Anon (no.delete@this.spam.com) on October 14, 2021 11:08 am wrote:
> rwessel (rwessel.delete@this.yahoo.com) on October 14, 2021 9:01 am wrote:
> > One of the concerns raised in earlier discussions was moving
> > a running copy to a different core (possibly to
> > a different machine in a cluster, or such). It appears
> > this punts the problem of underlying different native
> > move sizes to the OS fixing up an exception if the parameters end up incorrectly aligned on the new core.
>
> Or the combination of two cores supporting different move sizes will be unsupported, is
> the combination of two cores with different cache line sizes supported to start with?

I think, original ARM intention was to do big.LITTLE only with equal cache line sizes.
But then came Samsung and integrated 128B Mongoose with 64B Cortex-A53. And since it was Samsung, Google, wearing its Android cap, couldn't say them "Go away" as they would probably do to lesser partners.

But since Samsung no longer makes cores of their own, and QC is not back in cores business yet, now is probably a good moment to tighten the rules.