Home » Technology » Terminal: Homebrew is Apple Silicon Mac ready

Terminal: Homebrew is Apple Silicon Mac ready

Homebrew ad be ready for Macs Apple Silicon. Until then, the hugely popular package manager on macOS worked thanks to Rosetta 2 on newer Macs, with a working setup, but a bit of a DIY:

Homebrew previously required a specific configuration to be installed on an Apple Silicon Mac.

The newly released version 3.0.0 is optimized for all Macs, those that still rely on an Intel processor and the x86 architecture, and those lucky enough to run on an Apple M1 processor and the ARM architecture. The installer script will take care of installing the right version according to your Mac, you don’t need to worry about it anymore. From a user’s perspective, the experience should now be the same regardless of the Mac.

This is good news, although you have to distinguish between the package manager and the packages it installs. Homebrew is Apple Silicon optimized, but not everything it can install yet. The site that references these packages has been updated to report all those that are compiled for ARM and all that are still content to run on the x86 architecture. These packages can still be installed by Homebrew, but they will be treated differently so that Rosetta 2 can run them.

Concretely, Homebrew will now store packages in two different locations on Apple Silicon Macs: /opt/homebrew for ARM packages, /usr/local for x86_64 packages. This management, which is more complex than on an Intel Mac, should however have no impact for the user, it is the manager who takes care of everything, including a priori to move a package from one location to another if it is updated for the new platform.

If you had Homebrew installed on an Apple Silicon Mac, you can update it as always with brew update. Importantly, if you had created aliases to use the manager with Rosetta as I advised in the previous article, consider removing them to benefit from the native versions.

– .

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.