Johannes Blaschke

Thanks @garrettwrong -- for some reason I didn't see #26 even though it's plainly there. A little more context: most of the parallelism that we're interested in is data-parallel. We're...

Should we continue this discussion in #26? Or do you want to keep this one around?

Yea @janden -- since we have a specific use case in mind (data-parallel workflows managed by python), I'll start there and keep everything in my fork. I'll keep y'all posted...

I've put together a simple solution to multi-gpu support here: #71 Since we're controlling our workflows from python, this PR is trying to be as minimal as possible.

Yea @lu1and10 -- I just put some more guards into the makefile to skip the single-precision translation units

It looks like the problem is that the tests use the single-prec code ... If someone wants to put `#ifdef`s around those, please go ahead. Otherwise I'll fix that once...

@ahbarnett I have picked this up again, as we're running into this from time-to-time. I used the terminology from the FFTW3 docs: https://www.fftw.org/fftw3_doc/Installation-on-Unix.html where they use `./configure --enable-single` to build...

I think one of the main differences is the workflow: `JWTs.jl` wants a URI (either on a server or a file) to a json payload containing the private key params....

I think this would be a good basis for more complex amrex types. Since torch and python don't have a standardize framework for expressing amr, this is (in my opinion)...

Thanks @ax3l that list is a good starting point. I would vote for the python buffer protocol strategy as a starting point. This seems to work well PyCUDA also. We...