> As I said, f_pos is changed without holding BKL in the VFS already. Therefore IFF the VFS paths that change it can be invoked in parallel. > even if the driver tries to protect f_pos by holding the BKL it is racing > against concurrent read()/write() anyway on f_pos. It may simply be a matter of internal consistency. Anyway I've now made a list of the offenders so we can simply fix/delete them and continue. -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
