{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "No registered converter was able to produce a C++ rvalue of type unsigned int from this Python object of type numpy.int64",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[1], line 25\u001b[0m\n\u001b[1;32m     22\u001b[0m drv\u001b[38;5;241m.\u001b[39mmemcpy_htod(y_gpu, y)\n\u001b[1;32m     24\u001b[0m \u001b[38;5;66;03m# 调用函数\u001b[39;00m\n\u001b[0;32m---> 25\u001b[0m \u001b[43madd_func\u001b[49m\u001b[43m(\u001b[49m\u001b[43mn\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mx_gpu\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43my_gpu\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mblock\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m256\u001b[39;49m\u001b[43m,\u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m,\u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mgrid\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mn\u001b[49m\u001b[38;5;241;43m/\u001b[39;49m\u001b[38;5;241;43m/\u001b[39;49m\u001b[38;5;241;43m256\u001b[39;49m\u001b[43m,\u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     27\u001b[0m \u001b[38;5;66;03m# 将结果复制回CPU\u001b[39;00m\n\u001b[1;32m     28\u001b[0m drv\u001b[38;5;241m.\u001b[39mmemcpy_dtoh(y, y_gpu)\n",
      "File \u001b[0;32m~/anaconda3/envs/pycuda_3_10/lib/python3.10/site-packages/pycuda/driver.py:502\u001b[0m, in \u001b[0;36m_add_functionality.<locals>.function_call\u001b[0;34m(func, *args, **kwargs)\u001b[0m\n\u001b[1;32m    498\u001b[0m     \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtime\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m time\n\u001b[1;32m    500\u001b[0m     start_time \u001b[38;5;241m=\u001b[39m time()\n\u001b[0;32m--> 502\u001b[0m \u001b[43mfunc\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_launch_kernel\u001b[49m\u001b[43m(\u001b[49m\u001b[43mgrid\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mblock\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43marg_buf\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mshared\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m)\u001b[49m\n\u001b[1;32m    504\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m post_handlers \u001b[38;5;129;01mor\u001b[39;00m time_kernel:\n\u001b[1;32m    505\u001b[0m     Context\u001b[38;5;241m.\u001b[39msynchronize()\n",
      "\u001b[0;31mTypeError\u001b[0m: No registered converter was able to produce a C++ rvalue of type unsigned int from this Python object of type numpy.int64"
     ]
    }
   ],
   "source": [
    "import pycuda.autoinit\n",
    "import pycuda.driver as drv\n",
    "import numpy as np\n",
    "\n",
    "# 加载PTX文件\n",
    "mod = drv.module_from_file(\"test.ptx\")\n",
    "\n",
    "# 获取函数\n",
    "add_func = mod.get_function(\"add\")\n",
    "\n",
    "# 创建数据\n",
    "n = np.uint32(100)  # Convert to unsigned int\n",
    "x = np.random.rand(n).astype(np.float32)\n",
    "y = np.random.rand(n).astype(np.float32)\n",
    "\n",
    "# 分配内存\n",
    "x_gpu = drv.mem_alloc(x.nbytes)\n",
    "y_gpu = drv.mem_alloc(y.nbytes)\n",
    "\n",
    "# 将数据复制到GPU\n",
    "drv.memcpy_htod(x_gpu, x)\n",
    "drv.memcpy_htod(y_gpu, y)\n",
    "\n",
    "# 调用函数\n",
    "add_func(x_gpu, y_gpu, block=(256,1,1), grid=(n//256,1))\n",
    "\n",
    "# 将结果复制回CPU\n",
    "drv.memcpy_dtoh(y, y_gpu)\n",
    "\n",
    "# 检查结果\n",
    "print(y)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pycuda_3_10",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}