#11705: Port Sage to SUSE Linux Power 7 (ppc64).
---------------------------------------------------+------------------------
Reporter: was | Owner: drkirkby
Type: enhancement | Status: new
Priority: major | Milestone: sage-5.0
Component: porting | Resolution:
Keywords: sd32 sd35.5 | Work issues:
Report Upstream: N/A | Reviewers:
Authors: Paul Zimmermann, Jeroen Demeyer | Merged in:
Dependencies: #12829, #12832 | Stopgaps:
---------------------------------------------------+------------------------
Comment (by leif):
Replying to [comment:86 zimmerma]:
> I tried configuring GMP-ECM with {{{-enable-assert}}}, nothing wrong
happens.
> I have no further idea to investigate. Any suggestion out there?
>
> Of course a possible workaround would be to configure with {{{--disable-
asm-redc}}} on
> powerpc64, but it would be nice to find the reason of this problem. Is
there any other
> powerpc on which we could test?
>
> What is strange is that the bug occurs **inside** libecm.a when called
from within Sage,
> and does not occur when called from the binary GMP-ECM (with the same
libecm.a). Maybe a memory corruption related to the ecm_params structure
which is not really defined in libecm.pyx?
The pyx just contains a type definition, and passes a null pointer to
`ecm_factor()` instead of some structure. Segfaults also happen outside
of Sage; see below.
[[BR]]
> Is it possible to link libecm dynamically to Sage?
Of course, you just have to build a shared library (`ECM_EXTRA_OPTS
="--enable-shared"`); see below.
[[BR]]
This is IMHO an upstream (or probably compiler/assembler etc.) bug.
Note that GMP-ECM's test suite segfaults in its first test when configured
with `--enable-shared` (and asm-redc enabled, which is the default). Same
for a trivial stand-alone program just resembling Sage's doctest.
The segfaults vanish in both cases if I in addition pass `--disable-asm-
redc`. So Sage just exposes the problem, it doesn't cause it.
[FWIW, Sage's `ecmfactor()` apparently works for even numbers, as well as
interestingly 1025, while all other odd numbers between 950 and 1050 cause
a segfault.]
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/11705#comment:99>
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 post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
