anon2 (anon.delete@this.anon.com) on September 22, 2021 4:09 pm wrote:
> Linus Torvalds (torvalds.delete@this.linux-foundation.org) on September 22, 2021 11:50 am wrote:
> > sr (nobody.delete@this.nowhere.com) on September 22, 2021 6:55 am wrote:
> > >
> > > Why would segments need different page tables? What would
> > > be point of doing so instead of different processes?
> >
> > A segment (or whatever you want to call it: as mentioned,
> > people don't like some of the historical connotations,
> > and depending on which part of the problem you care about you might want to call it a "capability pointer")
> > is much more interesting if it gives you access to other peoples address spaces.
>
> PowerPC does this. Maybe it's good for a special case of database with big shared memory buffer cache
> segment that is mapped by a lot of processes. Aside from that, how interesting is it really?

That's basically all the PowerPC facility is for. Well really just for having enormous pages and saving on translation. Capabilities are something quite different and much more powerful but require good support for their use. A recent example is
https://www.cl.cam.ac.uk/research/security/ctsrd/cheri/
and Arm have produced some chips with this for research purposes
https://developer.arm.com/architectures/cpu-architecture/a-profile/morello
I wouldn't be surprisedd if Arm eventually produce an Armv10 with this or somethingvery like it in.