Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Muchun Song <songmuchun@bytedance.com>
To:     akpm@linux-foundation.org, willy@infradead.org, jgg@ziepe.ca,
        jhubbard@nvidia.com, william.kucharski@oracle.com,
        dan.j.williams@intel.com, jack@suse.cz
Cc:     linux-mm@kvack.org, linux-kernel@vger.kernel.org,
        linux-fsdevel@vger.kernel.org, nvdimm@lists.linux.dev,
        Muchun Song <songmuchun@bytedance.com>, stable@vger.kernel.org
Subject: [PATCH v2] mm: fix missing wake-up event for FSDAX pages
Date:   Tue,  5 Jul 2022 20:35:32 +0800
Message-Id: <20220705123532.283-1-songmuchun@bytedance.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

FSDAX page refcounts are 1-based, rather than 0-based: if refcount is
1, then the page is freed.  The FSDAX pages can be pinned through GUP,
then they will be unpinned via unpin_user_page() using a folio variant
to put the page, however, folio variants did not consider this special
case, the result will be to miss a wakeup event (like the user of
__fuse_dax_break_layouts()).  Since FSDAX pages are only possible get
by GUP users, so fix GUP instead of folio_put() to lower overhead.

Fixes: d8ddc099c6b3 ("mm/gup: Add gup_put_folio()")
Suggested-by: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Cc: <stable@vger.kernel.org>
---
v2:
 - Fix GUP instead of folio_put() suggested by Matthew.

 include/linux/mm.h | 14 +++++++++-----
 mm/gup.c           |  6 ++++--
 mm/memremap.c      |  6 +++---
 3 files changed, 16 insertions(+), 10 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 517f9deba56f..b324c9fa2940 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1157,23 +1157,27 @@ static inline bool is_zone_movable_page(const struct page *page)
 #if defined(CONFIG_ZONE_DEVICE) && defined(CONFIG_FS_DAX)
 DECLARE_STATIC_KEY_FALSE(devmap_managed_key);
 
-bool __put_devmap_managed_page(struct page *page);
-static inline bool put_devmap_managed_page(struct page *page)
+bool __put_devmap_managed_page_refs(struct page *page, int refs);
+static inline bool put_devmap_managed_page_refs(struct page *page, int refs)
 {
 	if (!static_branch_unlikely(&devmap_managed_key))
 		return false;
 	if (!is_zone_device_page(page))
 		return false;
-	return __put_devmap_managed_page(page);
+	return __put_devmap_managed_page_refs(page, refs);
 }
-
 #else /* CONFIG_ZONE_DEVICE && CONFIG_FS_DAX */
-static inline bool put_devmap_managed_page(struct page *page)
+static inline bool put_devmap_managed_page_refs(struct page *page, int refs)
 {
 	return false;
 }
 #endif /* CONFIG_ZONE_DEVICE && CONFIG_FS_DAX */
 
+static inline bool put_devmap_managed_page(struct page *page)
+{
+	return put_devmap_managed_page_refs(page, 1);
+}
+
 /* 127: arbitrary random number, small enough to assemble well */
 #define folio_ref_zero_or_close_to_overflow(folio) \
 	((unsigned int) folio_ref_count(folio) + 127u <= 127u)
diff --git a/mm/gup.c b/mm/gup.c
index 4e1999402e71..965ba755023f 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -87,7 +87,8 @@ static inline struct folio *try_get_folio(struct page *page, int refs)
 	 * belongs to this folio.
 	 */
 	if (unlikely(page_folio(page) != folio)) {
-		folio_put_refs(folio, refs);
+		if (!put_devmap_managed_page_refs(&folio->page, refs))
+			folio_put_refs(folio, refs);
 		goto retry;
 	}
 
@@ -176,7 +177,8 @@ static void gup_put_folio(struct folio *folio, int refs, unsigned int flags)
 			refs *= GUP_PIN_COUNTING_BIAS;
 	}
 
-	folio_put_refs(folio, refs);
+	if (!put_devmap_managed_page_refs(&folio->page, refs))
+		folio_put_refs(folio, refs);
 }
 
 /**
diff --git a/mm/memremap.c b/mm/memremap.c
index f0955785150f..58b20c3c300b 100644
--- a/mm/memremap.c
+++ b/mm/memremap.c
@@ -509,7 +509,7 @@ void free_zone_device_page(struct page *page)
 }
 
 #ifdef CONFIG_FS_DAX
-bool __put_devmap_managed_page(struct page *page)
+bool __put_devmap_managed_page_refs(struct page *page, int refs)
 {
 	if (page->pgmap->type != MEMORY_DEVICE_FS_DAX)
 		return false;
@@ -519,9 +519,9 @@ bool __put_devmap_managed_page(struct page *page)
 	 * refcount is 1, then the page is free and the refcount is
 	 * stable because nobody holds a reference on the page.
 	 */
-	if (page_ref_dec_return(page) == 1)
+	if (page_ref_sub_return(page, refs) == 1)
 		wake_up_var(&page->_refcount);
 	return true;
 }
-EXPORT_SYMBOL(__put_devmap_managed_page);
+EXPORT_SYMBOL(__put_devmap_managed_page_refs);
 #endif /* CONFIG_FS_DAX */
-- 
2.11.0