Hello,
it took a while, but I managed to compile all of Sage 2.8.15 with gcc
4.3.0 20071130 (experimental). Thanks to Carl Witty who gave me a
couple pointers it happened much more smoothly than I had expected.
Also thanks to cartman who fixed the LinBox gcc 4.3 issue that I
should have fixed a months ago - once you see the fix it is so obvious
it is scary.
Anyway, with a bunch of patches Sage starts, but we have some doctest
failures:
devel/sage-main/sage/misc/functional.py
devel/sage-main/sage/libs/mwrank/mwrank.pyx
devel/sage-main/sage/libs/mwrank/interface.py
devel/sage-main/sage/schemes/elliptic_curves/ell_point.py
devel/sage-main/sage/schemes/elliptic_curves/lseries_ell.py
devel/sage-main/sage/schemes/elliptic_curves/ell_rational_field.py
devel/sage-main/sage/schemes/elliptic_curves/padics.py
devel/sage-main/sage/schemes/elliptic_curves/sha.py
devel/sage-main/sage/schemes/elliptic_curves/padic_lseries.py
devel/sage-main/sage/schemes/elliptic_curves/ell_tate_curve.py
Those are all cause by mwrank. The issue seems to be the same issue as
Justin Walker reported in the 2.8.15 thread in sage-support.
Interestingly nobody ever reported it and I now finally have a
testcase that also produces the issue on sage.math under valgrind. I
have send John Cremona two valgrind traces and hope that it will help
to fix the issues.
Here are my short notes about the port. It is slash & burn style, so
it will take some effort to clean up the patches and send them
upstream and integrate them cleanly. Some of those already have trac
tickets, so once I get some sleep I will open tickets for the
remainder and start cleaning up the patches.
Cheers,
Michael
creomona: various header fixes, patch going to John
gmp:
* extern inline workaround from OSX
* uncomment std::FILE -> causes trouble - need clean fix
* abort() in t-locale.cc -> #include <cstdlib>
givario: givaromm.C -> add #include <cstring>
linbox:
* linbox/gmp++/gmp++.h -> include "limits.h"
* linbox/linbox/ffpack/ffpack.h:1037 rename double argument
mwrank: same as cremona, is going to get nuked anyway.
mpfi: tests fail to link, had to remove them from configure.ac to make
the library compile.
mpfr: have file issue -> libfplll
singular:
* ndbm.cc:30:8: error: extra tokens at end of #endif directive ->
insert comment
* globals.h:
#include <iostream>
#include <fstream>
#include <iomanip>
* using namespace std; in term_ordering.h, LLL.c, matrix.c,
matrix.h, test_data.h
* matrix.h, testdata.h: #include <fstream>
libfplll:
* src/main.cpp: <cstring>
* nr.cpp: mpz* string function HAVE_FILE issues in gmp, mpfr
sage:
cc1plus: warning: command line option "-Wstrict-prototypes" is valid
for Ada/C/ObjC but not for C++
sage/rings/real_rqdf.cpp:51:1: error: "INLINE" redefined
In file included from /tmp/Work/sage-2.8.15/local//include/pari/pari.h:
67,
from sage/rings/real_rqdf.cpp:46:
/tmp/Work/sage-2.8.15/local//include/pari/parisys.h:48:1: error: this
is the location of the previous definition
error: command 'gcc' failed with exit status 1
sage: There was an error installing modified sage library code.
Quick fix: change inline in parisys.h:48 to match Cython's
gcc -fno-strict-aliasing -DNDEBUG -g -O3 -Wall -Wstrict-prototypes
-fPIC -I/tmp/Work/sage-2.8.15/local//include
-I/tmp/Work/sage-2.8.15/local//include/csage
-I/tmp/Work/sage-2.8.15/devel//sage/sage/ext
-I/tmp/Work/sage-2.8.15/local/include/python2.5 -c sage/combinat/
partitions_c.cc -o build/temp.linux-x86_64-2.5/sage/combinat/
partitions_c.o -w
cc1plus: warning: command line option "-Wstrict-prototypes" is valid
for Ada/C/ObjC but not for C++
sage/combinat/partitions_c.cc:141:13: error: missing terminating '
character
sage/combinat/partitions_c.cc:307: error: explicit template
specialization cannot have a storage class
sage/combinat/partitions_c.cc:308: error: explicit template
specialization cannot have a storage class
remove "static inline". "static" itself might be enough.
fix #error string
#include <cstring>
lcalc:
Lriemannsiegel.cc:38: #include <cstring>
Lgamma.h:257 #include <cstring>
flintqs:
QS.cpp:1566: error: second argument of 'int main(int, unsigned
char**)' should be 'char **'
QS.cpp: In function 'int main(int, unsigned char**)':
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~----------~----~----~----~------~----~------~--~---
