[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-mm-commits
Subject:    + mm-gup-introduce-pin_user_page.patch added to mm-unstable branch
From:       Andrew Morton <akpm () linux-foundation ! org>
Date:       2022-08-31 23:06:39
Message-ID: 20220831230640.2E867C433C1 () smtp ! kernel ! org
[Download RAW message or body]


The patch titled
     Subject: mm/gup: introduce pin_user_page()
has been added to the -mm mm-unstable branch.  Its filename is
     mm-gup-introduce-pin_user_page.patch

This patch will shortly appear at
     https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-gup-introduce-pin_user_page.patch


This patch will later appear in the mm-unstable branch at
    git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

Before you just go and hit "reply", please:
   a) Consider who else should be cc'ed
   b) Prefer to cc a suitable mailing list as well
   c) Ideally: find the original patch on the mailing list and do a
      reply-to-all to that, adding suitable additional cc's

*** Remember to use Documentation/process/submit-checklist.rst when testing your code \
***

The -mm tree is included into linux-next via the mm-everything
branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
and is updated there every 2-3 working days

------------------------------------------------------
From: John Hubbard <jhubbard@nvidia.com>
Subject: mm/gup: introduce pin_user_page()
Date: Tue, 30 Aug 2022 21:18:38 -0700

pin_user_page() is an externally-usable version of try_grab_page(), but
with semantics that match get_page(), so that it can act as a drop-in
replacement for get_page(). Specifically, pin_user_page() has a void
return type.

pin_user_page() elevates a page's refcount using FOLL_PIN rules. This
means that the caller must release the page via unpin_user_page().

Link: https://lkml.kernel.org/r/20220831041843.973026-3-jhubbard@nvidia.com
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Anna Schumaker <anna@kernel.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Darrick J. Wong <djwong@kernel.org>
Cc: David Hildenbrand <david@redhat.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Logan Gunthorpe <logang@deltatee.com>
Cc: Miklos Szeredi <miklos@szeredi.hu>
Cc: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 include/linux/mm.h |    1 
 mm/gup.c           |   50 +++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 51 insertions(+)

--- a/include/linux/mm.h~mm-gup-introduce-pin_user_page
+++ a/include/linux/mm.h
@@ -1943,6 +1943,7 @@ long pin_user_pages_remote(struct mm_str
 long get_user_pages(unsigned long start, unsigned long nr_pages,
 			    unsigned int gup_flags, struct page **pages,
 			    struct vm_area_struct **vmas);
+void pin_user_page(struct page *page);
 long pin_user_pages(unsigned long start, unsigned long nr_pages,
 		    unsigned int gup_flags, struct page **pages,
 		    struct vm_area_struct **vmas);
--- a/mm/gup.c~mm-gup-introduce-pin_user_page
+++ a/mm/gup.c
@@ -3240,6 +3240,56 @@ long pin_user_pages(unsigned long start,
 }
 EXPORT_SYMBOL(pin_user_pages);
 
+/**
+ * pin_user_page() - apply a FOLL_PIN reference to a file-backed page that the
+ * caller already owns.
+ *
+ * @page: the page to be pinned.
+ *
+ * pin_user_page() elevates a page's refcount using FOLL_PIN rules. This means
+ * that the caller must release the page via unpin_user_page().
+ *
+ * pin_user_page() is intended as a drop-in replacement for get_page(). This
+ * provides a way for callers to do a subsequent unpin_user_page() on the
+ * affected page. However, it is only intended for use by callers (file systems,
+ * block/bio) that have a file-backed page. Anonymous pages are not expected nor
+ * supported, and will generate a warning.
+ *
+ * pin_user_page() may also be thought of as an externally-usable version of
+ * try_grab_page(), but with semantics that match get_page(), so that it can act
+ * as a drop-in replacement for get_page().
+ *
+ * IMPORTANT: The caller must release the page via unpin_user_page().
+ *
+ */
+void pin_user_page(struct page *page)
+{
+	struct folio *folio = page_folio(page);
+
+	WARN_ON_ONCE(folio_ref_count(folio) <= 0);
+
+	/*
+	 * This function is only intended for file-backed callers, who already
+	 * have a page reference.
+	 */
+	WARN_ON_ONCE(PageAnon(page));
+
+	/*
+	 * Similar to try_grab_page(): be sure to *also*
+	 * increment the normal page refcount field at least once,
+	 * so that the page really is pinned.
+	 */
+	if (folio_test_large(folio)) {
+		folio_ref_add(folio, 1);
+		atomic_add(1, folio_pincount_ptr(folio));
+	} else {
+		folio_ref_add(folio, GUP_PIN_COUNTING_BIAS);
+	}
+
+	node_stat_mod_folio(folio, NR_FOLL_PIN_ACQUIRED, 1);
+}
+EXPORT_SYMBOL(pin_user_page);
+
 /*
  * pin_user_pages_unlocked() is the FOLL_PIN variant of
  * get_user_pages_unlocked(). Behavior is the same, except that this one sets
_

Patches currently in -mm which might be from jhubbard@nvidia.com are

mm-change-release_pages-to-use-unsigned-long-for-npages.patch
mm-gup-introduce-pin_user_page.patch
block-add-dio_w_-wrappers-for-pin-unpin-user-pages.patch
iov_iter-new-iov_iter_pin_pages-routines.patch
block-bio-fs-convert-most-filesystems-to-pin_user_pages_fast.patch
nfs-direct-io-convert-to-foll_pin-pages.patch
fuse-convert-direct-io-paths-to-use-foll_pin.patch


[prev in list] [next in list] [prev in thread] [next in thread]