[MITgcm-devel] Changing _R4/_R8 macros
Dimitris Menemenlis
menemenlis at jpl.nasa.gov
Mon Apr 27 23:05:00 EDT 2009
JM isn't there some rule relating # of CVS commits and beer owed? Are you trying to cheat? ...
What you suggest is reasonable. I will update affected contrib stuff after your check in..
Dimitris Menemenlis
818-625-6498
-----Original Message-----
From: Jean-Michel Campin <jmc at ocean.mit.edu>
Date: Monday, Apr 27, 2009 3:46 pm
Subject: [MITgcm-devel] Changing _R4/_R8 macros
To: "mitgcm-devel at mitgcm.org" <mitgcm-devel at mitgcm.org>Reply-To: "MITgcm-devel at mitgcm.org" <MITgcm-devel at mitgcm.org>
Hi,
I bumped (once more) into this problem when I tried:
_GLOBAL_SUM_R4( tmpR4 , myThid )
with tmpR4 declared real*4 ...
I think it would be good to have a consistent set of macros for
_EXCH_ and _GLOBAL_SUM/MAX_ , using the suffix:
_R4 when applied to real*4 variable
_R8 when applied to real*8 variable
_RS when applied to real(_RS) variable
_RL when applied to real(_RL) variable
Unfortunately, it involves changing many files (~170).
I still have few details to fix, but will be ready to check-in
those changes sometime tomorrow.
Should I disable the sending of CVS-commit emails ?
I could do so for all check-in with the same (standard) cvs-commit
message:
"change macros (EXCH & GLOBAL_SUM/MAX) sufix _R4/_R8 to _RS/_RL
when applied to _RS/_RL variable"
and then turn it back for the few special cases with
a more specific cvs-commit message.
Comments ?
Cheers,
Jean-Michel
PS: if someone wants to update a new piece of code, here is
what I was using to make "standard" changes:
cat > modif.sed <<EOF
s/_GLOBAL_SUM_R4/_GLOBAL_SUM_RS/g
s/_GLOBAL_SUM_R8/_GLOBAL_SUM_RL/g
s/_GLOBAL_MAX_R4/_GLOBAL_MAX_RS/g
s/_GLOBAL_MAX_R8/_GLOBAL_MAX_RL/g
s/_EXCH_XY_R4/_EXCH_XY_RS/g
s/_EXCH_XY_R8/_EXCH_XY_RL/g
s/_EXCH_XYZ_R4/_EXCH_XYZ_RS/g
s/_EXCH_XYZ_R8/_EXCH_XYZ_RL/g
EOF
sed -f modif.sed file_to_update > updated_file
_______________________________________________
MITgcm-devel mailing list
MITgcm-devel at mitgcm.org
http://mitgcm.org/mailman/listinfo/mitgcm-devel
More information about the MITgcm-devel
mailing list