I don't believe this bug is in MPIR, but in Flint. You should be pulling from https://github.com/wbhart/flint2/ in the flint2 source tree.
Since you are now subscribed to flint-devel, we could continue the discussion over there if you like. Bill. On 23 November 2015 at 20:34, Dann Corbit <dcor...@connx.com> wrote: > I pulled from : > > https://github.com/wbhart/mpir > > And I notice below that I do not see any compile. (So existing objects > appear to have been used). > > Was there a check-in or did I pull from the wrong repository? > > > > *From:* mpir-devel@googlegroups.com [mailto:mpir-devel@googlegroups.com] *On > Behalf Of *Dann Corbit > *Sent:* Monday, November 23, 2015 11:31 AM > *To:* 'mpir-devel@googlegroups.com' <mpir-devel@googlegroups.com> > *Subject:* RE: [mpir-devel] Open source high precision math tools built > under 64 bit Windows operating systems state of the union > > > > DCorbit@dcorbit MINGW64 /f/math/flint2-trunk > > $ make check MOD=fmpz_lll > > make[1]: Entering directory '/f/math/flint2-trunk' > > make[1]: Nothing to be done for 'static'. > > make[1]: Leaving directory '/f/math/flint2-trunk' > > make[1]: Entering directory '/f/math/flint2-trunk/fmpz_lll' > > wrapper_with_removal_knapsack....PASS > > lll_mpf_with_removal.... > > This application has requested the Runtime to terminate it in an unusual > way. > > Please contact the application's support team for more information. > > FAIL (randintrel): basis matrices not equal! > > [[12 -62 -52 50 -32 -70 59 -82 -124 0 -18] > > [-106 -8 32 4 -102 -111 -63 9 -50 -54 -37] > > [59 107 -18 140 92 83 13 -6 -80 -77 0] > > [-52 73 -16 -31 80 -188 -1 47 -57 -25 -6] > > [38 9 92 -49 3 132 -68 -44 -47 37 92] > > [-163 -59 -70 54 44 -26 -60 43 85 63 52] > > [134 136 -67 29 -18 74 -11 146 -24 53 87] > > [58 -32 75 119 50 94 64 151 -18 -51 126] > > [34 34 -78 -153 -47 73 80 -143 173 -155 32] > > [-58 64 2 -134 -159 139 225 69 -125 204 -118] > > ][[8212269030201924110349129 5109674321441250958607765 > 1494481417875653033370726 -3126612440029326740763271 > 51429522477502228951637 3668135059057144339165983 > -1222207475347700051823571 -82 -124 0 -18] > > [17733886805887267123209168 -560817913328917789156032 > -164028448303425332568953 343164780003218788620738 -5644703686555121748298 > -402600189408710965161118 134144722904015859924009 9 -50 -54 -37] > > [19051207754227777873488604 373878608885945193185394 > 109352298868950221851030 -228776520002145858915019 3763135791036748585363 > 268400126272473976370620 -89429815269343907473720 -6 -80 -77 0] > > [16249366647065029671069516 -2928715769606570672178631 > -856593007806776738397760 1792082740016809230647187 > -29477897029787862255024 -2102467655801046144520393 > 700533552943193932914667 47 -57 -25 -6] > > [-12712607907404507225851694 2741776465163598076439377 > 801916858372301627665048 -1677694480015736301019069 27596329134269489271044 > 1968267592664809156941446 -655818645308521979257954 -44 -47 37 92] > > [-18886497648316966397456468 -2679463363682607209657108 > -783691475227476590922561 1639565060015378656777141 > -26969139835763363403391 -1923534238286063495126026 > 640913676096964661378769 43 85 63 52] > > [4375423774225741805660394 -9097712816224666341565513 > -2660905939144455400668437 5566895320052215904776669 > -91569637581894213385283 -6531069739296866750041668 > 2176125504887368384344694 146 -24 53 87] > > [13285581601783327693783123 -9409278323629620668734582 > -2752032854868580586711161 5757542420054004121559664 > -94705584074424838242400 -6754736511190595063461373 > 2250650350945154972781690 151 -18 -51 126] > > [-21960865427861023718955812 8910773511781693745946670 > 2606229789709980290427972 -5452507060051142975981787 > 89688069686375837813100 6396869676160629763057969 > -2131410597252696431846262 -143 173 -155 32] > > [9888124790847165658516615 -4299604002188369711180535 > -1257551436992927552303989 2630929980024677380049973 > -43276061596922611342621 -3086601452133450723399302 > 1028442875597454921359986 69 -125 204 -118] > > ]../Makefile.subdirs:91: recipe for target > '../build/fmpz_lll/test/t-lll_mpf_with_removal.exe_RUN' failed > > make[1]: *** [../build/fmpz_lll/test/t-lll_mpf_with_removal.exe_RUN] Error > 3 > > make[1]: Leaving directory '/f/math/flint2-trunk/fmpz_lll' > > Makefile:221: recipe for target 'check' failed > > make: *** [check] Error 2 > > > > DCorbit@dcorbit MINGW64 /f/math/flint2-trunk > > > > > > *From:* mpir-devel@googlegroups.com [mailto:mpir-devel@googlegroups.com > <mpir-devel@googlegroups.com>] > *Sent:* Monday, November 23, 2015 11:09 AM > *To:* mpir-devel <mpir-devel@googlegroups.com> > *Subject:* Re: [mpir-devel] Open source high precision math tools built > under 64 bit Windows operating systems state of the union > > > > Dear Dan, > > > > thanks very much for these. This at least tells us that the same problem > exists in lll_mpf. Oddly enough the failure is in the second of the tests > for fmpz_lll, not the first, so it's looking more and more like a corner > case. But even if I run it for 100 times longer on Linux it doesn't fail, > so it is certainly a puzzler. > > > > Anyhow, I have just gone through every single line of code in the LLL > implementation and removed as many potential Windows issues as I could. I > am not completely convinced any of them could cause this issue, but it is > definitely worth trying again. I'm pretty confident that I've corrected all > the likely culprits. After this point we are just going to have to do some > extended debugging, which will probably have to wait until we get the test > suite working again on our MinGW installation. > > > > Anyway, could you try pulling the latest changes and make check > MOD=fmpz_lll in case we have made some progress. > > > > Thanks very much for your efforts! > > > > Bill. > > > > On 23 November 2015 at 19:45, Dann Corbit <dcor...@connx.com> wrote: > > Sequence of make check operations attached > > > > *From:* mpir-devel@googlegroups.com [mailto:mpir-devel@googlegroups.com] > *Sent:* Friday, November 20, 2015 2:14 PM > > > *To:* mpir-devel <mpir-devel@googlegroups.com> > *Subject:* Re: [mpir-devel] Open source high precision math tools built > under 64 bit Windows operating systems state of the union > > > > Thanks. > > > > We think we have a fix on the way for the Howell form bug. However I don't > see the problem at all for the fmpz_lll bug. > > > > Something you could do which would help us greatly is the following: > > > > make check MOD=d_vec > > make check MOD=mpf_vec > > make check MOD=mpfr_vec > > make check MOD=d_mat > > make check MOD=mpf_mat > > make check MOD=mpfr_mat > > make check MOD=fmpz_lll:heuristic_dot > > make check MOD=fmpz_lll:lll_d > > make check MOD=fmpz_lll:lll_d_heuristic > > make check MOD=fmpz_lll:lll_d_heuristic_with_removal > > make check MOD=fmpz_lll:lll_d_with_removal > > make check MOD=fmpz_lll:lll_mpf > > make check MOD=fmpz_lll:lll_mpf_with_removal > > make check MOD=fmpz_lll:lll > > > > and let us know which ones report errors and what the errors are. > > > > At this stage there's around 9000 lines of code in which the error could > be hiding and nothing is jumping out at me. It all looks Windows safe at > this stage. > > > > Bill. > > > > > > On 20 November 2015 at 23:03, Dann Corbit <dcor...@connx.com> wrote: > > wrapper_with_removal_knapsack....PASS > > lll_mpf_with_removal.... > > This application has requested the Runtime to terminate it in an unusual > way. > > Please contact the application's support team for more information. > > FAIL (randintrel): basis matrices not equal! > > [[12 -62 -52 50 -32 -70 59 -82 -124 0 -18] > > [-106 -8 32 4 -102 -111 -63 9 -50 -54 -37] > > [59 107 -18 140 92 83 13 -6 -80 -77 0] > > [-52 73 -16 -31 80 -188 -1 47 -57 -25 -6] > > [38 9 92 -49 3 132 -68 -44 -47 37 92] > > [-163 -59 -70 54 44 -26 -60 43 85 63 52] > > [134 136 -67 29 -18 74 -11 146 -24 53 87] > > [58 -32 75 119 50 94 64 151 -18 -51 126] > > [34 34 -78 -153 -47 73 80 -143 173 -155 32] > > [-58 64 2 -134 -159 139 225 69 -125 204 -118] > > ][[8212269030201924110349129 5109674321441250958607765 > 1494481417875653033370726 -3126612440029326740763271 > 51429522477502228951637 3668135059057144339165983 > -1222207475347700051823571 -82 -124 0 -18] > > [17733886805887267123209168 -560817913328917789156032 > -164028448303425332568953 343164780003218788620738 -5644703686555121748298 > -402600189408710965161118 134144722904015859924009 9 -50 -54 -37] > > [19051207754227777873488604 373878608885945193185394 > 109352298868950221851030 -228776520002145858915019 3763135791036748585363 > 268400126272473976370620 -89429815269343907473720 -6 -80 -77 0] > > [16249366647065029671069516 -2928715769606570672178631 > -856593007806776738397760 1792082740016809230647187 > -29477897029787862255024 -2102467655801046144520393 > 700533552943193932914667 47 -57 -25 -6] > > [-12712607907404507225851694 2741776465163598076439377 > 801916858372301627665048 -1677694480015736301019069 27596329134269489271044 > 1968267592664809156941446 -655818645308521979257954 -44 -47 37 92] > > [-18886497648316966397456468 -2679463363682607209657108 > -783691475227476590922561 1639565060015378656777141 > -26969139835763363403391 -1923534238286063495126026 > 640913676096964661378769 43 85 63 52] > > [4375423774225741805660394 -9097712816224666341565513 > -2660905939144455400668437 5566895320052215904776669 > -91569637581894213385283 -6531069739296866750041668 > 2176125504887368384344694 146 -24 53 87] > > [13285581601783327693783123 -9409278323629620668734582 > -2752032854868580586711161 5757542420054004121559664 > -94705584074424838242400 -6754736511190595063461373 > 2250650350945154972781690 151 -18 -51 126] > > [-21960865427861023718955812 8910773511781693745946670 > 2606229789709980290427972 -5452507060051142975981787 > 89688069686375837813100 6396869676160629763057969 > -2131410597252696431846262 -143 173 -155 32] > > [9888124790847165658516615 -4299604002188369711180535 > -1257551436992927552303989 2630929980024677380049973 > -43276061596922611342621 -3086601452133450723399302 > 1028442875597454921359986 69 -125 204 -118] > > ]../Makefile.subdirs:91: recipe for target > '../build/fmpz_lll/test/t-lll_mpf_with_removal.exe_RUN' failed > > make[1]: *** [../build/fmpz_lll/test/t-lll_mpf_with_removal.exe_RUN] Error > 3 > > make[1]: Leaving directory '/f/math/flint2-trunk/fmpz_lll' > > Makefile:211: recipe for target 'check' failed > > make: *** [check] Error 2 > > > > DCorbit@dcorbit MINGW64 /f/math/flint2-trunk > > $ > > > > > > *From:* mpir-devel@googlegroups.com [mailto:mpir-devel@googlegroups.com] > *Sent:* Friday, November 20, 2015 12:27 PM > *To:* mpir-devel <mpir-devel@googlegroups.com> > *Subject:* Re: [mpir-devel] Open source high precision math tools built > under 64 bit Windows operating systems state of the union > > > > I've removed two further Windows bugs. This may have caused the fmpz_lll > bug you reported. But I can't be sure. If you can test it we would be > grateful. > > > > The howell_form bug is just a memory allocation issue and we are working > on that. I don't know when it will be fixed as the person who is > responsible for the code has gone home for the weekend. > > > > Bill. > > -- > > You received this message because you are subscribed to the Google Groups > "mpir-devel" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to mpir-devel+unsubscr...@googlegroups.com. > To post to this group, send email to mpir-devel@googlegroups.com. > Visit this group at http://groups.google.com/group/mpir-devel. > For more options, visit https://groups.google.com/d/optout. > > > > -- > You received this message because you are subscribed to the Google Groups > "mpir-devel" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to mpir-devel+unsubscr...@googlegroups.com. > To post to this group, send email to mpir-devel@googlegroups.com. > Visit this group at http://groups.google.com/group/mpir-devel. > For more options, visit https://groups.google.com/d/optout. > > -- > You received this message because you are subscribed to the Google Groups > "mpir-devel" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to mpir-devel+unsubscr...@googlegroups.com. > To post to this group, send email to mpir-devel@googlegroups.com. > Visit this group at http://groups.google.com/group/mpir-devel. > For more options, visit https://groups.google.com/d/optout. > > > > -- > You received this message because you are subscribed to the Google Groups > "mpir-devel" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to mpir-devel+unsubscr...@googlegroups.com. > To post to this group, send email to mpir-devel@googlegroups.com. > Visit this group at http://groups.google.com/group/mpir-devel. > For more options, visit https://groups.google.com/d/optout. > > -- > You received this message because you are subscribed to the Google Groups > "mpir-devel" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to mpir-devel+unsubscr...@googlegroups.com. > To post to this group, send email to mpir-devel@googlegroups.com. > Visit this group at http://groups.google.com/group/mpir-devel. > For more options, visit https://groups.google.com/d/optout. > > -- > You received this message because you are subscribed to the Google Groups > "mpir-devel" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to mpir-devel+unsubscr...@googlegroups.com. > To post to this group, send email to mpir-devel@googlegroups.com. > Visit this group at http://groups.google.com/group/mpir-devel. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "mpir-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to mpir-devel+unsubscr...@googlegroups.com. To post to this group, send email to mpir-devel@googlegroups.com. Visit this group at http://groups.google.com/group/mpir-devel. For more options, visit https://groups.google.com/d/optout.
