Proposal: Drop Intel-based Macbook dual boot release criterion

Project Discussion
, , ,
1

Hi everyone, this is a proposed change for Fedora Release Criteria. It is related to a wide set of changes that Fedora Quality team need to perform this cycle, which are summarized here, so please feel free to read that for more background information and general overview, thanks.

Proposal: No longer consider dual-boot functionality on an Intel-based Macbook to be release-blocking.

You can see the current release criterion here, and (expand the footnotes sections) it reads:

The installer must be able to install into free space alongside an existing macOS installation, install and configure a bootloader that will boot Fedora.
This criterion only covers Mac devices with an Intel x86_64 processor.

This criterion is being obsoleted by time, because Apple stopped producing Intel-based Macs many years ago (replaced them by “Apple silicon” M* processors, which are not supported by Fedora at this moment). The last Macbooks which were reasonably usable with Fedora are 2017 models, which still contain the T1 security chip (newer models have a T2 chip, and their internal keyboard and touchpad don’t work with Fedora kernel). System updates support for 2017 models ends this year, older models are already obsolete. This means that users of pre-2017 models likely already switched to Fedora, if they wanted, users of 2017 models might do it this year, and there are no future users in this regard, because their hardware is not well supported by Fedora.

The Quality team used to test Intel-based Mac dual-boot regularly. We lost the test hardware eventually, but still asked our community to participate in this testing, and tried to help resolve any issues. But the era of Intel-based Macs (at least those supported by Fedora) has come to an end, and it no longer makes sense to keep it release-blocking and devote significant testing time to this, in our eyes. The importance of it is inevitably converging to zero. (Asahi Linux might change the story for Apple silicon Macs, but that’s a future discussion once Fedora can offer the same functionality).

Important note: If you’re not very familiar with the release criteria process, please read this. Removing a release criterion doesn’t mean removing the feature. Even after this change, dual-boot installation to an Intel-based Mac will still work. The difference is that if a problem is found, it will not be considered critical enough to block the release of the next upcoming Fedora. Instead, it will be resolved as any other standard bug.

10 Likes
2

I think the arguments presented are reasonable. +1

3

macOS 26 (releasing this year) is the last version of macOS to support Intel Macs, period. And even that is with a small subset of existing Intel Macs. I’m not sure I’d like to drop the dual-boot release blocking criterion this year, but absolutely next year it probably should be.

1 Like
4

i agree with Neal on this one, since the intel macs will still be supported in the next OSX version (so one more year), we should probably wait and drop it as a release criterion after Apple officially drops support

5

You’re speaking about Macs with T2 chips. But those are unusable on Fedora (as described above), at least if you intend to really use it as a laptop - and otherwise it’s a super niche use case.

Macs with T1 chips (2017 and earlier) are only supported until macOS 13:
MacBook Pro (Intel-based) - Wikipedia
and their security updates should stop this year (2016 models were cut off last year).

So the last few remaining users (with 2017 models) might switch to Fedora now. If F43 support turns out broken (not that likely scenario), they can install F42 (that’s still supported, and functional) and upgrade.

Don’t get me wrong, it would be nice to keep the release-blocking status for at least one more release. But we don’t have the hardware nor the capacity (now), and so I think this is one the best options when we need to cut something, given the circumstances and timelines (ending software support from Apple).

1 Like
6

That’s fair. I’m fine with this, it’s not like we’re dropping the software for correctly registering with the Mac EFI firmware. And the GRUB Mac OS X chainloader doesn’t work anyway.

1 Like
7

+1 from me.

I can confirm from personal experience that Fedora cannot be used OOTB on T2 Intel Macs. And in case of MacBook Pros I am not even sure how well those had been tested against Fedora in the past, given that upon installation keyboard and touchpad don’t work at all.

This was pointed out during F41 Intel-based Mac dual boot testing, when Fedora QA reached out to the community for running some tests:

As a side note, I am still able to run Fedora 42 on an old 2008 Intel MacBook.

1 Like
8

It also works for me on my 2009 and 2014 MacBook Pros. :wink:

However, the battery is completely dead on them, so repairing them for use is not really feasible.

2 Likes
9

I’m a 2018 Intel MacMini owner. The current macOS release macOS 15 “Sequoia” is the last release of MacOS my MacMini and many other Intel Macs will receive. When I look at the previous releases, old MacOS versions always received security updates for at least two more years. So my Mac’s end of life will be most likely in September 2027.

Meanwhile, dual booting the machine will be a legit migration path to Linux for some Mac users. So I personally would not remove the release criterion until macOS 15 “Sequoia” is officially declared dead.

According to this guide, installing Fedora should not be too hard at least on Desktop / All-in-One Macs.

But that’s just my end user’s perspective.

10

If you notice, those instructions specifically require using third-party packages that are not supported by Fedora as well as an external mouse and keyboard. I don’t think that’s a decent argument for maintaining Mac dual-boot support as release-blocking, but it’s still an option that people can use.

2 Likes
11

Mac Mini from 2018 seems to have a T2 chip, so it’s in the same category as the Macbooks discussed above. Of course, Mac Minis don’t have a touchpad and internal keyboard, so that’s not an issue, but the t2linux project list other issues as well (look at “upstream” - that’s Fedora). It might be good enough for many users, sure. But Mac Mini users will already be a small segment compared to Macbook users, and the upstream issues will discourage some users, so we’re likely talking about a really small amount of users who might possibly want to switch to Fedora in upcoming years.

1 Like
12

+1. I don’t see a compelling argument to keep this hardware as release-blocking.

13

Should this proposal pass, will there still be a release blocking validation test for the (non-dual boot) installation of Fedora on Intel Macs? Or will Intel Macs be treated just like any other hardware?

14

There never was a test or release criterion for non-dual boot installation on Macs. Only dual-boot. So if this is accepted, Intel Macs are just regular hardware with no special status.

1 Like
15

This seems reasonable to me. I used to run Fedora on a 2015 Intel-based Macbook (until the battery swole up and that was that) – I didn’t care then if it dual-booted macOS, and I certainly wouldn’t now that Intel-based macs are losing current OS support from Apple.

16

I agree with this proposal.

17

+1 , though I still have a 2015 MacBook Air that Fedora works fine on :stuck_out_tongue:

18

+1 to this proposal; I’ve used the feature in the past, and really appreciated it, but it doesn’t need to be a release blocker anymore given that it targets a specific set of legacy (5-19 y.o.) hardware and requires substantial manual testing.

If there are community members that need it, and we break it, we will hear about it and fix it. I imagine it might be similar to booting on 32-bit EFI: We changed how images were built for F42 and broke it, people noticed, and we fixed it for the next release.

As a final idea, it might make sense to have a “supported boot cases” docs page or test matrix where we keep the various special or legacy boot cases (Mactel, 64-on-EFI32, some BIOS cases, etc.) known so that the community can still know about what’s supported and how to report things being broken.

19

Yes makes sense +1

I don’t think my 2008 or 2012 MacBook Pros would argue with that either.

Way more M series out there it seems.

20

+1, I don’t see any reason for this to continue to be release blocking.