Jan Ciesko
Jan Ciesko
Propagate padding when `create_mirror_view` and source view has `Kokkos::AllowPadding`. Case in point: ``` auto alloc = Kokkos::view_alloc("test", Kokkos::AllowPadding); auto v = Kokkos::View(alloc, 17, 4); auto h_v = Kokkos::create_mirror_view(v); ASSERT_EQ(v.stride(1), h_v.stride(1));...
We do not error out early enough and thus allocate over a wrong size. Example ``` #include using view_t_1 = Kokkos::View; using view_t_2 = Kokkos::View; int main(int argc, char *argv[])...
This tracks current issues for ULT- and PartCom support in OMPI. ULT Support: - [x] Compilation fails on current master @ HEAD when configured to --with-threads={qthreads,argobots} with ` ../../opal/mca/threads/qthreads/threads_qthreads.h:29:10: fatal...
This makes Qthreads compile with fast context switching implemented in asm.
This adds generated html man files making quick API lookup more convenient.
A faster alternative to ucontext on Apple M1 hw is desirable. Ine path is porting the current asm aarch64 implementation to Apple's object format (Mach-O). https://developer.apple.com/library/archive/documentation/Performance/Conceptual/CodeFootprint/Articles/MachOOverview.html Does ABT support fast...
- This replaces the use of opal atomics with the use of qthread's atomics in the opal/thread MCA if ompi is configured with qthreads. - This resolves the deadlock reported...
As reported by @ronawho > I'm seeing failures for a couple types of tests. > > One failure for tests that sleep, [here](https://github.com/chapel-lang/chapel/blob/4c48d6191fa2ea46e2dcb8964a6b2d2327f865a4/runtime/src/tasks/qthreads/tasks-qthreads.c#L950-L983) it seems like when doing sleeps from...
Currently [this test app](https://github.com/sandialabs/MPI-Partix/blob/main/apps/ultcorrectness1.cpp) hangs with executions with multiple streams and when OMPI is configured to use a ULT lib. To support this case, we would need ULT-aware threading primitives...