-
-
Notifications
You must be signed in to change notification settings - Fork 660
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Support for Apple Silicon #780
Comments
I'm curious about this too, but more for when a version installable from Homebrew itself is released, as that is how I used to have this installed. For the time being I've copied the x86_64 version from a backup, but longterm want exa installed via Homebrew again. |
@alexclst homebrew for ARM64 is already working great. I only have one formula I need the intel version for, and that's Exa. |
Yeah, same here. Homebrew working fine, but Exa the one remaining formula. |
The extent of the issue is unclear to me. Is exa compiling fine on macOS ARM64? Do we need to do something in this repository regarding the Howebrew formula? I suppose this issue is also asking for a build on GitHub releases too; I’ve no idea how this works so I don’t know if it’s possible. |
After installing rust, libgit2, and cmake (all via Apple Silicon homebrew) I was able to compile exa from source no problem at all. However the homebrew formula is not working. I'm seeing
I'm guessing this is the source of the problem, maybe related to how homebrew uses superenv, but I'm no homebrew build formula expert. I'm plugging away at it and will update here if I make any progress. |
I investigated. Support for aarch64-apple-darwin has been added in Rust 1.49.0 released 2020-12-31. The Homebrew formula is still using Rust 1.48.0, but Homebrew was already using Rust 1.49.0 beta instead of 1.48.0 for macOS ARM. It seems update to Rust stable 1.49.0 is ongoing at Howebrew/homebrew-core#68089; let’s see if that fixes the Homebrew problem or not. |
Switching to Rust 1.49 on its own doesn't seem to be enough to fix build failures for exa: Homebrew/homebrew-core#68089 (comment) |
So I was able to compile manually using Rust 1.49. I have a stable ARM exa binary on my M1 Mac. Is this then something to do with the build script in brew? |
By compile manually, do you mean using Anyway, the explanation for why it doesn't work for |
@carlocab got it. Yep that's what I meant. So this should resolve in time. And certainly when GHC becomes ARM64 compatible? |
@ariasuni - I see there hasn't been a release in over a year, is it likely that it will take a long time for a release with the updated cargo.lock file? |
Does |
@carlocab - nah because GHC 8.8 fails as x86 only. Also --HEAD wouldn't be updated unless manually removed and installed/reinstalled. I see lots of dev work on Exa, but very few releases. |
Yes, of course, I meant after removing the Anyway, that said, let's see if pulling in a newer |
I've taken the latest |
Above PR has been merged. Though it would still be nice to see a proper new release of exa that actually supports Apple Silicon, rather than having to patch the latest stable release to get it. |
@carlocab - thanks you're a star 👍🏼 |
@carlocab - if/when exa does release a new update, will that then flow down via homebrew, or require a new update by someone on homebrew side to reverse the formula change? |
Not sure what you mean. When there is a new version of exa, the Homebrew formula will be updated in the same way any other formula is updated. If the new version uses a new enough lock file, then the lock file patch can be removed. |
Since Homebrow is providing exa 0.10.0 for Apple Silicon, I believe this issue can be closed. |
Hello. I tried to install exa via asdf on an m1 and got the following error.
It seems like if this project published a |
If you're in a pinch...
|
Is there any plan to support Apple Silicon/ARM64?
The text was updated successfully, but these errors were encountered: