Re: [PATCH RFC 01/14] mm/memory: drop highest_memmap_pfn sanity check in vm_normal_page()

[Oscar Salvador]

On Mon, Jun 23, 2025 at 04:04:01PM +0200, David Hildenbrand wrote:
> Hi Oscar,

Hi David,

> 
> > I'm confused, I'm missing something here.
> > Before this change we would return NULL if e.g: pfn > highest_memmap_pfn, but
> > now we just print the warning and call pfn_to_page() anyway.
> > AFAIK, pfn_to_page() doesn't return NULL?
> 
> You're missing that vm_normal_page_pmd() was created as a copy from
> vm_normal_page() [history of the sanity check above], but as we don't have
> (and shouldn't have ...) print_bad_pmd(), we made the code look like this
> would be something that can just happen.

Ok I see.

> 
> "
> Do the same in vm_normal_page_pmd(), where we don't even report a
> problem at all ...
> "
> 
> So we made something that should never happen a runtime sanity check without
> ever reporting a problem ...

all clear now, thanks :-)



-- 
Oscar Salvador
SUSE Labs