[MITgcm-devel] no _d in header files (exf_constants.h)

[Chris Hill]

That's interesting! 

I think the proper fix is to change CPP_EEMACROS.h (which has an erroneous
_d definition) and make line read.

CPP = cat $< | cpp  -traditional -P | $(TOOLSDIR)/set64bitConst.sh

 but that will need a quick check on every experiment and fixing in
genmake2.

 Too late on Friday to do that, so I suggest you check in exf_constants.h
changed for now. 

 Where are you?

Chris

> -----Original Message-----
> From: mitgcm-devel-bounces at mitgcm.org 
> [mailto:mitgcm-devel-bounces at mitgcm.org] On Behalf Of Patrick Heimbach
> Sent: Friday, May 07, 2004 5:40 PM
> To: MITgcm-devel at mitgcm.org; Chris Hill
> Cc: MITgcm-devel at mitgcm.org
> Subject: RE: [MITgcm-devel] no _d in header files (exf_constants.h)
> 
> Chris, 
> 
> I am using it.
> Here the excerpt from my Makefile:
> 
> CPP = cat $< | $(TOOLSDIR)/set64bitConst.sh |  cpp  -traditional -P
> 
> So maybe set64bitConst.sh just needs fixing.
> But it looks sort of OK.
> 
> Let me know.
> I'm tempted to check in the modified exf_constants.h it's 
> really the only header that does use '_d'.
> 
> 
> Quoting Chris Hill <cnh at MIT.EDU>:
> 
> > It is much better to use "set64bitConst.sh" because then when you 
> > build on an NEC (for example) you can easily switch _d -> 
> e. Not using 
> > _d is how you start rumours that the code doesn't vectorize etc...
> > 
> > Do you know if "set64bitConst.sh" is being called in the 
> makefile or 
> > whether it is being left out all together.
> > 
> > Chris
> > 
> > > -----Original Message-----
> > > From: mitgcm-devel-bounces at mitgcm.org
> > 
> > > [mailto:mitgcm-devel-bounces at mitgcm.org] On Behalf Of Patrick 
> > > Heimbach
> > > Sent: Friday, May 07, 2004 5:09 PM
> > > To: mitgcm-devel at mitgcm.org
> > > Subject: [MITgcm-devel] no _d in header files (exf_constants.h)
> > > 
> > > Hi all,
> > > 
> > > a reminder that TAMC (as opposed to TAF) cannot handle blanks in 
> > > parameter statements of the form (1. d 0).
> > > Therefore, ' _d ' schould be avoided in .h files to maintain TAMC 
> > > compatibility.
> > > 
> > > I know Chris mentioned the set64bitConst.sh script which ought to 
> > > remove the blanks, but it doesn't, at least not on slough (SunOS 
> > > 5.8) where TAMC is installed.
> > > 
> > > The good news is that only one file seems to be affected, 
> > > exf_constants.h (tsss, those Greek-Canadians...) and 
> > > verification/carbon/ successfully ran with TAMC (it 
> doesn't use exf).
> > > 
> > > -p.
> > > 
> > > 
> > > _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
> > > Patrick Heimbach     Massachusetts Institute of Technology
> > > FON: +1/617/253-5259                    EAPS, Room 54-1518
> > > FAX: +1/617/253-4464               77 Massachusetts Avenue
> > > mailto:heimbach at mit.edu                 Cambridge MA 02139
> > > http://www.mit.edu/~heimbach/                          USA
> > > 
> > > _______________________________________________
> > > MITgcm-devel mailing list
> > > MITgcm-devel at mitgcm.org
> > > http://dev.mitgcm.org/mailman/listinfo/mitgcm-devel
> > > 
> > 
> > _______________________________________________
> > MITgcm-devel mailing list
> > MITgcm-devel at mitgcm.org
> > http://dev.mitgcm.org/mailman/listinfo/mitgcm-devel
> > 
> 
> 
> _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
> Patrick Heimbach     Massachusetts Institute of Technology
> FON: +1/617/253-5259                    EAPS, Room 54-1518
> FAX: +1/617/253-4464               77 Massachusetts Avenue
> mailto:heimbach at mit.edu                 Cambridge MA 02139
> http://www.mit.edu/~heimbach/                          USA
> 
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://dev.mitgcm.org/mailman/listinfo/mitgcm-devel
>