Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[OpenMP] Use half of available logical processors for collapse tests #88319

Merged
merged 6 commits into from
Apr 19, 2024
Merged

[OpenMP] Use half of available logical processors for collapse tests #88319

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
68c3976
Use lower max threads to reduce the testing time on AIX.
xingxue-ibm Apr 10, 2024
302b83d
Merge branch 'llvm:main' into main
xingxue-ibm Apr 17, 2024
8f4349e
Merge branch 'llvm:main' into main
xingxue-ibm Apr 18, 2024
49617a8
Addressed comments.
xingxue-ibm Apr 11, 2024
2f4f44b
Addressed comments.
xingxue-ibm Apr 15, 2024
9b53422
Addressed comments.
xingxue-ibm Apr 18, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 16 additions & 6 deletions openmp/runtime/test/worksharing/for/collapse_test.inc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,9 +18,9 @@
#define MAX_THREADS 256

#if defined VERBOSE
#define PRINTF printf
#define PRINTF(...) printf(__VA_ARGS__)
#else
#define PRINTF
#define PRINTF(...)
#endif

LOOP_TYPE0 iLB, iUB;
Expand Down Expand Up @@ -106,12 +106,21 @@ int test() {
unsigned scalarCount = 0;
unsigned uselessThreadsOpenMP = 0;
unsigned usefulThreadsOpenMP = 0;
unsigned chunkSizesOpenmp[MAX_THREADS] = {0};

unsigned num_threads = omp_get_max_threads();
// Use half of the available threads/logical processors.
unsigned num_threads = omp_get_max_threads() / 2;

// Make sure num_threads is not 0 after the division in case
// omp_get_max_threads() returns 1.
if (num_threads == 0)
num_threads = 1;

if (num_threads > MAX_THREADS)
num_threads = MAX_THREADS;
omp_set_num_threads(num_threads);

unsigned long *chunkSizesOpenmp =
(unsigned long *)malloc(sizeof(unsigned long) * num_threads);
memset(chunkSizesOpenmp, 0, sizeof(unsigned long) * num_threads);

// count iterations and allocate space
LOOP { ++trueCount; }
Expand All @@ -129,10 +138,10 @@ int test() {
// perform and record OpenMP iterations and thread use
#pragma omp parallel num_threads(num_threads)
{
unsigned gtid = omp_get_thread_num();
#pragma omp for collapse(3) private(i, j, k)
LOOP {
unsigned count;
unsigned gtid = omp_get_thread_num();
#pragma omp atomic update
++chunkSizesOpenmp[gtid];
#pragma omp atomic capture
Expand Down Expand Up @@ -197,5 +206,6 @@ int test() {
// clean up space
FreeSpace(openmpSpace);
FreeSpace(scalarSpace);
free(chunkSizesOpenmp);
return pass;
}
Loading