设备软件

CUDA 线程CUDA Thread

线程是线程组层次结构中的最低层级（顶部、左侧），并被映射到流式多处理器的核心上。改编自 NVIDIA 的 CUDA 复习：CUDA 编程模型和 NVIDIA CUDA C++ 编程指南中的图表。

_执行线程_（简称"线程"）是 GPUs 编程的最小单位，也是 CUDA 编程模型中线程层次结构的基础和原子单位。线程拥有独立的寄存器，但除此之外几乎没有其他资源。

SASS 和 PTX 程序都以线程为执行目标。相比之下，POSIX 环境中的典型 C 程序以进程为目标，而进程本身是一个或多个线程的集合。与 POSIX 线程不同，CUDA 线程不用于执行系统调用。

与 CPU 上的线程类似，GPU 线程可以拥有私有的指令指针/程序计数器。但是出于性能考虑，GPU 程序通常被编写为让一个线程束中的所有线程共享相同的指令指针，并以锁步方式执行指令（另请参阅线程束调度器）。

同样类似于 CPU 上的线程，GPU 线程在全局内存中拥有栈，用于存储溢出的寄存器和函数调用栈，但高性能内核通常限制这两者的使用。

单个 CUDA 核心执行来自单个线程的指令。

英文原文(Modal GPU Glossary)

of a Streaming Multiprocessor. Modified from diagrams in NVIDIA's CUDA Refresher: The CUDA Programming Model and the NVIDIA CUDA C++ Programming Guide.](themed-image://cuda-programming-model.svg)

A _thread of execution_ (or "thread" for short) is the lowest unit of programming for GPUs, the base and atom of the CUDA programming model's thread hierarchy. A thread has its own registers, but little else.

Both SASS and PTX programs target threads. Compare this to a typical C program in a POSIX environment, which targets a process, itself a collection of one or more threads. Unlike POSIX threads, CUDA threads are not used to make syscalls.

Like a thread on a CPU, a GPU thread can have a private instruction pointer/program counter. However, for performance reasons, GPU programs are generally written so that all the threads in a warp share the same instruction pointer, executing instructions in lock-step (see also Warp Scheduler).

Also like threads on CPUs, GPU threads have stacks in global memory for storing spilled registers and a function call stack, but high-performance kernels generally limit use of both.

A single CUDA Core executes instructions from a single thread.

相关词条