GPU / CUDA核心是SIMD的吗？ [英] Are GPU/CUDA cores SIMD ones?

问题描述

让我们采用 nVidia Fermi计算体系结构。它说：

第一个基于Fermi的GPU，实现了30亿个晶体管，具有多达512个CUDA内核。 CUDA内核每个线程的每个时钟执行一个浮点数或整数指令。 512个CUDA内核以16个SM（每个32个内核）进行组织。

The first Fermi based GPU, implemented with 3.0 billion transistors, features up to 512 CUDA cores. A CUDA core executes a floating point or integer instruction per clock for a thread. The 512 CUDA cores are organized in 16 SMs of 32 cores each.

[...]

每个CUDA处理器都具有完整的流水线整数算法。逻辑单元（ALU）和浮点单元（FPU）。

Each CUDA processor has a fully pipelined integer arithmetic logic unit (ALU) and floating point unit (FPU).

[...]

在Fermi中，新设计的整数ALU支持完整的整数。所有指令的32位精度，与标准编程语言要求一致。还优化了整数ALU，以有效支持64位和扩展的精度运算。 V

In Fermi, the newly designed integer ALU supports full 32-bit precision for all instructions, consistent with standard programming language requirements. The integer ALU is also optimized to efficiently support 64-bit and extended precision operations. V

据我所知，我不清楚的是，GPU在所谓的 warps 中执行线程，每个经线由〜32个线程组成。每个扭曲仅分配给一个核心（是吗？）。这是否意味着单个SM的32个内核中的每个内核都是SIMD处理器，其中单个指令处理 32个数据部分？如果是这样，那么为什么我们说一个扭曲中有32个线程，而不是一个SIMD线程？为什么有时将内核称为标量处理器，而不是矢量处理器？

From what I know, and what is unclear for me, is that GPUs execute the threads in so called warps, each warp consists of ~32 threads. Each warp is assigned to only one core (is that true?). So does that mean, that each of the 32 cores of a single SM is a SIMD processor, where a single instruction handles 32 data portions ? If so, then why we say there are 32 threads in a warp, not a single SIMD thread? Why cores are sometimes referred to as scalar processors, not vector processors ?

推荐答案

每个扭曲仅分配给一个核（是吗？）。

Each warp is assigned to only one core (is that true?).

不，不是这样。扭曲是32个执行线程的逻辑组合。要从一个warp执行一条指令，warp调度程序通常必须调度 32 执行单元（或 cores，尽管 core的定义有些松散）。

No, it's not true. A warp is a logical assembly of 32 threads of execution. To execute a single instruction from a single warp, the warp scheduler must usually schedule 32 execution units (or "cores", although the definition of a "core" is somewhat loose).

内核实际上是标量处理器，而不是矢量处理器。 warp调度程序将32个内核（或执行单元）编组为一条跨32个线程的单条指令，这就是 SIMT绰号的来源。

Cores are in fact scalar processors, not vector processors. 32 cores (or execution units) are marshalled by the warp scheduler to execute a single instruction, across 32 threads, which is where the "SIMT" moniker comes from.

这篇关于GPU / CUDA核心是SIMD的吗？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！