[OpenMP] Fix stack overflow for test bug54082.c

[tianshilei1992]

Summary

When N is 1024, int result[N][N] is obviously large stack that Windows cannot support...

Fix #60326.

[jdoerfert]

LG

[mstorsjo]

Thanks for the fix! It does indeed seem to fix the issue.

Do you happen to know why this worked before, but now suddenly stopped working after D142297 - was the stack usage so close to the edges of what's allowed, that any minor tweak pushed it over the limit, or did the change in D142297 affect the amount of stack used in a bigger way? (I presume that the main int result[N][N] array here only gets allocated once on the stack, and the OpenMP threads just refer to that? I presume the change didn't cause the array to be allocated in more places than before?)

[tianshilei1992]

In D142684#4084862, @mstorsjo wrote:

Thanks for the fix! It does indeed seem to fix the issue.

Do you happen to know why this worked before, but now suddenly stopped working after D142297 - was the stack usage so close to the edges of what's allowed, that any minor tweak pushed it over the limit, or did the change in D142297 affect the amount of stack used in a bigger way? (I presume that the main int result[N][N] array here only gets allocated once on the stack, and the OpenMP threads just refer to that? I presume the change didn't cause the array to be allocated in more places than before?)

There is no issue. That test case was added in that patch. It didn’t pass before…

[mstorsjo]

In D142684#4085354, @tianshilei1992 wrote:

In D142684#4084862, @mstorsjo wrote:

Thanks for the fix! It does indeed seem to fix the issue.

Do you happen to know why this worked before, but now suddenly stopped working after D142297 - was the stack usage so close to the edges of what's allowed, that any minor tweak pushed it over the limit, or did the change in D142297 affect the amount of stack used in a bigger way? (I presume that the main int result[N][N] array here only gets allocated once on the stack, and the OpenMP threads just refer to that? I presume the change didn't cause the array to be allocated in more places than before?)

There is no issue. That test case was added in that patch. It didn’t pass before…

Oh, I see - doh. Thanks for explaining and sorry for the noise!

openmp/runtime/test/parallel/bug54082.c

 // This test is adapted from test_parallel_for_allocate.c in SOLLVE V&V.
 // https://github.com/SOLLVE/sollve_vv/blob/master/tests/5.0/parallel_for/test_parallel_for_allocate.c
 // RUN: %libomp-compile-and-run
 #include <omp.h>
 
 #include <assert.h>
 #include <stdlib.h>
 
-#define N 1024
+#define N 128
 
 int main(int argc, char *argv[]) {
   int errors = 0;
   int *x;
   int result[N][N];
   int successful_alloc = 0;
 
   omp_memspace_handle_t x_memspace = omp_default_mem_space;