I’d like to know the status of the ARM port of Haiku. Reading the activity report from June 2021, it seems work si being done by Han Pengfei but I’m not sure what’s the target hardware and what’s the progress of the port.
I’d like to offer my help if a Raspberry Pi 4 (and Qemu on x64) Is enough for that.
I’m more of a frontend developer but have some knowledge of C/C++ and am very much interested in hardware so maybe I can be of some help.
Check out the thread on the topic and see if there’s a way for you to contribute. My bet is that the current work is being done in an emulator, and that they’re probably not at the hardware stage yet. But support for RPi seems to be in demand, and would be beneficial. It’s just a matter of who has time for what.
arm (32 bit): it have mostly working EFI boot loader and it currently crash on early kernel initialization, kernel serial output is working. It has partially working interrupt controller driver and MMU page table implementation that was done on previous attepmts on ARM. It should be possible to adjust them so it will work again.
“The core Haiku team throws $1000 bucks in hardware at a developer to finish a RISC V port that no one is going to use”.
Meanwhile, the low hanging fruit for a practical Haiku hardware port out there is for the millions of Raspberry Pi 4’s already sold and in use. For $125 USD, you can buy a Pi 400 kit that only needs a monitor to work for a full-fledged computer.
I have no doubt X512 is smart enough to get the ARM port working on a Pi 4 in the same amount of time it took him for the RISC V. X512, virtually no one cares about RISC V from a user level point of view. Meanwhile, MILLIONS care about what can run on a Raspberry Pi.
For once, can’t someone with high skill level to move the Haiku Project forward at the hardware level become a master of the obvious?
Hey, I’m one of those 'virtually no one’s - I have a few Raspberry Pis lying around even, but the hardware is very proprietary - it’s not about what’s available to hand, IMO, but the future possibilities of a platform.
The board X512 has, for example, allowed reuse of the Radeon drivers, since it has a PCI express slot - I doubt the GPU of the Raspberry Pi will be anything like as easy to get going.
Nobody asked Haiku Inc to fund them a rpi 4 for porting, so I don’t see why discussion boards care that the Inc approved the one request it actually did get.
It seems very strange to me to complain about a platform getting support.
Redmond said the group now has 435 companies in the RISC-V Foundation — 44 of those are chip companies, nine are input-output companies, 32 are research groups at universities, 25 are software companies, 31 are services firms, and the rest are from various industries. Commercial tool providers are supporting the group. In an interview with VentureBeat, Redmond said such growth hasn’t happened with other architectures in years.
And not one of them has stepped up to work on a port of Haiku
How is that possible?
If someone is interested in making progress on the ARM port, submits some work making progress on it, and request funding for a Raspberry Pi, I’m sure Haiku inc would fund it. And even if they don’t, probably other users would (that is how I got several variants of SPARC hardware donated to me).
It is quite a heavy lift, and a lot of RPi users probably aren’t at the level of tinkering with boot loaders and kernels, and understanding those types of errors.
I’m pretty sure the RISC-V port helped finding & fixing bugs on other platforms as well. So even though they may be less users out there it’s already proved useful.
Anyway, I’ll install Haiku in a VM: I like the idea of working on ARM port in from Haiku.
If each one of these users donated $1 to Haiku, I’m certain the RPi would get done
Haiku is a volunteer project, and developers invest their own free time in the project, which they take away from friends and family and other leisure activities. Dev volunteers will scratch whatever dev itch they may have.