Saieiei wrote:

> Could you add a positive test where the omp.task in_reduction is actually 
> nested inside an enclosing omp.taskgroup task_reduction?

@tblah added a nested case with explicit `omp.task in_reduction` inside an 
enclosing `omp.taskgroup task_reduction`.

The test checks that the enclosing taskgroup emits `__kmpc_taskred_init` / 
`__kmpc_taskgroup`, while the outlined task body only does 
`__kmpc_task_reduction_get_th_data(gtid, null, orig)`. It also checks that the 
task body uses the returned private pointer and does not register its own 
reduction.


https://github.com/llvm/llvm-project/pull/202611
_______________________________________________
llvm-branch-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

Reply via email to