News for hackers. Stuff that matters.

A comment on Building Elixir from source segfaults on macOS on Apple Silicon while compiling. Parent comment

Re:Building Elixir from source segfaults on macOS on Apple S... by oefrha on Monday July 13, 2020 @02:39PM (#23824070)
Any numpy (which uses gfortran) dependent is on that list. I suspect some could partially work by sacrificing some functionality, or maybe just run slower (disclaimer: I have no idea of numpy internals).
[ Parent | on HN ]
Re:Building Elixir from source segfaults on macOS on Apple S... by kalleboo on Tuesday July 14, 2020 @02:19AM (#23829093)
Interestingly at WWDC Apple said that they are contributing patches to NumPy to get it to work on Apple Silicon
[ Parent | on HN ]
Re:Building Elixir from source segfaults on macOS on Apple S... by vmchale on Monday July 13, 2020 @03:18PM (#23824494)
Wouldn't that entail swapping out all the underlying Fortran that makes NumPy so lovely?
[ Parent | on HN ]
Re:Building Elixir from source segfaults on macOS on Apple S... by dnautics on Monday July 13, 2020 @02:49PM (#23824184)
does numpy leverage blas?
[ Parent | on HN ]
Re:Building Elixir from source segfaults on macOS on Apple S... by oefrha on Monday July 13, 2020 @03:15PM (#23824469)
Yes. https://github.com/numpy/numpy/blob/62fa23c44fb49c1d238e1de4...
[ Parent | on HN ]
Re:Building Elixir from source segfaults on macOS on Apple S... by dnautics on Monday July 13, 2020 @05:47PM (#23825935)
well, that's probably where the fortran dependency could creep in, no?
[ Parent | on HN ]
Re:Building Elixir from source segfaults on macOS on Apple S... by sacheendra on Monday July 13, 2020 @03:15PM (#23824464)
yes
[ Parent | on HN ]