Andrey Prokopenko

Closing in favor of #728.

This indeed got rid of the warnings such as ``` warning: linking module '/opt/sycl/lib/clang/14.0.0/../../clc/remangled-l64-signed_char.libspirv-nvptx64--nvidiacl.bc': Linking two modules of different target triples: '/opt/sycl/lib/clang/14.0.0/../../clc/remangled-l64-signed_char.libspirv-nvptx64--nvidiacl.bc' is 'nvptx64-unknown-nvidiacl' whereas '/var/jenkins/workspace/ArborX_PR-617/test/tstDetailsAlgorithms.cpp' is 'nvptx64-nvidia-cuda' ``` However,...

> I am seeing Ah, you are right. I think that's coming from the install build, as you would also need to update the `post` section.

> I'm trying to argue more in the direction of us wanting `-Werror` to be clean. I would not want to enable `-Werror` in the builds. The reason is that...

> I just try to say that the warning checker should detect all warnings just as `-Werror` is supposed to. If we want to ignore/allow specific warnings, we should do...

I really dislike how `view_alloc` just spread all over the place. But I suppose this is the only way to do it.

Just FYI, we had #460. It was in complete state, except for the fact that the Serial was slower.

@Shihab-Shahriar Thanks for asking! I think you are right, and looking around I don't see the scripts to produce the plots from the `.m4` or `.tex` files. @dalg24 I don't...

We may also want to clean the full `BatchedQueries` struct. We only use a single function out of it. And we don't need a struct really, just need a namespace.

> What is the motivation here? Did we see any problems or is this purely about being more specific? There is no pressing need. This is simply because it's really...