On 18 April 2010 06:00, Minh Nguyen <nguyenmi...@gmail.com> wrote:
> Hi folks,
>
> Building Sage 4.4.alpha0 on t2.math failed with the following error message:
>
> gcc -fno-strict-aliasing -DNDEBUG -g -fwrapv -O3 -Wall
> -Wstrict-prototypes -fPIC
> -I/home/mvngu/sandbox/sage-4.4.alpha0-t2.math/local//include
> -I/home/mvngu/sandbox/sage-4.4.alpha0-t2.math/local//include/csage
> -I/home/mvngu/sandbox/sage-4.4.alpha0-t2.math/devel//sage/sage/ext
> -I/home/mvngu/sandbox/sage-4.4.alpha0-t2.math/local/include/python2.6
> -c sage/plot/plot3d/base.c -o
> build/temp.solaris-2.10-sun4v-2.6/sage/plot/plot3d/base.o -w
> sage/plot/plot3d/base.c: In function
> ‘__pyx_pf_4sage_4plot_6plot3d_4base_point_list_bounding_box’:
> sage/plot/plot3d/base.c:18661: error: ‘INFINITY’ undeclared (first use
> in this function)
> sage/plot/plot3d/base.c:18661: error: (Each undeclared identifier is
> reported only once
> sage/plot/plot3d/base.c:18661: error: for each function it appears in.)
> error: command 'gcc' failed with exit status 1
> sage: There was an error installing modified sage library code.
>
> ERROR installing SAGE
>
> real    167m42.848s
> user    140m38.556s
> sys     7m36.294s
> sage: An error occurred while installing sage-4.4.alpha0
>
> So the build error is triggered when trying to build the Sage library.

Which is one reason I think library code should be tested on Solaris
too. In this case,

http://hackage.haskell.org/trac/ghc/ticket/2929

might be relevant.

The C macro INFINITY is apparently defined by the C99 standard. On
Solaris they take a strict interpretation of this and only define it
conditionally:

#if defined(_STDC_C99) || _XOPEN_SOURCE - 0 >= 600 || defined(__C99FEATURES__)

I've not tried it, but if the library wants to use c99  features, it
should add the -std=c99 option.

Dave

-- 
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Reply via email to