A CPU without the user-hostile features but still able to run the massive existing software base would be ideal.
And who would make the tens of billions of investment to build a fab for that?
On the other side RISCV instruction set is far simpler, being a RISC instruction set it decides to not have advanced optimizations in the processor (even better, none at all) and leave the optimization work to the compiler, that not only simplifies the processor, but also reduces the surface of attack of the processor (Meltown, Spectre, and all these attacks are just impossible on RISCV!). Of course that has a performance penalty, but since you simplify the processor you can just put more core in the saved space right?
What I'm referring to is the expiration of patents from the P6 era, which would mean all the uop-based stuff is now free to implement.
What a lot of the RISC hype doesn't understand is the huge value in backwards compatibility --- you can have your "100% free" world but it'll forever remain niche. We need to accommodate the proprietary world if we want any chance of freedom winning; and not try to divide the world of computing.
And they will stay at that level, is the parent's point, which is 25+ years before today and thus affordable to clone and fabricate. It's not about 486 or for that matter 8086 beinh difficult...
Anyone that wants to be 100% sure of the supply chain has to move away from globalization.
I still cannot understand why the US and other countries with high tech allowed themselves to become so absolutely dependent on China.
When their own companies killed of local manufacturing and went to China out of greed and increased profits then governments should have stepped in on strategic grounds, so why didn't they?
It was obvious to me some 25 years ago and I've no monopoly on this insight so they must have been well aware that this would happen. In essence, these countries have been shooting themselves in the feet for decades.
The question is why.
Edit: I can remember the time when the US military required certain components, chips etc. to be able to be second-sourced from multiple manufacturers before they were incorporated into equipment. Does anyone else remember this?
Clearly governments, not only the US, have been aware of the problem for many decades and have chosen to do nothing about it. Moreover, what brought them to abandon this once good policy in the first instance?
Otherwise with x86 is more complex: you can choose between Intel and AMD (that has bought the license for the x86 instruction set - not something cheap to get), and both of them had their backdoor processor inside the computer (at least on Intel there are ways to disable it, as far as I know with AMD is more difficult if not impossible to do).
Only the base RISC V is guaranteed thanks extensions.
Also you are forgetting that just like Android and ARM, there are other forces at play that don't make it as easy in practice as FOSS advocates wish for.
I remembered hearing that same line when I bought a Raspberry Pi in 2012. "It's useless! You can't run x86 software on it, so what's the point?"
Flash-forwards a decade and now Graviton instances are blowing up like nothing else in the industry. RISC-V is in a very similar position to ARM 10 years ago; the groundwork has been laid, standards have been ratified and base packages/several kernels work perfectly fine on it. The only difference is that ARM is more expensive to license and is less flexible.
> Only the base RISC V is guaranteed thanks extensions.
Yeah. Is that a problem? The situation on ARM is equally bad if not worse (frequent iterations end up throwing even relatively recent CPU models under the bus), and the reason why RISC-V divided itself into extensions is so that you didn't have to start from scratch when John RISC decides to add in 3 new floating point instructions. It's a pretty damn good compromise if you ask me, and it certainly doesn't have any bearing on software availability; RISC-V programs run on RISC-V processors. ARM does not have that same liberty.
There are plenty of genuine constraints for RISC-V (the majority of them in the manufacturing/mass production side of things, now), but the majority of these software issues have been solved and taped out years ago.
The problem is dreamers thinking RISC V will be any different than other CPUs in the industry when big players come playing.
LLVM, contributions level at the scale of Linux kernel, C++20 support now lags behind everyone else.
When money comes to play, the rainbows and flowers eventually turn into wall street yuppies.
Surely you can't think the architecture itself is the differentiator. x86 didn't have all of this security 20 years ago, give engineers a few years of time to throw some locks on a risc-v chip and it'll be Enterprise Ready™ in no time.