sage-4.8.alpha5

24 views
Skip to first unread message

William Stein

unread,
Dec 14, 2011, 1:12:50 AM12/14/11
to sage-release, Martin Albrecht
Hi,

I tried building sage-4.8.alpha5 on Ubuntu 11.10 64-bit, and was
disappointed when I try to start it and get a Segmentation fault
immediately on startup. Here is the traceback [5] below, which I
guess suggests an issue with Martin Albrecht's new matrix_modn code
that uses Linbox directly [1], which was evidently merged into
4.8.alpha3. To test this, I made a clone back to right before this
4260 was included (using [2]), and did "sage -br" and then "sage -ba".
Then I hit [3]. Then I rebuilt singular again, and hit [4], which
made me see that "sage -ba" doesn't work right -- it leaves old code
in devel/sage/build. ARGH. Anyways. So I manually deleted
devel/sage/build completely and do "sage -ba" again, and Sage actually
starts. So evidently "sage -ba" is broken.

To summarize: we will not release sage-4.8 until it builds out of the
box on at least Ubuntu 11.10.

[1] http://trac.sagemath.org/sage_trac/ticket/4260

[2] sage-4.8.alpha5$ ./sage -clone -r 16293 pre4260
sage-4.8.alpha5$ ./sage -ba

[3] "ImportError:
/home/wstein/sage/sage-4.8.alpha5/local/lib/libsingular.so: undefined
symbol: UP"

[4] "ImportError:
/home/wstein/sage/sage-4.8.alpha5/local/lib/python2.6/site-packages/sage/matrix/matrix_modn_dense_double.so:
undefined symbol: _ZNK7IntegercvlEv"

[5] Program received signal SIGSEGV, Segmentation fault.
0x0000000000012336 in ?? ()
(gdb) bt
#0 0x0000000000012336 in ?? ()
#1 0x00007fffdf226612 in __Pyx_check_binary_version () at
sage/matrix/matrix_modn_dense_float.cpp:17227
#2 initmatrix_modn_dense_float () at
sage/matrix/matrix_modn_dense_float.cpp:15808
#3 0x00007ffff7b1c625 in _PyImport_LoadDynamicModule (
name=0x7ffffffc81c0 "sage.matrix.matrix_modn_dense_float",
pathname=0x7ffffffc70f0
"/home/wstein/sage/sage-4.8.alpha5/local/lib/python2.6/site-packages/sage/matrix/matrix_modn_dense_float.so",
fp=<optimized out>) at ./Python/importdl.c:53
#4 0x00007ffff7b1a6bc in import_submodule (mod=0x17df750,
subname=0x7ffffffc81cc "matrix_modn_dense_float",
fullname=0x7ffffffc81c0 "sage.matrix.matrix_modn_dense_float") at
Python/import.c:2589
#5 0x00007ffff7b1a93f in load_next (mod=0x17df750, altmod=0x17df750,
p_name=<optimized out>,
buf=0x7ffffffc81c0 "sage.matrix.matrix_modn_dense_float",
p_buflen=0x7ffffffc81b0)
at Python/import.c:2409
--
William Stein
Professor of Mathematics
University of Washington
http://wstein.org

Jeroen Demeyer

unread,
Dec 14, 2011, 3:35:37 AM12/14/11
to sage-r...@googlegroups.com
On 2011-12-14 07:12, William Stein wrote:
> To summarize: we will not release sage-4.8 until it builds out of the
> box on at least Ubuntu 11.10.
Have you tried a different gcc version? That's an obvious thing to try
if you get segmentation faults.

Martin Albrecht

unread,
Dec 14, 2011, 6:30:55 AM12/14/11
to sage-r...@googlegroups.com
On Wednesday 14 December 2011, William Stein wrote:
> Hi,

Hi,

> [5] Program received signal SIGSEGV, Segmentation fault.
> 0x0000000000012336 in ?? ()
> (gdb) bt
> #0 0x0000000000012336 in ?? ()
> #1 0x00007fffdf226612 in __Pyx_check_binary_version () at
> sage/matrix/matrix_modn_dense_float.cpp:17227
> #2 initmatrix_modn_dense_float () at
> sage/matrix/matrix_modn_dense_float.cpp:15808

If I understand this backtrace correctly it crashes in the initialisation of
the module ("__Pyx_check_binary_version ()"). Is there any 11.10 box where I
can ssh into?

> #3 0x00007ffff7b1c625 in _PyImport_LoadDynamicModule (
> name=0x7ffffffc81c0 "sage.matrix.matrix_modn_dense_float",
> pathname=0x7ffffffc70f0
> "/home/wstein/sage/sage-4.8.alpha5/local/lib/python2.6/site-packages/sage/m
> atrix/matrix_modn_dense_float.so", fp=<optimized out>) at
> ./Python/importdl.c:53
> #4 0x00007ffff7b1a6bc in import_submodule (mod=0x17df750,
> subname=0x7ffffffc81cc "matrix_modn_dense_float",
> fullname=0x7ffffffc81c0 "sage.matrix.matrix_modn_dense_float") at
> Python/import.c:2589
> #5 0x00007ffff7b1a93f in load_next (mod=0x17df750, altmod=0x17df750,
> p_name=<optimized out>,
> buf=0x7ffffffc81c0 "sage.matrix.matrix_modn_dense_float",
> p_buflen=0x7ffffffc81b0)
> at Python/import.c:2409

Cheers,
Martin

--
name: Martin Albrecht
_pgp: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x8EF0DC99
_otr: 47F43D1A 5D68C36F 468BAEBA 640E8856 D7951CCF
_www: http://martinralbrecht.wordpress.com/
_jab: martinr...@jabber.ccc.de

Francois Bissey

unread,
Dec 14, 2011, 6:55:32 AM12/14/11
to sage-r...@googlegroups.com
On Wed, 14 Dec 2011 11:30:55 Martin Albrecht wrote:
> On Wednesday 14 December 2011, William Stein wrote:
> > Hi,
>
> Hi,
>
> > [5] Program received signal SIGSEGV, Segmentation fault.
> > 0x0000000000012336 in ?? ()
> > (gdb) bt
> > #0 0x0000000000012336 in ?? ()
> > #1 0x00007fffdf226612 in __Pyx_check_binary_version () at
> > sage/matrix/matrix_modn_dense_float.cpp:17227
> > #2 initmatrix_modn_dense_float () at
> > sage/matrix/matrix_modn_dense_float.cpp:15808
>
> If I understand this backtrace correctly it crashes in the initialisation of
> the module ("__Pyx_check_binary_version ()"). Is there any 11.10 box where
> I can ssh into?
>
> > #3 0x00007ffff7b1c625 in _PyImport_LoadDynamicModule (
> >
> > name=0x7ffffffc81c0 "sage.matrix.matrix_modn_dense_float",
> > pathname=0x7ffffffc70f0
> >
> > "/home/wstein/sage/sage-4.8.alpha5/local/lib/python2.6/site-packages/sag
> > e/m atrix/matrix_modn_dense_float.so", fp=<optimized out>) at

> > ./Python/importdl.c:53
> > #4 0x00007ffff7b1a6bc in import_submodule (mod=0x17df750,
> >
> > subname=0x7ffffffc81cc "matrix_modn_dense_float",
> > fullname=0x7ffffffc81c0 "sage.matrix.matrix_modn_dense_float")
> > at
> >
> > Python/import.c:2589
> > #5 0x00007ffff7b1a93f in load_next (mod=0x17df750, altmod=0x17df750,
> > p_name=<optimized out>,
> >
> > buf=0x7ffffffc81c0 "sage.matrix.matrix_modn_dense_float",
> >
> > p_buflen=0x7ffffffc81b0)
> >
> > at Python/import.c:2409
>
I got that exact same problem the first time I built 4.8.alpha3 in sage-on-
gentoo. matrix_modn_dense_float and matrix_modn_dense_double should
both depend on givaro as well as linbox. That solved the issue for us and
I told Burcin about it but I haven't had time to open a ticket for it.
See
https://github.com/cschwan/sage-on-gentoo/issues/108
for how thing unfolded in sage-on-gentoo

Francois

This email may be confidential and subject to legal privilege, it may
not reflect the views of the University of Canterbury, and it is not
guaranteed to be virus free. If you are not an intended recipient,
please notify the sender immediately and erase all copies of the message
and any attachments.

Please refer to http://www.canterbury.ac.nz/emaildisclaimer for more
information.

Francois Bissey

unread,
Dec 14, 2011, 7:10:11 AM12/14/11
to sage-r...@googlegroups.com

On Thu, 15 Dec 2011 00:55:32 Francois Bissey wrote:
> On Wed, 14 Dec 2011 11:30:55 Martin Albrecht wrote:
> > On Wednesday 14 December 2011, William Stein wrote:
> > > Hi,
> >
> > Hi,
> >
> > > [5] Program received signal SIGSEGV, Segmentation fault.
> > > 0x0000000000012336 in ?? ()
> > > (gdb) bt
> > > #0  0x0000000000012336 in ?? ()
> > > #1  0x00007fffdf226612 in __Pyx_check_binary_version () at
> > > sage/matrix/matrix_modn_dense_float.cpp:17227
> > > #2  initmatrix_modn_dense_float () at
> > > sage/matrix/matrix_modn_dense_float.cpp:15808
> >
> > If I understand this backtrace correctly it crashes in the
> > initialisation of the module ("__Pyx_check_binary_version ()"). Is
> > there any 11.10 box where I can ssh into?
> >
> > > #3  0x00007ffff7b1c625 in _PyImport_LoadDynamicModule (
> > >
> > >     name=0x7ffffffc81c0 "sage.matrix.matrix_modn_dense_float",
> > >     pathname=0x7ffffffc70f0
> > >
> > > "/home/wstein/sage/sage-4.8.alpha5/local/lib/python2.6/site-packages

> > > /sag e/m atrix/matrix_modn_dense_float.so", fp=<optimized out>) at


> > > ./Python/importdl.c:53
> > > #4  0x00007ffff7b1a6bc in import_submodule (mod=0x17df750,
> > >
> > >     subname=0x7ffffffc81cc "matrix_modn_dense_float",
> > >     fullname=0x7ffffffc81c0
> > >     "sage.matrix.matrix_modn_dense_float")
> > >     at
> > >
> > > Python/import.c:2589
> > > #5  0x00007ffff7b1a93f in load_next (mod=0x17df750,
> > > altmod=0x17df750,
> > > p_name=<optimized out>,
> > >
> > >     buf=0x7ffffffc81c0 "sage.matrix.matrix_modn_dense_float",
> > >
> > > p_buflen=0x7ffffffc81b0)
> > >
> > >     at Python/import.c:2409
>
> I got that exact same problem the first time I built 4.8.alpha3 in sage-on-
> gentoo. matrix_modn_dense_float and matrix_modn_dense_double should
> both depend on givaro as well as linbox. That solved the issue for us and
> I told Burcin about it but I haven't had time to open a ticket for it.
> See
> https://github.com/cschwan/sage-on-gentoo/issues/108
> for how thing unfolded in sage-on-gentoo
>

I got a proper sage patch for it too since I was planning to open a ticket.
See attachment.

modn_givaro.patch

Dan Drake

unread,
Dec 14, 2011, 10:03:37 PM12/14/11
to sage-r...@googlegroups.com
On Thu, 15 Dec 2011 at 01:10AM +1300, Francois Bissey wrote:
> I got a proper sage patch for it too since I was planning to open a ticket.
> See attachment.

Your patch allows 4.8.alpha4 to start and pass doctests in sage/matrix
on my Ubuntu 11.10 machine. Other doctests are running, but I think
everything is okay.

Dan

--
--- Dan Drake
----- http://mathsci.kaist.ac.kr/~drake
-------

signature.asc

Volker Braun

unread,
Dec 15, 2011, 5:48:14 AM12/15/11
to sage-r...@googlegroups.com, Martin Albrecht
Reply all
Reply to author
Forward
0 new messages