How to start?

[leifbk]

Originally Posted by garo

ric beat me to the response there! Also, leif can you tell me what the error message was while compiling? I have successfully compiled decomp.c on Solaris and Linux. Did you use the -lm flag to tell it to use the math library?

First, I must say that I'm no 1337 C haxx0r

I changed the includes because of undeclared atoi() and isdigit():

#include <stdio.h>
#include <stdlib.h>
#include <ctype.h>
#include <math.h>
#include <malloc.h>
#include <memory.h>
#include <string.h>

Then I got "DECOMP.C:371: error: `main' must return `int'" and I replaced void main() with int main().

Now I get:

leif@balapapa lmh $ gcc -lm DECOMP.C -o decomp
DECOMP.C: In function `int isprime(long unsigned int)':
DECOMP.C:26: warning: converting to `long unsigned int' from `double'
/tmp/ccsu5CHE.o(.eh_frame+0x11): undefined reference to `__gxx_personality_v0'
collect2: ld returned 1 exit status

What now?

[garo]

This is what my includes look like: These are the original DECOMP.C includes.

#include <stdio.h>
#include<stdlib.h>
#include <math.h>
#include <malloc.h>
#include <memory.h>
#include <string.h>
#include <ctype.h>

And this is what compiling gives me:

gcc -o a.out DECOMP.C -lm
DECOMP.C: In function `int isprime(long unsigned int)':
DECOMP.C:25: warning: assignment to `long unsigned int' from `double'
DECOMP.C:25: warning: argument to `long unsigned int' from `double'

BTW, both atoi and isdigit are standard C library functions so I don't understand why you should have trouble if stdlib is included.

[leifbk]

Okay. I'm at work now, so I'll make a fresh start. I cut and pasted your includes and commented out the originals:

leif@bombadil lmh $ gcc -o decomp DECOMP.C -lm
DECOMP.C: In function `int isprime(long unsigned int)':
DECOMP.C:36: warning: converting to `long unsigned int' from `double'
DECOMP.C: At global scope:
DECOMP.C:381: error: `main' must return `int'

Then I make the change from "void main" to "int main":

leif@bombadil lmh $ gcc -o decomp DECOMP.C -lm
DECOMP.C: In function `int isprime(long unsigned int)':
DECOMP.C:36: warning: converting to `long unsigned int' from `double'
/tmp/cc01U5sU.o(.eh_frame+0x11): undefined reference to `__gxx_personality_v0'
collect2: ld returned 1 exit status

As you probably know, Gentoo is a "from source" distribution. Thus, my gcc (both here and at home) is in almost daily use for patching and upgrading, and hardly chokes on anything.

I'm suspecting a gcc version problem here:

leif@bombadil lmh $ gcc --version
gcc (GCC) 3.3.5  (Gentoo Linux 3.3.5-r1, ssp-3.3.2-3, pie-8.7.7.1)
Copyright (C) 2003 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

I suppose that there's a way to set a flag for backwards compatibility?



regards, Leif.

[garo]

Correct. A google search revealed the following:

Edit your Makefile and add "-lstdc++" as shown below to the end of the
variable named "STATIC_LIBS"

STATIC_LIBS= $(GD_STATIC) $(OGR_STATIC) -lstdc++

Since you are not using a makefile just add the -lstdc++ flag to the compile line.

[leifbk]

Originally Posted by garo

Correct. [...] Since you are not using a makefile just add the -lstdc++ flag to the compile line.

Yezz!!!

It still won't accept a "void main", so I have to change that to "int main", but that's trivial.

Besides, both stdlib.h and ctype.h are missing from the version in decomp.zip and have to be added manually.

Humble wish: May I suggest that somebody with a little more knowledge about C than what I've got, updates the source so it compiles cleanly with recent versions of gcc?

regards, Leif.

[Citrix]

Can Prime95 work on +1 numbers. I just want to find some factors for run.

Citrix

[Riza]

Can anybody write me, what a factoring work, I can let do at a

P 200 @ 340 MB RAM and

K6 400 @ 512 MB RAM?

Please write, what is to copy in Worktodo.ini.

The result I send to?

Thanks

[antiroach]

Is it worthwhile for AMD (athlon xp) processors to factor numbers upto 2^64? If not when does it become inefficient for Athlon XP processors to perform this type of factoring?

[1997rj7]

Athlon XPs are very efficient up to 2^64. They are not good above 2^64 because they lack SSE2.

[antiroach]

i've done work at LMH before when the software still had a prime.ini file. Im currently using the newest (?) version of mprime (25.3.2).
When i use the FactorOverride=xx command in the prime.txt file it doesnt seem to take effect and it tries to factor the number upto 70 bits. what am i doing wrong?

[tmorrow]

Hmm! Version 25.3 is very much an alpha version still. George is distributing it to test the v4 and v5 servers on trial factoring above 79.3 million and on multicore machines and odd internet settings. Refer to http://www.mersenneforum.org/showthread.php?t=8415 and http://www.mersenneforum.org/showthread.php?t=8431. That client may be specially configured to factor those high exponents (just guessing).

I've only had experience with v24.14 which seems to work fine and is still the recommended stable client.

Just out of curiosity, what exponent were you trial factoring?