num_threads 与 omp_set_num_threads 与 OMP_NUM_THREADS 之间的区别 [英] Difference between num_threads vs. omp_set_num_threads vs OMP_NUM_THREADS
问题描述
我对在代码的并行部分指定线程数的方法感到非常困惑.我知道我可以使用:
I am quite confused about the ways to specify the number of threads in parallel part of a code. I know I can use:
- 环境变量 OMP_NUM_THREADS
- 函数 omp_set_num_threads(int)
- num_threads(int) in
#pragma omp parallel for num_threads(NB_OF_THREADS)
到目前为止我收集到的前两个是等价的.但是第三个呢?有人可以提供更详细的差异说明吗,我在互联网上找不到任何关于 1/2 和 3 之间差异的信息.
What I have gathered so far the first two are equivalent. But what about the third one? Can someone provide a more detailed exposition of the difference, I could not find any information in the internet regarding the difference between 1/2 and 3.
推荐答案
OMP_NUM_THREADS
和 omp_set_num_threads()
不是等价的.环境变量仅用于设置控制组中最大线程数的nthreads-var ICV(内部控制变量)的初始值.omp_set_num_threads()
可用于随时更改 nthreads-var 的值(当然在任何并行区域之外)并影响所有后续并行区域.因此设置一个值,例如n
,到OMP_NUM_THREADS
相当于在遇到第一个并行区域之前调用omp_set_num_threads(n)
.
OMP_NUM_THREADS
and omp_set_num_threads()
are not equivalent. The environment variable is only used to set the initial value of the nthreads-var ICV (internal control variable) which controls the maximum number of threads in a team. omp_set_num_threads()
can be used to change the value of nthreads-var at any time (outside of any parallel regions, of course) and affects all subsequent parallel regions. Therefore setting a value, e.g. n
, to OMP_NUM_THREADS
is equivalent to calling omp_set_num_threads(n)
before the very first parallel region is encountered.
确定并行区域内线程数的算法在OpenMP 规范 可在 OpenMP 网站上免费获得:
The algorithm to determine the number of threads in a parallel region is very clearly described in the OpenMP specification that is available freely on the OpenMP website:
ifnum_threads
子句存在
然后让 ThreadsRequested 成为num_threads
子句表达式的值;
then let ThreadsRequested be the value of the num_threads
clause expression;
else let ThreadsRequested = nthreads-var 的第一个元素的值;
else let ThreadsRequested = value of the first element of nthreads-var;
在规范的ICV Override Relationships部分中列出了设置nthreads-var的不同方式的优先级:
That priority of the different ways to set nthreads-var is listed in the ICV Override Relationships part of the specification:
num_threads
子句和 omp_set_num_threads()
覆盖了 OMP_NUM_THREADS
环境变量的值和 OMP_NUM_THREADS
环境变量的初始值.em>nthreads-var ICV.
The
num_threads
clause andomp_set_num_threads()
override the value of theOMP_NUM_THREADS
environment variable and the initial value of the first element of the nthreads-var ICV.
翻译成人类语言,即:
OMP_NUM_THREADS
(如果存在)最初指定线程数;- 对
omp_set_num_threads()
的调用会覆盖OMP_NUM_THREADS
的值; num_threads
子句的存在覆盖了其他两个值.
OMP_NUM_THREADS
(if present) specifies initially the number of threads;- calls to
omp_set_num_threads()
override the value ofOMP_NUM_THREADS
; - the presence of the
num_threads
clause overrides both other values.
实际使用的线程数还受是否启用动态团队规模的影响(dyn-var ICV 可通过 OMP_DYNAMIC
和/或 omp_set_dynamic() 设置
)、线程限制是否由 thread-limit-var 强加(可通过 OMP_THREAD_LIMIT
设置),以及是否嵌套并行(OMP_NESTED
/omp_set_nested()
) 是否启用.
The actual number of threads used is also affected by whether dynamic team sizes are enabled (dyn-var ICV settable via OMP_DYNAMIC
and/or omp_set_dynamic()
), by whether a thread limit is imposed by thread-limit-var (settable via OMP_THREAD_LIMIT
), as well as by whether nested parallelism (OMP_NESTED
/ omp_set_nested()
) is enabled or not.
这篇关于num_threads 与 omp_set_num_threads 与 OMP_NUM_THREADS 之间的区别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!