Chris Siefert

We know about this and it is scheduled to get looked at.

No. Because we can't merge a PR to disable it.

Well, seeing as I can't merge into the GenConfig repo, I can't exactly do that either. Maybe you can.

> The trilinos spack recipe sets `gotype=long` by default, so that's why you're getting `-DTpetra_INST_INT_LONG:BOOL=ON`. That's horrific. Who decided to enable a capability that isn't tested by default?

Lesson: Never rely on somebody else's spack settings for software as complicated as Trilinos (or even Hypre). It won't set what you want.

@sethrj Dunno where you heard that. Trilinos has no plans to hard-code long long for global types.

@trilinos/framework We could definitely update that. Currently, the newest CUDA build we're testing nightly is with CUDA10.1/ GCC 7.3.1 / OpenMPI 4.0.3. CUDA 11 should work, as should newer GCCs...

TpetraCore_MultiVector_LocalViewTests_MPI_4 is supposed to throw exceptions. That's the _point_. It's testing error cases. As far as I can tell from the output, the test is running correctly. The problem is,...

And on the other machine the reproducer won't even configure correctly. Time to file another TRILINOSHD ticket...

> But that test does not seem to check that those lines are even printed. It is only checking for `End Result: TEST PASSED`. So the test would still pass...