Hello,
I'm using Sage to do some symbolic computations.
What I am basically trying to do is to compute sums made of terms of
the form :
cst_{i_0,...,i_d)*b_0^{i_0}*...*b_d^{i_d}*2^{-b_0-...-b_d}
where the b_i are variables
(i.e. a product of an exponential with b_i in the exponent by a
multivariate polynomial in b_i)
using functions (especially symbolic_sum) from sage.calculus.calculus.
I am regularly calling expand() and simplify() functions as well as
some custom functions using substitute() to get a more compact
expression and avoid redundant terms.
Everything works fine as long as the sums are not too "complicated",
but Sage segfaults when they become too big.
Running Sage with the -gdb flag, gives different errors according to
which of my functions I am calling, but both errors seem to be related
to a "compare" function in Pynac (SIGSEGV or SIGBUS):
First error message :
Program received signal SIGSEGV, Segmentation fault.
GiNaC::power::compare (this=0x3281c20, other=...) at power.cpp:899
899 power.cpp: Aucun fichier ou dossier de ce type.
in power.cpp
Second error message :
Program received signal SIGBUS, Bus error.
GiNaC::mul::compare (this=0x83490e0, other=...) at mul.cpp:1174
1174 mul.cpp: Aucun fichier ou dossier de ce type.
in mul.cpp
My box is running a Debian unstable with packages from the
experimental repository.
My kernel version is 2.6.35, compiled by myself.
My cpu is a Intel(R) Core(TM)2 Quad CPU Q6600.
The error is happening with Sage 4.5.2 and 4.4.4.
I compiled both of them myself with gcc (Debian 4.4.4-11) 4.4.5
20100824 (prerelease).
I also tried the precompiled sage-4.5.2-linux-64bit-ubuntu_10.04_lts-
x86_64-Linux which gives me the same result.
I didn't manage to reproduce the bug with a short and simple piece of
code.
However I can mail the source code and instructions to reproduce the
bug to anyone who can help.
I will also provide any information needed to investigate this
problem.
I am currently trying to compile Sage on another box to do another
test.
Any help will be appreciated, and thanks for a such a great piece of
software.
Best regards,
Jean-Pierre Flori
--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
