FactorTroll II x64 - Page 3

Batalov · 2016-02-23, 01:00

Quote:

Originally Posted by mrft2

And since I never will post the source code then as quoted the tool will never be downloadable from this forum?
If so then there is no reason to have this thread.

1. If a person came up to you in the street gave you a pill and said: "Here, try this pill. I guarantee - it will make you happy" will you drink it? He would also add "I will never tell you what is in it".
2. If the same happened in a club (or a public library), wouldn't you be surprised (or in fact glad) to see that person escorted from the premises?

So, you want the forum to be the target for scorn (or damages, -- slim but not entirely non-existent possibility) in exchange for the great honor of disseminating a program that nobody can be sure what it does? Well, gee, thanks, man!

If it makes you happier, three malware posts were already invisibly deleted from the forum this week (and a similar amount every week). Yours passed someone else's approval threshold. You should consider yourself lucky!

Batalov · 2016-02-23, 01:09

Quote:

Originally Posted by bsquared

Msieve doesn't natively support multi-threading for either siqs or ecm (i.e., -t 8 runs the same as -t 1).

True, but:

Quote:

Originally Posted by msieve

The linear algebra code used in the quadratic- and number field sieve is multithread aware, and the entire library is supposed to be multithread-safe. The [included] demo application [msieve[.exe]] has only one job: to act as a delivery vehicle for integers to be factored.

Just needs to be tweaked a bit and it will be m/t. But then again, this will be just a yayafu! (and yafu is better!)

Quote:

Originally Posted by bsquared

But it's not clear whether the OP modified msieve's source or if his programs runs msieve in a multithreaded wrapper somehow (I have a self-preservation instinct, I guess, and haven't downloaded the program to test it).

Same here. Pros: maybe, just maybe, see someone else's closed source (and GUI'd, meh) yet another "yeat another factoring utility"

. Cons: may need to spend an evening restoring from backup and otherwise cleaning up possible maware. Let me think... very tempting... not.

jasonp · 2016-02-23, 01:52

The Msieve source is public domain and thus has no licensing conditions on reuse or redistribution. YAFU is the same way. There are tiny parts of it that are BSD-licensed, but that doesn't forbid redistribution either.

Does the Miracl ECM implementation have a fast stage 2? GMP-ECM does, and it makes a big difference when the ECM bounds increase in size. Unfortunately the GMP-ECM library is not technically multithread-safe, I think it uses a sieve of eratosthenes that has some global state. You could actually do the community a great service if you could fix that and Msieve does have a GMP-ECM driver. Alternately, YAFU can call multiple copies of the GMP-ECM demo binary.

(Miracl is open source for non-commercial use, though IIRC it is not GPLed)

YAFU can also run the number field sieve automatically, and making Msieve do that would take an amount of work equivalent to what YAFU already has :)

bsquared · 2016-02-23, 02:01

Quote:

Originally Posted by jasonp

Does the Miracl ECM implementation have a fast stage 2? GMP-ECM does, and it makes a big difference when the ECM bounds increase in size.

IIRC, no. It uses prime pairing and some other tricks and has a fast underlying bignum library but has no FFT-based continuation.

Quote:

Originally Posted by jasonp

Unfortunately the GMP-ECM library is not technically multithread-safe, I think it uses a sieve of eratosthenes that has some global state. You could actually do the community a great service if you could fix that and Msieve does have a GMP-ECM driver. Alternately, YAFU can call multiple copies of the GMP-ECM demo binary.

Yep :) The unsafeness of the library is why yafu only allows multithreading with the external binary.

Quote:

Originally Posted by jasonp

(Miracl is open source for non-commercial use, though IIRC it is not GPLed)

The miracl.h header file on github has this:

Quote:

Originally Posted by miracl.h

/***************************************************************************
*
Copyright 2013 CertiVox UK Ltd. *
*
This file is part of CertiVox MIRACL Crypto SDK. *
*
The CertiVox MIRACL Crypto SDK provides developers with an *
extensive and efficient set of cryptographic functions. *
For further information about its features and functionalities please *
refer to http://www.certivox.com *
*
* The CertiVox MIRACL Crypto SDK is free software: you can *
redistribute it and/or modify it under the terms of the *
GNU Affero General Public License as published by the *
Free Software Foundation, either version 3 of the License, *
or (at your option) any later version. *
*
* The CertiVox MIRACL Crypto SDK is distributed in the hope *
that it will be useful, but WITHOUT ANY WARRANTY; without even the *
implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. *
See the GNU Affero General Public License for more details. *
*
* You should have received a copy of the GNU Affero General Public *
License along with CertiVox MIRACL Crypto SDK. *
If not, see <http://www.gnu.org/licenses/>. *
*
You can be released from the requirements of the license by purchasing *
a commercial license. Buying such a license is mandatory as soon as you *
develop commercial activities involving the CertiVox MIRACL Crypto SDK *
without disclosing the source code of your own applications, or shipping *
the CertiVox MIRACL Crypto SDK with a closed source product. *
*
***************************************************************************/

LaurV · 2016-02-23, 03:10

Quote:

Originally Posted by Dubslow

That's not entirely the whole truth: there is in fact another program that's effectively hosted on this forum which, to the best of my knowledge, hasn't had source posted: http://mersenneforum.org/showthread.php?t=17915

Wrong. Did you followed the links in your link? I was one who strongly argued with Scott for those source codes. They are available on the download page, and ever were since. OTOH, it is true that I use the exe provided by Scott, as I was lazy to compile the newest versions (in fact, I use an order version, if it runs do not fix it). I however used to build it myself for a while, and the sources were available, in fact Scott sent me the sources long before putting them on web, because I was such a big mouth...

Antonio · 2016-02-23, 08:53

I may be missing something, BUT, could someone please explain how publishing source code implies the lack of malware being attached to an executable, which may just have the same name?
It could be that the source and the executable provided have no common base, so would be a deliberate infection attempt, or the originators computer may be infected and the executable is accidentally carrying a malware payload.

axn · 2016-02-23, 09:28

We can always build the executable from source. More importantly, looking at the source, we can see if there is code corresponding to claimed functionality. That helps with the trust factor.

LaurV · 2016-02-23, 09:43

Quote:

Originally Posted by Antonio

how publishing source code implies ...

It does not. Publishing a right source code shows that the author really worked on the problem, and it gives to the more skeptical guy a way to compile his own exe file. Other people don't care, they will use the exe with or without source code, or they will not use the exe, with or without the source code. If I compile some stuff by myself and get the same binary like the public one, I have a reason to trust that guy's binaries in the future. The golden rule is "don't run binaries from the web". But we all know this is "too restrictive"

henryzz · 2016-02-23, 16:20

Quote:

Originally Posted by LaurV

Wrong. Did you followed the links in your link? I was one who strongly argued with Scott for those source codes. They are available on the download page, and ever were since. OTOH, it is true that I use the exe provided by Scott, as I was lazy to compile the newest versions (in fact, I use an order version, if it runs do not fix it). I however used to build it myself for a while, and the sources were available, in fact Scott sent me the sources long before putting them on web, because I was such a big mouth...

Posting binaries for .net programs is equivalent to posting source unless you go though loops to hide the source as you can decompile.

ixfd64 · 2016-02-23, 20:43

The screenshot links are returning a 404 error.

Batalov · 2016-02-23, 21:14

Looks like it was malware, then. Promptly removed.

2016-02-23, 01:52	#25
jasonp Tribal Bullet Oct 2004 3,541 Posts	The Msieve source is public domain and thus has no licensing conditions on reuse or redistribution. YAFU is the same way. There are tiny parts of it that are BSD-licensed, but that doesn't forbid redistribution either. Does the Miracl ECM implementation have a fast stage 2? GMP-ECM does, and it makes a big difference when the ECM bounds increase in size. Unfortunately the GMP-ECM library is not technically multithread-safe, I think it uses a sieve of eratosthenes that has some global state. You could actually do the community a great service if you could fix that and Msieve does have a GMP-ECM driver. Alternately, YAFU can call multiple copies of the GMP-ECM demo binary. (Miracl is open source for non-commercial use, though IIRC it is not GPLed) YAFU can also run the number field sieve automatically, and making Msieve do that would take an amount of work equivalent to what YAFU already has :) Last fiddled with by jasonp on 2016-02-23 at 01:55

2016-02-23, 08:53	#28
Antonio "Antonio Key" Sep 2011 UK 3²×59 Posts	I may be missing something, BUT, could someone please explain how publishing source code implies the lack of malware being attached to an executable, which may just have the same name? It could be that the source and the executable provided have no common base, so would be a deliberate infection attempt, or the originators computer may be infected and the executable is accidentally carrying a malware payload.

2016-02-23, 09:28	#29
axn Jun 2003 5051₁₀ Posts	We can always build the executable from source. More importantly, looking at the source, we can see if there is code corresponding to claimed functionality. That helps with the trust factor.

2016-02-23, 20:43	#32
ixfd64 Bemusing Prompter "Danny" Dec 2002 California 100101010110₂ Posts	The screenshot links are returning a 404 error.

2016-02-23, 21:14	#33
Batalov "Serge" Mar 2008 Phi(4,2^7658614+1)/2 10010100000101₂ Posts	Looks like it was malware, then. Promptly removed.