| 题号 | 主题 | 文章 |
|——|——|——|
| 256 | 写一个线程安全的memory pool，支持allocate和… | [256-写一个线程安全的memory-pool，支持allocate和free.md](/ai-infra-interview-305/19-CPP%E4%B8%8ECUDA%E7%BC%96%E7%A8%8B/256-%E5%86%99%E4%B8%80%E4%B8%AA%E7%BA%BF%E7%A8%8B%E5%AE%89%E5%85%A8%E7%9A%84memory-pool%EF%BC%8C%E6%94%AF%E6%8C%81allocate%E5%92%8Cfree.html) |
| 257 | CUDA的shared memory使用注意事项？bank… | 257-CUDA的__shared__-memory使用注意事项.md |
| 258 | 实现一个ring buffer用于CPU-GPU异步数据传输 | 258-实现一个ring-buffer用于CPU-GPU异步数据传输.md |
| 259 | CUDA stream和event的同步机制？cudaStre… | [259-CUDA-stream和event的同步机制.md](/ai-infra-interview-305/19-CPP%E4%B8%8ECUDA%E7%BC%96%E7%A8%8B/259-CUDA-stream%E5%92%8Cevent%E7%9A%84%E5%90%8C%E6%AD%A5%E6%9C%BA%E5%88%B6.html) |
| 260 | thrust库的使用经验？transform_reduce？ | [260-thrust库的使用经验.md](/ai-infra-interview-305/19-CPP%E4%B8%8ECUDA%E7%BC%96%E7%A8%8B/260-thrust%E5%BA%93%E7%9A%84%E4%BD%BF%E7%94%A8%E7%BB%8F%E9%AA%8C.html) |
| 261 | CUTLASS的gemm调用示例？Epilogue定制？ | [261-CUTLASS的gemm调用示例.md](/ai-infra-interview-305/19-CPP%E4%B8%8ECUDA%E7%BC%96%E7%A8%8B/261-CUTLASS%E7%9A%84gemm%E8%B0%83%E7%94%A8%E7%A4%BA%E4%BE%8B.html) |
| 262 | NCCL的all-reduce实现？ring算法代码走读？ | [262-NCCL的all-reduce实现.md](/ai-infra-interview-305/19-CPP%E4%B8%8ECUDA%E7%BC%96%E7%A8%8B/262-NCCL%E7%9A%84all-reduce%E5%AE%9E%E7%8E%B0.html) |
| 263 | zero-copy memory的使用场景？cudaHostAll… | 263-zero-copy-memory的使用场景.md |
| 264 | 多GPU的peer-to-peer访问？cudaDeviceEna… | [264-多GPU的peer-to-peer访问.md](/ai-infra-interview-305/19-CPP%E4%B8%8ECUDA%E7%BC%96%E7%A8%8B/264-%E5%A4%9AGPU%E7%9A%84peer-to-peer%E8%AE%BF%E9%97%AE.html) |
| 265 | CUDA Graph的捕获条件？哪些操作不能捕获？ | [265-CUDA-Graph的捕获条件.md](/ai-infra-interview-305/19-CPP%E4%B8%8ECUDA%E7%BC%96%E7%A8%8B/265-CUDA-Graph%E7%9A%84%E6%8D%95%E8%8E%B7%E6%9D%A1%E4%BB%B6.html) |
| 266 | template metaprogramming在CUDA优化中的应… | [266-template-metaprogramming在CUDA优化中的应用.md](/ai-infra-interview-305/19-CPP%E4%B8%8ECUDA%E7%BC%96%E7%A8%8B/266-template-metaprogramming%E5%9C%A8CUDA%E4%BC%98%E5%8C%96%E4%B8%AD%E7%9A%84%E5%BA%94%E7%94%A8.html) |
| 267 | RAII模式管理CUDA资源？unique_ptr with c… | [267-RAII模式管理CUDA资源.md](/ai-infra-interview-305/19-CPP%E4%B8%8ECUDA%E7%BC%96%E7%A8%8B/267-RAII%E6%A8%A1%E5%BC%8F%E7%AE%A1%E7%90%86CUDA%E8%B5%84%E6%BA%90.html) |
| 268 | CUDA error处理的最佳实践？CUDA_CHECK宏？ | [268-CUDA-error处理的最佳实践.md](/ai-infra-interview-305/19-CPP%E4%B8%8ECUDA%E7%BC%96%E7%A8%8B/268-CUDA-error%E5%A4%84%E7%90%86%E7%9A%84%E6%9C%80%E4%BD%B3%E5%AE%9E%E8%B7%B5.html) |
| 269 | warp shuffle指令的使用？__shfl_sync？ | [269-warp-shuffle指令的使用.md](/ai-infra-interview-305/19-CPP%E4%B8%8ECUDA%E7%BC%96%E7%A8%8B/269-warp-shuffle%E6%8C%87%E4%BB%A4%E7%9A%84%E4%BD%BF%E7%94%A8.html) |
| 270 | cooperative groups的高级用法？grid_grou… | 270-cooperative-groups的高级用法.md |