Re: [PATCH v2 2/3] mm: Allow architectures to request 'old' entries when prefaulting

On Mon, Jan 11, 2021 at 02:37:42PM +, Will Deacon wrote: > On Mon, Jan 11, 2021 at 05:25:33PM +0300, Kirill A. Shutemov wrote: > > On Fri, Jan 08, 2021 at 05:15:16PM +, Will Deacon wrote: > > > diff --git a/mm/filemap.c b/mm/filemap.c > > > index c1f2dc89b8a7..0fb9d1714797 100644 > > > ---

Re: [PATCH v2 2/3] mm: Allow architectures to request 'old' entries when prefaulting

On Mon, Jan 11, 2021 at 05:25:33PM +0300, Kirill A. Shutemov wrote: > On Fri, Jan 08, 2021 at 05:15:16PM +, Will Deacon wrote: > > diff --git a/mm/filemap.c b/mm/filemap.c > > index c1f2dc89b8a7..0fb9d1714797 100644 > > --- a/mm/filemap.c > > +++ b/mm/filemap.c > > @@ -3051,14 +3051,18 @@

Re: [PATCH v2 2/3] mm: Allow architectures to request 'old' entries when prefaulting

On Fri, Jan 08, 2021 at 05:15:16PM +, Will Deacon wrote: > diff --git a/mm/filemap.c b/mm/filemap.c > index c1f2dc89b8a7..0fb9d1714797 100644 > --- a/mm/filemap.c > +++ b/mm/filemap.c > @@ -3051,14 +3051,18 @@ vm_fault_t filemap_map_pages(struct vm_fault *vmf, > if

[PATCH v2 2/3] mm: Allow architectures to request 'old' entries when prefaulting

Commit 5c0a85fad949 ("mm: make faultaround produce old ptes") changed the "faultaround" behaviour to initialise prefaulted PTEs as 'old', since this avoids vmscan wrongly assuming that they are hot, despite having never been explicitly accessed by userspace. The change has been shown to benefit