File: nvptx-merged-loop.c

package info (click to toggle)
gcc-riscv64-unknown-elf 8.3.0.2019.08%2Bdfsg-1
  • links: PTS, VCS
  • area: main
  • in suites: bullseye
  • size: 680,956 kB
  • sloc: ansic: 3,237,715; cpp: 896,882; ada: 772,854; f90: 144,254; asm: 68,788; makefile: 67,456; sh: 29,743; exp: 28,045; objc: 15,273; fortran: 11,885; python: 7,369; pascal: 5,375; awk: 3,725; perl: 2,872; yacc: 316; xml: 311; ml: 285; lex: 198; haskell: 122
file content (30 lines) | stat: -rw-r--r-- 635 bytes parent folder | download | duplicates (4) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
 
/* { dg-do link } */
/* { dg-require-effective-target offload_nvptx } */
/* { dg-options "-fopenacc -O2 -foffload=-fdump-rtl-mach\\ -dumpbase\\ nvptx-merged-loop.c\\ -Wa,--no-verify" } */

#define N (32*32*32+17)
void __attribute__ ((noinline)) Foo (int *ary)
{
  int ix;

#pragma acc parallel num_workers(32) vector_length(32) copyout(ary[0:N])
  {
    /* Loop partitioning should be merged.  */
#pragma acc loop worker vector
    for (unsigned ix = 0; ix < N; ix++)
      {
	ary[ix] = ix;
      }
  }
}

int main ()
{
  int ary[N];

  Foo (ary);

  return 0;
}   

/* { dg-final { scan-rtl-dump "Merging loop .* into " "mach" } } */