Custom Query (964 matches)
Results (37 - 39 of 964)
Ticket | Owner | Reporter | Resolution | Summary |
---|---|---|---|---|
#24 | Fixed | BL_TEST case suffering from bit rot | ||
Description |
The BL_TEST case appears to lack some preprocessor macros that it needs. The macro ANA_WWAVE is defined, but there is a compiler error on the source in ana_wwave.h. See below. Should this case be updated or deleted? The error: in ana_wwave.h the following code is specified #if defined BL_TEST wdir=210.0_r8*deg2rad DO j=JstrR,JendR DO i=IstrR,IendR Hwave(i,j)=0.5_r8 Dwave(i,j)=wdir Pwave(i,j)=8.0_r8 END DO END DO #elif ... however Hwave is available only if WAVES_HEIGHT is defined (which it isn't), Dwave is available only if BBL_MODEL or NEARSHORE_MELLOR is defined (which they aren't), and Pwave appears to have been replaced by Pwave_top and Pwave_bot. |
|||
#33 | Fixed | Remove references to CANYON_A and CANYON_B in ana_*.h | ||
Description |
ROMS used to have CANYON_A and CANYON_B test cases, the former being 2D and the latter 3D (I think). Now there is just one preprocessor symbol (CANYON) used for these cases, with corresponding header file canyon.h, and the dimensionality is controlled by SOLVE3D. However there are still references to CANYON_A and CANYON_B in ROMS/Functionals/ana_*.h. I have replaced them with CANYON throughout. See attached files. |
|||
#41 | Fixed | Minor changes to platform-specific make files | ||
Description |
Attached are modified versions of 3 files: Compilers/CYGWIN-gfortran.mk Compilers/Linux-gfortran.mk Compilers/Linux-ifort.mk Changes are CYGWIN-gfortran.mk: add OpenMP-related settings, as Cygwin Gfortran now supports OpenMP; remove warnings flags for USE_DEBUG, as the length of my_fflags in mod_strings.F is too long in some situations, causing a syntax error, and no-one ever pays attention to warnings anyway. Linux-gfortran.mk: removing warning flags as above Linux-ifort.mk: add -traceback flag when USE_DEBUG is set. |