Custom Search
-- Jason Swails Quantum Theory Project, University of Florida Ph.D. Graduate Student 352-392-4032 On Nov 9, 2011, at 10:08 PM, case <case.biomaps.rutgers.edu> wrote: > On Wed, Nov 09, 2011, Michael Sternberg wrote: >> >> Speaking of which - I noted a few instances where "LOCALFLAGS = >> $(FREEFORMAT_FLAG)" was set but is was not actually passed on to the >> Fortran compiler because there were rule overrides either in the >> Makefile or in ./depend. This was on the following platform: > > What *should* be happening is that > > (a) FFLAGS and FOPTFLAGS both include $(LOCALFLAGS) (in config.h), > (b) LOCALFLAGS is set in the sander Makefile, > (c) the compilation rules in sander's depend script use FFLAGS or FOPTFLAGS > > Final result is that the free-format flag gets passed to all compiles. > > This certainly happens on my machine, and must be happening on everyone else's > as well, as we have been using this scheme for many years. If the patch you > suggest were needed, no one would have been able to compile sander. > > You patch suggests that parts (b) and (c) are fine on your computer, so > you should examine your config.h file, which should have lines like this: > > FFLAGS= -O0 $(LOCALFLAGS) $(AMBERBUILDFLAGS) > FOPTFLAGS= -O3 $(LOCALFLAGS) $(AMBERBUILDFLAGS) > > [some options will be different for intel, but $(LOCALFLAGS) should be there.] > > If indeed your config.h file does not conform to this expectation, we need to > find out how you made it. You environment looks very vanilla to me. As an > experiment, try asking for gnu compilers rather than intel, just to see if > that makes any difference. > > ....dac > > > _______________________________________________ > AMBER mailing list > AMBER.ambermd.org > http://lists.ambermd.org/mailman/listinfo/amber _______________________________________________ AMBER mailing list AMBER.ambermd.org http://lists.ambermd.org/mailman/listinfo/amberReceived on Wed Nov 09 2011 - 22:00:02 PST