On Fri, Jun 09, 2017 at 02:23:51PM -0700, Dan Williams wrote:
> On Wed, Jun 7, 2017 at 1:48 PM, Ross Zwisler
> <ross.zwisler@xxxxxxxxxxxxxxx> wrote:
> > To be able to use the common 4k zero page in DAX we need to have our PTE
> > fault path look more like our PMD fault path where a PTE entry can be
> > marked as dirty and writeable as it is first inserted, rather than waiting
> > for a follow-up dax_pfn_mkwrite() => finish_mkwrite_fault() call.
> >
> > Right now we can rely on having a dax_pfn_mkwrite() call because we can
> > distinguish between these two cases in do_wp_page():
> >
> > case 1: 4k zero page => writable DAX storage
> > case 2: read-only DAX storage => writeable DAX storage
> >
> > This distinction is made by via vm_normal_page(). vm_normal_page() returns
> > false for the common 4k zero page, though, just as it does for DAX ptes.
> > Instead of special casing the DAX + 4k zero page case, we will simplify our
> > DAX PTE page fault sequence so that it matches our DAX PMD sequence, and
> > get rid of dax_pfn_mkwrite() completely.
> >
> > This means that insert_pfn() needs to follow the lead of insert_pfn_pmd()
> > and allow us to pass in a 'mkwrite' flag. If 'mkwrite' is set insert_pfn()
> > will do the work that was previously done by wp_page_reuse() as part of the
> > dax_pfn_mkwrite() call path.
> >
> > Signed-off-by: Ross Zwisler <ross.zwisler@xxxxxxxxxxxxxxx>
> > ---
> > include/linux/mm.h | 9 +++++++--
> > mm/memory.c | 21 ++++++++++++++-------
> > 2 files changed, 21 insertions(+), 9 deletions(-)
> >
> > diff --git a/include/linux/mm.h b/include/linux/mm.h
> > index b892e95..11e323a 100644
> > --- a/include/linux/mm.h
> > +++ b/include/linux/mm.h
> > @@ -2294,10 +2294,15 @@ int vm_insert_pfn(struct vm_area_struct *vma, unsigned long addr,
> > unsigned long pfn);
> > int vm_insert_pfn_prot(struct vm_area_struct *vma, unsigned long addr,
> > unsigned long pfn, pgprot_t pgprot);
> > -int vm_insert_mixed(struct vm_area_struct *vma, unsigned long addr,
> > - pfn_t pfn);
> > +int vm_insert_mixed_mkwrite(struct vm_area_struct *vma, unsigned long addr,
> > + pfn_t pfn, bool mkwrite);
>
> Are there any other planned public users of vm_insert_mixed_mkwrite()
> that would pass false? I think not.
>
> > int vm_iomap_memory(struct vm_area_struct *vma, phys_addr_t start, unsigned long len);
> >
> > +static inline int vm_insert_mixed(struct vm_area_struct *vma,
> > + unsigned long addr, pfn_t pfn)
> > +{
> > + return vm_insert_mixed_mkwrite(vma, addr, pfn, false);
> > +}
>
> ...in other words instead of making the distinction of
> vm_insert_mixed_mkwrite() and vm_insert_mixed() with extra flag
> argument just move the distinction into mm/memory.c directly.
>
> So, the prototype remains the same as vm_insert_mixed()
>
> int vm_insert_mixed_mkwrite(struct vm_area_struct *vma, unsigned long
> addr, pfn_t pfn);
>
> ...and only static insert_pfn(...) needs to change.
My usage of vm_insert_mixed_mkwrite() in fs/dax.c needs the mkwrite flag to be
there. From dax_insert_mapping():
return vm_insert_mixed_mkwrite(vma, vaddr, pfn,
vmf->flags & FAULT_FLAG_WRITE);
So, yes, we could do what you suggest, but then that code becomes:
if (vmf->flags & FAULT_FLAG_WRITE)
vm_insert_mixed_mkwrite(vma, vaddr, pfn);
else
vm_insert_mixed(vma, vaddr, pfn);
And vm_insert_mixed_mkwrite() and vm_insert_mixed() are redundant with only
the insert_pfn() line differing? This doesn't seem better...unless I'm
missing something?
The way it is, vm_insert_mixed_mkwrite() also closely matches
insert_pfn_pmd(), which we use in the PMD case and which also takes a 'write'
boolean which works the same as our newly added 'mkwrite'.
