Valentin Churavy

For the CPU I had long hoped to use SIMD.jl or a compiler pass to perform vectorization. Would a subgroupsize of 1 be legal?

I talked with Simone and I think expressing this as an Index projection similar to `LinearIndicies` and `CartesianIndices` is the right direction.

Is there documentation on the Flange ABI? The original post mentioned scipy. Users like Julia and Python need to know the ABI in order to call fortran applications. If the...

``` level=warn name=parca-agent ts=2023-07-13T15:52:32.55032116Z caller=batch_remote_write_client.go:139 msg="batch write client failed to send profiles" count=0 err="rpc error: code = DeadlineExceeded desc = context deadline exceeded" level=warn name=parca-agent ts=2023-07-13T15:57:32.551395853Z caller=batch_remote_write_client.go:139 msg="batch write client...

I think we determined that this was a case of the endpoint not accepting the data (due to the account not being fully active) Having a better error message here...

The benefit is that the user can choose to not install CUDA and just use the CPU implementation of Oceananigans, or use AMDGPU instead of CUDA. So installing Oceanigans will...

> I don't know if you are able to export things from the extension... No that is one of the limitation, extensions can't export new things.

With `JULIA_DEBUG=Documenter`: ``` ┌ Debug: Evaluating @docs block: │ Base.ScopedValues.ScopedValue │ Base.ScopedValues.with │ Base.ScopedValues.@with │ Base.isassigned(::ScopedValue) │ Base.ScopedValues.get └ @ Documenter ~/src/julia/doc/deps/packages/Documenter/1HwWe/src/expander_pipeline.jl:370 ... ```

What is `typeof(inters)`. But the likely solution is something like: ``` f += sum(ntuple(Val(N)) do inter_i @inline force_gpu(inters[inter_i], dr, coord_i, coord_j, atoms[i], atoms[j], boundary, special) end) ``` Important is that...

If you can get a backtrace from where the hang is coming from there might be something we can intuit from it.