Paul Tiede

@thautwarm thank you! It looks like that fixed everything. Thank you for looking into this so quickly!

I'm traveling right now, but I should be able to look at this in a couple of days. I think all the changes make sense and I'm sorry for missing...

@sethaxen I set it up to use a separate step size and mass matrix for each tempering level. Using a global adaptation would probably increase the communication overhead and we...

Magic! The new version seems to have fixed this. I'll close it. Thanks for fixing this.

Ok excellent, I was hoping I understood the docs. Is there a way that Enzyme could potentially alert the user to this during compilation? The docs suggest I should see...

That is weird. For fixed inputs I get the following: ```julia using Enzyme function inplace_map(out, inner, a, b) map!(out, inner, a, b) do vi, ai, bi ai*vi*bi end return nothing...

Here is my status. I think I am using the current version of Enzyme. ```julia Status `~/Research/EnzymeTest/Project.toml` [7da242da] Enzyme v0.11.12 `https://github.com/EnzymeAD/Enzyme.jl.git#main ``` and `st -m` ```julia Status `~/Research/EnzymeTest/Manifest.toml` ⌅ [79e6a3ab]...

The easiest is probably SSH. I messaged you in slack.

Here is the latest map file ```julia using Enzyme Enzyme.API.printall!(true) @noinline function my_map_n!(dest::AbstractArray, As) n = length(As[1]) i = 0 while i < 70 i+=1 @inbounds dest[i] = @inbounds As[1][i]...

New version of issue ```julia using Enzyme Enzyme.API.printall!(true) @noinline function my_map_n!(dest, As) n = length(As[1]) i = 0 ptr = Base.unsafe_convert(Ptr{Float64}, As[1]) d = Base.unsafe_convert(Ptr{Float64}, dest) while i < n...