#14501: Fix memory allocation problems in data_structures_pyx.pxi
---------------------------------+------------------------------------------
Reporter: dcoudert | Owner: joyner
Type: defect | Status: needs_review
Priority: major | Milestone: sage-5.10
Component: group theory | Resolution:
Keywords: | Work issues:
Report Upstream: N/A | Reviewers:
Authors: David Coudert | Merged in:
Dependencies: | Stopgaps:
---------------------------------+------------------------------------------
Old description:
> The following test causes a segfault on my computer (Intel(R) Xeon(R) CPU
> W3520 @ 2.67GHz / 4GB ram / linux)
> {{{
> sage: G = graphs.CompleteGraph(10)
> sage: GG = G.line_graph().line_graph().line_graph()
> sage: H = GG.relabel(inplace=False)
> sage: GG.is_isomorphic(H)
> ...
> sage: line 135: 3438 Segmentation fault (core dumped)
> "$SAGE_ROOT/spkg/bin/sage" "$@"
> }}}
> Tracking the segfault, I ended up in the {{{SC_new}}} of
> {{{sage/groups/perm_gps/partn_ref/data_structures_pyx.pxi}}}
>
> This patch fix some assignments performed before testing memory
> allocations and other similar stuff, but does not fix my problem.
>
> Apply:
> * trac_14501.2.patch
New description:
The following test causes a segfault on my computer (Intel(R) Xeon(R) CPU
W3520 @ 2.67GHz / 4GB ram / linux)
{{{
sage: G = graphs.CompleteGraph(10)
sage: GG = G.line_graph().line_graph().line_graph()
sage: H = GG.relabel(inplace=False)
sage: GG.is_isomorphic(H)
...
sage: line 135: 3438 Segmentation fault (core dumped)
"$SAGE_ROOT/spkg/bin/sage" "$@"
}}}
Tracking the segfault, I ended up in the {{{SC_new}}} of
{{{sage/groups/perm_gps/partn_ref/data_structures_pyx.pxi}}} and also in
the {{{(de)allocate_dc_work_space}}} methods of {{{double_coset.pyx}}}
file.
This patch fix some assignments performed before testing memory
allocations and other similar stuff. Finally, I end up with a proper
MemoryError and so the problem is solved
Apply:
* trac_14501.patch
--
Comment (by dcoudert):
> In the case at hand, 10 lines after the malloc that you relaced with a
calloc the whole vector is overwritten with data that makes sense,
contrary to 0s.
The point is that if the {{{SC_dealloc}}} function is called before
assigning something that make sense to SC.generators[i], then sage_free is
called with unknown argument.
I did further corrections in this file plus in the {{{double_coset.pyx}}}
file.
In both files we had calls to free methods on unassigned pointers.
Overall these modification solve the problem and now the method raises a
MemoryError!
Further cleaning should certainly be done in these files to prevent
similar errors.
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/14501#comment:6>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-trac?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
