If this goes through, then it would be one of the biggest acquisitions in tech history. Of course, there are likely to be many ARM licensees that may oppose this such as the following:
Apple
AMD
Samsung (AMD’s mobile partner)
Intel (surprisingly)
Qualcomm
Apple was initially approached, but turned the offer down due to them recognizing that the licensing business would be incompatible with them.
Bad news IMO. Nvidia are pretty terrible for open source… And support. They simply drop some products after a year or two without releasing any more binary blobs to support them. And they don’t reply to questions asking for advice on anything Linux related in their forums. Just my experience with tegra and hybrid graphics products. Although I’d hope that ARM would retain their own personality even with different owners.
ARM does not provide any tool or drivers, they provide an instruction set and implementations of it.
But in any case if you want opensource you should be looking into sparc, risc-V and POWER, not ARM
Actually ARM provide both compilers (gcc, clang etc) and graphics drivers (mali) and other tools. They have a massive compiler team in Cambridge for starters. I’m not talking open hardware, or even open source, just ongoing support for more than one or two point revisions of the Linux kernel would be nice, which is what I’ve seen happen with for example with NVidia T20 and T30.
Well, yes they are. It’s their ethical standopoint that worries me most. Let alone their official, recent, statements about China and Chineese Gov (Microsoft perhaps it’s a bit better here i must admit). Gone are the days of “Don’t be evil” I guess. They’re simply too big for not being evil, that’s a fact
I’m more a PowerPC guy I would say, waiting for this. As a matter of facts, I would have no idea on how and where pick up a SPARC based machine today
Maybe it’s only me, but I still fail to see the end-goal for trying to port Haiku into architectures that no average Jane will use (architectures except x86, x86_64, and ARM). It’s never going to be more than a proof-of-concept, why bother?
Probably I’ll hear back arguments like optimisation for old hardware and stuff, but we all know optimisation is not bound to these hardware.
Anyway, whatever. Time is yours to spend. Number of open tickets still do increase though.
Developers having fun is the main thing here. It’s anice learning experience to know how the hardware is different and what impact it has on our code.
That may come in useful later too. If ARM indeed gets bought and their licensing policy changes, people may turn to other architectures. Mips may make a comeback. PowerPC and sparc may become relevant again. Risc-V may take off sooner than expected. Or some major problem like spectre/meltdown may leave 68000 as the only safe architecture. Who knows what will happen in the next 20 years?
I feel what you are saying. Haiku on a modern Power processor notebook would be a rather pleasing opportunity to glimpse into a future that never was. One that could have been, if Microsoft and Intel hadn’t crushed all their rivals in the 1990s.
