the calls to it had different number of arguments in different places.
some compilers obviously don't care since the same code compiled
on my PCs and at NCSA just fine, but the BG compiler apparently insists
that you be consistent...
carlos
On 9/25/07, Kim F Wong <kfwong.hec.utah.edu> wrote:
>
> Ross & Carlos,
>
> Can you tell me why evb_io_debug and evb_gfchk are croaking on bluegene
> ... that is, are there any programming habits I need to break? The
> evb_io_debug was for my own edification and can be commented out but the
> evb_gfchk is integral to Berny Schelgel's distributed Gaussian EVB
> approach. Sorry the codes are messy. I will clean up the debugging soon.
>
> -Kim
>
>
> Ross Walker wrote:
> > Hi Carlos,
> >
> > I fixed the calls for evb_io_debug in evb_gfchk.f and now things go
> further
> > but that file is now crashing the compiler:
> >
> > /users/rcw/amber10/src % make parallel
> > Starting installation of Amber10 (parallel) at Tue Sep 25 13:48:01 PDT
> 2007.
> > cd sander; make parallel
> > make[1]: Entering directory `/users/rcw/amber10/src/sander'
> > ./checkparconf
> > /opt/ibmcmp/xlf/bg/10.1/exe/cpp -P -DNMLEQ -DCLINK_PLAIN -DXLF90
> > -I/bgl/BlueLight/ppcfloor/bglsys/include/ -DPOE -DMPI -DUSE_MPI_IN_PLACE
> > evb_gfchk.f > _evb_gfchk.f
> > mpxlf90 -qarch=440 -qtune=440 -c -qfixed -c -qfree=f90 -o evb_gfchk.o
> > _evb_gfchk.f
> > ** evb_gfchk === End of Compilation 1 ===
> > ** evb_gfchk_alloc === End of Compilation 2 ===
> > ** evb_gfchk_dealloc === End of Compilation 3 ===
> > EXCEPTION terminates program: no handler for exception.
> > Exception of type: UNDIAGNOSED at 100D0710. Regs:
> > 101743D0 FFFFCC40 4004BD20 10379060 00000000 00001228 00000000 00000013
> > 1028A4C8 00000000 103792D8 0000496C 102A2EE8 1023D528 10230000 10230000
> > 0000006A 0000006A 00000000 FFFFD538 00000050 FFFFCDE8 10230000 0000065A
> > FFFFCDE4 FFFFCC70 00001978 0000120B 10230000 102943F8 1029461C 10230000
> > resume: cr: lr: ctr: xer: fpscr: msr: mq:
> > 100D0714 28024842 101743D0 100CE088 00000000 00000000 0000D032 00000000
> > Traceback:
> > Line ? Disp 100D0710 ExtendedCompareExpr Dsa:
> FFFFCC40
> > Line ? Disp 101743CC ScReplaceMATMUL Dsa:
> FFFFCC60
> > Line ? Disp 1015CB20 ProcessExpr Dsa:
> FFFFCE70
> > Line ? Disp 1015CD48 ProcessExpr Dsa:
> FFFFCFE0
> > Line ? Disp 1015CB4C ProcessExpr Dsa:
> FFFFD150
> > Line ? Disp 1015CD1C ProcessExpr Dsa:
> FFFFD2C0
> > Line ? Disp 1015DC88 ScProcessExpr Dsa:
> FFFFD430
> > Line ? Disp 100AC048 ScProcessStmt Dsa:
> FFFFD470
> > Line ? Disp 1004CAD8 StmtWalk Dsa:
> FFFFD4D0
> > Line ? Disp 100A65D4 Scalarizer Dsa:
> FFFFD530
> > Line ? Disp 100418F8 InvokeScalarizer Dsa:
> FFFFD580
> > Line ? Disp 10041BE8 astidrvr Dsa:
> FFFFD5B0
> > Line ? Disp 10002724 main Dsa:
> FFFFD610
> > Line ? Disp 0FCF4D10 generic_start_main Dsa:
> FFFFD6D0
> > --- End of call chain ---
> > 1501-511 Compilation failed for file _evb_gfchk.f.
> > make[1]: *** [evb_gfchk.o] Error 217
> > make[1]: Leaving directory `/users/rcw/amber10/src/sander'
> > make: *** [parallel] Error 2
> >
> > Can you try it on your system - with the updated cvs and see if you see
> the
> > same problem and if you do I'll escalate it to IBM.
> >
> > All the best
> > Ross
> >
> > /\
> > \/
> > |\oss Walker
> >
> > | HPC Consultant and Staff Scientist |
> > | San Diego Supercomputer Center |
> > | Tel: +1 858 822 0854 | EMail:- ross.rosswalker.co.uk |
> > | http://www.rosswalker.co.uk | PGP Key available on request |
> >
> > Note: Electronic Mail is not secure, has no guarantee of delivery, may
> not
> > be read every day, and should not be used for urgent or sensitive
> issues.
> >
> >
> >
> >
>
>
Received on Wed Sep 26 2007 - 06:07:25 PDT
