Rami

An afterthought: having written the pullback by hand a few times, is `enzyme` effectively saving `pp/pc` for every `it in 1:nt`?

Looking at the code, the pattern ``` @sync for i in pids @async remotecall_fetch(**do_work**,i,...) ``` is common (and natural). So this may happen anywhere where `**do_work**` is heavy. I guess...

This is what `Array(d::DArray)` does. So it is already [implemented](https://github.com/JuliaParallel/DistributedArrays.jl/blob/b98875d16c6810d33498405d5d852a3acee19b38/src/darray.jl#L570).

Could you please fix similar to give the right distribution while you are at it :) This doesn't work: ``` julia> dd=drand((10,10),workers(),[2,1]); julia> ds=similar(dd); julia> ds.indices 1x2 Array{Tuple{UnitRange{Int64},UnitRange{Int64}},2}: (1:10, 1:5)...

How would this be different than `pmap`?

First, thanks for the reply! 1- same directory 2- Works in Julia REPL 3- The `@` is part of the filename, but I had already tried with another filename with...

``` julia Base.repl_cmd(`sustripsrc.rsf`,STDOUT) sustrip: subcalls.c: efopen: fopen failed Base.repl_cmd(@cmd("sustripsrc.rsf@"),STDOUT) sustrip: subcalls.c: efopen: fopen failed ```

Ok, some advance here, the command has a optional argument `par=/dev/tty`. Without it, it prints the parameters to screen (which is what python, and julia REPL do). Once I added...

@JobJob same error for above command

It gives: ```julia cmd =`sustripsrc.rsf@` shell = Base.shell_split(get(ENV,"JULIA_SHELL",get(ENV,"SHELL","/bin/sh"))) run(`$shell -i -c "($(Base.shell_escape(cmd))) && true"`) bash: no job control in this shell sustrip: subcalls.c: efopen: fopen failed ``` I tried with...