On Fri, May 26, 2023 at 10:41:40PM +0100, David Howells wrote:
> Make pin_user_pages*() leave a ZERO_PAGE unpinned if it extracts a pointer
> to it from the page tables and make unpin_user_page*() correspondingly
> ignore a ZERO_PAGE when unpinning. We don't want to risk overrunning a
> zero page's refcount as we're only allowed ~2 million pins on it -
> something that userspace can conceivably trigger.
>
> Add a pair of functions to test whether a page or a folio is a ZERO_PAGE.
>
> Signed-off-by: David Howells <dhowells@xxxxxxxxxx>
> cc: Christoph Hellwig <hch@xxxxxxxxxxxxx>
> cc: David Hildenbrand <david@xxxxxxxxxx>
> cc: Lorenzo Stoakes <lstoakes@xxxxxxxxx>
> cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
> cc: Jens Axboe <axboe@xxxxxxxxx>
> cc: Al Viro <viro@xxxxxxxxxxxxxxxxxx>
> cc: Matthew Wilcox <willy@xxxxxxxxxxxxx>
> cc: Jan Kara <jack@xxxxxxx>
> cc: Jeff Layton <jlayton@xxxxxxxxxx>
> cc: Jason Gunthorpe <jgg@xxxxxxxxxx>
> cc: Logan Gunthorpe <logang@xxxxxxxxxxxx>
> cc: Hillf Danton <hdanton@xxxxxxxx>
> cc: Christian Brauner <brauner@xxxxxxxxxx>
> cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
> cc: linux-fsdevel@xxxxxxxxxxxxxxx
> cc: linux-block@xxxxxxxxxxxxxxx
> cc: linux-kernel@xxxxxxxxxxxxxxx
> cc: linux-mm@xxxxxxxxx
> ---
>
> Notes:
> ver #3)
> - Move is_zero_page() and is_zero_folio() to mm.h for dependency reasons.
> - Add more comments and adjust the docs.
>
> ver #2)
> - Fix use of ZERO_PAGE().
> - Add is_zero_page() and is_zero_folio() wrappers.
> - Return the zero page obtained, not ZERO_PAGE(0) unconditionally.
>
> Documentation/core-api/pin_user_pages.rst | 6 +++++
> include/linux/mm.h | 26 +++++++++++++++++--
> mm/gup.c | 31 ++++++++++++++++++++++-
> 3 files changed, 60 insertions(+), 3 deletions(-)
>
> diff --git a/Documentation/core-api/pin_user_pages.rst b/Documentation/core-api/pin_user_pages.rst
> index 9fb0b1080d3b..d3c1f6d8c0e0 100644
> --- a/Documentation/core-api/pin_user_pages.rst
> +++ b/Documentation/core-api/pin_user_pages.rst
> @@ -112,6 +112,12 @@ pages:
> This also leads to limitations: there are only 31-10==21 bits available for a
> counter that increments 10 bits at a time.
>
> +* Because of that limitation, special handling is applied to the zero pages
> + when using FOLL_PIN. We only pretend to pin a zero page - we don't alter its
> + refcount or pincount at all (it is permanent, so there's no need). The
> + unpinning functions also don't do anything to a zero page. This is
> + transparent to the caller.
> +
Great! Cheers. :)
> * Callers must specifically request "dma-pinned tracking of pages". In other
> words, just calling get_user_pages() will not suffice; a new set of functions,
> pin_user_page() and related, must be used.
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index 27ce77080c79..3c2f6b452586 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -1910,6 +1910,28 @@ static inline bool page_needs_cow_for_dma(struct vm_area_struct *vma,
> return page_maybe_dma_pinned(page);
> }
>
> +/**
> + * is_zero_page - Query if a page is a zero page
> + * @page: The page to query
> + *
> + * This returns true if @page is one of the permanent zero pages.
> + */
> +static inline bool is_zero_page(const struct page *page)
> +{
> + return is_zero_pfn(page_to_pfn(page));
> +}
> +
> +/**
> + * is_zero_folio - Query if a folio is a zero page
> + * @folio: The folio to query
> + *
> + * This returns true if @folio is one of the permanent zero pages.
> + */
> +static inline bool is_zero_folio(const struct folio *folio)
> +{
> + return is_zero_page(&folio->page);
> +}
> +
> /* MIGRATE_CMA and ZONE_MOVABLE do not allow pin pages */
> #ifdef CONFIG_MIGRATION
> static inline bool is_longterm_pinnable_page(struct page *page)
> @@ -1920,8 +1942,8 @@ static inline bool is_longterm_pinnable_page(struct page *page)
> if (mt == MIGRATE_CMA || mt == MIGRATE_ISOLATE)
> return false;
> #endif
> - /* The zero page may always be pinned */
> - if (is_zero_pfn(page_to_pfn(page)))
> + /* The zero page can be "pinned" but gets special handling. */
> + if (is_zero_page(page))
> return true;
>
> /* Coherent device memory must always allow eviction. */
> diff --git a/mm/gup.c b/mm/gup.c
> index bbe416236593..ad28261dcafd 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -51,7 +51,8 @@ static inline void sanity_check_pinned_pages(struct page **pages,
> struct page *page = *pages;
> struct folio *folio = page_folio(page);
>
> - if (!folio_test_anon(folio))
> + if (is_zero_page(page) ||
> + !folio_test_anon(folio))
> continue;
> if (!folio_test_large(folio) || folio_test_hugetlb(folio))
> VM_BUG_ON_PAGE(!PageAnonExclusive(&folio->page), page);
> @@ -131,6 +132,13 @@ struct folio *try_grab_folio(struct page *page, int refs, unsigned int flags)
> else if (flags & FOLL_PIN) {
> struct folio *folio;
>
> + /*
> + * Don't take a pin on the zero page - it's not going anywhere
> + * and it is used in a *lot* of places.
> + */
> + if (is_zero_page(page))
> + return page_folio(page);
> +
> /*
> * Can't do FOLL_LONGTERM + FOLL_PIN gup fast path if not in a
> * right zone, so fail and let the caller fall back to the slow
> @@ -180,6 +188,8 @@ struct folio *try_grab_folio(struct page *page, int refs, unsigned int flags)
> static void gup_put_folio(struct folio *folio, int refs, unsigned int flags)
> {
> if (flags & FOLL_PIN) {
> + if (is_zero_folio(folio))
> + return;
> node_stat_mod_folio(folio, NR_FOLL_PIN_RELEASED, refs);
> if (folio_test_large(folio))
> atomic_sub(refs, &folio->_pincount);
> @@ -224,6 +234,13 @@ int __must_check try_grab_page(struct page *page, unsigned int flags)
> if (flags & FOLL_GET)
> folio_ref_inc(folio);
> else if (flags & FOLL_PIN) {
> + /*
> + * Don't take a pin on the zero page - it's not going anywhere
> + * and it is used in a *lot* of places.
> + */
> + if (is_zero_page(page))
> + return 0;
> +
> /*
> * Similar to try_grab_folio(): be sure to *also*
> * increment the normal page refcount field at least once,
> @@ -3079,6 +3096,9 @@ EXPORT_SYMBOL_GPL(get_user_pages_fast);
> *
> * FOLL_PIN means that the pages must be released via unpin_user_page(). Please
> * see Documentation/core-api/pin_user_pages.rst for further details.
> + *
> + * Note that if a zero_page is amongst the returned pages, it will not have
> + * pins in it and unpin_user_page() will not remove pins from it.
> */
> int pin_user_pages_fast(unsigned long start, int nr_pages,
> unsigned int gup_flags, struct page **pages)
> @@ -3110,6 +3130,9 @@ EXPORT_SYMBOL_GPL(pin_user_pages_fast);
> *
> * FOLL_PIN means that the pages must be released via unpin_user_page(). Please
> * see Documentation/core-api/pin_user_pages.rst for details.
> + *
> + * Note that if a zero_page is amongst the returned pages, it will not have
> + * pins in it and unpin_user_page*() will not remove pins from it.
> */
> long pin_user_pages_remote(struct mm_struct *mm,
> unsigned long start, unsigned long nr_pages,
> @@ -3143,6 +3166,9 @@ EXPORT_SYMBOL(pin_user_pages_remote);
> *
> * FOLL_PIN means that the pages must be released via unpin_user_page(). Please
> * see Documentation/core-api/pin_user_pages.rst for details.
> + *
> + * Note that if a zero_page is amongst the returned pages, it will not have
> + * pins in it and unpin_user_page*() will not remove pins from it.
> */
> long pin_user_pages(unsigned long start, unsigned long nr_pages,
> unsigned int gup_flags, struct page **pages,
> @@ -3161,6 +3187,9 @@ EXPORT_SYMBOL(pin_user_pages);
> * pin_user_pages_unlocked() is the FOLL_PIN variant of
> * get_user_pages_unlocked(). Behavior is the same, except that this one sets
> * FOLL_PIN and rejects FOLL_GET.
> + *
> + * Note that if a zero_page is amongst the returned pages, it will not have
> + * pins in it and unpin_user_page*() will not remove pins from it.
> */
> long pin_user_pages_unlocked(unsigned long start, unsigned long nr_pages,
> struct page **pages, unsigned int gup_flags)
>
All looks good, thanks for adding comments/updating doc!
Reviewed-by: Lorenzo Stoakes <lstoakes@xxxxxxxxx>
