I suspect you have an installation of openblas, or/and some other blas/lapack, in /usr/local - and this results in scipy being unable to link it correctly.
(or perhaps it's not /usr/local, but some other location known to pkg-config) numpy and scipy accumulated ~20 years of layer upon layer of old outdated code for recognising BLAS/LAPACK libraries, it's such a nightmare to debug. On Sun, Feb 5, 2023 at 10:36 AM 'Nasser M. Abbasi' via sage-release <sage-release@googlegroups.com> wrote: > > Here is the output for the new command you posted > > >ldd > >local/var/lib/sage/venv-python3.10/lib/python3.10/site-packages/numpy/core/_multiarray_umath.cpython-310-x86_64-linux-gnu.so > linux-vdso.so.1 (0x00007ffd883d4000) > libcblas.so.3 => /usr/lib/libcblas.so.3 (0x00007f25a6532000) > libopenblas.so.3 => /usr/lib/libopenblas.so.3 (0x00007f25a5260000) > libm.so.6 => /usr/lib/libm.so.6 (0x00007f25a5178000) > libc.so.6 => /usr/lib/libc.so.6 (0x00007f25a4f91000) > /usr/lib64/ld-linux-x86-64.so.2 (0x00007f25a6a5b000) > libgomp.so.1 => /usr/lib/libgomp.so.1 (0x00007f25a4f48000) > libgcc_s.so.1 => /usr/lib/libgcc_s.so.1 (0x00007f25a4f28000) > > > > I've also put in same folder here a file called log_folders.tar.gz which is > backup of the logs/ folder in my build tree of 9.8 rc0 so it will have the > logs you asked for. > > any other information needed please let me know. > --Nasser > > > On Sunday, February 5, 2023 at 4:17:44 AM UTC-6 Dima Pasechnik wrote: >> >> On Sun, Feb 5, 2023 at 9:58 AM Dima Pasechnik <dim...@gmail.com> wrote: >> > >> > On Sun, Feb 5, 2023 at 9:04 AM 'Nasser M. Abbasi' via sage-release >> > <sage-r...@googlegroups.com> wrote: >> > > >> > > >> > > Here is the output >> > > >> > > >ldd >> > > >/home/me/TMP/sage-9.8.rc0/local/var/lib/sage/venv-python3.10/lib/python3.10/site-packages/scipy/special/_ufuncs.cpython-310-x86_64-linux-gnu.so >> > > linux-vdso.so.1 (0x00007ffc53be9000) >> > > libgfortran.so.5 => /usr/lib/libgfortran.so.5 (0x00007fc232878000) >> > > libm.so.6 => /usr/lib/libm.so.6 (0x00007fc232790000) >> > > libmvec.so.1 => /usr/lib/libmvec.so.1 (0x00007fc232696000) >> > > libgcc_s.so.1 => /usr/lib/libgcc_s.so.1 (0x00007fc232676000) >> > > libc.so.6 => /usr/lib/libc.so.6 (0x00007fc23248f000) >> > > libquadmath.so.0 => /usr/lib/../lib/libquadmath.so.0 (0x00007fc232444000) >> > > /usr/lib64/ld-linux-x86-64.so.2 (0x00007fc232ce1000) >> > >> > here is how it looks on a non-broken installation. >> > >> > $ ldd >> > local/var/lib/sage/venv-python3.10/lib/python3.10/site-packages/scipy/special/_ufuncs.cpython-310-x86_64-linux-gnu.so >> > linux-vdso.so.1 (0x00007ffe6b935000) >> > libopenblas.so.0 => /lib64/libopenblas.so.0 (0x00007fe561400000) >> > libgfortran.so.5 => /lib64/libgfortran.so.5 (0x00007fe561000000) >> > libm.so.6 => /lib64/libm.so.6 (0x00007fe563865000) >> > libmvec.so.1 => /lib64/libmvec.so.1 (0x00007fe561303000) >> > libgcc_s.so.1 => /lib64/libgcc_s.so.1 (0x00007fe563845000) >> > libc.so.6 => /lib64/libc.so.6 (0x00007fe560c00000) >> > libquadmath.so.0 => /lib64/libquadmath.so.0 (0x00007fe560fb8000) >> > /lib64/ld-linux-x86-64.so.2 (0x00007fe563ad3000) >> > >> > As you see, in your case libopenblas.so is not there, and you have an >> > error due to this. >> > Why scipy builds such a broken module, I don't know. I'd say it's a scipy >> > bug. >> > >> > Please post logs/pkgs/numpy* and logs/pkgs/scipy* >> >> please also check that >> >> ldd >> local/var/lib/sage/venv-python3.10/lib/python3.10/site-packages/numpy/core/_multiarray_umath.cpython-310-x86_64-linux-gnu.so >> >> has openblas in its output. >> >> > >> > >> > > > >> > > >> > > "You should add fpylll at the end of the line in >> > > >> > > build/pkgs/sagemath_doc_html/dependencies >> > > >> > > and run ./configure and make again" >> > > >> > > Ok thanks. Will give this a try. >> > > >> > > "sage in a different directory which is in your PATH typically breaks >> > > things" >> > > >> > > but it is NOT in the path. This is the reason I do >> > > >> > > unset SAGE_ROOT >> > > unset SAGE_LOCAL >> > > >> > > before building any new release. I make sure current installed sagemath >> > > (in different folder) is not visible. >> > > >> > > --Nasser >> > > >> > > On Saturday, February 4, 2023 at 9:15:14 AM UTC-6 Dima Pasechnik wrote: >> > >> >> > >> On Sat, Feb 4, 2023 at 1:51 PM 'Nasser M. Abbasi' via sage-release >> > >> <sage-r...@googlegroups.com> wrote: >> > >> > >> > >> > "Was it an incremental build. If so, >> > >> > make distclean >> > >> > and >> > >> > build again from scratch." >> > >> > >> > >> > It was not an incremental build. I extracted the gzip file and did >> > >> > the steps I showed. So I do not need to do "make distclean" since >> > >> > everything is new. >> > >> > >> > >> > But I repeated now everything again using "make distclean" and got >> > >> > same error. >> > >> > >> > >> > "Probably the problem is linking with openblas. Do you have a number >> > >> > of openblas versions accessible to the linker?" >> > >> > >> > >> > I have no idea. What command should I use to check? >> > >> >> > >> run >> > >> >> > >> ldd >> > >> /home/me/TMP/sage-9.8.rc0/local/var/lib/sage/venv-python3.10/lib/python3.10/site-packages/scipy/special/_ufuncs.cpython-310-x86_64-linux-gnu.so >> > >> >> > >> Post its output >> > >> - there should be an openblas mentioned, the question is which one. >> > >> >> > >> As far as fpylll is concerned, you've hit the bug fixed in not yet >> > >> merged >> > >> https://trac.sagemath.org/ticket/34940 >> > >> >> > >> You should add fpylll at the end of the line in >> > >> >> > >> build/pkgs/sagemath_doc_html/dependencies >> > >> >> > >> and run ./configure and make again >> > >> >> > >> >> > >> >> > >> >> > >> > >> > >> > I am just following the instruction to download and install sagemath. >> > >> > >> > >> > I have giac and maxima and fricas and texlive already installed on >> > >> > this virtual box already. I also have sagemath 9.7 installed in >> > >> > different folder. >> > >> > >> > >> > --Nasser >> > >> > >> > >> > >> > >> > On Saturday, February 4, 2023 at 5:27:59 AM UTC-6 Dima Pasechnik >> > >> > wrote: >> > >> >> >> > >> >> Hello, >> > >> >> Was it an incremental build. If so, >> > >> >> make distclean >> > >> >> and >> > >> >> build again from scratch. >> > >> >> >> > >> >> Anyhow, if you look in your istall.log you'll see problems with a >> > >> >> number of packages, e.g. fpylll, scipy. >> > >> >> >> > >> >> [sagemath_doc_html-none] [manifolds] ImportError: >> > >> >> /home/me/TMP/sage-9.8.rc0/local/var/lib/sage/venv-python3.10/lib/python3.10/site-packages/scipy/special/_ufuncs.cpython-310-x86_64-linux-gnu.so: >> > >> >> undefined symbol: dstevr_ >> > >> >> >> > >> >> Probably the problem is linking with openblas. Do you have a number >> > >> >> of openblas versions accessible to the linker? >> > >> >> >> > >> >> >> > >> >> >> > >> >> By the way, >> > >> >> what are you trying to achieve with exporting >> > >> >> MAXIMA ? >> > >> >> Sage does not support using external Maxima at the moment, >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> On Sat, 4 Feb 2023, 08:12 'Nasser M. Abbasi' via sage-release, >> > >> >> <sage-r...@googlegroups.com> wrote: >> > >> >>> >> > >> >>> I get this error on Linux virtual box running >> > >> >>> >> > >> >>> lsb_release -a >> > >> >>> LSB Version: n/a >> > >> >>> Distributor ID: ManjaroLinux >> > >> >>> Description: Manjaro Linux >> > >> >>> Release: 22.0.0 >> > >> >>> Codename: Sikaris >> > >> >>> >> > >> >>> After extracting the gzip file, these are the commands I did >> > >> >>> >> > >> >>> unset SAGE_ROOT >> > >> >>> unset SAGE_LOCAL >> > >> >>> export GIAC=/usr/local/bin/giac >> > >> >>> export MAXIMA=/usr/local/bin/maxima >> > >> >>> ./configure --with-system-python=no >> > >> >>> make >> > >> >>> >> > >> >>> At the end I see >> > >> >>> >> > >> >>> ------------- >> > >> >>> sagemath_doc_html-none] File >> > >> >>> "/home/me/TMP/sage-9.8.rc0/src/sage_docbuild/sphinxbuild.py", line >> > >> >>> 263, in raise_errors >> > >> >>> [sagemath_doc_html-none] raise OSError(self._error) >> > >> >>> [sagemath_doc_html-none] OSError: >> > >> >>> /home/me/TMP/sage-9.8.rc0/src/sage/manifolds/chart.py:docstring of >> > >> >>> sage.manifolds.chart.RealChart.plot:104: WARNING: Exception >> > >> >>> occurred in plotting chart-1 >> > >> >>> [sagemath_doc_html-none] >> > >> >>> [sagemath_doc_html-none] Note: incremental documentation builds >> > >> >>> sometimes cause spurious >> > >> >>> [sagemath_doc_html-none] error messages. To be certain that these >> > >> >>> are real errors, run >> > >> >>> [sagemath_doc_html-none] "make doc-clean doc-uninstall" first and >> > >> >>> try again. >> > >> >>> [sagemath_doc_html-none] make[6]: *** [Makefile:28: >> > >> >>> doc-inventory--reference-manifolds] Error 1 >> > >> >>> [sagemath_doc_html-none] make[5]: *** [Makefile:45: >> > >> >>> doc-inventory-reference] Error 2 >> > >> >>> make[4]: *** [Makefile:3262: sagemath_doc_html-SAGE_DOCS-no-deps] >> > >> >>> Error 2 >> > >> >>> make[3]: *** [Makefile:3262: >> > >> >>> /home/me/TMP/sage-9.8.rc0/local/var/lib/sage/installed/sagemath_doc_html-none] >> > >> >>> Error 2 >> > >> >>> make[2]: *** [Makefile:2766: all-start] Error 2 >> > >> >>> make[2]: Leaving directory '/home/me/TMP/sage-9.8.rc0/build/make' >> > >> >>> >> > >> >>> real 71m30.833s >> > >> >>> user 103m7.355s >> > >> >>> sys 5m48.306s >> > >> >>> *************************************************************** >> > >> >>> Error building Sage. >> > >> >>> >> > >> >>> The following package(s) may have failed to build (not necessarily >> > >> >>> during this run of 'make all-start'): >> > >> >>> >> > >> >>> It is safe to delete any log files and build directories, but they >> > >> >>> contain information that is helpful for debugging build problems. >> > >> >>> WARNING: If you now run 'make' again, the build directory of the >> > >> >>> same version of the package will, by default, be deleted. Set the >> > >> >>> environment variable SAGE_KEEP_BUILT_SPKGS=yes to prevent this. >> > >> >>> ---------------------------- >> > >> >>> >> > >> >>> The following folder contains config.log and install.log files >> > >> >>> >> > >> >>> Thank you >> > >> >>> --Nasser >> > >> >>> >> > >> >>> >> > >> >>> >> > >> >>> >> > >> >>> >> > >> >>> >> > >> >>> On Friday, February 3, 2023 at 3:40:20 AM UTC-6 Dima Pasechnik >> > >> >>> wrote: >> > >> >>>> >> > >> >>>> On Fri, Feb 3, 2023 at 7:33 AM Sébastien Labbé <sla...@gmail.com> >> > >> >>>> wrote: >> > >> >>>> > >> > >> >>>> > >> > >> >>>> > >> > >> >>>> > On Thursday, February 2, 2023 at 7:37:27 PM UTC+1 >> > >> >>>> > matthia...@gmail.com wrote: >> > >> >>>> > On Thursday, February 2, 2023 at 9:30:30 AM UTC-8 Sébastien >> > >> >>>> > Labbé wrote: >> > >> >>>> > ./bootstrap: 9: m4: Permission denied >> > >> >>>> > make: *** [Makefile:365 : configure] Erreur 1 >> > >> >>>> > >> > >> >>>> > Do you have "." in PATH? >> > >> >>>> > >> > >> >>>> > >> > >> >>>> > No >> > >> >>>> > >> > >> >>>> > What does "which m4" say? >> > >> >>>> > >> > >> >>>> > >> > >> >>>> > It returns nothing! Thanks for the hint. I found the problem: >> > >> >>>> > when reading the sage README to install sage from scratch on >> > >> >>>> > this new machine, I installed the basic stuff as follows: >> > >> >>>> > >> > >> >>>> > $ sudo apt install gcc gfortran g++ make m4 perl ranlib tar bc >> > >> >>>> > [...] >> > >> >>>> > E: Impossible de trouver le paquet ranlib >> > >> >>>> > >> > >> >>>> > But this returned an error on ranlib and I assumed it had >> > >> >>>> > installed the other packages in the list but no, it did not >> > >> >>>> > (gcc, gfortran, etc. were not installed as well). >> > >> >>>> > >> > >> >>>> > Is ranlib really needed for sage? Can we remove it from the >> > >> >>>> > README? >> > >> >>>> >> > >> >>>> ranlib is a standard tool, and is used by various build systems. On >> > >> >>>> Debian it's a part of binutils package. >> > >> >>>> >> > >> >>>> README is not platform-specific, and it does not tell you to >> > >> >>>> install >> > >> >>>> *package* called ranlib. >> > >> >>>> However, we should add there an advice to look for what packages >> > >> >>>> to install in >> > >> >>>> build/pkgs/_prereq/distros/ >> > >> >>>> (e.g. for Debian it's in build/pkgs/_prereq/distros/debian.txt) >> > >> >>>> >> > >> >>>> (Once our GitHub is live in few days, it's a very quick PR to do >> > >> >>>> :-)) >> > >> >>>> >> > >> >>>> >> > >> >>>> >> > >> >>>> > >> > >> >>>> > >> > >> >>>> > >> > >> >>>> > -- >> > >> >>>> > 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 sage-release...@googlegroups.com. >> > >> >>>> > To view this discussion on the web visit >> > >> >>>> > https://groups.google.com/d/msgid/sage-release/d6aecaee-279a-45f8-a645-36483d72a382n%40googlegroups.com. >> > >> >>> >> > >> >>> -- >> > >> >>> 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 sage-release...@googlegroups.com. >> > >> >>> >> > >> >>> To view this discussion on the web visit >> > >> >>> https://groups.google.com/d/msgid/sage-release/4a516a44-b761-4f02-ac10-e121cadc54d3n%40googlegroups.com. >> > >> > >> > >> > -- >> > >> > 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 sage-release...@googlegroups.com. >> > >> > To view this discussion on the web visit >> > >> > https://groups.google.com/d/msgid/sage-release/4d86cb2d-8c03-4101-904e-21a1a1ce8b93n%40googlegroups.com. >> > > >> > > -- >> > > 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 sage-release...@googlegroups.com. >> > > To view this discussion on the web visit >> > > https://groups.google.com/d/msgid/sage-release/2660c6b9-57cd-4d75-b0f8-97636bbfaf71n%40googlegroups.com. > > -- > 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 sage-release+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/sage-release/f362b0db-0f36-4c03-9949-473954352652n%40googlegroups.com. -- 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 sage-release+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-release/CAAWYfq2%2By59k9ryk4_hvZiBgJF9rTMQTu8Bu%3Do%2B49%2BG25ua3zw%40mail.gmail.com.
