sage-5.2-r2 failed to build with gcc-4.7.1

[zasdfgbnm]

x86_64-pc-linux-gnu-g++ -pthread -Os -pipe -march=native -mtune=native -fno-strict-aliasing -fPIC -I/usr//include -I/usr/include/csage -Isage/ext -I/usr/include/python2.7 -c sage/rings/number_field/number_field_element_quadratic.cpp -o build/temp.linux-x86_64-2.7/sage/rings/number_field/number_field_element_quadratic.o
In file included from /usr/include/givaro/givpoly1factor.h:176:0,
                 from /usr/include/givaro/givgfq.h:23,
                 from sage/rings/finite_rings/element_givaro.cpp:279:
/usr/include/givaro/givpoly1proot.inl: In instantiation of ‘Poly1FactorDom::Rep& Poly1FactorDom::give_prim_root(Poly1FactorDom::Rep&, const Rep&) const [with Domain = GFqDom; Tag = Dense; RandIter = GivRandom; Poly1FactorDom::Rep = givvector >]’:
/usr/include/givaro/givgfq.inl:1052:3:   required from ‘GFqDom::GFqDom(GFqDom::UTT, GFqDom::UTT, const std::vector::unsigned_type>&) [with TT = int; GFqDom::UTT = unsigned int; typename Signed_Trait::unsigned_type = unsigned int]’
sage/rings/finite_rings/element_givaro.cpp:3721:84:   required from here
/usr/include/givaro/givpoly1proot.inl:294:15: error: ‘degree’ was not declared in this scope, and no declarations were found by argument-dependent lookup at the point of instantiation [-fpermissive]
/usr/include/givaro/givpoly1proot.inl:294:15: note: declarations in dependent base ‘Poly1Dom, Dense>’ are not found by unqualified lookup
/usr/include/givaro/givpoly1proot.inl:294:15: note: use ‘this->degree’ instead
/usr/include/givaro/givpoly1proot.inl:301:9: error: ‘init’ was not declared in this scope, and no declarations were found by argument-dependent lookup at the point of instantiation [-fpermissive]
/usr/include/givaro/givpoly1proot.inl:301:9: note: declarations in dependent base ‘Poly1Dom, Dense>’ are not found by unqualified lookup
/usr/include/givaro/givpoly1proot.inl:301:9: note: use ‘this->init’ instead
/usr/include/givaro/givpoly1proot.inl:311:9: error: ‘init’ was not declared in this scope, and no declarations were found by argument-dependent lookup at the point of instantiation [-fpermissive]
/usr/include/givaro/givpoly1proot.inl:311:9: note: declarations in dependent base ‘Poly1Dom, Dense>’ are not found by unqualified lookup
/usr/include/givaro/givpoly1proot.inl:311:9: note: use ‘this->init’ instead
x86_64-pc-linux-gnu-gcc -pthread -Os -pipe -march=native -mtune=native -fno-strict-aliasing -fPIC -I/usr//include -I/usr/include/csage -Isage/ext -I/usr/include/python2.7 -c sage/rings/number_field/number_field_morphisms.c -o build/temp.linux-x86_64-2.7/sage/rings/number_field/number_field_morphisms.o
x86_64-pc-linux-gnu-gcc -pthread -shared -Wl,-O1 -Wl,--as-needed -Os -pipe -march=native -mtune=native -fno-strict-aliasing build/temp.linux-x86_64-2.7/sage/rings/number_field/number_field_morphisms.o -L/usr//lib -L/usr/lib64 -lcsage -lstdc++ -lntl -lpari -lgmp -lpython2.7 -o build/lib.linux-x86_64-2.7/sage/rings/number_field/number_field_morphisms.so
x86_64-pc-linux-gnu-gcc -pthread -Os -pipe -march=native -mtune=native -fno-strict-aliasing -fPIC -I/usr//include -I/usr/include/csage -Isage/ext -I/usr/include/python2.7 -c sage/rings/number_field/totallyreal.c -o build/temp.linux-x86_64-2.7/sage/rings/number_field/totallyreal.o
x86_64-pc-linux-gnu-g++ -pthread -shared -Wl,-O1 -Wl,--as-needed -Os -pipe -march=native -mtune=native -fno-strict-aliasing build/temp.linux-x86_64-2.7/sage/rings/finite_rings/element_ntl_gf2e.o -L/usr//lib -L/usr/lib64 -lcsage -lntl -lgmp -lstdc++ -lntl -lpari -lgmp -lpython2.7 -o build/lib.linux-x86_64-2.7/sage/rings/finite_rings/element_ntl_gf2e.so
x86_64-pc-linux-gnu-gcc -pthread -Os -pipe -march=native -mtune=native -fno-strict-aliasing -fPIC -I/usr//include -I/usr/include/csage -Isage/ext -I/usr/include/python2.7 -c sage/rings/number_field/totallyreal_data.c -o build/temp.linux-x86_64-2.7/sage/rings/number_field/totallyreal_data.o
x86_64-pc-linux-gnu-g++ -pthread -shared -Wl,-O1 -Wl,--as-needed -Os -pipe -march=native -mtune=native -fno-strict-aliasing build/temp.linux-x86_64-2.7/sage/rings/number_field/number_field_element_quadratic.o -L/usr//lib -L/usr/lib64 -lcsage -lntl -lgmp -lstdc++ -lntl -lpari -lgmp -lpython2.7 -o build/lib.linux-x86_64-2.7/sage/rings/number_field/number_field_element_quadratic.so
x86_64-pc-linux-gnu-gcc -pthread -Os -pipe -march=native -mtune=native -fno-strict-aliasing -fPIC -I/usr//include -I/usr/include/csage -Isage/ext -I/usr/include/python2.7 -c sage/rings/padics/local_generic_element.c -o build/temp.linux-x86_64-2.7/sage/rings/padics/local_generic_element.o
x86_64-pc-linux-gnu-gcc -pthread -shared -Wl,-O1 -Wl,--as-needed -Os -pipe -march=native -mtune=native -fno-strict-aliasing build/temp.linux-x86_64-2.7/sage/rings/padics/local_generic_element.o -L/usr//lib -L/usr/lib64 -lcsage -lstdc++ -lntl -lpari -lgmp -lpython2.7 -o build/lib.linux-x86_64-2.7/sage/rings/padics/local_generic_element.so
x86_64-pc-linux-gnu-gcc -pthread -Os -pipe -march=native -mtune=native -fno-strict-aliasing -fPIC -I/usr//include -I/usr/include/csage -Isage/ext -I/usr/include/python2.7 -c sage/rings/padics/padic_base_coercion.c -o build/temp.linux-x86_64-2.7/sage/rings/padics/padic_base_coercion.o
In file included from sage/rings/padics/padic_base_coercion.c:285:0:
/usr/include/csage/gmp_globals.h:6:0: warning: "EXTERN" redefined [enabled by default]
In file included from sage/rings/padics/padic_base_coercion.c:282:0:
/usr/include/csage/ntl_wrap.h:34:0: note: this is the location of the previous definition
x86_64-pc-linux-gnu-gcc -pthread -shared -Wl,-O1 -Wl,--as-needed -Os -pipe -march=native -mtune=native -fno-strict-aliasing build/temp.linux-x86_64-2.7/sage/rings/number_field/totallyreal.o -L/usr//lib -L/usr/lib64 -lcsage -lpari -lgmp -lstdc++ -lntl -lpari -lgmp -lpython2.7 -o build/lib.linux-x86_64-2.7/sage/rings/number_field/totallyreal.so
x86_64-pc-linux-gnu-gcc -pthread -shared -Wl,-O1 -Wl,--as-needed -Os -pipe -march=native -mtune=native -fno-strict-aliasing build/temp.linux-x86_64-2.7/sage/rings/number_field/totallyreal_data.o -L/usr//lib -L/usr/lib64 -lcsage -lgmp -lstdc++ -lntl -lpari -lgmp -lpython2.7 -o build/lib.linux-x86_64-2.7/sage/rings/number_field/totallyreal_data.so
x86_64-pc-linux-gnu-gcc -pthread -Os -pipe -march=native -mtune=native -fno-strict-aliasing -fPIC -I/usr//include -I/usr/include/csage -Isage/ext -I/usr/include/python2.7 -c sage/rings/padics/padic_base_generic_element.c -o build/temp.linux-x86_64-2.7/sage/rings/padics/padic_base_generic_element.o
x86_64-pc-linux-gnu-gcc -pthread -Os -pipe -march=native -mtune=native -fno-strict-aliasing -fPIC -I/usr//include -I/usr/include/csage -Isage/ext -I/usr/include/python2.7 -c sage/rings/padics/padic_capped_absolute_element.c -o build/temp.linux-x86_64-2.7/sage/rings/padics/padic_capped_absolute_element.o
In file included from sage/rings/padics/padic_capped_absolute_element.c:285:0:
/usr/include/csage/gmp_globals.h:6:0: warning: "EXTERN" redefined [enabled by default]
In file included from sage/rings/padics/padic_capped_absolute_element.c:282:0:
/usr/include/csage/ntl_wrap.h:34:0: note: this is the location of the previous definition
x86_64-pc-linux-gnu-gcc -pthread -shared -Wl,-O1 -Wl,--as-needed -Os -pipe -march=native -mtune=native -fno-strict-aliasing build/temp.linux-x86_64-2.7/sage/rings/padics/padic_base_generic_element.o -L/usr//lib -L/usr/lib64 -lcsage -lgmp -lstdc++ -lntl -lpari -lgmp -lpython2.7 -o build/lib.linux-x86_64-2.7/sage/rings/padics/padic_base_generic_element.so
x86_64-pc-linux-gnu-gcc -pthread -Os -pipe -march=native -mtune=native -fno-strict-aliasing -fPIC -I/usr//include -I/usr/include/csage -Isage/ext -I/usr/include/python2.7 -c sage/rings/padics/padic_capped_relative_element.c -o build/temp.linux-x86_64-2.7/sage/rings/padics/padic_capped_relative_element.o
In file included from sage/rings/padics/padic_capped_relative_element.c:289:0:
/usr/include/csage/gmp_globals.h:6:0: warning: "EXTERN" redefined [enabled by default]
In file included from sage/rings/padics/padic_capped_relative_element.c:282:0:
/usr/include/csage/ntl_wrap.h:34:0: note: this is the location of the previous definition
sage/rings/padics/padic_capped_relative_element.c: In function ‘__pyx_f_4sage_5rings_6padics_29padic_capped_relative_element_26pAdicCappedRelativeElement__set_from_mpz_both’:
sage/rings/padics/padic_capped_relative_element.c:6458:221: warning: passing argument 2 of ‘__gmpz_cmp’ from incompatible pointer type [enabled by default]
In file included from sage/rings/padics/padic_capped_relative_element.c:280:0:
/usr/include/gmp.h:755:20: note: expected ‘mpz_srcptr’ but argument is of type ‘struct __mpz_struct (*)[1]’
sage/rings/padics/padic_capped_relative_element.c: In function ‘__pyx_f_4sage_5rings_6padics_29padic_capped_relative_element_26pAdicCappedRelativeElement__set_mpz_into’:
sage/rings/padics/padic_capped_relative_element.c:8148:218: warning: passing argument 3 of ‘__gmpz_mul’ from incompatible pointer type [enabled by default]
In file included from sage/rings/padics/padic_capped_relative_element.c:280:0:
/usr/include/gmp.h:988:21: note: expected ‘mpz_srcptr’ but argument is of type ‘struct __mpz_struct (*)[1]’
sage/rings/padics/padic_capped_relative_element.c: In function ‘__pyx_f_4sage_5rings_6padics_29padic_capped_relative_element_26pAdicCappedRelativeElement__set_mpq_into’:
sage/rings/padics/padic_capped_relative_element.c:8227:214: warning: passing argument 2 of ‘__gmpz_set’ from incompatible pointer type [enabled by default]
In file included from sage/rings/padics/padic_capped_relative_element.c:280:0:
/usr/include/gmp.h:1073:21: note: expected ‘mpz_srcptr’ but argument is of type ‘struct __mpz_struct (*)[1]’
sage/rings/padics/padic_capped_relative_element.c:8288:242: warning: passing argument 3 of ‘__gmpz_mul’ from incompatible pointer type [enabled by default]
In file included from sage/rings/padics/padic_capped_relative_element.c:280:0:
/usr/include/gmp.h:988:21: note: expected ‘mpz_srcptr’ but argument is of type ‘struct __mpz_struct (*)[1]’
x86_64-pc-linux-gnu-gcc -pthread -shared -Wl,-O1 -Wl,--as-needed -Os -pipe -march=native -mtune=native -fno-strict-aliasing build/temp.linux-x86_64-2.7/sage/rings/padics/padic_base_coercion.o -L/usr//lib -L/usr/lib64 -lcsage -lgmp -lstdc++ -lntl -lpari -lgmp -lpython2.7 -o build/lib.linux-x86_64-2.7/sage/rings/padics/padic_base_coercion.so
x86_64-pc-linux-gnu-g++ -pthread -Os -pipe -march=native -mtune=native -fno-strict-aliasing -fPIC -I/usr//include -I/usr/include/csage -Isage/ext -I/usr/include/python2.7 -c sage/rings/padics/padic_ext_element.cpp -o build/temp.linux-x86_64-2.7/sage/rings/padics/padic_ext_element.o
x86_64-pc-linux-gnu-gcc -pthread -shared -Wl,-O1 -Wl,--as-needed -Os -pipe -march=native -mtune=native -fno-strict-aliasing build/temp.linux-x86_64-2.7/sage/rings/finite_rings/integer_mod.o -L/usr//lib -L/usr/lib64 -lcsage -lgmp -lstdc++ -lntl -lpari -lgmp -lpython2.7 -o build/lib.linux-x86_64-2.7/sage/rings/finite_rings/integer_mod.so
error: command 'x86_64-pc-linux-gnu-g++' failed with exit status 1
x86_64-pc-linux-gnu-gcc -pthread -Os -pipe -march=native -mtune=native -fno-strict-aliasing -fPIC -I/usr//include -I/usr/include/csage -Isage/ext -I/usr/include/python2.7 -c sage/rings/padics/padic_fixed_mod_element.c -o build/temp.linux-x86_64-2.7/sage/rings/padics/padic_fixed_mod_element.o
 * ERROR: sci-mathematics/sage-5.2-r2 failed (compile phase):
 *   Building failed
 * 
 * Call stack:
In file included from sage/rings/padics/padic_fixed_mod_element.c:285:0:
/usr/include/csage/gmp_globals.h:6:0: warning: "EXTERN" redefined [enabled by default]
In file included from sage/rings/padics/padic_fixed_mod_element.c:282:0:
/usr/include/csage/ntl_wrap.h:34:0: note: this is the location of the previous definition

[kiwifb]

Is this a hardened set up and when did you sync the overlay? Some "fixes" to givaro where made last week and we need to be sure you got them in.

[zasdfgbnm]

$ eix givaro
[I] sci-libs/givaro [1]
Available versions: (~)3.2.16-r1^m M3.3.6^m M3.7.0^m {{static-libs}}
Installed versions: 3.2.16-r1^m(11:26:36 PM 08/16/2012)(-static-libs)
Homepage: http://ljk.imag.fr/CASYS/LOGICIELS/givaro/
Description: Givaro is a C++ library for arithmetic and algebraic computations

[1] "sage-on-gentoo" /var/lib/layman/sage-on-gentoo

[zasdfgbnm]

Is 3.2.16-r1 the version with the problem fixed?

[zasdfgbnm]

I sync my overlay just now.

[kiwifb]

Presumably fixed. I am quite unable to review any c++ templating code but another user worked hard at it. It is unfortunate that we still seem to have trouble with it.
I may have to make an executive decision and adopt sage's patches for good or bad until we can switch to givaro-3.7.0, this is now a sage 5.4 item and not a 5.3 to my disappointment.

[jauhien]

Hm. I'm becoming really tired of this package. I've thought its 'make check' should discover all issues with gcc-4.7. Will try to find a solution for finding all this problems other then reading through all givaro sources.
Until it will be fixed I suggest zasdfgbnm set -fpermissive in CXXFLAGS. This should help. Btw, zasdfgbnm, if you are using gcc-4.7 in gentoo that is masked you should be an experienced user and may be you can look for solution (the problem really is simple: degree and other similar stuff should be changed to this->degree, but the problem is where it should be changed, with the last patch I was sure, I've changed it anywhere it was needed, but I see I was wrong).

[jauhien]

Oh, yeah. It seems I've understood where was the problem with the last patch. May be veelai had wrong version of givaro, because previous patching in givaro was done without changing revision number, so portage did not update it on veelai's computer after it was patched in overlay and so previous patch really worked for sage (but not for givaro's 'make check').
After veelai reported issue I rechecked givaro packaged, but only with its built in 'make check' and fixed all the problems found. But I see now some parts of code are not covered by 'make check', and those parts sage uses. So, there was problems fixed in previous patch, that are not fixed by the patch I've made using 'make check'. I'm looking for this problems now and will add fix to the current patch. I hope after this it will work.
Summary is: I will test everything next time.

[jauhien]

I've found 3 places that need fix:
/usr/include/givaro/givpoly1proot.inl:294:15: note: use ‘this->degree’ instead
/usr/include/givaro/givpoly1proot.inl:301:9: note: use ‘this->init’ instead
/usr/include/givaro/givpoly1proot.inl:311:9: note: use ‘this->init’ instead

Will do it in few hours.

[kiwifb]

I am very grateful that you are looking at this. I understand the basic concept of c++ but most code that use a decent amount of features of the language look like spaghetti to me.

[kiwifb]

Thanks for the pull. Can you test zasdfgbnm?

[jauhien]

Frankly speaking it was my fault: I have not checked all the previous fixes important for sage was included in new patch. I checked it now. sage should compile without errors, and I hope it's the last problem with givaro, sage and gcc-4.7. )
There is one point more: afaik there are other packages dependent on givaro, I'll check them in a few time, may be there are still problems with them. So please do not close issue until I recheck this.
And yes, zasdfgbnm, check this fix please.

[jauhien]

I've found only these packages depending on givaro on my computer:
jauhien@jpiatlicki givaro % equery depends givaro

• These packages depend on givaro:
  sci-libs/linbox-1.1.6-r7 (=sci-libs/givaro-3.2_)
  sci-mathematics/sage-5.2-r2 (=sci-libs/givaro-3.2_)

I've tested sage once again with version of givaro ebuild synced from overlay. I've tested linbox too. Both sage and linbox do compile.

zasdfgbnm: please test not only whether sage compiles, but also please recompile linbox, so it also will be tested.

[zasdfgbnm]

Both sage and linbox compile successfully

[kiwifb]

thank you! Closing now.