From: Namjae Jeon <[email protected]>
FALLOC_FL_INSERT_RANGE command is the opposite command of
FALLOC_FL_COLLAPSE_RANGE that is needed for advertisers or someone who want to
add some data in the middle of file. FALLOC_FL_INSERT_RANGE will create space
for writing new data within a file after shifting extents to right as given
length. and this command also has same limitation as FALLOC_FL_COLLAPSE_RANGE,
that is block boundary and use ftruncate(2) for crosses EOF.
Signed-off-by: Namjae Jeon <[email protected]>
Signed-off-by: Ashish Sangwan <[email protected]>
---
fs/open.c | 8 ++++++--
include/linux/falloc.h | 6 ++++++
include/uapi/linux/falloc.h | 17 +++++++++++++++++
3 files changed, 29 insertions(+), 2 deletions(-)
diff --git a/fs/open.c b/fs/open.c
index 33f9cbf..b724cc0 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -231,8 +231,7 @@ int vfs_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
return -EINVAL;
/* Return error if mode is not supported */
- if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE |
- FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_ZERO_RANGE))
+ if (mode & ~FALLOC_FL_SUPPORTED_MASK)
return -EOPNOTSUPP;
/* Punch hole and zero range are mutually exclusive */
@@ -250,6 +249,11 @@ int vfs_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
(mode & ~FALLOC_FL_COLLAPSE_RANGE))
return -EINVAL;
+ /* Insert range should only be used exclusively. */
+ if ((mode & FALLOC_FL_INSERT_RANGE) &&
+ (mode & ~FALLOC_FL_INSERT_RANGE))
+ return -EINVAL;
+
if (!(file->f_mode & FMODE_WRITE))
return -EBADF;
diff --git a/include/linux/falloc.h b/include/linux/falloc.h
index 3159168..9961110 100644
--- a/include/linux/falloc.h
+++ b/include/linux/falloc.h
@@ -21,4 +21,10 @@ struct space_resv {
#define FS_IOC_RESVSP _IOW('X', 40, struct space_resv)
#define FS_IOC_RESVSP64 _IOW('X', 42, struct space_resv)
+#define FALLOC_FL_SUPPORTED_MASK (FALLOC_FL_KEEP_SIZE | \
+ FALLOC_FL_PUNCH_HOLE | \
+ FALLOC_FL_COLLAPSE_RANGE | \
+ FALLOC_FL_ZERO_RANGE | \
+ FALLOC_FL_INSERT_RANGE)
+
#endif /* _FALLOC_H_ */
diff --git a/include/uapi/linux/falloc.h b/include/uapi/linux/falloc.h
index d1197ae..3e445a7 100644
--- a/include/uapi/linux/falloc.h
+++ b/include/uapi/linux/falloc.h
@@ -41,4 +41,21 @@
*/
#define FALLOC_FL_ZERO_RANGE 0x10
+/*
+ * FALLOC_FL_INSERT_RANGE is use to insert space within the file size without
+ * overwriting any existing data. The contents of the file beyond offset are
+ * shifted towards right by len bytes to create a hole. As such, this
+ * operation will increase the size of the file by len bytes.
+ *
+ * Different filesystems may implement different limitations on the granularity
+ * of the operation. Most will limit operations to filesystem block size
+ * boundaries, but this boundary may be larger or smaller depending on
+ * the filesystem and/or the configuration of the filesystem or file.
+ *
+ * Attempting to insert space using this flag at OR beyond the end of
+ * the file is considered an illegal operation - just use ftruncate(2) or
+ * fallocate(2) with mode 0 for such type of operations.
+ */
+#define FALLOC_FL_INSERT_RANGE 0x20
+
#endif /* _UAPI_FALLOC_H_ */
--
1.7.9.5
On Sun, Feb 22, 2015 at 12:45:43AM +0900, Namjae Jeon wrote:
> From: Namjae Jeon <[email protected]>
>
> FALLOC_FL_INSERT_RANGE command is the opposite command of
> FALLOC_FL_COLLAPSE_RANGE that is needed for advertisers or someone who want to
> add some data in the middle of file. FALLOC_FL_INSERT_RANGE will create space
> for writing new data within a file after shifting extents to right as given
> length. and this command also has same limitation as FALLOC_FL_COLLAPSE_RANGE,
> that is block boundary and use ftruncate(2) for crosses EOF.
>
> Signed-off-by: Namjae Jeon <[email protected]>
> Signed-off-by: Ashish Sangwan <[email protected]>
Folks, I have merged this into the xfs tree for the 4.1 merge
window. It is under the stable branch "fallocate-insert-range"
in the tree here:
git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs.git
an dis currently included in the for-next branch that is included in
the linux-next tree. I have not included the ext4 kernel support
patch in this branch (patch 3) because I have not seen any reviews
from ext4 developers for it.
Ted, if you want to include that patch in you ext4 tree, feel free
to base it on the above branch, or if you give your sob or
reviewed-by ok I can pull it into my tree. I'll leave it up to you
to determine how you want to deal with this.
Cheers,
Dave.
--
Dave Chinner
[email protected]