On Friday, January 9, 2015 at 12:30:16 PM UTC+1, Jean-Pierre Flori wrote: > > > > On Friday, January 9, 2015 at 9:49:52 AM UTC+1, Sebastien Gouezel wrote: >> >> The webpage http://trac.sagemath.org/wiki/Cygwin64Port indicates that >> sage can now be built under cygwin64 almost out of the box. So, I tried, >> but I failed... >> >> Here are the problems I encountered, maybe someone with better computer >> skills can see what is going on. My sage version is sage 6.4beta5 with >> #17365 pulled. Very recent cygwin64 > > >> - First, pyzmq did not compile, because of undefined references in >> libzmq.a. It seems that the flag -no-undefined was not passed to ld >> while building libzmq.a (despite #17333). I hacked the makefile to make >> sure that the flag was passed when compiling zeromq, this solved the >> issue. >> >> Note that "-no-undefined" is a libtool flag, not an ld one. > I'll try to have a look next week asI'll be in the wild for more than a > month after that. > Ok, I screwed up that one. Have a look at: * http://git.sagemath.org/sage.git/tree/build/pkgs/zeromq/patches/build_system/cygwin_sharedlib.patch?id=f1ae029d18d5ded509461f9b092e3f48ad0d5e64#n26 The if/endif part is at a wrong place...
> > - Then, while compiling sage proper, I get the following error >> >> gcc -I/usr/include/ncurses -fno-strict-aliasing -g -O2 -DNDEBUG -g >> -fwrapv -O3 -Wall -I/home/Sebastien/sage/local/include >> -I/home/Sebastien/sage/local/include/csage -I/home/Sebastien/sage/src >> -I/home/Sebastien/sage/src/sage/ext >> -I/home/Sebastien/sage/local/include/python2.7 -c >> build/cythonized/sage/graphs/distances_all_pairs.c -o >> build/temp.cygwin-1.7.33-x86_64-2.7/build/cythonized/sage/graphs/distances_all_pairs.o >> >> >> -fno-strict-aliasing -w -fno-tree-dominator-opts >> >> gcc -shared -Wl,--enable-auto-image-base >> -L/home/Sebastien/sage/local/lib >> build/temp.cygwin-1.7.33-x86_64-2.7/build/cythonized/sage/graphs/distances_all_pairs.o >> >> >> -L/home/Sebastien/sage/local/lib >> -L/home/Sebastien/sage/local/lib/python2.7/config >> -L/home/Sebastien/sage/local/lib -lcsage -lpython2.7 -o >> build/lib.cygwin-1.7.33-x86_64-2.7/sage/graphs/distances_all_pairs.dll >> >> build/temp.cygwin-1.7.33-x86_64-2.7/build/cythonized/sage/graphs/distances_all_pairs.o: >> >> >> dans la fonction « >> __pyx_f_4sage_6graphs_19distances_all_pairs_bitset_clear »: >> >> /home/Sebastien/sage/src/build/cythonized/sage/graphs/distances_all_pairs.c:2010: >> >> >> référence indéfinie vers « __imp___gmpn_zero » >> >> /home/Sebastien/sage/src/build/cythonized/sage/graphs/distances_all_pairs.c:2010:(.text+0x1493): >> >> >> relocalisation tronquée pour concorder avec la taille: R_X86_64_PC32 >> vers le symbole indéfini __imp___gmpn_zero >> >> /home/Sebastien/sage/src/build/cythonized/sage/graphs/distances_all_pairs.c:2010: >> >> >> référence indéfinie vers « __imp___gmpn_zero » >> >> /home/Sebastien/sage/src/build/cythonized/sage/graphs/distances_all_pairs.c:2010:(.text+0xba8e): >> >> >> relocalisation tronquée pour concorder avec la taille: R_X86_64_PC32 >> vers le symbole indéfini __imp___gmpn_zero >> >> That is easy to fix. > Just add gmp to the linked libraries in module_list.py for these files. > People should really pay attention to what they use in their code, Linux > is just too nice. > -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at http://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
