On Monday 31 January 2005 14:52, Salvatore Filippone wrote: > Hi there, > I have been chasing a bug for a coupple of days, and have found the > following behaviour: > > integer, pointer :: itmp(:) ! =>null() > ........ > > itmp => b%ia1 > b%ia1 => b%ia2 > b%ia2 => itmp Does -fno-strict-aliasing fix your problem? You may be hitting PR16898. Paul