gpgpu-rs

gpgpu-rs 是基于 wgpu，采用 Rust 实现的简易 GPU 计算库。

如需开始使用gpgpu，先创建[Framework]实例然后按照下面的示例进行操作。

向量 A B 相乘并将结果存入 C 的例子：

 use gpgpu::*;

 fn main() -> GpuResult<()> {
    let fw = Framework::default();
    
    // Original CPU data
    let cpu_data = (0..10000).into_iter().collect::<Vec<u32>>();

    // GPU buffer creation
    let buf_a = GpuBuffer::from_slice(&fw, &cpu_data);       // Input
    let buf_b = GpuBuffer::from_slice(&fw, &cpu_data);       // Input
    let buf_c = GpuBuffer::<u32>::new(&fw, cpu_data.len());  // Output

    // Shader load from SPIR-V binary file
    let shader_module = utils::shader::from_spirv_file(&fw, "<SPIR-V shader path>")?;
    //  or from a WGSL source file
    let shader_module = utils::shader::from_wgsl_file(&fw, "<WGSL shader path>")?;    

    // Descriptor set creation
    let desc_set = DescriptorSet::default()
        .bind_buffer(&buf_a, GpuBufferUsage::ReadOnly)
        .bind_buffer(&buf_b, GpuBufferUsage::ReadOnly)
        .bind_buffer(&buf_c, GpuBufferUsage::ReadWrite);
    
    // Kernel creation and enqueuing
    fw.create_kernel_builder(&shader_module, "main")   // Entry point
        .add_descriptor_set(desc_set)                      
        .build()
        .enqueue(cpu_data.len() as u32, 1, 1);         // Enqueuing, not very optimus ��

    let output = buf_c.read()?;                        // Read back C from GPU
    for (a, b) in cpu_data.into_iter().zip(output) {
        assert_eq!(a.pow(2), b);
    }

    Ok(())
}

使用 WGSL 编写的着色器程序

// Vector type definition. Used for both input and output
[[block]]
struct Vector {
    data: [[stride(4)]] array<u32>;
};

// A, B and C vectors
[[group(0), binding(0)]] var<storage, read>  a: Vector;
[[group(0), binding(1)]] var<storage, read>  b: Vector;
[[group(0), binding(2)]] var<storage, read_write> c: Vector;

[[stage(compute), workgroup_size(1)]]
fn main([[builtin(global_invocation_id)]] global_id: vec3<u32>) {
    c.data[global_id.x] = a.data[global_id.x] * b.data[global_id.x];
}