[Bug libgomp/87834] New: nvptx offloading: "WARNING: program timed out" for libgomp.fortran/target2.f90 execution test at -O0, -O1

[tschwinge at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=87834

            Bug ID: 87834
           Summary: nvptx offloading: "WARNING: program timed out" for
                    libgomp.fortran/target2.f90 execution test at -O0, -O1
           Product: gcc
           Version: 9.0
            Status: UNCONFIRMED
          Keywords: openmp
          Severity: normal
          Priority: P3
         Component: libgomp
          Assignee: unassigned at gcc dot gnu.org
          Reporter: tschwinge at gcc dot gnu.org
                CC: cesar at gcc dot gnu.org, jakub at gcc dot gnu.org,
                    vries at gcc dot gnu.org
  Target Milestone: ---
            Target: nvptx-none

I'm aware of PR81691 "libgomp.fortran/target2.f90 fails for nvptx at -O0 and
-O1", but after r264397 "[nvptx] Remove use of CUDA unified memory in libgomp"
they now also run into "WARNING: program timed out", which is annoying.

    PASS: libgomp.fortran/target2.f90   -O0  (test for excess errors)
    {+WARNING: program timed out.+}
    FAIL: libgomp.fortran/target2.f90   -O0  execution test
    PASS: libgomp.fortran/target2.f90   -O1  (test for excess errors)
    {+WARNING: program timed out.+}
    FAIL: libgomp.fortran/target2.f90   -O1  execution test
    PASS: libgomp.fortran/target2.f90   -O2  (test for excess errors)
    PASS: libgomp.fortran/target2.f90   -O2  execution test

I have not yet analyzed what's causing this, but I have some ideas about
pending patches that might cure it.