[PATCH v3 00/18] Allow readpage to return a locked page

[PATCH v3 00/18] Allow readpage to return a locked page

[Matthew Wilcox (Oracle)]

I've dropped the conversion of readpage implementations to synchronous
from this patchset.  I realised I'd neglected the requirement for making
the sleep killable, and that turns out to be more convoluted to fix.

So these patches add:
 - An error-path bugfix for cachefiles.
 - The ability for the filesystem to tell the caller of ->readpage
   that the read was successful and the page was not unlocked.  This is
   a performance improvement for some scenarios that I think are rare.
 - Mildly improved error handling for ext4.

v2: https://lore.kernel.org/linux-fsdevel/20201009143104.22673-1-willy@infradead.org/
v1: https://lore.kernel.org/linux-fsdevel/20200917151050.5363-1-willy@infradead.org/
Matthew Wilcox (Oracle) (18):
  cachefiles: Handle readpage error correctly
  swap: Call aops->readahead if appropriate
  fs: Add AOP_UPDATED_PAGE return value
  mm/filemap: Inline wait_on_page_read into its one caller
  9p: Tell the VFS that readpage was synchronous
  afs: Tell the VFS that readpage was synchronous
  ceph: Tell the VFS that readpage was synchronous
  cifs: Tell the VFS that readpage was synchronous
  cramfs: Tell the VFS that readpage was synchronous
  ecryptfs: Tell the VFS that readpage was synchronous
  ext4: Tell the VFS that readpage was synchronous
  ext4: Return error from ext4_readpage
  fuse: Tell the VFS that readpage was synchronous
  hostfs: Tell the VFS that readpage was synchronous
  jffs2: Tell the VFS that readpage was synchronous
  ubifs: Tell the VFS that readpage was synchronous
  udf: Tell the VFS that readpage was synchronous
  vboxsf: Tell the VFS that readpage was synchronous

 Documentation/filesystems/locking.rst |  7 +++---
 Documentation/filesystems/vfs.rst     | 21 +++++++++++------
 fs/9p/vfs_addr.c                      |  6 ++++-
 fs/afs/file.c                         |  3 ++-
 fs/buffer.c                           | 15 +++++++-----
 fs/cachefiles/rdwr.c                  |  9 ++++++++
 fs/ceph/addr.c                        |  9 ++++----
 fs/cifs/file.c                        |  8 +++++--
 fs/cramfs/inode.c                     |  5 ++--
 fs/ecryptfs/mmap.c                    | 11 +++++----
 fs/ext4/inline.c                      |  9 +++++---
 fs/ext4/readpage.c                    | 24 +++++++++++--------
 fs/fuse/file.c                        |  2 ++
 fs/hostfs/hostfs_kern.c               |  2 ++
 fs/jffs2/file.c                       |  6 +++--
 fs/ubifs/file.c                       | 16 ++++++++-----
 fs/udf/file.c                         |  3 +--
 fs/vboxsf/file.c                      |  2 ++
 include/linux/fs.h                    |  5 ++++
 mm/filemap.c                          | 33 +++++++++++++--------------
 mm/page_io.c                          | 27 ++++++++++++++++++++--
 mm/readahead.c                        |  3 ++-
 22 files changed, 151 insertions(+), 75 deletions(-)

-- 
2.28.0

[PATCH v3 01/18] cachefiles: Handle readpage error correctly

[Matthew Wilcox (Oracle)]

If ->readpage returns an error, it has already unlocked the page.

Fixes: 5e929b33c393 ("CacheFiles: Handle truncate unlocking the page we're reading")
Cc: stable@vger.kernel.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 fs/cachefiles/rdwr.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/fs/cachefiles/rdwr.c b/fs/cachefiles/rdwr.c
index 3080cda9e824..8bda092e60c5 100644
--- a/fs/cachefiles/rdwr.c
+++ b/fs/cachefiles/rdwr.c
@@ -121,7 +121,7 @@ static int cachefiles_read_reissue(struct cachefiles_object *object,
 		_debug("reissue read");
 		ret = bmapping->a_ops->readpage(NULL, backpage);
 		if (ret < 0)
-			goto unlock_discard;
+			goto discard;
 	}
 
 	/* but the page may have been read before the monitor was installed, so
@@ -138,6 +138,7 @@ static int cachefiles_read_reissue(struct cachefiles_object *object,
 
 unlock_discard:
 	unlock_page(backpage);
+discard:
 	spin_lock_irq(&object->work_lock);
 	list_del(&monitor->op_link);
 	spin_unlock_irq(&object->work_lock);
-- 
2.28.0

[PATCH v3 02/18] swap: Call aops->readahead if appropriate

[Matthew Wilcox (Oracle)]

Some filesystems have a synchronous readpage and an asynchronous
readahead.  Call the readahead operation if we're trying to do swap
readahead.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 mm/page_io.c | 22 ++++++++++++++++++++--
 1 file changed, 20 insertions(+), 2 deletions(-)

diff --git a/mm/page_io.c b/mm/page_io.c
index e485a6e8a6cd..faf5ccb42946 100644
--- a/mm/page_io.c
+++ b/mm/page_io.c
@@ -367,6 +367,24 @@ int __swap_writepage(struct page *page, struct writeback_control *wbc,
 	return ret;
 }
 
+static int mapping_readpage(struct file *file, struct address_space *mapping,
+		struct page *page, bool synchronous)
+{
+	struct readahead_control ractl = {
+		.file = file,
+		.mapping = mapping,
+		._index = page->index,
+		._nr_pages = 1,
+	};
+
+	if (!synchronous && mapping->a_ops->readahead) {
+		mapping->a_ops->readahead(&ractl);
+		return 0;
+	}
+
+	return mapping->a_ops->readpage(file, page);
+}
+
 int swap_readpage(struct page *page, bool synchronous)
 {
 	struct bio *bio;
@@ -395,9 +413,9 @@ int swap_readpage(struct page *page, bool synchronous)
 
 	if (data_race(sis->flags & SWP_FS)) {
 		struct file *swap_file = sis->swap_file;
-		struct address_space *mapping = swap_file->f_mapping;
 
-		ret = mapping->a_ops->readpage(swap_file, page);
+		ret = mapping_readpage(swap_file, swap_file->f_mapping,
+				page, synchronous);
 		if (!ret)
 			count_vm_event(PSWPIN);
 		goto out;
-- 
2.28.0

[PATCH v3 03/18] fs: Add AOP_UPDATED_PAGE return value

[Matthew Wilcox (Oracle)]

Allow synchronous ->readpage implementations to execute more
efficiently by skipping the re-locking of the page.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 Documentation/filesystems/locking.rst |  7 ++++---
 Documentation/filesystems/vfs.rst     | 21 ++++++++++++++-------
 fs/buffer.c                           | 15 +++++++++------
 fs/cachefiles/rdwr.c                  |  8 ++++++++
 include/linux/fs.h                    |  5 +++++
 mm/filemap.c                          | 15 +++++++++++++--
 mm/page_io.c                          |  7 ++++++-
 mm/readahead.c                        |  3 ++-
 8 files changed, 61 insertions(+), 20 deletions(-)

diff --git a/Documentation/filesystems/locking.rst b/Documentation/filesystems/locking.rst
index 64f94a18d97e..06a7a8bf2362 100644
--- a/Documentation/filesystems/locking.rst
+++ b/Documentation/filesystems/locking.rst
@@ -269,7 +269,7 @@ locking rules:
 ops			PageLocked(page)	 i_rwsem
 ======================	======================== =========
 writepage:		yes, unlocks (see below)
-readpage:		yes, unlocks
+readpage:		yes, may unlock
 writepages:
 set_page_dirty		no
 readahead:		yes, unlocks
@@ -294,8 +294,9 @@ swap_deactivate:	no
 ->write_begin(), ->write_end() and ->readpage() may be called from
 the request handler (/dev/loop).
 
-->readpage() unlocks the page, either synchronously or via I/O
-completion.
+->readpage() may return AOP_UPDATED_PAGE if the page is now Uptodate
+or 0 if the page will be unlocked asynchronously by I/O completion.
+If it returns -errno, it should unlock the page.
 
 ->readahead() unlocks the pages that I/O is attempted on like ->readpage().
 
diff --git a/Documentation/filesystems/vfs.rst b/Documentation/filesystems/vfs.rst
index ca52c82e5bb5..16248c299aaa 100644
--- a/Documentation/filesystems/vfs.rst
+++ b/Documentation/filesystems/vfs.rst
@@ -643,7 +643,7 @@ set_page_dirty to write data into the address_space, and writepage and
 writepages to writeback data to storage.
 
 Adding and removing pages to/from an address_space is protected by the
-inode's i_mutex.
+inode's i_rwsem held exclusively.
 
 When data is written to a page, the PG_Dirty flag should be set.  It
 typically remains set until writepage asks for it to be written.  This
@@ -757,12 +757,19 @@ cache in your filesystem.  The following members are defined:
 
 ``readpage``
 	called by the VM to read a page from backing store.  The page
-	will be Locked when readpage is called, and should be unlocked
-	and marked uptodate once the read completes.  If ->readpage
-	discovers that it needs to unlock the page for some reason, it
-	can do so, and then return AOP_TRUNCATED_PAGE.  In this case,
-	the page will be relocated, relocked and if that all succeeds,
-	->readpage will be called again.
+	will be Locked and !Uptodate when readpage is called.  Ideally,
+	the filesystem will bring the page Uptodate and return
+	AOP_UPDATED_PAGE.  If the filesystem encounters an error, it
+	should unlock the page and return a negative errno without marking
+	the page Uptodate.  It does not need to mark the page as Error.
+	If the filesystem returns 0, this means the page will be unlocked
+	asynchronously by I/O completion.  The VFS will wait for the
+	page to be unlocked, so there is no advantage to executing this
+	operation asynchronously.
+
+	The filesystem can also return AOP_TRUNCATED_PAGE to indicate
+	that it had to unlock the page to avoid a deadlock.  The caller
+	will re-check the page cache and call ->readpage again.
 
 ``writepages``
 	called by the VM to write out pages associated with the
diff --git a/fs/buffer.c b/fs/buffer.c
index 50bbc99e3d96..1d5337517dcd 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -2856,14 +2856,17 @@ int nobh_truncate_page(struct address_space *mapping,
 	/* Ok, it's mapped. Make sure it's up-to-date */
 	if (!PageUptodate(page)) {
 		err = mapping->a_ops->readpage(NULL, page);
-		if (err) {
+		if (err < 0) {
 			put_page(page);
 			goto out;
-		}
-		lock_page(page);
-		if (!PageUptodate(page)) {
-			err = -EIO;
-			goto unlock;
+		} else if (err == 0) {
+			lock_page(page);
+			if (!PageUptodate(page)) {
+				err = -EIO;
+				goto unlock;
+			}
+		} else {
+			err = 0;
 		}
 		if (page_has_buffers(page))
 			goto has_buffers;
diff --git a/fs/cachefiles/rdwr.c b/fs/cachefiles/rdwr.c
index 8bda092e60c5..cc4cc535caf5 100644
--- a/fs/cachefiles/rdwr.c
+++ b/fs/cachefiles/rdwr.c
@@ -122,6 +122,10 @@ static int cachefiles_read_reissue(struct cachefiles_object *object,
 		ret = bmapping->a_ops->readpage(NULL, backpage);
 		if (ret < 0)
 			goto discard;
+		if (ret == AOP_UPDATED_PAGE) {
+			ret = 0;
+			goto unlock_discard;
+		}
 	}
 
 	/* but the page may have been read before the monitor was installed, so
@@ -285,6 +289,8 @@ static int cachefiles_read_backing_file_one(struct cachefiles_object *object,
 	ret = bmapping->a_ops->readpage(NULL, backpage);
 	if (ret < 0)
 		goto read_error;
+	if (ret == AOP_UPDATED_PAGE)
+		unlock_page(backpage);
 
 	/* set the monitor to transfer the data across */
 monitor_backing_page:
@@ -523,6 +529,8 @@ static int cachefiles_read_backing_file(struct cachefiles_object *object,
 		ret = bmapping->a_ops->readpage(NULL, backpage);
 		if (ret < 0)
 			goto read_error;
+		if (ret == AOP_UPDATED_PAGE)
+			unlock_page(backpage);
 
 		/* add the netfs page to the pagecache and LRU, and set the
 		 * monitor to transfer the data across */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 7519ae003a08..badf80e133fd 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -273,6 +273,10 @@ struct iattr {
  *  			reference, it should drop it before retrying.  Returned
  *  			by readpage().
  *
+ * @AOP_UPDATED_PAGE: The readpage method has brought the page Uptodate
+ * without releasing the page lock.  This is suitable for synchronous
+ * implementations of readpage.
+ *
  * address_space_operation functions return these large constants to indicate
  * special semantics to the caller.  These are much larger than the bytes in a
  * page to allow for functions that return the number of bytes operated on in a
@@ -282,6 +286,7 @@ struct iattr {
 enum positive_aop_returns {
 	AOP_WRITEPAGE_ACTIVATE	= 0x80000,
 	AOP_TRUNCATED_PAGE	= 0x80001,
+	AOP_UPDATED_PAGE	= 0x80002,
 };
 
 #define AOP_FLAG_CONT_EXPAND		0x0001 /* called from cont_expand */
diff --git a/mm/filemap.c b/mm/filemap.c
index 1aaea26556cc..95b68ec1f22c 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -2254,8 +2254,13 @@ ssize_t generic_file_buffered_read(struct kiocb *iocb,
 		 * PG_error will be set again if readpage fails.
 		 */
 		ClearPageError(page);
-		/* Start the actual read. The read will unlock the page. */
+		/* Start the actual read. The read may unlock the page. */
 		error = mapping->a_ops->readpage(filp, page);
+		if (error == AOP_UPDATED_PAGE) {
+			unlock_page(page);
+			error = 0;
+			goto page_ok;
+		}
 
 		if (unlikely(error)) {
 			if (error == AOP_TRUNCATED_PAGE) {
@@ -2619,7 +2624,7 @@ vm_fault_t filemap_fault(struct vm_fault *vmf)
 	 */
 	if (unlikely(!PageUptodate(page)))
 		goto page_not_uptodate;
-
+page_ok:
 	/*
 	 * We've made it this far and we had to drop our mmap_lock, now is the
 	 * time to return to the upper layer and have it re-find the vma and
@@ -2654,6 +2659,8 @@ vm_fault_t filemap_fault(struct vm_fault *vmf)
 	ClearPageError(page);
 	fpin = maybe_unlock_mmap_for_io(vmf, fpin);
 	error = mapping->a_ops->readpage(file, page);
+	if (error == AOP_UPDATED_PAGE)
+		goto page_ok;
 	if (!error) {
 		wait_on_page_locked(page);
 		if (!PageUptodate(page))
@@ -2867,6 +2874,10 @@ static struct page *do_read_cache_page(struct address_space *mapping,
 			err = filler(data, page);
 		else
 			err = mapping->a_ops->readpage(data, page);
+		if (err == AOP_UPDATED_PAGE) {
+			unlock_page(page);
+			goto out;
+		}
 
 		if (err < 0) {
 			put_page(page);
diff --git a/mm/page_io.c b/mm/page_io.c
index faf5ccb42946..f141a171d09c 100644
--- a/mm/page_io.c
+++ b/mm/page_io.c
@@ -376,13 +376,18 @@ static int mapping_readpage(struct file *file, struct address_space *mapping,
 		._index = page->index,
 		._nr_pages = 1,
 	};
+	int ret;
 
 	if (!synchronous && mapping->a_ops->readahead) {
 		mapping->a_ops->readahead(&ractl);
 		return 0;
 	}
 
-	return mapping->a_ops->readpage(file, page);
+	ret = mapping->a_ops->readpage(file, page);
+	if (ret != AOP_UPDATED_PAGE)
+		return ret;
+	unlock_page(page);
+	return 0;
 }
 
 int swap_readpage(struct page *page, bool synchronous)
diff --git a/mm/readahead.c b/mm/readahead.c
index 3c9a8dd7c56c..f182fa731ab3 100644
--- a/mm/readahead.c
+++ b/mm/readahead.c
@@ -142,7 +142,8 @@ static void read_pages(struct readahead_control *rac, struct list_head *pages,
 		rac->_nr_pages = 0;
 	} else {
 		while ((page = readahead_page(rac))) {
-			aops->readpage(rac->file, page);
+			if (aops->readpage(rac->file, page) == AOP_UPDATED_PAGE)
+				unlock_page(page);
 			put_page(page);
 		}
 	}
-- 
2.28.0

[PATCH v3 04/18] mm/filemap: Inline wait_on_page_read into its one caller

[Matthew Wilcox (Oracle)]

Having this code inline helps the function read more easily.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 mm/filemap.c | 20 ++++----------------
 1 file changed, 4 insertions(+), 16 deletions(-)

diff --git a/mm/filemap.c b/mm/filemap.c
index 95b68ec1f22c..0ef06d515532 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -2834,18 +2834,6 @@ EXPORT_SYMBOL(filemap_page_mkwrite);
 EXPORT_SYMBOL(generic_file_mmap);
 EXPORT_SYMBOL(generic_file_readonly_mmap);
 
-static struct page *wait_on_page_read(struct page *page)
-{
-	if (!IS_ERR(page)) {
-		wait_on_page_locked(page);
-		if (!PageUptodate(page)) {
-			put_page(page);
-			page = ERR_PTR(-EIO);
-		}
-	}
-	return page;
-}
-
 static struct page *do_read_cache_page(struct address_space *mapping,
 				pgoff_t index,
 				int (*filler)(void *, struct page *),
@@ -2876,7 +2864,10 @@ static struct page *do_read_cache_page(struct address_space *mapping,
 			err = mapping->a_ops->readpage(data, page);
 		if (err == AOP_UPDATED_PAGE) {
 			unlock_page(page);
-			goto out;
+		} else if (!err) {
+			wait_on_page_locked(page);
+			if (!PageUptodate(page))
+				err = -EIO;
 		}
 
 		if (err < 0) {
@@ -2884,9 +2875,6 @@ static struct page *do_read_cache_page(struct address_space *mapping,
 			return ERR_PTR(err);
 		}
 
-		page = wait_on_page_read(page);
-		if (IS_ERR(page))
-			return page;
 		goto out;
 	}
 	if (PageUptodate(page))
-- 
2.28.0

[PATCH v3 05/18] 9p: Tell the VFS that readpage was synchronous

[Matthew Wilcox (Oracle)]

The 9p readpage implementation was already synchronous, so use
AOP_UPDATED_PAGE to avoid cycling the page lock.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Acked-by: Dominique Martinet <asmadeus@codewreck.org>
---
 fs/9p/vfs_addr.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c
index cce9ace651a2..506ca0ba2ec7 100644
--- a/fs/9p/vfs_addr.c
+++ b/fs/9p/vfs_addr.c
@@ -65,7 +65,7 @@ static int v9fs_fid_readpage(void *data, struct page *page)
 	SetPageUptodate(page);
 
 	v9fs_readpage_to_fscache(inode, page);
-	retval = 0;
+	return AOP_UPDATED_PAGE;
 
 done:
 	unlock_page(page);
@@ -280,6 +280,10 @@ static int v9fs_write_begin(struct file *filp, struct address_space *mapping,
 		goto out;
 
 	retval = v9fs_fid_readpage(v9inode->writeback_fid, page);
+	if (retval == AOP_UPDATED_PAGE) {
+		retval = 0;
+		goto out;
+	}
 	put_page(page);
 	if (!retval)
 		goto start;
-- 
2.28.0

[PATCH v3 06/18] afs: Tell the VFS that readpage was synchronous

[Matthew Wilcox (Oracle)]

The afs readpage implementation was already synchronous, so use
AOP_UPDATED_PAGE to avoid cycling the page lock.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 fs/afs/file.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/fs/afs/file.c b/fs/afs/file.c
index 371d1488cc54..4aa2ad3bea6a 100644
--- a/fs/afs/file.c
+++ b/fs/afs/file.c
@@ -367,7 +367,8 @@ int afs_page_filler(void *data, struct page *page)
 			BUG_ON(PageFsCache(page));
 		}
 #endif
-		unlock_page(page);
+		_leave(" = AOP_UPDATED_PAGE");
+		return AOP_UPDATED_PAGE;
 	}
 
 	_leave(" = 0");
-- 
2.28.0

[PATCH v3 07/18] ceph: Tell the VFS that readpage was synchronous

[Matthew Wilcox (Oracle)]

The ceph readpage implementation was already synchronous, so use
AOP_UPDATED_PAGE to avoid cycling the page lock.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
---
 fs/ceph/addr.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index 6ea761c84494..b2bf8bf7a312 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -291,10 +291,11 @@ static int ceph_do_readpage(struct file *filp, struct page *page)
 static int ceph_readpage(struct file *filp, struct page *page)
 {
 	int r = ceph_do_readpage(filp, page);
-	if (r != -EINPROGRESS)
-		unlock_page(page);
-	else
-		r = 0;
+	if (r == -EINPROGRESS)
+		return 0;
+	if (r == 0)
+		return AOP_UPDATED_PAGE;
+	unlock_page(page);
 	return r;
 }
 
-- 
2.28.0

[PATCH v3 08/18] cifs: Tell the VFS that readpage was synchronous

[Matthew Wilcox (Oracle)]

The cifs readpage implementation was already synchronous, so use
AOP_UPDATED_PAGE to avoid cycling the page lock.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 fs/cifs/file.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index be46fab4c96d..533b151a9143 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -4537,7 +4537,8 @@ static int cifs_readpage_worker(struct file *file, struct page *page,
 	/* send this page to the cache */
 	cifs_readpage_to_fscache(file_inode(file), page);
 
-	rc = 0;
+	kunmap(page);
+	return AOP_UPDATED_PAGE;
 
 io_error:
 	kunmap(page);
@@ -4677,7 +4678,10 @@ static int cifs_write_begin(struct file *file, struct address_space *mapping,
 		 * an error, we don't need to return it. cifs_write_end will
 		 * do a sync write instead since PG_uptodate isn't set.
 		 */
-		cifs_readpage_worker(file, page, &page_start);
+		int err = cifs_readpage_worker(file, page, &page_start);
+
+		if (err == AOP_UPDATED_PAGE)
+			goto out;
 		put_page(page);
 		oncethru = 1;
 		goto start;
-- 
2.28.0

[PATCH v3 09/18] cramfs: Tell the VFS that readpage was synchronous

[Matthew Wilcox (Oracle)]

The cramfs readpage implementation was already synchronous, so use
AOP_UPDATED_PAGE to avoid cycling the page lock.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 fs/cramfs/inode.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c
index 912308600d39..7a642146c074 100644
--- a/fs/cramfs/inode.c
+++ b/fs/cramfs/inode.c
@@ -916,15 +916,14 @@ static int cramfs_readpage(struct file *file, struct page *page)
 	flush_dcache_page(page);
 	kunmap(page);
 	SetPageUptodate(page);
-	unlock_page(page);
-	return 0;
+	return AOP_UPDATED_PAGE;
 
 err:
 	kunmap(page);
 	ClearPageUptodate(page);
 	SetPageError(page);
 	unlock_page(page);
-	return 0;
+	return -EIO;
 }
 
 static const struct address_space_operations cramfs_aops = {
-- 
2.28.0

[PATCH v3 10/18] ecryptfs: Tell the VFS that readpage was synchronous

[Matthew Wilcox (Oracle)]

The ecryptfs readpage implementation was already synchronous, so use
AOP_UPDATED_PAGE to avoid cycling the page lock.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 fs/ecryptfs/mmap.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/fs/ecryptfs/mmap.c b/fs/ecryptfs/mmap.c
index 019572c6b39a..dee35181d789 100644
--- a/fs/ecryptfs/mmap.c
+++ b/fs/ecryptfs/mmap.c
@@ -219,12 +219,13 @@ static int ecryptfs_readpage(struct file *file, struct page *page)
 		}
 	}
 out:
-	if (rc)
-		ClearPageUptodate(page);
-	else
-		SetPageUptodate(page);
-	ecryptfs_printk(KERN_DEBUG, "Unlocking page with index = [0x%.16lx]\n",
+	ecryptfs_printk(KERN_DEBUG, "Returning page with index = [0x%.16lx]\n",
 			page->index);
+	if (!rc) {
+		SetPageUptodate(page);
+		return AOP_UPDATED_PAGE;
+	}
+	ClearPageUptodate(page);
 	unlock_page(page);
 	return rc;
 }
-- 
2.28.0

[PATCH v3 11/18] ext4: Tell the VFS that readpage was synchronous

[Matthew Wilcox (Oracle)]

The ext4 inline data readpage implementation was already synchronous,
so use AOP_UPDATED_PAGE to avoid cycling the page lock.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 fs/ext4/inline.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c
index 75c97bca0815..2a489243e4de 100644
--- a/fs/ext4/inline.c
+++ b/fs/ext4/inline.c
@@ -490,7 +490,8 @@ static int ext4_read_inline_page(struct inode *inode, struct page *page)
 	zero_user_segment(page, len, PAGE_SIZE);
 	SetPageUptodate(page);
 	brelse(iloc.bh);
-
+	if (ret >= 0)
+		return AOP_UPDATED_PAGE;
 out:
 	return ret;
 }
@@ -514,12 +515,14 @@ int ext4_readpage_inline(struct inode *inode, struct page *page)
 	else if (!PageUptodate(page)) {
 		zero_user_segment(page, 0, PAGE_SIZE);
 		SetPageUptodate(page);
+		ret = AOP_UPDATED_PAGE;
 	}
 
 	up_read(&EXT4_I(inode)->xattr_sem);
 
-	unlock_page(page);
-	return ret >= 0 ? 0 : ret;
+	if (ret < 0)
+		unlock_page(page);
+	return ret;
 }
 
 static int ext4_convert_inline_data_to_extent(struct address_space *mapping,
-- 
2.28.0

[PATCH v3 12/18] ext4: Return error from ext4_readpage

[Matthew Wilcox (Oracle)]

The error returned from ext4_map_blocks() was being discarded, leading
to the generic -EIO being returned to userspace.  Now ext4 can return
more precise errors.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 fs/ext4/readpage.c | 24 ++++++++++++++----------
 1 file changed, 14 insertions(+), 10 deletions(-)

diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c
index f014c5e473a9..00a024f3a954 100644
--- a/fs/ext4/readpage.c
+++ b/fs/ext4/readpage.c
@@ -237,7 +237,7 @@ int ext4_mpage_readpages(struct inode *inode,
 	sector_t blocks[MAX_BUF_PER_PAGE];
 	unsigned page_block;
 	struct block_device *bdev = inode->i_sb->s_bdev;
-	int length;
+	int length, err = 0;
 	unsigned relative_block = 0;
 	struct ext4_map_blocks map;
 	unsigned int nr_pages = rac ? readahead_count(rac) : 1;
@@ -301,14 +301,9 @@ int ext4_mpage_readpages(struct inode *inode,
 				map.m_lblk = block_in_file;
 				map.m_len = last_block - block_in_file;
 
-				if (ext4_map_blocks(NULL, inode, &map, 0) < 0) {
-				set_error_page:
-					SetPageError(page);
-					zero_user_segment(page, 0,
-							  PAGE_SIZE);
-					unlock_page(page);
-					goto next_page;
-				}
+				err = ext4_map_blocks(NULL, inode, &map, 0);
+				if (err < 0)
+					goto err;
 			}
 			if ((map.m_flags & EXT4_MAP_MAPPED) == 0) {
 				fully_mapped = 0;
@@ -395,6 +390,15 @@ int ext4_mpage_readpages(struct inode *inode,
 		} else
 			last_block_in_bio = blocks[blocks_per_page - 1];
 		goto next_page;
+	set_error_page:
+		err = -EIO;
+	err:
+		if (rac) {
+			SetPageError(page);
+			zero_user_segment(page, 0, PAGE_SIZE);
+		}
+		unlock_page(page);
+		goto next_page;
 	confused:
 		if (bio) {
 			submit_bio(bio);
@@ -410,7 +414,7 @@ int ext4_mpage_readpages(struct inode *inode,
 	}
 	if (bio)
 		submit_bio(bio);
-	return 0;
+	return err;
 }
 
 int __init ext4_init_post_read_processing(void)
-- 
2.28.0

[PATCH v3 13/18] fuse: Tell the VFS that readpage was synchronous

[Matthew Wilcox (Oracle)]

The fuse readpage implementation was already synchronous, so use
AOP_UPDATED_PAGE to avoid cycling the page lock.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 fs/fuse/file.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 6611ef3269a8..7aa5626bc582 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -850,6 +850,8 @@ static int fuse_readpage(struct file *file, struct page *page)
 
 	err = fuse_do_readpage(file, page);
 	fuse_invalidate_atime(inode);
+	if (!err)
+		return AOP_UPDATED_PAGE;
  out:
 	unlock_page(page);
 	return err;
-- 
2.28.0

[PATCH v3 14/18] hostfs: Tell the VFS that readpage was synchronous

[Matthew Wilcox (Oracle)]

The hostfs readpage implementation was already synchronous, so use
AOP_UPDATED_PAGE to avoid cycling the page lock.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Acked-by: Richard Weinberger <richard@nod.at>
---
 fs/hostfs/hostfs_kern.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index c070c0d8e3e9..c49221c09c4b 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -455,6 +455,8 @@ static int hostfs_readpage(struct file *file, struct page *page)
  out:
 	flush_dcache_page(page);
 	kunmap(page);
+	if (!ret)
+		return AOP_UPDATED_PAGE;
 	unlock_page(page);
 	return ret;
 }
-- 
2.28.0

[PATCH v3 15/18] jffs2: Tell the VFS that readpage was synchronous

[Matthew Wilcox (Oracle)]

The jffs2 readpage implementation was already synchronous, so use
AOP_UPDATED_PAGE to avoid cycling the page lock.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Acked-by: Richard Weinberger <richard@nod.at>
---
 fs/jffs2/file.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c
index f8fb89b10227..959a74027041 100644
--- a/fs/jffs2/file.c
+++ b/fs/jffs2/file.c
@@ -116,15 +116,17 @@ int jffs2_do_readpage_unlock(void *data, struct page *pg)
 	return ret;
 }
 
-
 static int jffs2_readpage (struct file *filp, struct page *pg)
 {
 	struct jffs2_inode_info *f = JFFS2_INODE_INFO(pg->mapping->host);
 	int ret;
 
 	mutex_lock(&f->sem);
-	ret = jffs2_do_readpage_unlock(pg->mapping->host, pg);
+	ret = jffs2_do_readpage_nolock(pg->mapping->host, pg);
 	mutex_unlock(&f->sem);
+	if (!ret)
+		return AOP_UPDATED_PAGE;
+	unlock_page(pg);
 	return ret;
 }
 
-- 
2.28.0

[PATCH v3 16/18] ubifs: Tell the VFS that readpage was synchronous

[Matthew Wilcox (Oracle)]

The ubifs readpage implementation was already synchronous, so use
AOP_UPDATED_PAGE to avoid cycling the page lock.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Acked-by: Richard Weinberger <richard@nod.at>
---
 fs/ubifs/file.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index b77d1637bbbc..82633509c45e 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -772,7 +772,6 @@ static int ubifs_do_bulk_read(struct ubifs_info *c, struct bu_info *bu,
 	if (err)
 		goto out_warn;
 
-	unlock_page(page1);
 	ret = 1;
 
 	isize = i_size_read(inode);
@@ -892,11 +891,16 @@ static int ubifs_bulk_read(struct page *page)
 
 static int ubifs_readpage(struct file *file, struct page *page)
 {
-	if (ubifs_bulk_read(page))
-		return 0;
-	do_readpage(page);
-	unlock_page(page);
-	return 0;
+	int err;
+
+	err = ubifs_bulk_read(page);
+	if (err == 0)
+		err = do_readpage(page);
+	if (err < 0) {
+		unlock_page(page);
+		return err;
+	}
+	return AOP_UPDATED_PAGE;
 }
 
 static int do_writepage(struct page *page, int len)
-- 
2.28.0

[PATCH v3 17/18] udf: Tell the VFS that readpage was synchronous

[Matthew Wilcox (Oracle)]

The udf inline data readpage implementation was already synchronous,
so use AOP_UPDATED_PAGE to avoid cycling the page lock.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: Jan Kara <jack@suse.cz>
---
 fs/udf/file.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/fs/udf/file.c b/fs/udf/file.c
index 628941a6b79a..52bbe92d7c43 100644
--- a/fs/udf/file.c
+++ b/fs/udf/file.c
@@ -61,9 +61,8 @@ static int udf_adinicb_readpage(struct file *file, struct page *page)
 {
 	BUG_ON(!PageLocked(page));
 	__udf_adinicb_readpage(page);
-	unlock_page(page);
 
-	return 0;
+	return AOP_UPDATED_PAGE;
 }
 
 static int udf_adinicb_writepage(struct page *page,
-- 
2.28.0

[PATCH v3 18/18] vboxsf: Tell the VFS that readpage was synchronous

[Matthew Wilcox (Oracle)]

The vboxsf inline data readpage implementation was already synchronous,
so use AOP_UPDATED_PAGE to avoid cycling the page lock.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 fs/vboxsf/file.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/fs/vboxsf/file.c b/fs/vboxsf/file.c
index c4ab5996d97a..c2a144e5cb5a 100644
--- a/fs/vboxsf/file.c
+++ b/fs/vboxsf/file.c
@@ -228,6 +228,8 @@ static int vboxsf_readpage(struct file *file, struct page *page)
 	}
 
 	kunmap(page);
+	if (!err)
+		return AOP_UPDATED_PAGE;
 	unlock_page(page);
 	return err;
 }
-- 
2.28.0

Re: [PATCH v3 11/18] ext4: Tell the VFS that readpage was synchronous

[Theodore Y. Ts'o]

On Fri, Oct 16, 2020 at 05:04:36PM +0100, Matthew Wilcox (Oracle) wrote:
> The ext4 inline data readpage implementation was already synchronous,
> so use AOP_UPDATED_PAGE to avoid cycling the page lock.
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>

Acked-by: Theodore Ts'o <tytso@mit.edu>

Re: [PATCH v3 12/18] ext4: Return error from ext4_readpage

[Theodore Y. Ts'o]

On Fri, Oct 16, 2020 at 05:04:37PM +0100, Matthew Wilcox (Oracle) wrote:
> The error returned from ext4_map_blocks() was being discarded, leading
> to the generic -EIO being returned to userspace.  Now ext4 can return
> more precise errors.
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>

This change is independent of the synchronous readpage changes,
correct?  Or am I missing something?

Cheers,

					- Ted

Re: [PATCH v3 12/18] ext4: Return error from ext4_readpage

[Matthew Wilcox]

On Sun, Oct 18, 2020 at 10:25:57AM -0400, Theodore Y. Ts'o wrote:
> On Fri, Oct 16, 2020 at 05:04:37PM +0100, Matthew Wilcox (Oracle) wrote:
> > The error returned from ext4_map_blocks() was being discarded, leading
> > to the generic -EIO being returned to userspace.  Now ext4 can return
> > more precise errors.
> > 
> > Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> 
> This change is independent of the synchronous readpage changes,
> correct?  Or am I missing something?

It's a step along the way.  If you want to queue it up independently of
the other changes, I see no problem with that.  The requirement to make
a synchronous ->readpage killable is making the conversion quite thorny
and I'm not sure I'm going to get it done this merge window.

Re: [PATCH v3 13/18] fuse: Tell the VFS that readpage was synchronous

[Miklos Szeredi]

On Fri, Oct 16, 2020 at 6:04 PM Matthew Wilcox (Oracle)
<willy@infradead.org> wrote:
>
> The fuse readpage implementation was already synchronous, so use
> AOP_UPDATED_PAGE to avoid cycling the page lock.
>
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>

Acked-by: Miklos Szeredi <mszeredi@redhat.com>

Thanks,
Miklos