跨 I/O 集线器的 CUDA 点对点 [英] CUDA Peer-to-Peer across I/O hubs

问题描述

是否存在 SBIOS 条目或其他配置更改，使点对点能够跨连接 I/O 集线器(或套接字，如果 CPU 集成了 I/O 集线器)的 QPI 链路为 CUDA 工作- Sandy Bridge 或更高)?

Is there an SBIOS entry or other configuration change that will enable peer-to-peer to work for CUDA across the QPI links that connect I/O hubs (or sockets, in the case of CPUs that integrate the I/O hub - Sandy Bridge and higher)?

推荐答案

没有.QPI 链路的协议并未完全涵盖 PCIE 协议的所有特性，尤其是 P2P 协议使用的某些特性.

No. The QPI link has a protocol which does not entirely cover all features of the PCIE protocol, and in particular some features used by the P2P protocol.

英特尔数据表中记录了具体差异 这里.

A specific difference is documented in an intel datasheet here.

IOH 不支持从 PCI Express 为远程点对点 MMIO 事务启用非连续字节.这是对 PCI Express 标准要求的额外限制，以防止与英特尔 QuickPath 互连不兼容."(第 135 页)

"The IOH does not support non-contiguous byte enables from PCI Express for remote peer-to-peer MMIO transactions. This is an additional restriction over the PCI Express standard requirements to prevent incompatibility with Intel QuickPath Interconnect." (page 135)

因此 P2P 需要在两个设备之间使用连续的 PCIE 结构.两个设备都需要在同一个 PCIE 根联合体上.NVIDIA 在 CUDA 4.0 时间范围内公布了这一特殊要求，当时 GPUDirect v2.0 (Peer-to-Peer) 首次引入.

So P2P requires a continuous PCIE fabric between the two devices. Both devices need to be on the same PCIE root complex. This particular requirement was publicized by NVIDIA in the CUDA 4.0 timeframe when GPUDirect v2.0 (Peer-to-Peer) was first introduced.

请注意，一般来说，P2P 支持可能因 GPU 或 GPU 系列而异.在一种 GPU 类型或 GPU 系列上运行 P2P 的能力并不一定表明它可以在另一种 GPU 类型或系列上运行，即使在相同的系统/设置中也是如此.GPU P2P 支持的最终决定因素是提供的通过 cudaDeviceCanAccessPeer 查询运行时的工具.P2P 支持也会因系统和其他因素而异.此处的任何陈述均不保证任何特定设置中的任何特定 GPU 都支持 P2P.

Note that in general, P2P support may vary by GPU or GPU family. The ability to run P2P on one GPU type or GPU family does not necessarily indicate it will work on another GPU type or family, even in the same system/setup. The final determinant of GPU P2P support are the tools provided that query the runtime via cudaDeviceCanAccessPeer. P2P support can vary by system and other factors as well. No statements made here are a guarantee of P2P support for any particular GPU in any particular setup.

这篇关于跨 I/O 集线器的 CUDA 点对点的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！