[MITgcm-devel] testreport and cross compiling

[Ed Hill]

On Tue, 12 Feb 2008 10:47:08 +0100 Martin Losch <Martin.Losch at awi.de>
wrote:

> To follow up on this: I have replace in genmake2 all lines (there  
> were 3)
> make genmake_tc_1.o >> genmake_warnings 2>&1
> with
>      COMM="$CC $CFLAGS -c genmake_tc_1.c"
>      echo $COMM >> genmake_warnings
>      $COMM >> genmake_warnings 2>&1
> in anology to the fortran command. With this change, my problems on  
> the SX8 are solved, both for genmake and for testreport.
> 
> If there is no particular reason for using the make-command for  
> compiling the c-test-codes, I would like to check in this change to  
> genmake2. May I?


Hi Martin,

Good catch!

I think you're right that it is better to use the more explicit
"$CC $CFLAGS ..." syntax rather than the make invocation.  I would
vote that you check it into CVS.  And it makes sense that it took a
cross-compiling system such as your SX to trigger this problem.

Also, if you do check it into CVS then it *may* temporarily break
testreport on a few systems where the default CC=cc does not point to a
usable C compiler.  In those cases, we can simply ask people to define
CC to point to a valid C compiler.

Ed

-- 
Edward H. Hill III, PhD  |  ed at eh3.com  |  http://eh3.com/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://mitgcm.org/pipermail/mitgcm-devel/attachments/20080212/b45b8ba5/attachment.sig>