[PATCH v2 3/4] arm64: mm: Don't remap pgtables for allocate vs populate

[Mark Rutland]

On Thu, Apr 11, 2024 at 03:57:04PM +0100, Ryan Roberts wrote:
> On 11/04/2024 15:48, Mark Rutland wrote:
> > On Thu, Apr 11, 2024 at 02:37:49PM +0100, Ryan Roberts wrote:
> >> On 11/04/2024 14:02, Mark Rutland wrote:
> >>> but the logic remains fairly simple, and I suspect the overhead for late
> >>> allocations might not matter since the bulk of late changes are non-allocating.
> >>
> >> Its just the fixmap overhead that remains...
> > 
> > True; my thinking there is that almost all of the later changes are for smaller
> > ranges than the linear map (~10s of MB vs GBs in your test data), so I'd expect
> > the overhead of those to be dominated by the cost of mappin the linear map.
> > 
> > The only big exception is arch_add_memory(), but memory hotplug is incredibly
> > rare, and we're not making it massively slower than it already was...
> 
> What about something like coco guest mem (or whatever its called). Isn't that
> scrubbed out of the linear map? So if a coco VM is started with GBs of memory,
> could that be a real case we want to optimize?

I think that's already handled -- the functions we have to carve portions out
of the linear map use apply_to_page_range(), which doesn't use the fixmap. See
set_memory_*() and set_direct_map_*() in arch/arm64/mm/pageattr.c.

Note that apply_to_page_range() does what its name implies and *only* handles
mappings at page granularity. Hence not using that for
mark_linear_text_alias_ro() and mark_rodata_ro() which need to be able to
handle blocks.

Mark.