[Bug libgomp/122280] target teams distribute parallel for collapse(2) yields different results in a matmul than separate loops (one with omp target teams distribute the second with omp parallel for) on nvptx target. Clang compiles the code correctly

[schulz.benjamin at googlemail dot com via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122280

--- Comment #1 from Benjamin Schulz <schulz.benjamin at googlemail dot com> ---
Created attachment 62556
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62556&action=edit
mathdemonstrations-xnvptx-none.ii.tar.gz

Above, I attached the source of the test applications suite. Here, I attached a
precompiled .ii file