On Fri, Nov 27, 2020 at 9:36 PM Tobias Weiss <tobwefo...@web.de> wrote:
>
> I still have a question: Keeping in mind Tumbleweed is a rolling release of
> Opensuse - is it clever to build a fixed binary? Will this binary persist
> with the rolling software updates?
we regularly test SageMath on Gentoo, which is a rolling release system, too.
Needless to say, once in a while one might need to rebuild something,
if a library SageMath depends on was updated
in an incompatible way.
>
> Maybe that's why the dependencies of source and binary package don't work?
no, it's not they don't work, it's a design choice, leading to fewer updates;
often only a binary package changes, then if it is a dynamic library
then nothing needs to be rebuilt.
(perhaps - I am not sure here; perhaps it's just a historical choice
that stayed)
> I will test these dependencies tomorrow at all, so that I don't tell you
> wrong things and the dependencies work at all ...
>
>
> Tobias
>
> Tobias Weiss schrieb am Freitag, 27. November 2020 um 21:32:15 UTC+1:
>>
>> OK, so I stop the current build. Then I install for every dependency both -
>> devel and binary package - and retry the build tomorrow.
>>
>> Then it don't makes sence, that the binary is not installed automatically
>> when I install the source package - but this is an Opensuse problem. Once,
>> it was like this ...
>>
>>
>> Thank you!
>>
>> Tobias
>>
>> dim...@gmail.com schrieb am Freitag, 27. November 2020 um 21:05:45 UTC+1:
>>>
>>> On Fri, Nov 27, 2020 at 7:21 PM Tobias Weiss <tobwe...@web.de> wrote:
>>> >
>>> > ... I thought there must be a reason I can install the development
>>> > package without binary package. But no it makes no sense to me. Can this
>>> > be the reason of both complile errors?
>>> >
>>> > Tobias
>>> >
>>> > Tobias Weiss schrieb am Freitag, 27. November 2020 um 20:15:24 UTC+1:
>>> >>
>>> >> ... how does compiling work? Needs the compiler also the binary of some
>>> >> library or only the library-devel package with the code? For most
>>> >> packages I only installed the devel package ...
>>> I don't know much about OpenSUSE, but some distributions typically
>>> have <library>-dev packages dependent on <library> packages, and
>>> some (e.g. Fedora) do not, i.e. you always need to install both
>>> <library>-dev (or <library>-devel) and <library>.
>>> It seems to me that OpenSUSE is an rpm-based distro, thus it would be
>>> like Fedora, and require <library>-devel and <library> in some cases.
>>>
>>>
>>> If you just run ./configure
>>> it will print at the end some messages about packages that are
>>> recommended to be installed (after guessing, hopefully correctly
>>> the kind of Linux system it is on)
>>>
>>>
>>> >>
>>> >> Tobias
>>> >>
>>> >> Tobias Weiss schrieb am Freitag, 27. November 2020 um 19:30:49 UTC+1:
>>> >>>
>>> >>> I didn't want to find out, what is the problem with sagelib. The error
>>> >>> was from the system packages. So I wanted no conflicts between system
>>> >>> wide packages and the packages from the tarball. So I configured like
>>> >>> this:
>>> >>>
>>> >>> ./configure --with-system-arb=no --with-system-boost=no
>>> >>> --with-system-brial=no --with-system-cddlib=no --with-system-cliquer=no
>>> >>> --with-system-ecm=no --with-system-eclib=no
>>> >>> --with-system-fflas_ffpack=no --
>>> >>> with-system-flint=no --with-system-flintqs=no --with-system-fplll=no
>>> >>> --with-system-freetype=no --with-system-gc=no --with-system-giac=no
>>> >>> --with-system-givaro=no --with-system-glpk=no --with-system-gmp=no
>>> >>> --with-system-gsl=no
>>> >>> --with-system-iconv=no --with-system-iml=no --with-system-isl=no
>>> >>> --with-system-lcalc=no --with-system-libbraiding=no
>>> >>> --with-system-libatomic_ops=no --with-system-libffi=no
>>> >>> --with-system-libgd=no --with-system-libsemigroups=
>>> >>> no --with-system-lrcalc=no --with-system-m4rie=no --with-system-m4ri=no
>>> >>> --with-system-mpc=no --with-system-mpfi=no --with-system-mpfr=no
>>> >>> --with-system-mpir=no --with-system-ninja_build=no --with-system-ntl=no
>>> >>> --with-system-o
>>> >>> penblas=no --with-system-pandoc=no --with-system-pari=no
>>> >>> --with-system-pari_galdata=no --with-system-pari_seadata_small=no
>>> >>> --with-system-pcre=no --with-system-perl_term_readline_gnu=no
>>> >>> --with-system-planarity=no --with-syste
>>> >>> m-ppl=no --with-system-rw=no --with-system-readline=no
>>> >>> --with-system-suitesparse=no --with-system-symmetrica=no
>>> >>> --with-system-yasm=no --with-system-zeromq=no --with-system-zn_poly=no
>>> >>> --with-system-pynac=no
>>> >>>
>>> >>> Sure, I wanted to use gcc from system. But it was also compiled newly.
>>> >>>
>>> >>> Shall I reproduce the error with sagelib, when I have time?
>>> >>>
>>> >>>
>>> >>> Tobias
>>> >>>
>>> >>> dim...@gmail.com schrieb am Freitag, 27. November 2020 um 18:49:10
>>> >>> UTC+1:
>>> >>>>
>>> >>>> On Fri, Nov 27, 2020 at 5:44 PM Tobias Weiss <tobwe...@web.de> wrote:
>>> >>>> >
>>> >>>> > OK! I just uninstalled flint and did configure without system
>>> >>>> > packages again (after make clean). I can tell you if it works
>>> >>>> > tomorrow or later, because of compile time ...
>>> >>>> >
>>> >>>> > Well, I first tried to compile with system packages, but this made
>>> >>>> > an error, when making sagelib - beacause of pari, I think.
>>> >>>>
>>> >>>> We very much welcome reports of failures like this, as we strive to
>>> >>>> unvendor as much as possible.
>>> >>>> Skipping all the system packages is surely a huge overkill - e.g. why
>>> >>>> on Earth would you need to build gcc
>>> >>>> on a regular Linux system?
>>> >>>>
>>> >>>>
>>> >>>> >
>>> >>>> >
>>> >>>> > Thank you! I hope this works,
>>> >>>> > Tobias
>>> >>>> >
>>> >>>> > dim...@gmail.com schrieb am Freitag, 27. November 2020 um 16:09:05
>>> >>>> > UTC+1:
>>> >>>> >>
>>> >>>> >>
>>> >>>> >>
>>> >>>> >> On Fri, 27 Nov 2020, 14:21 Tobias, <tobwe...@web.de> wrote:
>>> >>>> >>>
>>> >>>> >>> Hallo Everybody,
>>> >>>> >>>
>>> >>>> >>> on OpenSuse Tumbleweed I try building sage 9.2 without most of
>>> >>>> >>> packages from
>>> >>>> >>> system. The make command stopped compiling pynac because "the
>>> >>>> >>> programm needs
>>> >>>> >>> libflint", but flint was already compiled by make command.
>>> >>>> >>>
>>> >>>> >>>
>>> >>>> >>>
>>> >>>> >>> Then I wanted to start make command again with pynac, but make
>>> >>>> >>> skipped pynac
>>> >>>> >>> and started with the next package.
>>> >>>> >>>
>>> >>>> >>>
>>> >>>> >>> What should I do? I would probably restart building and configure
>>> >>>> >>> with --with-
>>> >>>> >>> system-flint=yes and --with-system-flintqs=yes
>>> >>>> >>> But it took so long to come here ...
>>> >>>> >>>
>>> >>>> >>>
>>> >>>> >>> What would you say?
>>> >>>> >>
>>> >>>> >>
>>> >>>> >> I guess you have flint installed systemwide, as well, and pynac
>>> >>>> >> gets confused as to which flint to use.
>>> >>>> >>
>>> >>>> >>>
>>> >>>> >>>
>>> >>>> >>> Kind regards,
>>> >>>> >>> Tobias
>>> >>>> >>>
>>> >>>> >>> from log file
>>> >>>> >>> > sage-9.2/local/var/tmp/sage/build/pynac-0.7.26.sage-2020-04-03.p0/src/config
>>> >>>> >>> >.log
>>> >>>> >>>
>>> >>>> >>> > configure:17041: gcc -o conftest -g -O2
>>> >>>> >>> > -Wl,-rpath-link,/home/tobias/bin/sage-9.2/local/lib
>>> >>>> >>> > -L/home/tobias/bin/sage-9.2/local/lib
>>> >>>> >>> > -Wl,-rpath,/home/tobias/bin/sage-9.2/local/lib conftest.c
>>> >>>> >>> > -lflint -lgmp
>>> >>>> >>> > >&5
>>> >>>> >>> > /usr/lib64/gcc/x86_64-suse-linux/10/../../../../x86_64-suse-linux/bin/ld:
>>> >>>> >>> > warning: libgmp.so.23, needed by
>>> >>>> >>> > /home/tobias/bin/sage-9.2/local/lib/libflint.so, may conflict
>>> >>>> >>> > with
>>> >>>> >>> > libgmp.so.10
>>> >>>> >>> > /usr/lib64/gcc/x86_64-suse-linux/10/../../../../x86_64-suse-linux/bin/ld:
>>> >>>> >>> > _ZN3NTL9zz_pEInfoE: TLS definition in /usr/lib64/libntl.so.43
>>> >>>> >>> > section .tbss
>>> >>>> >>> > mismatches non-TLS reference in
>>> >>>> >>> > /home/tobias/bin/sage-9.2/local/lib/libflint.so
>>> >>>> >>> > /usr/lib64/gcc/x86_64-suse-linux/10/../../../../x86_64-suse-linux/bin/ld:
>>> >>>> >>> > /usr/lib64/libntl.so.43: error adding symbols: bad value
>>> >>>> >>> > collect2: error:
>>> >>>> >>> > ld returned 1 exit status
>>> >>>> >>> > configure:17041: $? = 1
>>> >>>> >>> >
>>> >>>> >>> > configure: failed program was:
>>> >>>> >>> > | /* confdefs.h */
>>> >>>> >>> > | #define PACKAGE_NAME "pynac"
>>> >>>> >>> > | #define PACKAGE_TARNAME "pynac"
>>> >>>> >>> > | #define PACKAGE_VERSION "0.7.26.sage-2020-04-03"
>>> >>>> >>> > | #define PACKAGE_STRING "pynac 0.7.26.sage-2020-04-03"
>>> >>>> >>> > | #define PACKAGE_BUGREPORT "<pynac...@googlegroups.com>"
>>> >>>> >>> > | #define PACKAGE_URL ""
>>> >>>> >>> > | #define PACKAGE "pynac"
>>> >>>> >>> > | #define VERSION "0.7.26.sage-2020-04-03"
>>> >>>> >>> > | #define ARCHIVE_VERSION 3
>>> >>>> >>> > | #define ARCHIVE_AGE 0
>>> >>>> >>> > | #define STDC_HEADERS 1
>>> >>>> >>> > | #define HAVE_SYS_TYPES_H 1
>>> >>>> >>> > | #define HAVE_SYS_STAT_H 1
>>> >>>> >>> > | #define HAVE_STDLIB_H 1
>>> >>>> >>> > | #define HAVE_STRING_H 1
>>> >>>> >>> > | #define HAVE_MEMORY_H 1
>>> >>>> >>> > | #define HAVE_STRINGS_H 1
>>> >>>> >>> > | #define HAVE_INTTYPES_H 1
>>> >>>> >>> > | #define HAVE_STDINT_H 1
>>> >>>> >>> > | #define HAVE_UNISTD_H 1
>>> >>>> >>> > | #define HAVE_DLFCN_H 1
>>> >>>> >>> > | #define LT_OBJDIR ".libs/"
>>> >>>> >>> > | #define HAVE_GMP_H 1
>>> >>>> >>> > | #define HAVE_FLINT_FMPQ_POLY_H 1
>>> >>>> >>> > | /* end confdefs.h. */
>>> >>>> >>> > |
>>> >>>> >>> > | /* Override any GCC internal prototype to avoid an error.
>>> >>>> >>> > |
>>> >>>> >>> > | Use char because int might match the return type of a GCC
>>> >>>> >>> > | builtin and then its argument prototype would still apply. */
>>> >>>> >>> > |
>>> >>>> >>> > | #ifdef __cplusplus
>>> >>>> >>> > | extern "C"
>>> >>>> >>> > | #endif
>>> >>>> >>> > | char fmpq_get_mpz_frac ();
>>> >>>> >>> > | int
>>> >>>> >>> > | main ()
>>> >>>> >>> > | {
>>> >>>> >>> > | return fmpq_get_mpz_frac ();
>>> >>>> >>> > |
>>> >>>> >>> > | ;
>>> >>>> >>> > | return 0;
>>> >>>> >>> > |
>>> >>>> >>> > | }
>>> >>>> >>> >
>>> >>>> >>> > configure:17058: result: no
>>> >>>> >>> > configure:17065: error: This package needs libflint
>>> >>>> >>>
>>> >>>> >>> or from file
>>> >>>> >>> > sage-9.2/logs/pkgs/pynac-0.7.26.sage-2020-04-03.p0.log
>>> >>>> >>>
>>> >>>> >>>
>>> >>>> >>> > checking for library containing __gmpz_get_str... -lgmp
>>> >>>> >>> > checking flint/fmpq_poly.h usability... yes
>>> >>>> >>> > checking flint/fmpq_poly.h presence... yes
>>> >>>> >>> > checking for flint/fmpq_poly.h... yes
>>> >>>> >>> > checking for library containing fmpq_get_mpz_frac... no
>>> >>>> >>> > configure: error: This package needs libflint
>>> >>>> >>> > ****************************************************************************
>>> >>>> >>> > ****************************************************************************
>>> >>>> >>> > ************************************************************************
>>> >>>> >>> > Error configuring pynac-0.7.26.sage-2020-04-03.p0
>>> >>>> >>> > See the file
>>> >>>> >>> >
>>> >>>> >>> > /home/tobias/bin/sage-9.2/local/var/tmp/sage/build/pynac-0.7.26.sage-202
>>> >>>> >>> > 0-04-03.p0/src/config.log>
>>> >>>> >>> > for details.
>>> >>>> >>> > ****************************************************************************
>>> >>>> >>> > ****************************************************************************
>>> >>>> >>> > ************************************************************************
>>> >>>> >>>
>>> >>>> >>> --
>>> >>>> >>> You received this message because you are subscribed to the Google
>>> >>>> >>> Groups "sage-support" group.
>>> >>>> >>> To unsubscribe from this group and stop receiving emails from it,
>>> >>>> >>> send an email to sage-support...@googlegroups.com.
>>> >>>> >>> To view this discussion on the web visit
>>> >>>> >>> https://groups.google.com/d/msgid/sage-support/2410372.RVVYDCxLDo%40localhost.localdomain.
>>> >>>> >
>>> >>>> > --
>>> >>>> > You received this message because you are subscribed to the Google
>>> >>>> > Groups "sage-support" group.
>>> >>>> > To unsubscribe from this group and stop receiving emails from it,
>>> >>>> > send an email to sage-support...@googlegroups.com.
>>> >>>> > To view this discussion on the web visit
>>> >>>> > https://groups.google.com/d/msgid/sage-support/23ae3a10-bf65-424e-9051-b0e89fa1d1d6n%40googlegroups.com.
>>> >
>>> > --
>>> > You received this message because you are subscribed to the Google Groups
>>> > "sage-support" group.
>>> > To unsubscribe from this group and stop receiving emails from it, send an
>>> > email to sage-support...@googlegroups.com.
>>> > To view this discussion on the web visit
>>> > https://groups.google.com/d/msgid/sage-support/773dfae9-620c-456e-9953-7a94e5270c6bn%40googlegroups.com.
>
> --
> You received this message because you are subscribed to the Google Groups
> "sage-support" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-support+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sage-support/76e63d8e-6530-4c2d-9321-1fb8a1adb01cn%40googlegroups.com.
--
You received this message because you are subscribed to the Google Groups
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sage-support+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/sage-support/CAAWYfq36CHjpCU32y_15RB%3DgyYzo8wR331z3uLhWcC5UFEa1Aw%40mail.gmail.com.
