如何将一个浮点数组传递给金属内核函数？

import Metal

func computeWithMetal(floatArray: [Float]) {
    // 创建Metal设备对象和命令队列
    guard let device = MTLCreateSystemDefaultDevice(),
          let commandQueue = device.makeCommandQueue() else {
        return
    }
    
    // 创建Metal函数库对象和内核函数
    guard let library = try? device.makeDefaultLibrary(),
          let kernelFunction = library.makeFunction(name: "myKernelFunction") else {
        return
    }
    
    // 创建输入缓冲区
    let inputBuffer = device.makeBuffer(bytes: floatArray, length: floatArray.count * MemoryLayout<Float>.stride, options: [])
    
    // 创建输出缓冲区
    let outputBuffer = device.makeBuffer(length: floatArray.count * MemoryLayout<Float>.stride, options: [])
    
    // 创建计算任务
    guard let commandBuffer = commandQueue.makeCommandBuffer(),
          let computeEncoder = commandBuffer.makeComputeCommandEncoder() else {
        return
    }
    
    // 设置内核函数
    computeEncoder.setComputePipelineState(try! device.makeComputePipelineState(function: kernelFunction))
    
    // 设置内核函数参数
    computeEncoder.setBuffer(inputBuffer, offset: 0, index: 0)
    computeEncoder.setBuffer(outputBuffer, offset: 0, index: 1)
    
    // 执行内核函数
    let threadGroupSize = MTLSize(width: floatArray.count, height: 1, depth: 1)
    let threadGroups = MTLSize(width: 1, height: 1, depth: 1)
    computeEncoder.dispatchThreadgroups(threadGroups, threadsPerThreadgroup: threadGroupSize)
    
    // 完成计算任务
    computeEncoder.endEncoding()
    commandBuffer.commit()
    commandBuffer.waitUntilCompleted()
    
    // 从输出缓冲区中读取结果
    let outputData = NSData(bytesNoCopy: outputBuffer.contents(), length: floatArray.count * MemoryLayout<Float>.stride, freeWhenDone: false)
    let resultArray = outputData.toArray(type: Float.self)
    print("Result: \(resultArray)")
}

// 测试
let inputFloatArray: [Float] = [1.0, 2.0, 3.0, 4.0]
computeWithMetal(floatArray: inputFloatArray)