On Thu, Apr 26, 2018 at 12:10 PM, Vivek Goyal <vgoyal@xxxxxxxxxx> wrote: > When a fsync happens, it calls fsync on real underlying file. If a file > has been metadata only copied up on upper, then file containing data > is the real file and fsync will be called on lower data file. > > We should be issuing fsync on upper metacopy inode as well to make sure > upper inode has been made persistent on disk. This is papering over a bug in Miklos's patch. fsync should ONLY ever sync upper inode. same as a bug that was recently fixed in ovl_dir_fsync(). Thanks, Amir. -- 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
