[Openmp-commits] [PATCH] D28377: Fix a race in shutdown when tasking is used

Jonas Hahnfeld via Phabricator via Openmp-commits openmp-commits at lists.llvm.org
Fri Jan 6 00:30:13 PST 2017

Hahnfeld added a comment.

Thanks, this patch seems to solve the problem!

Comment at: runtime/src/kmp_runtime.cpp:4007-4012
+    if (__kmp_tasking_mode != tskm_immediate_exec)
+        // When tasking is possible, threads are not safe to reap until they are
+        // done tasking; this will be set when tasking code is exited in wait
+        this_thr->th.th_reap_state = KMP_NOT_SAFE_TO_REAP;
+    else  // no tasking --> always safe to reap
+        this_thr->th.th_reap_state = KMP_SAFE_TO_REAP;
Do we need this here or would it be enough to have the flag completely handled in `__kmp_execute_tasks_template`?

I don't know whether that would create a race on `th_reap_state`. If `__kmp_execute_tasks_template` is not guaranteed to be called at least once before a barrier finishes, why aren't there problems with multiple parallel regions? Each thread will have `th.th_reap_state = KMP_SAFE_TO_REAP` at the end of the first parallel region...

Comment at: runtime/src/kmp_wait_release.h:215-224
+                    else
+                        this_thr->th.th_reap_state = KMP_SAFE_TO_REAP;
                 else {
                     this_thr->th.th_task_team = NULL;
+                    this_thr->th.th_reap_state = KMP_SAFE_TO_REAP;
Resetting to `th.th_reap_state = KMP_SAFE_TO_REAP` could then be done at the end of `__kmp_execute_tasks_template`



More information about the Openmp-commits mailing list