Re: [PATCH v2] infiniband/mm: convert put_page() to put_user_page*()

[Jason Gunthorpe]

On Sun, May 26, 2019 at 04:06:31AM -0700, Matthew Wilcox wrote:
> On Fri, May 24, 2019 at 06:45:22PM -0700, john.hubbard@xxxxxxxxx wrote:
> > For infiniband code that retains pages via get_user_pages*(),
> > release those pages via the new put_user_page(), or
> > put_user_pages*(), instead of put_page()
> 
> I have no objection to this particular patch, but ...
> 
> > This is a tiny part of the second step of fixing the problem described
> > in [1]. The steps are:
> > 
> > 1) Provide put_user_page*() routines, intended to be used
> >    for releasing pages that were pinned via get_user_pages*().
> > 
> > 2) Convert all of the call sites for get_user_pages*(), to
> >    invoke put_user_page*(), instead of put_page(). This involves dozens of
> >    call sites, and will take some time.
> > 
> > 3) After (2) is complete, use get_user_pages*() and put_user_page*() to
> >    implement tracking of these pages. This tracking will be separate from
> >    the existing struct page refcounting.
> > 
> > 4) Use the tracking and identification of these pages, to implement
> >    special handling (especially in writeback paths) when the pages are
> >    backed by a filesystem. Again, [1] provides details as to why that is
> >    desirable.
> 
> I thought we agreed at LSFMM that the future is a new get_user_bvec()
> / put_user_bvec().  This is largely going to touch the same places as
> step 2 in your list above.  Is it worth doing step 2?

I think so, as these two conversions can run in parallel, whichever we
finish first, biovec or put_user_pages lets John progress to step #3

Jason