From: David Hildenbrand <david@redhat.com>
To: lizhe.67@bytedance.com, alex.williamson@redhat.com,
akpm@linux-foundation.org, peterx@redhat.com
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH v4 2/3] gup: introduce unpin_user_folio_dirty_locked()
Date: Tue, 17 Jun 2025 09:35:00 +0200 [thread overview]
Message-ID: <e09ed520-6b52-4b53-b6bb-dfee8f34be5b@redhat.com> (raw)
In-Reply-To: <20250617041821.85555-3-lizhe.67@bytedance.com>
On 17.06.25 06:18, lizhe.67@bytedance.com wrote:
> From: Li Zhe <lizhe.67@bytedance.com>
>
> Introduce a new interface, unpin_user_folio_dirty_locked(). This
> interface is similar to unpin_user_folio(), but it adds the
> capability to conditionally mark a folio as dirty. VFIO will utilize
> this interface to accelerate the VFIO DMA unmap process.
>
> Suggested-by: David Hildenbrand <david@redhat.com>
> Signed-off-by: Li Zhe <lizhe.67@bytedance.com>
> ---
> include/linux/mm.h | 2 ++
> mm/gup.c | 27 +++++++++++++++++++++------
> 2 files changed, 23 insertions(+), 6 deletions(-)
>
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index fdda6b16263b..242b05671502 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -1689,6 +1689,8 @@ void unpin_user_page_range_dirty_lock(struct page *page, unsigned long npages,
> bool make_dirty);
> void unpin_user_pages(struct page **pages, unsigned long npages);
> void unpin_user_folio(struct folio *folio, unsigned long npages);
> +void unpin_user_folio_dirty_locked(struct folio *folio,
> + unsigned long npins, bool make_dirty);
> void unpin_folios(struct folio **folios, unsigned long nfolios);
>
> static inline bool is_cow_mapping(vm_flags_t flags)
> diff --git a/mm/gup.c b/mm/gup.c
> index 84461d384ae2..15debead5f5b 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -360,12 +360,7 @@ void unpin_user_page_range_dirty_lock(struct page *page, unsigned long npages,
>
> for (i = 0; i < npages; i += nr) {
> folio = gup_folio_range_next(page, npages, i, &nr);
> - if (make_dirty && !folio_test_dirty(folio)) {
> - folio_lock(folio);
> - folio_mark_dirty(folio);
> - folio_unlock(folio);
> - }
> - gup_put_folio(folio, nr, FOLL_PIN);
> + unpin_user_folio_dirty_locked(folio, nr, make_dirty);
> }
> }
> EXPORT_SYMBOL(unpin_user_page_range_dirty_lock);
> @@ -435,6 +430,26 @@ void unpin_user_folio(struct folio *folio, unsigned long npages)
> }
> EXPORT_SYMBOL(unpin_user_folio);
>
> +/**
> + * unpin_user_folio_dirty_locked() - conditionally mark a folio
> + * dirty and unpin it
> + *
> + * @folio: pointer to folio to be released
> + * @npins: number of pins
> + * @make_dirty: whether to mark the folio dirty
> + *
> + * Mark the folio as being modified if @make_dirty is true. Then
> + * release npins of the folio.
> + */
> +void unpin_user_folio_dirty_locked(struct folio *folio,
> + unsigned long npins, bool make_dirty)
> +{
> + if (make_dirty && !folio_test_dirty(folio))
> + folio_mark_dirty_lock(folio);
> + gup_put_folio(folio, npins, FOLL_PIN);
> +}
> +EXPORT_SYMBOL_GPL(unpin_user_folio_dirty_locked);
> +
> /**
> * unpin_folios() - release an array of gup-pinned folios.
> * @folios: array of folios to be marked dirty and released.
Acked-by: David Hildenbrand <david@redhat.com>
--
Cheers,
David / dhildenb
next prev parent reply other threads:[~2025-06-17 7:35 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-06-17 4:18 [PATCH v4 0/3] optimize vfio_unpin_pages_remote() for large folio lizhe.67
2025-06-17 4:18 ` [PATCH v4 1/3] vfio/type1: batch vfio_find_vpfn() in function vfio_unpin_pages_remote() lizhe.67
2025-06-17 4:18 ` [PATCH v4 2/3] gup: introduce unpin_user_folio_dirty_locked() lizhe.67
2025-06-17 7:35 ` David Hildenbrand [this message]
2025-06-17 13:42 ` Jason Gunthorpe
2025-06-17 13:45 ` David Hildenbrand
2025-06-17 13:58 ` David Hildenbrand
2025-06-17 14:04 ` David Hildenbrand
2025-06-17 15:22 ` Jason Gunthorpe
2025-06-18 6:28 ` lizhe.67
2025-06-18 8:20 ` David Hildenbrand
2025-06-18 11:36 ` Jason Gunthorpe
2025-06-18 11:40 ` David Hildenbrand
2025-06-18 11:42 ` David Hildenbrand
2025-06-18 11:46 ` Jason Gunthorpe
2025-06-18 11:52 ` David Hildenbrand
2025-06-18 11:56 ` Jason Gunthorpe
2025-06-18 12:19 ` lizhe.67
2025-06-18 13:23 ` Jason Gunthorpe
2025-06-19 9:05 ` lizhe.67
2025-06-19 12:35 ` Jason Gunthorpe
2025-06-19 12:49 ` lizhe.67
2025-06-17 4:18 ` [PATCH v4 3/3] vfio/type1: optimize vfio_unpin_pages_remote() for large folio lizhe.67
2025-06-17 7:43 ` David Hildenbrand
2025-06-17 9:21 ` [PATCH v4 2/3] gup: introduce unpin_user_folio_dirty_locked() lizhe.67
2025-06-17 9:27 ` David Hildenbrand
2025-06-17 9:47 ` [PATCH v4 3/3] vfio/type1: optimize vfio_unpin_pages_remote() for large folio lizhe.67
2025-06-17 9:49 ` David Hildenbrand
2025-06-17 12:42 ` lizhe.67
2025-06-17 13:47 ` David Hildenbrand
2025-06-18 6:11 ` lizhe.67
2025-06-18 7:22 ` lizhe.67
2025-06-18 8:54 ` David Hildenbrand
2025-06-18 9:39 ` lizhe.67
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=e09ed520-6b52-4b53-b6bb-dfee8f34be5b@redhat.com \
--to=david@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=alex.williamson@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lizhe.67@bytedance.com \
--cc=peterx@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).