Re: [PATCH] mm: fix COW mapping handing in generic_access_phys

[Jason Gunthorpe <jgg@nvidia.com>]

On Wed, May 28, 2025 at 02:47:54PM -0300, Jason Gunthorpe wrote:
> On Wed, May 28, 2025 at 07:32:59PM +0200, David Hildenbrand wrote:
> > On 28.05.25 18:29, Jason Gunthorpe wrote:
> > > On Wed, May 28, 2025 at 12:06:07PM -0400, Peter Xu wrote:
> > > > #define VM_PFNMAP	0x00000400	/* Page-ranges managed without "struct page", just pure PFN */
> > > > 
> > > > I'm not confident to blame any driver yet to have those special cases for
> > > > VM_PFNMAP, because it only says "managed without struct page", it didn't
> > > > say "it must not contain struct page"..  Hence it hints the core mm "please
> > > > do not manage these mappings with struct page at all".  Still sounds fair
> > > > contract, even if not ideal.
> > > 
> > > I think it is pretty clear, if a VMA has VM_PFNMAP then nothing must
> > > ever try to obtain a struct page from any PTEs in it, for any reason,
> > > even if things in it might have a struct page. In practice it means
> > > nothing can call vm_normal_page() on a VM_PFNMAP.
> > 
> > No, not until we remove any COW mappings of VM_PFNMAP.
> 
> I stand by the statement, the COW mapping thing is broken. Just
> because it is wrong doesn't mean it is allowed to call
> vm_normal_page() on VM_PFNMAP.

I guess I got this backwards in my memory, the vm_normal_page() thing
is specifically to fix the COW nastyness on VM_PFNMAP.

So it is OK to call vm_normal_page() on both kinds of VMAs and it will
return NULL if you are not allowed to touch the struct page.

This is still what GUP should be doing, vm_normal_page() and if no
struct page comes back then failure. It should never try to get a
struct page some other way.

Yes, GUP fast will be inconsistent with GUP slow on non-special
arches, but that is unfixable without taking a performance hit.

Jason