Use vfs_copy_file_range() helper instead of calling do_splice_direct()
when copying up file data.
When copying up within the same fs, which supports copy_file_range(),
fs implementation can be more efficient then do_splice_direct().
vfs_copy_file_range() helper falls back to do_splice_direct() if
it cannot use the file system's copy_file_range() implementation.
A previous change added a vfs_clone_file_range() call before
the data copy loop, so this change is only effective for filesystems
that support copy_file_range() and *do not* support clone_file_range().
At the moment, there are no such filesystems in the kernel that
can be used as overlayfs upper, so I tested this change by disabling
the vfs_clone_file_range() call.
Tested correct behavior when lower and upper are on:
1. same ext4 (copy)
2. same xfs + reflink patches + mkfs.xfs (copy)
3. same xfs + reflink patches + mkfs.xfs -m reflink=1 (reflink)
4. different xfs + reflink patches + mkfs.xfs -m reflink=1 (copy)
For comparison, on my laptop, xfstest overlay/001 (copy up of large
sparse files) takes less than 1 second in the xfs reflink setup vs.
25 seconds on the rest of the setups.
Signed-off-by: Amir Goldstein <amir73il@xxxxxxxxx>
---
fs/overlayfs/copy_up.c | 10 +++++-----
fs/read_write.c | 1 +
2 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/fs/overlayfs/copy_up.c b/fs/overlayfs/copy_up.c
index ba039f8..a6d6bac 100644
--- a/fs/overlayfs/copy_up.c
+++ b/fs/overlayfs/copy_up.c
@@ -146,7 +146,6 @@ static int ovl_copy_up_data(struct path *old, struct path *new, loff_t len)
/* Can't clone, so now we try to copy the data */
error = 0;
- /* FIXME: copy up sparse files efficiently */
while (len) {
size_t this_len = OVL_COPY_UP_CHUNK_SIZE;
long bytes;
@@ -159,15 +158,16 @@ static int ovl_copy_up_data(struct path *old, struct path *new, loff_t len)
break;
}
- bytes = do_splice_direct(old_file, &old_pos,
- new_file, &new_pos,
- this_len, SPLICE_F_MOVE);
+ bytes = vfs_copy_file_range(old_file, old_pos,
+ new_file, new_pos,
+ this_len, 0);
if (bytes <= 0) {
error = bytes;
break;
}
- WARN_ON(old_pos != new_pos);
+ old_pos += bytes;
+ new_pos += bytes;
len -= bytes;
}
out:
diff --git a/fs/read_write.c b/fs/read_write.c
index 6975fe8..dfc083a 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -1515,6 +1515,7 @@ ssize_t vfs_copy_file_range(struct file *file_in, loff_t pos_in,
file_out->f_op->copy_file_range)
ret = file_out->f_op->copy_file_range(file_in, pos_in, file_out,
pos_out, len, flags);
+ /* FIXME: copy sparse file range efficiently */
if (ret == -EOPNOTSUPP)
ret = do_splice_direct(file_in, &pos_in, file_out, &pos_out,
len > MAX_RW_COUNT ? MAX_RW_COUNT : len, 0);
--
2.7.4
--
To unsubscribe from this list: send the line "unsubscribe linux-unionfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
