufferBindingType::Storage { read_only: true },
has_dynamic_offset: false,
min_binding_size: None,
},
count: None,
},
// binding 3: mlp_weight (storage read)
wgpu::BindGroupLayoutEntry {
binding: 3,
visibility: wgpu::ShaderStages::COMPUTE,
ty: wgpu::BindingType::Buffer {
ty: wgpu::BufferBindingType::Storage { read_only: true },
has_dynamic_offset: false,
min_binding_size: None,
},
count: None,
},
// binding 4: mlp_bias (storage read)
wgpu::BindGroupLayoutEntry {
binding: 4,
visibility: wgpu::ShaderStages::COMPUTE,
ty: wgpu::BindingType::Buffer {
ty: wgpu::BufferBindingType::Storage { read_only: true },
has_dynamic_offset: false,
min_binding_size: None,
},
count: None,
},
// binding 5: output (storage read_write)
wgpu::BindGroupLayoutEntry {
binding: 5,
visibility: wgpu::ShaderStages::COMPUTE,
ty: wgpu::BindingType::Buffer {
ty: wgpu::BufferBindingType::Storage { read_only: false },
has_dynamic_offset: false,
min_binding_size: None,
},
count: None,
},
// binding 6: params (uniform)
wgpu::BindGroupLayoutEntry {
binding: 6,
visibility: wgpu::ShaderStages::COMPUTE,
ty: wgpu::BindingType::Buffer {
ty: wgpu::BufferBindingType::Uniform,
has_dynamic_offset: false,
min_binding_size: None,
},
count: None,
},
],
});
// Create bind group
let bind_group = device.device.create_bind_group(&wgpu::BindGroupDescriptor {
label: Some("EdgeConv Bind Group"),
layout: &bind_group_layout,
entries: &[
wgpu::BindGroupEntry {
binding: 0,
resource: self.node_features.buffer().as_entire_binding(),
},
wgpu::BindGroupEntry {
binding: 1,
resource: self.edge_offsets.buffer().as_entire_binding(),
},
wgpu::BindGroupEntry {
binding: 2,
resource: self.edge_targets.buffer().as_entire_binding(),
},
wgpu::BindGroupEntry {
binding: 3,
resource: self.mlp_weight.buffer().as_entire_binding(),
},
wgpu::BindGroupEntry {
binding: 4,
resource: self.mlp_bias.buffer().as_entire_binding(),
},
wgpu::BindGroupEntry {
binding: 5,
resource: output_buffer.as_entire_binding(),
},
wgpu::BindGroupEntry {
binding: 6,
resource: params_buffer.as_entire_binding(),
},
],
});
// Compile shader
let shader = device.compile_shader(Self::wgsl_shader(), Some("EdgeConv"));
// Create pipeline
let pipeline_layout =
device
.device
.create_pipeline_layout(&wgpu::PipelineLayoutDescriptor {
label: Some("EdgeConv Pipeline Layout"),
bind_group_layouts: &[&bind_group_layout],
immediate_size: 0,
});
let pipeline = device
.device
.create_compute_pipeline(&wgpu::ComputePipelineDescriptor {
label: Some("EdgeConv Pipeline"),
layout: Some(&pipeline_layout),
module: &shader,
entry_point: Some("main"),
cache: None,
compilation_options: Default::default(),
});
// Encode and execute
let mut encoder = device.create_encoder_guarded(&wgpu::CommandEncoderDescriptor {
label: Some("EdgeConv Encoder"),
});
{
let mut pass = encoder.begin_compute_pass(&wgpu::ComputePassDescriptor {
label: Some("EdgeConv Pass"),
timestamp_writes: None,
});
pass.set_pipeline(&pipeline);
pass.set_bind_group(0, Some(&bind_group), &[]);
// Deep Debt Evolution: Capability-based dispatch
let caps = DeviceCapabilities::from_device(device);
let optimal_wg_size = caps.optimal_workgroup_size(WorkloadType::MatMul);
let workgroups = (num_nodes as u32).div_ceil(optimal_wg_size);
pass.dispatch_workgroups(workgroups, 1, 1);
}
device.submit_commands(Some(encoder.finish()));
// Create output tensor
Ok(Tensor::from_buffer(
output_buffer,
vec![num_nodes, output_dim],
device.clone(),
))
}
}
#[cfg(test)]
mod tests {
use super::*;
use crate::device::test_pool::get_test_device_if_f64_gpu_available;
#[tokio::test]
async fn test_edge_conv_basic() {
let Some(device) = get_test_device_if_f64_gpu_available().await else {
return;
};
let num_nodes = 5;
let feature_dim = 3;
let output_dim = 4;
let node_features = Tensor::from_vec_on(
vec![1.0; num_nodes * feature_dim],
vec![num_nodes, feature_dim],
device.clone(),
)
.await
.unwrap();
// CSR format: chain graph 0→1→2→3→4
// edge_offsets: [0, 1, 2, 3, 4, 4] (node 4 has no outgoing edges)
// edge_targets: [1, 2, 3, 4]
let edge_offsets = Tensor::from_vec_on(
vec![0.0, 1.0, 2.0, 3.0, 4.0, 4.0], // num_nodes + 1 entries
vec![num_nodes + 1],
device.clone(),
)
.await
.unwrap();
let edge_targets = Tensor::from_vec_on(
vec![1.0, 2.0, 3.0, 4.0], // 4 edges
vec![4],
device.clone(),
)
.await
.unwrap();
let mlp_weight = Tensor::from_vec_on(
vec![0.1; output_dim * 2 * feature_dim],
vec![output_dim, 2 * feature_dim],
device.clone(),
)
.await
.unwrap();
let mlp_bias = Tensor::from_vec_on(vec![0.0; output_dim], vec![output_dim], device.clone())
.await
.unwrap();
let result = EdgeConv::new(
node_features,
edge_offsets,
edge_targets,
mlp_weight,
mlp_bias,
)
.unwrap()
.execute()
.unwrap();
assert_eq!(result.shape(), &[num_nodes, output_dim]);
}
}