Re: [PATCH v4 09/23] mm/gup: introduce pin_user_pages*() and FOLL_PIN
On 11/13/19 3:22 PM, John Hubbard wrote:
> On 11/13/19 2:43 AM, Jan Kara wrote:
> ...
>> How does FOLL_PIN result in grabbing (at least normal, for now) page
>> reference?
>> I didn't find that anywhere in this patch but it is a prerequisite to
>> converting any user to pin_user_pages() interface, right?
>
>
> ohhh, I messed up on this intermediate patch: it doesn't quite stand alone as
> it should, as you noticed. To correct this, I can do one of the following:
>
> a) move the new pin*() routines into the later patch 16 ("mm/gup:
> track FOLL_PIN pages"), or
>
> b) do a temporary thing here, such as setting FOLL_GET and adding a TODO,
> within the pin*() implementations. And this switching it over to FOLL_PIN
> in patch 16.
>
> I'm thinking (a) is less error-prone, so I'm going with that unless someone
> points out that that is stupid. :)
>
OK, just to save anyone from wasting time reading the above: (a) is, in fact,
stupid, after all. ha. That is because pin_user_pages() is called in the
intervening patches.
So anyway, I'll work out an ordering to fix it up, it's not complicated.
thanks,
--
John Hubbard
NVIDIA
Re: [PATCH v4 09/23] mm/gup: introduce pin_user_pages*() and FOLL_PIN
On 11/13/19 2:43 AM, Jan Kara wrote:
...
How does FOLL_PIN result in grabbing (at least normal, for now) page reference?
I didn't find that anywhere in this patch but it is a prerequisite to
converting any user to pin_user_pages() interface, right?
ohhh, I messed up on this intermediate patch: it doesn't quite stand alone as
it should, as you noticed. To correct this, I can do one of the following:
a) move the new pin*() routines into the later patch 16 ("mm/gup:
track FOLL_PIN pages"), or
b) do a temporary thing here, such as setting FOLL_GET and adding a TODO,
within the pin*() implementations. And this switching it over to FOLL_PIN
in patch 16.
I'm thinking (a) is less error-prone, so I'm going with that unless someone
points out that that is stupid. :)
...
I was somewhat wondering about the number of functions you add here. So we
have:>
pin_user_pages()
pin_user_pages_fast()
pin_user_pages_remote()
and then longterm variants:
pin_longterm_pages()
pin_longterm_pages_fast()
pin_longterm_pages_remote()
and obviously we have gup like:
get_user_pages()
get_user_pages_fast()
get_user_pages_remote()
... and some other gup variants ...
I think we really should have pin_* vs get_* variants as they are very
different in terms of guarantees and after conversion, any use of get_*
variant in non-mm code should be closely scrutinized. OTOH pin_longterm_*
don't look *that* useful to me and just using pin_* instead with
FOLL_LONGTERM flag would look OK to me and somewhat reduce the number of
functions which is already large enough? What do people think? I don't feel
too strongly about this but wanted to bring this up.
Honza
Sounds just right to me, and I see that Dan and Ira also like it.
So I'll proceed with that.
thanks,
--
John Hubbard
NVIDIA
Re: [PATCH v4 09/23] mm/gup: introduce pin_user_pages*() and FOLL_PIN
On 11/13/19 10:59 AM, Ira Weiny wrote: On Tue, Nov 12, 2019 at 08:26:56PM -0800, John Hubbard wrote: Introduce pin_user_pages*() variations of get_user_pages*() calls, and also pin_longterm_pages*() variations. These variants all set FOLL_PIN, which is also introduced, and thoroughly documented. The pin_longterm*() variants also set FOLL_LONGTERM, in addition to FOLL_PIN: pin_user_pages() pin_user_pages_remote() pin_user_pages_fast() pin_longterm_pages() pin_longterm_pages_remote() pin_longterm_pages_fast() At some point in this conversation I thought we were going to put in "unpin_*" versions of these. Is that still in the plans? Why yes it is! :) Daniel Vetter and Jan Kara both already weighed in [1], in favor of "unpin_user_page*()", rather than "put_user_page*()". I'll change those names. [1] https://lore.kernel.org/r/20191113101210.gd6...@quack2.suse.cz thanks, -- John Hubbard NVIDIA
Re: [PATCH v4 09/23] mm/gup: introduce pin_user_pages*() and FOLL_PIN
On Tue, Nov 12, 2019 at 08:26:56PM -0800, John Hubbard wrote: > Introduce pin_user_pages*() variations of get_user_pages*() calls, > and also pin_longterm_pages*() variations. > > These variants all set FOLL_PIN, which is also introduced, and > thoroughly documented. > > The pin_longterm*() variants also set FOLL_LONGTERM, in addition > to FOLL_PIN: > > pin_user_pages() > pin_user_pages_remote() > pin_user_pages_fast() > > pin_longterm_pages() > pin_longterm_pages_remote() > pin_longterm_pages_fast() At some point in this conversation I thought we were going to put in "unpin_*" versions of these. Is that still in the plans? Ira
Re: [PATCH v4 09/23] mm/gup: introduce pin_user_pages*() and FOLL_PIN
On Wed, Nov 13, 2019 at 2:43 AM Jan Kara wrote:
>
> On Tue 12-11-19 20:26:56, John Hubbard wrote:
> > Introduce pin_user_pages*() variations of get_user_pages*() calls,
> > and also pin_longterm_pages*() variations.
> >
> > These variants all set FOLL_PIN, which is also introduced, and
> > thoroughly documented.
> >
> > The pin_longterm*() variants also set FOLL_LONGTERM, in addition
> > to FOLL_PIN:
> >
> > pin_user_pages()
> > pin_user_pages_remote()
> > pin_user_pages_fast()
> >
> > pin_longterm_pages()
> > pin_longterm_pages_remote()
> > pin_longterm_pages_fast()
> >
> > All pages that are pinned via the above calls, must be unpinned via
> > put_user_page().
> >
> > The underlying rules are:
> >
> > * These are gup-internal flags, so the call sites should not directly
> > set FOLL_PIN nor FOLL_LONGTERM. That behavior is enforced with
> > assertions, for the new FOLL_PIN flag. However, for the pre-existing
> > FOLL_LONGTERM flag, which has some call sites that still directly
> > set FOLL_LONGTERM, there is no assertion yet.
> >
> > * Call sites that want to indicate that they are going to do DirectIO
> > ("DIO") or something with similar characteristics, should call a
> > get_user_pages()-like wrapper call that sets FOLL_PIN. These wrappers
> > will:
> > * Start with "pin_user_pages" instead of "get_user_pages". That
> > makes it easy to find and audit the call sites.
> > * Set FOLL_PIN
> >
> > * For pages that are received via FOLL_PIN, those pages must be returned
> > via put_user_page().
> >
> > Thanks to Jan Kara and Vlastimil Babka for explaining the 4 cases
> > in this documentation. (I've reworded it and expanded upon it.)
> >
> > Reviewed-by: Mike Rapoport # Documentation
> > Reviewed-by: Jérôme Glisse
> > Cc: Jonathan Corbet
> > Cc: Ira Weiny
> > Signed-off-by: John Hubbard
>
> Thanks for the documentation. It looks great!
>
> > diff --git a/mm/gup.c b/mm/gup.c
> > index 83702b2e86c8..4409e84dff51 100644
> > --- a/mm/gup.c
> > +++ b/mm/gup.c
> > @@ -201,6 +201,10 @@ static struct page *follow_page_pte(struct
> > vm_area_struct *vma,
> > spinlock_t *ptl;
> > pte_t *ptep, pte;
> >
> > + /* FOLL_GET and FOLL_PIN are mutually exclusive. */
> > + if (WARN_ON_ONCE((flags & (FOLL_PIN | FOLL_GET)) ==
> > + (FOLL_PIN | FOLL_GET)))
> > + return ERR_PTR(-EINVAL);
> > retry:
> > if (unlikely(pmd_bad(*pmd)))
> > return no_page_table(vma, flags);
>
> How does FOLL_PIN result in grabbing (at least normal, for now) page
> reference?
> I didn't find that anywhere in this patch but it is a prerequisite to
> converting any user to pin_user_pages() interface, right?
>
> > +/**
> > + * pin_user_pages_fast() - pin user pages in memory without taking locks
> > + *
> > + * Nearly the same as get_user_pages_fast(), except that FOLL_PIN is set.
> > See
> > + * get_user_pages_fast() for documentation on the function arguments,
> > because
> > + * the arguments here are identical.
> > + *
> > + * FOLL_PIN means that the pages must be released via put_user_page().
> > Please
> > + * see Documentation/vm/pin_user_pages.rst for further details.
> > + *
> > + * This is intended for Case 1 (DIO) in
> > Documentation/vm/pin_user_pages.rst. It
> > + * is NOT intended for Case 2 (RDMA: long-term pins).
> > + */
> > +int pin_user_pages_fast(unsigned long start, int nr_pages,
> > + unsigned int gup_flags, struct page **pages)
> > +{
> > + /* FOLL_GET and FOLL_PIN are mutually exclusive. */
> > + if (WARN_ON_ONCE(gup_flags & FOLL_GET))
> > + return -EINVAL;
> > +
> > + gup_flags |= FOLL_PIN;
> > + return internal_get_user_pages_fast(start, nr_pages, gup_flags,
> > pages);
> > +}
> > +EXPORT_SYMBOL_GPL(pin_user_pages_fast);
>
> I was somewhat wondering about the number of functions you add here. So we
> have:
>
> pin_user_pages()
> pin_user_pages_fast()
> pin_user_pages_remote()
>
> and then longterm variants:
>
> pin_longterm_pages()
> pin_longterm_pages_fast()
> pin_longterm_pages_remote()
>
> and obviously we have gup like:
> get_user_pages()
> get_user_pages_fast()
> get_user_pages_remote()
> ... and some other gup variants ...
>
> I think we really should have pin_* vs get_* variants as they are very
> different in terms of guarantees and after conversion, any use of get_*
> variant in non-mm code should be closely scrutinized. OTOH pin_longterm_*
> don't look *that* useful to me and just using pin_* instead with
> FOLL_LONGTERM flag would look OK to me and somewhat reduce the number of
> functions which is already large enough? What do people think? I don't feel
> too strongly about this but wanted to bring this up.
I'd vote for FOLL_LONGTERM should obviate the need for
{get,pin}_user_pages_longterm(). It's a property that is passed by the
call site, not an internal flag.
Re: [PATCH v4 09/23] mm/gup: introduce pin_user_pages*() and FOLL_PIN
> > +/**
> > + * pin_user_pages_fast() - pin user pages in memory without taking locks
> > + *
> > + * Nearly the same as get_user_pages_fast(), except that FOLL_PIN is set.
> > See
> > + * get_user_pages_fast() for documentation on the function arguments,
> > because
> > + * the arguments here are identical.
> > + *
> > + * FOLL_PIN means that the pages must be released via put_user_page().
> > Please
> > + * see Documentation/vm/pin_user_pages.rst for further details.
> > + *
> > + * This is intended for Case 1 (DIO) in
> > Documentation/vm/pin_user_pages.rst. It
> > + * is NOT intended for Case 2 (RDMA: long-term pins).
> > + */
> > +int pin_user_pages_fast(unsigned long start, int nr_pages,
> > + unsigned int gup_flags, struct page **pages)
> > +{
> > + /* FOLL_GET and FOLL_PIN are mutually exclusive. */
> > + if (WARN_ON_ONCE(gup_flags & FOLL_GET))
> > + return -EINVAL;
> > +
> > + gup_flags |= FOLL_PIN;
> > + return internal_get_user_pages_fast(start, nr_pages, gup_flags, pages);
> > +}
> > +EXPORT_SYMBOL_GPL(pin_user_pages_fast);
>
> I was somewhat wondering about the number of functions you add here. So we
> have:
>
> pin_user_pages()
> pin_user_pages_fast()
> pin_user_pages_remote()
>
> and then longterm variants:
>
> pin_longterm_pages()
> pin_longterm_pages_fast()
> pin_longterm_pages_remote()
>
> and obviously we have gup like:
> get_user_pages()
> get_user_pages_fast()
> get_user_pages_remote()
> ... and some other gup variants ...
>
> I think we really should have pin_* vs get_* variants as they are very
> different in terms of guarantees and after conversion, any use of get_*
> variant in non-mm code should be closely scrutinized. OTOH pin_longterm_*
> don't look *that* useful to me and just using pin_* instead with
> FOLL_LONGTERM flag would look OK to me and somewhat reduce the number of
> functions which is already large enough? What do people think? I don't feel
> too strongly about this but wanted to bring this up.
I'm a bit concerned with the function explosion myself. I think what you
suggest is a happy medium. So I'd be ok with that.
Ira
Re: [PATCH v4 09/23] mm/gup: introduce pin_user_pages*() and FOLL_PIN
On Tue 12-11-19 20:26:56, John Hubbard wrote:
> Introduce pin_user_pages*() variations of get_user_pages*() calls,
> and also pin_longterm_pages*() variations.
>
> These variants all set FOLL_PIN, which is also introduced, and
> thoroughly documented.
>
> The pin_longterm*() variants also set FOLL_LONGTERM, in addition
> to FOLL_PIN:
>
> pin_user_pages()
> pin_user_pages_remote()
> pin_user_pages_fast()
>
> pin_longterm_pages()
> pin_longterm_pages_remote()
> pin_longterm_pages_fast()
>
> All pages that are pinned via the above calls, must be unpinned via
> put_user_page().
>
> The underlying rules are:
>
> * These are gup-internal flags, so the call sites should not directly
> set FOLL_PIN nor FOLL_LONGTERM. That behavior is enforced with
> assertions, for the new FOLL_PIN flag. However, for the pre-existing
> FOLL_LONGTERM flag, which has some call sites that still directly
> set FOLL_LONGTERM, there is no assertion yet.
>
> * Call sites that want to indicate that they are going to do DirectIO
> ("DIO") or something with similar characteristics, should call a
> get_user_pages()-like wrapper call that sets FOLL_PIN. These wrappers
> will:
> * Start with "pin_user_pages" instead of "get_user_pages". That
> makes it easy to find and audit the call sites.
> * Set FOLL_PIN
>
> * For pages that are received via FOLL_PIN, those pages must be returned
> via put_user_page().
>
> Thanks to Jan Kara and Vlastimil Babka for explaining the 4 cases
> in this documentation. (I've reworded it and expanded upon it.)
>
> Reviewed-by: Mike Rapoport # Documentation
> Reviewed-by: Jérôme Glisse
> Cc: Jonathan Corbet
> Cc: Ira Weiny
> Signed-off-by: John Hubbard
Thanks for the documentation. It looks great!
> diff --git a/mm/gup.c b/mm/gup.c
> index 83702b2e86c8..4409e84dff51 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -201,6 +201,10 @@ static struct page *follow_page_pte(struct
> vm_area_struct *vma,
> spinlock_t *ptl;
> pte_t *ptep, pte;
>
> + /* FOLL_GET and FOLL_PIN are mutually exclusive. */
> + if (WARN_ON_ONCE((flags & (FOLL_PIN | FOLL_GET)) ==
> + (FOLL_PIN | FOLL_GET)))
> + return ERR_PTR(-EINVAL);
> retry:
> if (unlikely(pmd_bad(*pmd)))
> return no_page_table(vma, flags);
How does FOLL_PIN result in grabbing (at least normal, for now) page reference?
I didn't find that anywhere in this patch but it is a prerequisite to
converting any user to pin_user_pages() interface, right?
> +/**
> + * pin_user_pages_fast() - pin user pages in memory without taking locks
> + *
> + * Nearly the same as get_user_pages_fast(), except that FOLL_PIN is set. See
> + * get_user_pages_fast() for documentation on the function arguments, because
> + * the arguments here are identical.
> + *
> + * FOLL_PIN means that the pages must be released via put_user_page(). Please
> + * see Documentation/vm/pin_user_pages.rst for further details.
> + *
> + * This is intended for Case 1 (DIO) in Documentation/vm/pin_user_pages.rst.
> It
> + * is NOT intended for Case 2 (RDMA: long-term pins).
> + */
> +int pin_user_pages_fast(unsigned long start, int nr_pages,
> + unsigned int gup_flags, struct page **pages)
> +{
> + /* FOLL_GET and FOLL_PIN are mutually exclusive. */
> + if (WARN_ON_ONCE(gup_flags & FOLL_GET))
> + return -EINVAL;
> +
> + gup_flags |= FOLL_PIN;
> + return internal_get_user_pages_fast(start, nr_pages, gup_flags, pages);
> +}
> +EXPORT_SYMBOL_GPL(pin_user_pages_fast);
I was somewhat wondering about the number of functions you add here. So we
have:
pin_user_pages()
pin_user_pages_fast()
pin_user_pages_remote()
and then longterm variants:
pin_longterm_pages()
pin_longterm_pages_fast()
pin_longterm_pages_remote()
and obviously we have gup like:
get_user_pages()
get_user_pages_fast()
get_user_pages_remote()
... and some other gup variants ...
I think we really should have pin_* vs get_* variants as they are very
different in terms of guarantees and after conversion, any use of get_*
variant in non-mm code should be closely scrutinized. OTOH pin_longterm_*
don't look *that* useful to me and just using pin_* instead with
FOLL_LONGTERM flag would look OK to me and somewhat reduce the number of
functions which is already large enough? What do people think? I don't feel
too strongly about this but wanted to bring this up.
Honza
--
Jan Kara
SUSE Labs, CR
