Patrick Chase (patrickjchase.delete@this.gmail.com) on May 20, 2013 9:29 am wrote:
[snip]
> I think you can objectively state that the A15 has server-oriented features, most notably ARM's version
> of PAE (can't remember what they called it). They wouldn't have put that in if A15 had been targeted
> purely as a mobile application processor, so there appears to be at least some underlying grain of
> truth to David's argument. It's a matter of temporal order (which came first) and degree.

ARM's Virtualization Extention (which tends to integrate with the Large Physical Address Extension) also seems to be a server-oriented feature.

While virtualization might be useful in some embedded systems to allow a single core to be used by more than one operating system without software changes, such would presumably exclude replacing Cortex-M (because of area/power costs?) and Cortex-R (because of timing predictability issues?) cores and would likely require support for reasonably fine-grained multithreading (which is lacking in A15). Also, while a core designed for such a use would need to be higher performance than the multiple cores it would be replacing, A15 might be a little too high in performance. The market for such use of virtualization would also probably be rather small--insufficient to justify development effort (especially [I think] with no multithreading support in the ISA or even in implementations).

> What I actually suspect happened is that A15 started out targeted as a mobile application processor, but then
> microservers happened and ARM tried to stretch it in that direction during development. PAE in particular
> seemed to be a late-ish addition. They would have been silly if they hadn't done that, so this isn't any sort
> of negative comment towards/about ARM. This is one case where I agree 100% with their execution.

It has difficult to tell how much LPAE was a stop-gap measure with ARM realizing that a 32-bit physical address space was becoming close to insufficient even for tablets. ARM seems to have been very late in announcing its 64-bit ISA; providing LPAE might have facilitated delaying such announcements and ISA details (possibly to avoid an Osborne Effect, possibly to avoid rushing a 64-bit ISA into market and being stuck supporting any wrong choices for years to come, possibly [but unlikely] to have a more continual flow of press releases, possibly [silly guessing can sometimes be a kind of fun] to leverage commitments among major licensees with earlier access to documentation and perhaps even the ISA definition phase).