[PATCH v2 0/2] zonefs: Fix data corruption and error return

[PATCH v2 0/2] zonefs: Fix data corruption and error return

[Damien Le Moal]

Patch 1 fixes a potential data corruption when the last page of the
inode is cached and a sync append direct write partially modify the
data of that page on disk. The second patch avoids propagating to user
space the -ENOTBLK error potentially returned by iomap_dio_rw().

Changes from v1:
 - Reworked patch 1 commit message and use an exact end argument for the
   call to invalidate_inode_pages2_range().
 - Added patch 2.

Damien Le Moal (2):
  zonefs: Always invalidate last cached page on append write
  zonefs: Do not propagate iomap_dio_rw() ENOTBLK error to user space

 fs/zonefs/file.c | 28 ++++++++++++++++++++++++++--
 1 file changed, 26 insertions(+), 2 deletions(-)

-- 
2.39.2

[PATCH v2 1/2] zonefs: Always invalidate last cached page on append write

[Damien Le Moal]

When a direct append write is executed, the append offset may correspond
to the last page of a sequential file inode which might have been cached
already by buffered reads, page faults with mmap-read or non-direct
readahead. To ensure that the on-disk and cached data is consistant for
such last cached page, make sure to always invalidate it in
zonefs_file_dio_append(). If the invalidation fails, return -EBUSY to
userspace to differentiate from IO errors.

This invalidation will always be a no-op when the FS block size (device
zone write granularity) is equal to the page size (e.g. 4K).

Reported-by: Hans Holmberg <Hans.Holmberg@wdc.com>
Fixes: 02ef12a663c7 ("zonefs: use REQ_OP_ZONE_APPEND for sync DIO")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
---
 fs/zonefs/file.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/fs/zonefs/file.c b/fs/zonefs/file.c
index 617e4f9db42e..c6ab2732955e 100644
--- a/fs/zonefs/file.c
+++ b/fs/zonefs/file.c
@@ -382,6 +382,7 @@ static ssize_t zonefs_file_dio_append(struct kiocb *iocb, struct iov_iter *from)
 	struct zonefs_zone *z = zonefs_inode_zone(inode);
 	struct block_device *bdev = inode->i_sb->s_bdev;
 	unsigned int max = bdev_max_zone_append_sectors(bdev);
+	pgoff_t start, end;
 	struct bio *bio;
 	ssize_t size = 0;
 	int nr_pages;
@@ -390,6 +391,19 @@ static ssize_t zonefs_file_dio_append(struct kiocb *iocb, struct iov_iter *from)
 	max = ALIGN_DOWN(max << SECTOR_SHIFT, inode->i_sb->s_blocksize);
 	iov_iter_truncate(from, max);
 
+	/*
+	 * If the inode block size (zone write granularity) is smaller than the
+	 * page size, we may be appending data belonging to the last page of the
+	 * inode straddling inode->i_size, with that page already cached due to
+	 * a buffered read or readahead. So make sure to invalidate that page.
+	 * This will always be a no-op for the case where the block size is
+	 * equal to the page size.
+	 */
+	start = iocb->ki_pos >> PAGE_SHIFT;
+	end = (iocb->ki_pos + iov_iter_count(from) - 1) >> PAGE_SHIFT;
+	if (invalidate_inode_pages2_range(inode->i_mapping, start, end))
+		return -EBUSY;
+
 	nr_pages = iov_iter_npages(from, BIO_MAX_VECS);
 	if (!nr_pages)
 		return 0;
-- 
2.39.2

[PATCH v2 2/2] zonefs: Do not propagate iomap_dio_rw() ENOTBLK error to user space

[Damien Le Moal]

The call to invalidate_inode_pages2_range() in __iomap_dio_rw() may
fail, in which case -ENOTBLK is returned and this error code is
propagated back to user space trhough iomap_dio_rw() ->
zonefs_file_dio_write() return chain. This error code is fairly obscure
and may confuse the user. Avoid this and be consistent with the behavior
of zonefs_file_dio_append() for similar invalidate_inode_pages2_range()
errors by returning -EBUSY to user space when iomap_dio_rw() returns
-ENOTBLK.

Suggested-by: Christoph Hellwig <hch@infradead.org>
Fixes: 8dcc1a9d90c1 ("fs: New zonefs file system")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
---
 fs/zonefs/file.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/fs/zonefs/file.c b/fs/zonefs/file.c
index c6ab2732955e..132f01d3461f 100644
--- a/fs/zonefs/file.c
+++ b/fs/zonefs/file.c
@@ -581,11 +581,21 @@ static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from)
 		append = sync;
 	}
 
-	if (append)
+	if (append) {
 		ret = zonefs_file_dio_append(iocb, from);
-	else
+	} else {
+		/*
+		 * iomap_dio_rw() may return ENOTBLK if there was an issue with
+		 * page invalidation. Overwrite that error code with EBUSY to
+		 * be consistent with zonefs_file_dio_append() return value for
+		 * similar issues.
+		 */
 		ret = iomap_dio_rw(iocb, from, &zonefs_write_iomap_ops,
 				   &zonefs_write_dio_ops, 0, NULL, 0);
+		if (ret == -ENOTBLK)
+			ret = -EBUSY;
+	}
+
 	if (zonefs_zone_is_seq(z) &&
 	    (ret > 0 || ret == -EIOCBQUEUED)) {
 		if (ret > 0)
-- 
2.39.2

Re: [PATCH v2 1/2] zonefs: Always invalidate last cached page on append write

[Christoph Hellwig]

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2 2/2] zonefs: Do not propagate iomap_dio_rw() ENOTBLK error to user space

[Christoph Hellwig]

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2 0/2] zonefs: Fix data corruption and error return

[Johannes Thumshirn]

Looks good,
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>

Re: [PATCH v2 0/2] zonefs: Fix data corruption and error return

[Hans Holmberg]

Retested with this series, and it solves the reported issue.

Tested-by: Hans Holmberg <hans.holmberg@wdc.com>

On Thu, Mar 30, 2023 at 10:46 AM Johannes Thumshirn
<Johannes.Thumshirn@wdc.com> wrote:
>
> Looks good,
> Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>