Hi, I am experiencing a gcc/g++ crash when compiling polymake (the polyhedron manipulation utility at http://www.math.tu-berlin.de/polymake/, download at http://www.math.tu-berlin.de/polymake/download_choice.html):
--- g++ -c -o facets_from_incidence.o -I../../../apps/polytope/include -I../../../modules/co mmon/include -I../../../modules/graph/include -I../../../lib/poly_client/include -I../../. ./lib/PTL/include -I../../../lib/PTL/include/std -I../../../lib/gmp_wrapper/include -march =pentium4 -Wall -ftemplate-depth-200 -DALL_PLAUSIBLE_CHECKS=0 -O0 -MD ../../../apps/polyt ope/src/facets_from_incidence.cc g++: Internal error: Segmentation fault (program cc1plus) Please submit a full bug report. See <URL:http://gcc.gnu.org/bugs.html> for instructions. make[2]: *** [facets_from_incidence.o] Error 1 make[1]: *** [do_all] Error 2 make[1]: Leaving directory `/cygdrive/d/install/Polymake/x/polymake-2.1.0/build/apps/polyt ope' make: *** [all] Error 2 --- Since cc1plus is pushing the total commit charge for the machine to 1.1GB at the point of the crash, I suspect this may be a memory-related problem. The cc1plus process is taking 560,116KB at the moment of the crash. As you can see from the above, it also happens when I disable optimizations (-O0). A successful build on a Red Hat Enterprise Linux 4 showed that gcc/g++ needs about 1.5GB to get through facets_from_incidence.cc I also investigated whether this is an inherent Windows limitation. As is well known, Windows XP allows at most 2GB of user-side memory. A test program I wrote successfully allocated 1.7GB in 10 blocks of 170MB each on this machine. Your help would be most appreciated. Sincerely, Dan
cygcheck.out
Description: Binary data
-- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
