Re: [PATCH 05/21] x86/mm/pat: mirror direct map changes to ASI

[Brendan Jackman]

On Tue Jan 20, 2026 at 4:37 PM UTC, Borislav Petkov wrote:
> On Thu, Oct 02, 2025 at 05:08:59PM +0000, Brendan Jackman wrote:
>> OK, then it probably just sounds wrong to me because I'm steeped in the
>> current jargon. For v2 I'll try just dropping "[un]restricted".
>
> So AFAIU, we have two address spaces - the full one - which has *everything*
> mapped in and the limited one, with things removed from it. Right?
>
> So calling the unrestricted "the full address space" makes sense to me. I.e.,
> it has *everything* in it.
>
> And then there's the subset of the full address space which has holes in it.
> Looking at Merriam Webster, it suggests those antonyms to "complete":
>
> partial, incomplete, reduced, abbreviated, diminished.
>
> Yahaa, they all make sense.
>
> "partial address space" sounds good to me. "Reduced" even better. So having
> the full and the reduced address space would make the nomenclature very easy,
> IMO. 
>
> Thoughts?

Full/partial sounds like good naming for the address spaces, but that
doesn't help with the issue that we have two related concepts that we
need jargon for:

1. Address spaces (current patch's terminology: restricted/unrestricted)

2. The property of some memory being mapped or unmapped in the
   "restricted"/"partial"/"sensitive" address space (current patch's
   terminology: sensitive/nonsensitive).

IIUC Dave's complaint wasn't that the word choices don't work for their
respective concepts. His issue was that we have separte jargon for 1 and
2 even though they are closely entangled concepts. Since I'm now
convinced that "sensitive/nonsensitive" makes sense for both, I think we
should stick to that.