Vyas Ramasubramani
Vyas Ramasubramani
In what contexts is this failing? I assume it's a statically linked case, so the symbols cannot have been (dynamically) loaded with `RTLD_LOCAL` in a way that would obscure them...
This PR has gone through a lot of churn so I haven't been keeping close track of it. @robertmaynard are there still cases that need resolving that you want help...
Where is that logic? I see that we're skipping `libcudf-examples` in cudf, for example, but not tests, and no equivalent key is specified for rmm.
Wow that is very hidden... nice find though, thanks!
It looks like it is in a vtable for primitive types as well. ``` /* "declarations.pyx":4 * * * cdef class A: #
Maybe conda packages weren't out yet when the builds failed? Wheel builds will pull the source of raft to rebuild the static lib (unless you changed something in how cuvs...
I don't know if you intended it to be in scope for this issue, but beyond implementing this on a per-column basis we would also benefit from supporting it as...
@jakirkham may be able to comment on how useful this optimization even is right now since [he added the relevant comment](https://github.com/rapidsai/rmm/commit/f50da49a5ad4e2623e76744d14978dfa178ee3f5).
I'm generally a fan of slimmer dependencies for any number of reasons including smaller environments, faster package solves, faster import times, less external sources of breakage, etc. It also makes...
Certainly, I agree with that. In fact I've pushed in that direction in other cases too where I think removing a dependency would be too hard and cost too much....