c - OpenMP 卸载到 nvptx 时 gcc 7 和 8 (debian) 的问题

Question

我都安装了 gcc-7 , gcc-8 , gcc-7-offload-nvptx和 gcc-8-offload-nvptx

我尝试同时使用两者来编译带卸载的简单 OpenMP 代码:

#include <omp.h>
#include <stdio.h>

int main(){
    #pragma omp target
    #pragma omp teams distribute parallel for
    for (int i=0; i<omp_get_num_threads(); i++)
        printf("%d in %d of %d\n",i,omp_get_thread_num(), omp_get_num_threads());
}

使用以下行(也使用 gcc-7):

gcc-8 code.c -fopenmp -foffload=nvptx-none

但是编译不通过，报如下错误:

/tmp/ccKESWcF.o: In function "main":
teste.c:(.text+0x50): undefined reference to "GOMP_target_ext"
/tmp/cc0iOH1Y.target.o: In function "init":
ccPXyu6Y.c:(.text+0x1d): undefined reference to "GOMP_offload_register_ver"
/tmp/cc0iOH1Y.target.o: In function "fini":
ccPXyu6Y.c:(.text+0x41): undefined reference to "GOMP_offload_unregister_ver"
collect2: error: ld returned 1 exit status

一些线索？

Answer 1

您的代码使用 -foffload=disable -fno-stack-protector 为我编译和运行与 gcc7和 gcc-7-offload-nvptx和 Ubuntu 17.10。

但在 GPU 上(没有 -foffload=disable )它无法编译。你不能调用 printf从GPU。相反，您可以这样做:

#include <omp.h>
#include <stdio.h>
#include <stdlib.h>

int main(){
  int nthreads;
  #pragma omp target teams map(tofrom:nthreads)
  #pragma omp parallel
  #pragma omp single
  nthreads = omp_get_num_threads();

  int *ithreads = malloc(sizeof *ithreads *nthreads);

  #pragma omp target teams distribute parallel for map(tofrom:ithreads[0:nthreads])
  for (int i=0; i<nthreads; i++) ithreads[i] = omp_get_thread_num();

  for (int i=0; i<nthreads; i++)
    printf("%d in %d of %d\n", i, ithreads[i], nthreads);

  free(ithreads);  
}

对我来说这个输出

0 in 0 of 8
1 in 0 of 8
2 in 0 of 8
3 in 0 of 8
4 in 0 of 8
5 in 0 of 8
6 in 0 of 8
7 in 0 of 8