Probably this code was syncing a lot more often then intended because
the do_sync variable wasn't set to zero.
Cc: stable@vger.kernel.org # v3.11+
Fixes: c62988ec0910 ('ceph: avoid meaningless calling ceph_caps_revoking if sync_mode == WB_SYNC_ALL.')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Ilya Dryomov <idryomov@redhat.com>
        int rc = 0;
        unsigned wsize = 1 << inode->i_blkbits;
        struct ceph_osd_request *req = NULL;
-       int do_sync;
+       int do_sync = 0;
        u64 truncate_size, snap_size;
        u32 truncate_seq;