Another Haiku monthly activity report, but this time around, there’s actually a big ticket item. Haiku has been in a pretty solid and stable state for a while now, so the activity reports have been dominated by fairly small, obscure changes, but during March a major milestone was reached for the ARM64 port.
smrobtzz contributed the bulk of the work, including fixes for building on macOS on ARM64, drivers for the Apple S5L UART, fixes to the kernel base address, clearing the frame pointer before entering the kernel, mapping physical memory correctly, the basics for userland, and more. SED4906 contributed some fixes to the bootloader page mapping, and
runtime_loader’s page-size checks.Combined, these changes allow the ARM64 port to get to the desktop in QEMU. There’s a forum thread, complete with screenshots, for anyone interested in following along.
↫ waddlesplash
While it’s only in QEMU, this is still a major achievement and paves the way for more people to work on the ARM64 port, possibly increasing its health. There’s tons of smaller changes and fixes all over the place, too, as usual, and the team mentions beta 6 isn’t quite ready yet, still. Don’t let that stop you from just downloading the latest nightly, though – Haiku is mature enough to use it.
I am surprised the ARM port took this long. They had RISC-V up and running on real hardware back in 2021.
https://www.haiku-os.org/blog/kallisti5/2021-11-07_booting_our_risc-v_images/
Maybe the documentation for RISC-V is more accessible and there are less boards out there to support.
Like anything else it requires people interested in doing the work and the core developers are not interested in ARM. That’s not really surprising.
@obrado
Haiku on Apple Silicon would be something. A kind of metaphorical return to its roots in a way.
If you’re going to want them to work on ARM targeting the option that is the least likely and the most issues isn’t helping the situation.
Don’t be so sure ; the contributor who’s been working on this has a stated goal of getting it working on their M1 laptop. Not as terrible of a goal as it might seem given all the reverse engineering work that’s been done by Asahi, and I might also note those have the advantage that they’re a lot more standardized in terms of hardware than the wild west of FDTs that is every other random ARM SoC.
@abrado
I “want” them to target RISC-V actually, specifically the new DeepComputing K3 based laptop. To date, I have only run Haiku on x86-64.
But the people doing the work want to run Haiku on Apple Silicon. The first line of their forum post is “I’ve been spending some time improving the arm64 port of Haiku with the goal of some day running Haiku on my M1 MacBook Air.”
The M1 MacBook Air seems like a realistic target to me. It is perhaps one of the better documented ARM platforms due to the tireless efforts of the Asahi Linux team.
When will it run on my Hobbit system?
@AndrewZ
Haiku supports 32 bit, so you could probably do it.
I am curious, did you say this as a joke because the first BeBox systems were Hobbit based? If so, very meta of you.
See the last paragraph of “History” on Wikipedia:
https://en.wikipedia.org/wiki/AT%26T_Hobbit
Yes, a joke. But I did just go down the rabbit/hobbit hole while thinking about BeBox BeOS on the Hobbit. Seems that BeOS was one of the few OS’s to run on it. Hobbit was not very successful.
Your joke becomes meta when considering the Hobbit CPU was designed to compete with the original ARM.
Would it be practical to target the Raspberry Pi in addition to the macs? There’s Linux for the pies, but at least back in the day that relied on binary blobs rather than documented hardware IIRC?
Haiku in one of those pie-in-a-keyboard things could be a pretty neat computer.…
@m_eiman
It has been worked on.
https://www.haiku-os.org/guides/building/port_status/