On a system recompiled with ./configure --with-system-givaro==no, ptestlong reports no new failures (= not already seen in previous versions). Mattias’ diagnosis seems exact and my workaround seems to work. Le samedi 18 décembre 2021 à 19:48:55 UTC+1, Emmanuel Charpentier a écrit :
> It seems to : while rebuilding from scratch, linbox compilation just > completed successfully. More after ptestlong… > > Le samedi 18 décembre 2021 à 19:10:25 UTC+1, Emmanuel Charpentier a écrit : > >> >> >> Le samedi 18 décembre 2021 à 19:01:43 UTC+1, [email protected] a >> écrit : >> >>> This is https://trac.sagemath.org/ticket/33042 >> >> >> Thanks a lot ! >> >> Is there a workaround ? I’m thinking of a way to ./configure Sage to >> force the use of “acceptable” versions. Would --with-system-givaro >> suffice ? >> >> >> On Saturday, December 18, 2021 at 6:11:58 AM UTC-8 [email protected] >>> wrote: >>> >>>> No such bloody luck. From the linbox log : >>>> >>>> libtool: compile: g++ -std=gnu++11 -DHAVE_CONFIG_H -I. -I../.. -I../.. >>>> -fopenmp -fabi-version=6 -O2 -Wall -g -DNDEBUG -U_LB_DEBUG >>>> -DDISABLE_COMMENTATOR -g -O2 -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 >>>> -mavx -mavx2 -mfma -c diophantine-solver.C -fPIC -DPIC -o >>>> .libs/diophantine-solver.o >>>> In file included from ../../linbox/solutions/methods.h:25, >>>> from ../../linbox/algorithms/rational-solver.h:41, >>>> from ../../linbox/algorithms/diophantine-solver.h:29, >>>> from diophantine-solver.C:26: >>>> ../../linbox/field/field-traits.h:88:12: error: redefinition of 'struct >>>> LinBox::ClassifyRing<Givaro::ZRing<Givaro::Integer> >' >>>> 88 | struct ClassifyRing<Givaro::IntegerDom> { >>>> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ >>>> ../../linbox/field/field-traits.h:84:12: note: previous definition of >>>> 'struct LinBox::ClassifyRing<Givaro::ZRing<Givaro::Integer> >' >>>> 84 | struct ClassifyRing<Givaro::ZRing<Givaro::Integer>> { >>>> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ >>>> In file included from ../../linbox/matrix/densematrix/blas-matrix.h:43, >>>> from ../../linbox/matrix/dense-matrix.h:79, >>>> from ../../linbox/solutions/methods.h:26, >>>> from ../../linbox/algorithms/rational-solver.h:41, >>>> from ../../linbox/algorithms/diophantine-solver.h:29, >>>> from diophantine-solver.C:26: >>>> ../../linbox/field/hom.h:274:42: error: invalid use of incomplete type >>>> 'LinBox::Hom<Givaro::QField<Givaro::Rational>, _Target>::Source' {aka >>>> 'class Givaro::QField<Givaro::Rational>'} >>>> 274 | typedef typename Source::Element SrcElt; >>>> | ^~~~~~~ >>>> In file included from ../../linbox/field/field-traits.h:28, >>>> from ../../linbox/solutions/methods.h:25, >>>> from ../../linbox/algorithms/rational-solver.h:41, >>>> from ../../linbox/algorithms/diophantine-solver.h:29, >>>> from diophantine-solver.C:26: >>>> /usr/include/givaro/givrational.h:49:11: note: declaration of >>>> 'LinBox::Hom<Givaro::QField<Givaro::Rational>, _Target>::Source' {aka >>>> 'class Givaro::QField<Givaro::Rational>'} >>>> 49 | class QField; >>>> | ^~~~~~ >>>> In file included from ../../linbox/matrix/densematrix/blas-matrix.h:43, >>>> from ../../linbox/matrix/dense-matrix.h:79, >>>> from ../../linbox/solutions/methods.h:26, >>>> from ../../linbox/algorithms/rational-solver.h:41, >>>> from ../../linbox/algorithms/diophantine-solver.h:29, >>>> from diophantine-solver.C:26: >>>> ../../linbox/field/hom.h:305:24: error: field '_source' has incomplete >>>> type 'LinBox::Hom<Givaro::QField<Givaro::Rational>, _Target>::Source' {aka >>>> 'Givaro::QField<Givaro::Rational>'} >>>> 305 | Source _source; >>>> | ^~~~~~~ >>>> In file included from ../../linbox/field/field-traits.h:28, >>>> from ../../linbox/solutions/methods.h:25, >>>> from ../../linbox/algorithms/rational-solver.h:41, >>>> from ../../linbox/algorithms/diophantine-solver.h:29, >>>> from diophantine-solver.C:26: >>>> /usr/include/givaro/givrational.h:49:11: note: declaration of >>>> 'LinBox::Hom<Givaro::QField<Givaro::Rational>, _Target>::Source' {aka >>>> 'class Givaro::QField<Givaro::Rational>'} >>>> 49 | class QField; >>>> | ^~~~~~ >>>> In file included from ../../linbox/matrix/densematrix/blas-matrix.h:43, >>>> from ../../linbox/matrix/dense-matrix.h:79, >>>> from ../../linbox/solutions/methods.h:26, >>>> from ../../linbox/algorithms/rational-solver.h:41, >>>> from ../../linbox/algorithms/diophantine-solver.h:29, >>>> from diophantine-solver.C:26: >>>> ../../linbox/field/hom.h: In member function >>>> 'LinBox::Hom<Givaro::QField<Givaro::Rational>, _Target>::Elt& >>>> LinBox::Hom<Givaro::QField<Givaro::Rational>, >>>> _Target>::image(LinBox::Hom<Givaro::QField<Givaro::Rational>, >>>> _Target>::Elt&, const SrcElt&)': >>>> ../../linbox/field/hom.h:281:45: error: request for member 'deno' in 's', >>>> which is of non-class type 'const SrcElt' {aka 'const int'} >>>> 281 | if (_source.isOne(s.deno())) { >>>> | ^~~~ >>>> ../../linbox/field/hom.h:282:57: error: request for member 'nume' in 's', >>>> which is of non-class type 'const SrcElt' {aka 'const int'} >>>> 282 | return _target.init(t,s.nume()); >>>> | ^~~~ >>>> ../../linbox/field/hom.h:284:50: error: request for member 'nume' in 's', >>>> which is of non-class type 'const SrcElt' {aka 'const int'} >>>> 284 | else if (_source.isOne(s.nume())) { >>>> | ^~~~ >>>> ../../linbox/field/hom.h:285:50: error: request for member 'deno' in 's', >>>> which is of non-class type 'const SrcElt' {aka 'const int'} >>>> 285 | _target.init(t,s.deno()); >>>> | ^~~~ >>>> ../../linbox/field/hom.h:289:55: error: request for member 'deno' in 's', >>>> which is of non-class type 'const SrcElt' {aka 'const int'} >>>> 289 | _target. init (tmp, s.deno()); >>>> | ^~~~ >>>> ../../linbox/field/hom.h:290:53: error: request for member 'nume' in 's', >>>> which is of non-class type 'const SrcElt' {aka 'const int'} >>>> 290 | _target. init (t, s.nume()); >>>> | ^~~~ >>>> ../../linbox/field/hom.h: In member function >>>> 'LinBox::Hom<Givaro::QField<Givaro::Rational>, _Target>::SrcElt& >>>> LinBox::Hom<Givaro::QField<Givaro::Rational>, >>>> _Target>::preimage(LinBox::Hom<Givaro::QField<Givaro::Rational>, >>>> _Target>::SrcElt&, const Elt&)': >>>> ../../linbox/field/hom.h:296:45: error: request for member 'nume' in 's', >>>> which is of non-class type 'LinBox::Hom<Givaro::QField<Givaro::Rational>, >>>> _Target>::SrcElt' {aka 'int'} >>>> 296 | _target. convert (s.nume(), t); >>>> | ^~~~ >>>> ../../linbox/field/hom.h:297:45: error: request for member 'nume' in 's', >>>> which is of non-class type 'LinBox::Hom<Givaro::QField<Givaro::Rational>, >>>> _Target>::SrcElt' {aka 'int'} >>>> 297 | _source. init (s, s.nume()); >>>> | ^~~~ >>>> ../../linbox/field/hom.h: At global scope: >>>> ../../linbox/field/hom.h:315:33: error: invalid use of incomplete type >>>> 'LinBox::Hom<Givaro::QField<Givaro::Rational>, >>>> Givaro::QField<Givaro::Rational> >::Domain' {aka 'class >>>> Givaro::QField<Givaro::Rational>'} >>>> 315 | typedef Domain::Element Elt; >>>> | ^~~~~~~ >>>> In file included from ../../linbox/field/field-traits.h:28, >>>> from ../../linbox/solutions/methods.h:25, >>>> from ../../linbox/algorithms/rational-solver.h:41, >>>> from ../../linbox/algorithms/diophantine-solver.h:29, >>>> from diophantine-solver.C:26: >>>> /usr/include/givaro/givrational.h:49:11: note: declaration of >>>> 'LinBox::Hom<Givaro::QField<Givaro::Rational>, >>>> Givaro::QField<Givaro::Rational> >::Domain' {aka 'class >>>> Givaro::QField<Givaro::Rational>'} >>>> 49 | class QField; >>>> | ^~~~~~ >>>> In file included from ../../linbox/matrix/densematrix/blas-matrix.h:43, >>>> from ../../linbox/matrix/dense-matrix.h:79, >>>> from ../../linbox/solutions/methods.h:26, >>>> from ../../linbox/algorithms/rational-solver.h:41, >>>> from ../../linbox/algorithms/diophantine-solver.h:29, >>>> from diophantine-solver.C:26: >>>> ../../linbox/field/hom.h:318:17: error: 'Elt' does not name a type >>>> 318 | Elt& image(Elt& t, const Elt& s) const { return >>>> t=s; } >>>> | ^~~ >>>> ../../linbox/field/hom.h:320:17: error: 'Elt' does not name a type >>>> 320 | Elt& preimage(Elt& s, const Elt& t) const { return >>>> s=t; } >>>> | ^~~ >>>> In file included from ../../linbox/algorithms/rational-reconstruction.h:40, >>>> from ../../linbox/algorithms/rational-solver.inl:38, >>>> from ../../linbox/algorithms/rational-solver.h:1050, >>>> from ../../linbox/algorithms/diophantine-solver.h:29, >>>> from diophantine-solver.C:26: >>>> ../../linbox/algorithms/rational-reconstruction-base.h: In member function >>>> 'bool LinBox::RReconstructionBase<Givaro::ZRing<Givaro::Integer> >>>> >::reconstructRational(LinBox::RReconstructionBase<Givaro::ZRing<Givaro::Integer> >>>> >::Element&, LinBox::RReconstructionBase<Givaro::ZRing<Givaro::Integer> >>>> >::Element&, const Element&, const Element&)': >>>> ../../linbox/algorithms/rational-reconstruction-base.h:337:41: error: >>>> 'LinBox::RReconstructionBase<Givaro::ZRing<Givaro::Integer> >::Ring' {aka >>>> 'class Givaro::ZRing<Givaro::Integer>'} has no member named >>>> 'reconstructRational' >>>> 337 | return >>>> _intRing.reconstructRational(a,b,x,m,a_bound,a_bound); >>>> | ^~~~~~~~~~~~~~~~~~~ >>>> ../../linbox/algorithms/rational-reconstruction-base.h: In member function >>>> 'bool LinBox::RReconstructionBase<Givaro::ZRing<Givaro::Integer> >>>> >::reconstructRational(LinBox::RReconstructionBase<Givaro::ZRing<Givaro::Integer> >>>> >::Element&, LinBox::RReconstructionBase<Givaro::ZRing<Givaro::Integer> >>>> >::Element&, const Element&, const Element&, const Element&)': >>>> ../../linbox/algorithms/rational-reconstruction-base.h:342:34: error: >>>> 'LinBox::RReconstructionBase<Givaro::ZRing<Givaro::Integer> >::Ring' {aka >>>> 'class Givaro::ZRing<Givaro::Integer>'} has no member named >>>> 'reconstructRational' >>>> 342 | >>>> _intRing.reconstructRational(a,b,x,m,a_bound); >>>> | ^~~~~~~~~~~~~~~~~~~ >>>> make[9]: *** [Makefile:685: diophantine-solver.lo] Error 1 >>>> >>>> Still stuck… >>>> >>>> Le samedi 18 décembre 2021 à 13:27:00 UTC+1, [email protected] >>>> a écrit : >>>> >>>>> I had this happen before. It's a bug a failed to report, but I suppose >>>>> with an easy fix (at least it was easy for me). I'm not exactly sure how >>>>> I >>>>> fixed it. Something appears to be broken with the build chain. >>>>> >>>>> Can you touch `src/sage/libs/flint/*` and recompile this part and see >>>>> if this fixes your issue? >>>>> >>>>> I had this problem when working on >>>>> https://trac.sagemath.org/ticket/32211. And I believe I fixed it by >>>>> touching the correct files indicated at startup and running `sage -b`. >>>>> >>>>> Jonathan >>>>> On 18.12.21 13:17, Emmanuel Charpentier wrote: >>>>> >>>>> On Debian testing running on core i7 + 16 GB RAM, an attempt at >>>>> upgrading Sage 9.5.beta7 compiled from a git tree to 9.5.beta8 initially >>>>> succeeded ; however, after a routine upgrade of Testing, I saw Sage >>>>> crashing at startup. An attempt to rebuild the Sage library failed, and >>>>> led >>>>> me to restort to "make distclean && make". This failed totally :see this >>>>> report <https://groups.google.com/g/sage-support/c/Vv7quhmFEao> on >>>>> sage-support. >>>>> >>>>> Attached : >>>>> the initial crash report after Debian upgrade ; >>>>> the installation log after make-distclean ; >>>>> the compilation log of linbox, pointed to by the error messagesof >>>>> install.log. >>>>> >>>>> I'm somewhat stuck... >>>>> -- >>>>> Emmanuel Charpentier >>>>> >>>>> -- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "sage-release" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to [email protected]. >>>>> To view this discussion on the web visit >>>>> https://groups.google.com/d/msgid/sage-release/aa10b1b927484eb0f369f722f9d0ca84dc1592f0.camel%40free.fr >>>>> >>>>> <https://groups.google.com/d/msgid/sage-release/aa10b1b927484eb0f369f722f9d0ca84dc1592f0.camel%40free.fr?utm_medium=email&utm_source=footer> >>>>> . >>>>> >>>>> -- You received this message because you are subscribed to the Google Groups "sage-release" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-release/fb5766bf-7a63-4027-ab14-760ef5f64d11n%40googlegroups.com.
