流多处理器，块和线程（CUDA） [英] Streaming multiprocessors, Blocks and Threads (CUDA)

问题描述

CUDA核心，流多处理器和块和线程的CUDA模型之间的关系是什么？

What is the relationship between a CUDA core, a streaming multiprocessor and the CUDA model of blocks and threads?

什么被映射到什么和什么是并行化， ？更有效，最大化块数或线程数？

What gets mapped to what and what is parallelized and how? and what is more efficient, maximize the number of blocks or the number of threads?

我目前的理解是，每个多处理器8个cuda内核。并且每个cuda核心将能够一次执行一个cuda块。

My current understanding is that there are 8 cuda cores per multiprocessor. and that every cuda core will be able to execute one cuda block at a time. and all the threads in that block are executed serially in that particular core.

这是否正确？

推荐答案

线程/块布局在 CUDA programming guide 。特别是，第4章陈述：

The thread / block layout is described in detail in the CUDA programming guide. In particular, chapter 4 states:

CUDA架构是围绕可扩展的多线程流多处理器（SM）阵列构建的。当主机CPU上的CUDA程序调用内核网格时，网格的块被枚举并分布到具有可用执行能力的多处理器。线程块的线程在一个多处理器上并发执行，多个线程块可以在一个多处理器上并发执行。

The CUDA architecture is built around a scalable array of multithreaded Streaming Multiprocessors (SMs). When a CUDA program on the host CPU invokes a kernel grid, the blocks of the grid are enumerated and distributed to multiprocessors with available execution capacity. The threads of a thread block execute concurrently on one multiprocessor, and multiple thread blocks can execute concurrently on one multiprocessor. As thread blocks terminate, new blocks are launched on the vacated multiprocessors.

每个SM包含8个CUDA内核，并且在任何一个时刻，重新执行32个线程的单个扭曲 - 因此需要4个时钟周期来为整个扭曲发布单个指令。你可以假设任何给定warp中的线程在锁步中执行，但要跨warp进行同步，您需要使用 __ syncthreads（）。

Each SM contains 8 CUDA cores, and at any one time they're executing a single warp of 32 threads - so it takes 4 clock cycles to issue a single instruction for the whole warp. You can assume that threads in any given warp execute in lock-step, but to synchronise across warps, you need to use __syncthreads().

这篇关于流多处理器，块和线程（CUDA）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！