Valentin Churavy

For the ccallable one likely needs https://github.com/maleadt/LLVM.jl/blob/dc7f828da0d219cca50916917bf861cfac140637/test/orc_tests.jl#L45 Alternatively we could use absolute symbol resolvers https://github.com/maleadt/LLVM.jl/blob/dc7f828da0d219cca50916917bf861cfac140637/test/orc_tests.jl#L169

> Should we parse julia +channel specially in the script to deal with this? That'd complicate argument parsing quite a bit. Probably, or recommend the user uses `JULIAUP_CHANNEL=1.10 ./mpiexecjl `

Does Metal support register spilling? What happens when you exceed the total numbers of registers available (on CUDA 255 iirc)

You will need to add a rule with https://github.com/JuliaGPU/Adapt.jl to define how to translate your struct across the CPU-GPU boundary. The `@adapt_strucuture` may be sufficient.

I don't think you won't get around using Enzyme internally, but you might be able to define a ChainRule to define something that DI can access. This is what we...

I gathered a backtrace for this: ``` parse_string at /workspace/srcdir/mpfr-4.2.1/src/strtofr.c:329 [inlined] mpfr_strtofr at /workspace/srcdir/mpfr-4.2.1/src/strtofr.c:958 typeparser at /home/vchuravy/.julia/packages/Parsers/6Nt9C/src/floats.jl:63 #26 at /home/vchuravy/.julia/packages/Parsers/6Nt9C/src/components.jl:375 [inlined] checkforsentinel at /home/vchuravy/.julia/packages/Parsers/6Nt9C/src/components.jl:239 [inlined] stripwhitespace at /home/vchuravy/.julia/packages/Parsers/6Nt9C/src/components.jl:86 findquoted at...

On the issue of safety. This can lead to "torn" updates. E.g. one thread updating re one updating im. Since you are doing an accumulate that should be fine. We...

>Is there something I can do to improve the info provided? gist.github.com is a good place for large logs.

Let me turn your question around :) What do you need for multi-GPU? The thing that is likely missing the most is https://github.com/JuliaGPU/KernelAbstractions.jl/issues/395 (contributions always welcome.) If you simply need...

> Hence, the functions of a package I am interested in are the ones to handle parallelism with: > - Halo / Boundaries management > - Efficient way to store...