On 9/9/19 4:49 AM, Matthew Bobrowski wrote:
For iomap direct IO write code path changes, we need to accommodate
for the case where the block mapping flags passed to ext4_map_blocks()
will result in m_flags having both EXT4_MAP_MAPPED and
EXT4_MAP_UNWRITTEN bits set. In order for the allocated unwritten
extents to be converted properly in the end_io handler, iomap->type
must be set to IOMAP_UNWRITTEN, so we need to reshuffle the
conditional statement in order to achieve this.
This change is a no-op for DAX code path as the block mapping flag
passed to ext4_map_blocks() when IS_DAX(inode) never results in
EXT4_MAP_MAPPED and EXT4_MAP_UNWRITTEN being set at once.
Signed-off-by: Matthew Bobrowski <mbobrowski@xxxxxxxxxxxxxx>
Looks good to me.
Reviewed-by: Ritesh Harjani <riteshh@xxxxxxxxxxxxx>
---
fs/ext4/inode.c | 17 ++++++++++++++---
1 file changed, 14 insertions(+), 3 deletions(-)
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 761ce6286b05..efb184928e51 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -3581,10 +3581,21 @@ static int ext4_iomap_begin(struct inode *inode, loff_t offset, loff_t length,
iomap->type = delalloc ? IOMAP_DELALLOC : IOMAP_HOLE;
iomap->addr = IOMAP_NULL_ADDR;
} else {
- if (map.m_flags & EXT4_MAP_MAPPED) {
- iomap->type = IOMAP_MAPPED;
- } else if (map.m_flags & EXT4_MAP_UNWRITTEN) {
+ /*
+ * Flags passed to ext4_map_blocks() for direct IO
+ * writes can result in m_flags having both
+ * EXT4_MAP_MAPPED and EXT4_MAP_UNWRITTEN bits set. In
+ * order for allocated unwritten extents to be
+ * converted to written extents in the end_io handler
+ * correctly, we need to ensure that the iomap->type
+ * is also set appropriately in that case. Thus, we
+ * need to check whether EXT4_MAP_UNWRITTEN is set
+ * first.
+ */
+ if (map.m_flags & EXT4_MAP_UNWRITTEN) {
iomap->type = IOMAP_UNWRITTEN;
+ } else if (map.m_flags & EXT4_MAP_MAPPED) {
+ iomap->type = IOMAP_MAPPED;
} else {
WARN_ON_ONCE(1);
return -EIO;
