Btrfs: don't update atime on RO subvolumes

Before the update_time inode operation was indroduced, it was
not possible to prevent updates of atime on RO subvolumes. VFS
was only able to check for RO on the mount, but did not know
anything about btrfs subvolumes.

btrfs_update_time does now check if the root is RO and skip
updating of times.

Signed-off-by: Alexander Block <ablock84@googlemail.com>

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index a7d1921ac76b8ee5d85d563c1ceed80f4a65a2dd..fcc65802f367dd16d066a45d2149761af0aae042 100644 (file)
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -4518,6 +4518,11 @@ int btrfs_dirty_inode(struct inode *inode)
 static int btrfs_update_time(struct inode *inode, struct timespec *now,
                             int flags)
 {
+       struct btrfs_root *root = BTRFS_I(inode)->root;
+
+       if (btrfs_root_readonly(root))
+               return -EROFS;
+
        if (flags & S_VERSION)
                inode_inc_iversion(inode);
        if (flags & S_CTIME)

diff --git a/fs/inode.c b/fs/inode.c
index c99163b1b31036ef68974c0c5dbc192f8f73f4da..033529ecd242b94ca91f1b1818fe51df7b8c4ba8 100644 (file)
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -1551,6 +1551,8 @@ void touch_atime(struct path *path)
         * Btrfs), but since we touch atime while walking down the path we
         * really don't care if we failed to update the atime of the file,
         * so just ignore the return value.
+        * We may also fail on filesystems that have the ability to make parts
+        * of the fs read only, e.g. subvolumes in Btrfs.
         */
        update_time(inode, &now, S_ATIME);
        mnt_drop_write(mnt);