Re: [PATCH v1 00/15] Keep track of GUPed pages in fs and block

From: Jerome Glisse
Date: Tue Apr 16 2019 - 15:57:56 EST


On Tue, Apr 16, 2019 at 10:28:40PM +0300, Boaz Harrosh wrote:
> On 16/04/19 22:12, Dan Williams wrote:
> > On Tue, Apr 16, 2019 at 11:59 AM Kent Overstreet
> > <kent.overstreet@xxxxxxxxx> wrote:
> <>
> > This all reminds of the failed attempt to teach the block layer to
> > operate without pages:
> >
> > https://lore.kernel.org/lkml/20150316201640.33102.33761.stgit@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx/
> >
>
> Exactly why I want to make sure it is just a [pointer | flag] and not any kind of pfn
> type. Let us please not go there again?
>
> >>
> >> Question though - why do we need a flag for whether a page is a GUP page or not?
> >> Couldn't the needed information just be determined by what range the pfn is not
> >> (i.e. whether or not it has a struct page associated with it)?
> >
> > That amounts to a pfn_valid() check which is a bit heavier than if we
> > can store a flag in the bv_pfn entry directly.
> >
> > I'd say create a new PFN_* flag, and make bv_pfn a 'pfn_t' rather than
> > an 'unsigned long'.
> >
>
> No, please please not. This is not a pfn and not a pfn_t. It is a page-ptr
> and a flag that says where/how to put_page it. IE I did a GUP on this page
> please do a PUP on this page instead of regular put_page. So no where do I mean
> pfn or pfn_t in this code. Then why?
>
> > That said, I'm still in favor of Jan's proposal to just make the
> > bv_page semantics uniform. Otherwise we're complicating this core
> > infrastructure for some yet to be implemented GPU memory management
> > capabilities with yet to be determined value. Circle back when that
> > value is clear, but in the meantime fix the GUP bug.
> >
>
> I agree there are simpler ways to solve the bugs at hand then
> to system wide separate get_user_page from get_page and force all put_user
> callers to remember what to do. Is there some Document explaining the
> all design of where this is going?
>

A very long thread on this:

https://lkml.org/lkml/2018/12/3/1128

especialy all the reply to this first one

There is also:

https://lkml.org/lkml/2019/3/26/1395
https://lwn.net/Articles/753027/

Cheers,
Jérôme