I think you should consider to upgrade to a more recent version of maxima,
actual is 5.18.1
I tried to factor some very big numbers and I had no problems.
This took about an half minute:
(%i5) factor(320982340942382308202876786876067868068706066060860);
(%o5) 2^2*3*5*7^2*127*5573*66321025235333753*2325893443359844534239163
Stefano
In data marted? 05 maggio 2009 17:26:48, dastew at sympatico.ca ha scritto:
: > I just bought a laptop (HP Pavilion 2G ram, AMD Turion X2) and put Ubuntu
> 9.04 on it.
>
>
> I then, using Synaptic package manager, installed Maxima ( xmaxima 5.13.0.
> It seemed to work OK, so I tried to factor a big number to see how fast it
> was going. At this point I got an error message, so Tried a smaller number
> and it factored OK.
>
> To make a long story sort , let me say that I tried many different tests,
> and I think there is a stack overflow problem.
>
> factor( big numbers) fails
>
> when I run the built in tests it fails on:
>
> rtestnset.mac at test
>
> map(lambda([x], apply("+",x)), integer_partitions(25));
> set(25)$
>
> but if I just run that test by itself , it passes the test.
>
>
> rtest15 problem 192 fails : it also uses factor.
>
>
> I have Ubuntu 8.04 on a desktop and ran all the same test on it and all
> worked fine. This was all done with xmaxima 5.13.0, same maxima in both
> machines.
>
> I then looked at what was new in Ubuntu 9.04, and found:
>
>
>
> Hardened Kernels
>
>
>
> Hardened kernels are modifications to the Linux kernel that add additional
> security measures. This could include:
>
> The randomization of ports,
> memory addresses, process ID's, and other information that is typically
> predictable. This can thwart off many types of common attacks.Identify and
> prevent buffer overflow attacks from resulting in compromise by killing
> compromised processes (PaX bundled with
> grsecurity, or Redhat's Exec-Shield combined with prelink
> randomization). Edgy and higher contain GCC stack protection enforced
> in most applications, but is unable to respond to several kinds of
> attacks that a kernel-layer enforcer could. Likewise, PaX and friends
> have weakness that GCC stack protection helps cover, so the two work
> great as a duo.Hiding information that Linux usually allows everyone to
> see, including all running processes on the system, load averages, CPU
> info, IP addresses, etc. Obscuring this information can help keep attackers
> "in the dark" so to speak.More aggressive enforcement of buffer overflow
> protection than what Ubuntu's standard gcc stack protector can do.Adding
> additional restrictions on the capabilities of regular users that prevent
> channels of attack.Additional permissions systems that allow finer-grained
> tuning of various aspects of Linux.
>
>
> particularly this statement.
> "Edgy and higher contain GCC stack protection enforced
> in most applications,"
>
>
> So now my questions;
>
> Has anybody else seen this problem?
>
> How do I debug this more to prove it is a stack problem?
>
> Can you monitor the stack from Ubuntu?
>
> It is not a memory problem, memory usage was at about 30% when the problem
> first show up, an I ran Octave and used up 50% of the memory and then ran
> Maxima again and it failed at exactly the same steps.
>
> Doug Stewart