adrien/zig-wgpu
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Fixed binding size limit error by chunking to max 1GB binding

Browse Source
This commit is contained in:
adrien 2026-05-17 17:52:07 +02:00
parent cfc1069309
commit a381c71550
2 changed files with 33 additions and 53 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

79
src/Mat.zig
View File

@ -40,7 +40,7 @@ pub fn zeros(gloc: *GpuAllocator, rows: usize, cols: usize) !Mat {
}
pub fn deinit(self: Mat) void {
self.buf.deinit(); // Automatically cleans tracking map & releases GPU memory
self.buf.deinit();
}
pub fn len(self: Mat) usize {
@ -58,34 +58,7 @@ pub fn add(self: Mat, gloc: *GpuAllocator, other: Mat) !Mat {
errdefer result.deinit();
const pipeline = try gloc.pipAdd();
try dispatch2in1out(gloc, pipeline, self.buf, other.buf, result.buf, self.byteSize(), self.len());
return result;
}
pub fn scale(self: Mat, gloc: *GpuAllocator, scalar: f32) !Mat {
const result = try Mat.zeros(gloc, self.rows, self.cols);
errdefer result.deinit();
const bytes = self.byteSize();
const n = self.len();
const uni_buf = try GpuBuffer.init(
gloc,
@sizeOf(f32),
c.WGPUBufferUsage_Uniform | c.WGPUBufferUsage_CopyDst,
);
defer uni_buf.deinit(); // Gracefully deinitializes locally
c.wgpuQueueWriteBuffer(gloc.queue, uni_buf.raw, 0, &scalar, @sizeOf(f32));
const pipeline = try gloc.pipScale();
const entries = [_]c.WGPUBindGroupEntry{
.{ .binding = 0, .buffer = self.buf.raw, .offset = 0, .size = bytes },
.{ .binding = 1, .buffer = result.buf.raw, .offset = 0, .size = bytes },
.{ .binding = 2, .buffer = uni_buf.raw, .offset = 0, .size = @sizeOf(f32) },
};
try submitPass(gloc, pipeline, &entries, n);
try dispatch2in1out(gloc, pipeline, self.buf, other.buf, result.buf, self.byteSize());
return result;
}
@ -144,29 +117,39 @@ fn dispatch2in1out(
buf_b: GpuBuffer,
buf_out: GpuBuffer,
bytes: u64,
n: usize,
) !void {
// 1. Create a 4-byte Uniform buffer to hold the u32 size
const info_buf = try GpuBuffer.init(
gloc,
@sizeOf(u32),
c.WGPUBufferUsage_Uniform | c.WGPUBufferUsage_CopyDst,
);
defer info_buf.deinit(); // Clean up immediately after the pass submits
const max_chunk_bytes: u64 = 1024 * 1024 * 1024; // 1 GB
// 2. Cast the usize 'n' to a u32 and write it to the GPU queue
const size_payload: u32 = @intCast(n);
c.wgpuQueueWriteBuffer(gloc.queue, info_buf.raw, 0, &size_payload, @sizeOf(u32));
var offset: u64 = 0;
while (offset < bytes) {
// Calculate bounds for the current chunk
const current_chunk_bytes = @min(max_chunk_bytes, bytes - offset);
const current_chunk_elements: u32 = @intCast(current_chunk_bytes / @sizeOf(f32));
// 3. Create the 4 entries matching your WGSL @binding() tags
const entries = [_]c.WGPUBindGroupEntry{
.{ .binding = 0, .buffer = buf_a.raw, .offset = 0, .size = bytes },
.{ .binding = 1, .buffer = buf_b.raw, .offset = 0, .size = bytes },
.{ .binding = 2, .buffer = buf_out.raw, .offset = 0, .size = bytes },
.{ .binding = 3, .buffer = info_buf.raw, .offset = 0, .size = @sizeOf(u32) }, // <--- The 4th binding!
};
// Create uniform buffer for this specific chunk's size
const info_buf = try GpuBuffer.init(
gloc,
@sizeOf(u32),
c.WGPUBufferUsage_Uniform | c.WGPUBufferUsage_CopyDst,
);
defer info_buf.deinit();
try submitPass(gloc, pipeline, &entries, n);
// Write the number of elements *in this chunk* to the uniform buffer
c.wgpuQueueWriteBuffer(gloc.queue, info_buf.raw, 0, &current_chunk_elements, @sizeOf(u32));
// Bind only the sub-slice for this chunk using `.offset` and `.size`
const entries = [_]c.WGPUBindGroupEntry{
.{ .binding = 0, .buffer = buf_a.raw, .offset = offset, .size = current_chunk_bytes },
.{ .binding = 1, .buffer = buf_b.raw, .offset = offset, .size = current_chunk_bytes },
.{ .binding = 2, .buffer = buf_out.raw, .offset = offset, .size = current_chunk_bytes },
.{ .binding = 3, .buffer = info_buf.raw, .offset = 0, .size = @sizeOf(u32) },
};
// Submit the pass for this specific chunk
try submitPass(gloc, pipeline, &entries, current_chunk_elements);
offset += current_chunk_bytes;
}
}
/// Create bind group, encode pass, submit.

7
src/main.zig
View File

@ -18,7 +18,8 @@ pub fn main(init: std.process.Init) !void {
4 * 1024 * 1024,
4 * 4 * 1024 * 1024,
4 * 4 * 4 * 1024 * 1024,
1024 * 1024 * 1024,
4 * 4 * 4 * 4 * 1024 * 1024,
4 * 4 * 4 * 4 * 2 * 1024 * 1024,
};
// Print table header
@ -52,10 +53,6 @@ pub fn main(init: std.process.Init) !void {
const sum = try a.add(&gloc, b);
defer sum.deinit();
// sum * 2
const scaled = try sum.scale(&gloc, 2.0);
defer scaled.deinit();
// Read back (allocating dynamically for read-back buffers too)
const out_sum = try allocator.alloc(f32, size);
defer allocator.free(out_sum);