It would be helpful if you could report specifically *what* large numbers
did and didn't cause problems. Also, you should probably update to a more
recent version of Maxima.
See below for some examples of Maxima successfully factoring rather large
numbers.
-s
(%i1) display2d:false$
(%i2) showtime:true$
Evaluation took 0.0000 seconds (0.0000 elapsed)
(%i3) factor(1000!); <<<<
very composite number
Evaluation took 0.0100 seconds (0.0100 elapsed)
(%o3)
2^994*3^498*5^249*7^164*11^98*13^81*17^61*19^54*23^44*29^35*31^33*37^27
*41^24*43^23*47^21*53^18*59^16*61^16*67^14*71^14*73^13*79^12*83^12
*89^11*97^10*101^9*103^9*107^9*109^9*113^8*127^7*131^7*137^7*139^7
*149^6*151^6*157^6*163^6*167^5*173^5*179^5*181^5*191^5*193^5*197^5
*199^5*211^4*223^4*227^4*229^4*233^4*239^4*241^4*251^3*257^3*263^3
*269^3*271^3*277^3*281^3*283^3*293^3*307^3*311^3*313^3*317^3*331^3
*337^2*347^2*349^2*353^2*359^2*367^2*373^2*379^2*383^2*389^2*397^2
*401^2*409^2*419^2*421^2*431^2*433^2*439^2*443^2*449^2*457^2*461^2
*463^2*467^2*479^2*487^2*491^2*499^2*503*509*521*523*541*547*557
*563*569*571*577*587*593*599*601*607*613*617*619*631*641*643*647
*653*659*661*673*677*683*691*701*709*719*727*733*739*743*751*757
*761*769*773*787*797*809*811*821*823*827*829*839*853*857*859*863
*877*881*883*887*907*911*919*929*937*941*947*953*967*971*977*983
*991*997
(%i4) factor(40!+1); <<<<<<< not very composite number
Evaluation took 3.3900 seconds (3.3900 elapsed)
(%o4) 41*59*277*217823*16558103*142410167827*2370686450613664429
(%i5) factor(50!+1);
bpv: 74195127103 <<<<<<<<<< hmm, I don't know what this means
Evaluation took 5.3100 seconds (5.3100 elapsed)
(%o5)
149*3989*74195127103*6854870037011*100612041036938568804690996722352077
On Tue, May 5, 2009 at 11:26 AM, <dastew at sympatico.ca> wrote:
>
> 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:
>
> 1. 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.
> 2. 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.
> 3. 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.
> 4. More aggressive enforcement of buffer overflow protection than what
> Ubuntu's standard gcc stack protector can do.
> 5. Adding additional restrictions on the capabilities of regular users
> that prevent channels of attack.
> 6. 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
>
>
>
>
>
>
>
> _______________________________________________
> Maxima mailing list
> Maxima at math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima
>
>