LIM3 compiler error with gfortran (gcc48) in limdiahsb

[clevy]

In the trunk revision 4862, running the SETTE tests when using gfortran compiler (gcc48), 
/Users/clairelevy/Documents/WORK/FORORCA1/NEMOGCM/CONFIG/ORCA2LIM3_LONG/BLD/ppsrc/nemo/limdiahsb.f90:88.26:

  zinda = MAX( 0.d0 , SIGN( 1.d0 , glob_sum( area(:,:) * tms(:,:) ) - epsi0
    1

Error: Generic function 'sign' at (1) is not consistent with a specific intrinsic interface
/Users/clairelevy/Documents/WORK/FORORCA1/NEMOGCM/CONFIG/ORCA2LIM3_LONG/BLD/ppsrc/nemo/limdiahsb.f90:164.26:

  zindb = MAX( 0.d0 , SIGN( 1.d0 , zbg_ivo - epsi06 ) )
    1

Error: Generic function 'sign' at (1) is not consistent with a specific intrinsic interface

In the trunk revision 4862, when using gfortran compiler,
/Users/clairelevy/Documents/WORK/FORORCA1/NEMOGCM/CONFIG/ORCA2LIM3_LONG/BLD/ppsrc/nemo/limdiahsb.f90:88.26:

zinda = MAX( 0.d0 , SIGN( 1.d0 , glob_sum( area(:,:) * tms(:,:) ) - epsi0

1

Error: Generic function 'sign' at (1) is not consistent with a specific intrinsic interface
/Users/clairelevy/Documents/WORK/FORORCA1/NEMOGCM/CONFIG/ORCA2LIM3_LONG/BLD/ppsrc/nemo/limdiahsb.f90:164.26:

zindb = MAX( 0.d0 , SIGN( 1.d0 , zbg_ivo - epsi06 ) )

1

Error: Generic function 'sign' at (1) is not consistent with a specific intrinsic interface

[timgraham]

I think this has been fixed as part of the 2014 merge. The above line now reads:

zinda = MAX( 0._wp , SIGN( 1._wp , glob_sum( area(:,:) * tms(:,:) ) - epsi0

I think the error might be due to the use of 1.d0 in the call to SIGN. Is anyone able to test this easily with the gfortran compiler to see if we can close the ticket?

[clem]

yes, I tested it and it is working with gfortran 4.8.2 (ubuntu OS)