mersenneforum.org  

Go Back   mersenneforum.org > Factoring Projects > Operazione Doppi Mersennes

Reply
 
Thread Tools
Old 2016-01-23, 16:19   #1
ET_
Banned
 
ET_'s Avatar
 
"Luigi"
Aug 2002
Team Italia

2·2,417 Posts
Cool Deep Sieving MM49 in parallel

Well, now that we have a new Mersenne prime, we may also add MM49 to the Search for Double Mersennes.

As someone already knows, I started sieving the possible Ks of MM49 to be used for the 2kp+1 potential factors of MM49.

The last version of the siever allows for parallel sieving, saving the new factors on a results file for further filtering.

On this first post I will include the source code of my siever (MMpsieve7f.c) and a bunch of ranges that need to be sieved for different primes.

Ranges:
MM49_0T_10T.zip - Luigi (ET_) - Completed
MM49_10T_20T.zip - firejuggler - Completed
MM49_20T_30T.zip - Luigi (ET_) - Completed
MM49_30T_40T.zip - Luigi (ET_) - Completed
MM49_40T_50T.zip - firejuggler - Completed
MM49_50T_60T.zip - firejuggler - Completed
MM49_60T_70T.zip - firejuggler - Completed
MM49_70T_80T.zip - Luigi (ET_) - Completed
MM49_80T_90T.zip - Luigi (ET_) - Completed
MM49_90T_100T.zip - Luigi (ET_) - Completed
MM49_100T_110T.zip - Luigi (ET_) - Completed
MM49_110T_120T.zip - Luigi (ET_) - Completed
MM49_120T_130T.zip - Luigi (ET_) - Completed
MM49_130T_140T.zip - Luigi (ET_) - Completed
MM49_140T_150T.zip - Luigi (ET_) - Completed
MM49_150T_160T.zip - Luigi (ET_) - Completed
MM49_160T_170T.zip - Luigi (ET_) - Completed
MM49_170T_180T.zip - Luigi (ET_) - Completed
MM49_180T_190T.zip - Luigi (ET_) - Completed
MM49_190T_200T.zip - Luigi (ET_) - Completed
MM49_200T_210T.zip - Luigi (ET_) - Completed
MM49_210T_220T.zip - Luigi (ET_) - Completed
MM49_220T_230T.zip - Luigi (ET_) - Completed
MM49_230T_240T.zip - Luigi (ET_) - Completed
MM49_240T_250T.zip - Luigi (ET_) - Completed
MM49_250T_260T.zip - firejuggler - Completed
MM49_260T_270T.zip - firejuggler - Completed
MM49_270T_280T.zip - Luigi (ET_) - Completed
MM49_280T_290T.zip - Luigi (ET_) - Completed
MM49_290T_300T.zip - Luigi (ET_) - Completed
MM49_300T_310T.zip - Luigi (ET_) - Completed
MM49_310T_320T.zip - Luigi (ET_) - Completed
MM49_320T_330T.zip - Luigi (ET_) - Completed
MM49_330T_340T.zip - Luigi (ET_) - Completed
MM49_340T_350T.zip - Luigi (ET_) - Completed
MM49_350T_360T.zip - Luigi (ET_) - Completed
MM49_360T_370T.zip - Luigi (ET_) - Completed
MM49_370T_380T.zip - Luigi (ET_) - Completed
MM49_380T_390T.zip - Luigi (ET_)
MM49_390T_400T.zip - Luigi (ET_)
MM49_400T_410T.zip - Luigi (ET_)
MM49_410T_420T.zip
MM49_420T_430T.zip
MM49_430T_440T.zip
MM49_440T_450T.zip
MM49_450T_460T.zip - Luigi (ET_)
MM49_460T_470T.zip - Luigi (ET_)
MM49_470T_480T.zip - Luigi (ET_)
MM49_480T_490T.zip - Luigi (ET_)
MM49_490T_500T.zip - Luigi (ET_)


The source code is here:
http://www.doublemersennes.org/sievi...MMpsieve7f.zip

How to launch the executable:
./MMpsieve7f max_range 4000000 input_file
where max_range is the range of p to achieve, and
input_file is the name of the downloaded file.

Example:
Code:
./MMsieve7f 70000000000000 4000000 MM49_60T_70T
Other files:
The executable will create a status file named input_file.out each 2G p's,
and a file named inlut_file.fact where the factors will be accumulated.
Once the run is completed, I need the *.fact file back.

Luigi

Last fiddled with by ET_ on 2016-03-03 at 12:20
ET_ is online now   Reply With Quote
Old 2016-01-23, 16:44   #2
firejuggler
 
firejuggler's Avatar
 
"Vincent"
Apr 2010
Over the rainbow

22×11×61 Posts
Default

and the compile under mingw fail... unless it is 64 bit only?
Attached Thumbnails
Click image for larger version

Name:	gccfail.jpg
Views:	288
Size:	43.6 KB
ID:	13766  

Last fiddled with by firejuggler on 2016-01-23 at 16:47
firejuggler is online now   Reply With Quote
Old 2016-01-23, 16:51   #3
firejuggler
 
firejuggler's Avatar
 
"Vincent"
Apr 2010
Over the rainbow

A7C16 Posts
Default

nevermind, managed to circumvent it

gcc -O2 -std=c99 -o MMpsieve.exe did it.

taking the 10T-20T file

Last fiddled with by firejuggler on 2016-01-23 at 17:40
firejuggler is online now   Reply With Quote
Old 2016-01-23, 18:19   #4
ET_
Banned
 
ET_'s Avatar
 
"Luigi"
Aug 2002
Team Italia

12E216 Posts
Default

Quote:
Originally Posted by firejuggler View Post
nevermind, managed to circumvent it

gcc -O2 -std=c99 -o MMpsieve.exe did it.

taking the 10T-20T file
Thank you
ET_ is online now   Reply With Quote
Old 2016-01-23, 20:43   #5
firejuggler
 
firejuggler's Avatar
 
"Vincent"
Apr 2010
Over the rainbow

22×11×61 Posts
Default

ok a few hours later I have a question : it went like this

Code:
MMpsieve7.exe - version 1.07f - compiled 2016-01-23
74207281
#sieved to k = 4000000/2, with primes up to 1316134912
#1316134912

 2548290 :      1316774618 - k =  1274145

 3275946 :      1316919498 - k =  1637973

 3617290 :      1317249468 - k =  1808645

  435418 :      1317261992 - k =   217709

 2243802 :      1317936900 - k =  1121901
..............

3231816 :      2146420832 - k =  1615908

3237120 :      2147061410 - k =  1618560

 961306 :      2147626442 - k =   480653

3034920 :      2149790928 - k =  1517460
it hasn't moved from this last line since the begining. is it normal? Or should I try under linux?
firejuggler is online now   Reply With Quote
Old 2016-01-23, 22:26   #6
firejuggler
 
firejuggler's Avatar
 
"Vincent"
Apr 2010
Over the rainbow

22×11×61 Posts
Default

and of course, the behavior under linux is different. and correct, I suppose
Code:
./MM49sieve  - version 1.07f - compiled 2016-01-23
74207281
#sieved to k = 4000000/2, with primes up to 10000000000000
#10000000000000
last_prime = 10004000440000
 3061378 :  10005885476274 - k =  1530689
last_prime = 10008000880000
and it write to a file... while under windows it stall there...
firejuggler is online now   Reply With Quote
Old 2016-01-24, 11:38   #7
ET_
Banned
 
ET_'s Avatar
 
"Luigi"
Aug 2002
Team Italia

2×2,417 Posts
Default

Quote:
Originally Posted by firejuggler View Post
and of course, the behavior under linux is different. and correct, I suppose
Code:
./MM49sieve  - version 1.07f - compiled 2016-01-23
74207281
#sieved to k = 4000000/2, with primes up to 10000000000000
#10000000000000
last_prime = 10004000440000
 3061378 :  10005885476274 - k =  1530689
last_prime = 10008000880000
and it write to a file... while under windows it stall there...
Odd...

The windows version starts with
Code:
#1316134912
.
Such number should be the maximum 64-bit p reached in the factoring level by the sieve: as you can see, the number is instead below 2^31. I guess you are using a 32-bit compile under MinGW, meaning you can't take advantage of uint64 elements, or I should create a conditional compile using a different declaration.

I suppose that such problem, and the \r\n coding for new lines has to do with the printout difference; sadly I don't have ready windows environments for the tests at the moment

The behaviour under Linux looks correct.


Luigi
ET_ is online now   Reply With Quote
Old 2016-01-25, 11:49   #8
firejuggler
 
firejuggler's Avatar
 
"Vincent"
Apr 2010
Over the rainbow

1010011111002 Posts
Default

Don't worry about the Windows executable. I have a Linux VM and it work.
firejuggler is online now   Reply With Quote
Old 2016-01-27, 14:05   #9
firejuggler
 
firejuggler's Avatar
 
"Vincent"
Apr 2010
Over the rainbow

1010011111002 Posts
Default

about to finish 10-20 T, will take 40-50.

Last fiddled with by firejuggler on 2016-01-27 at 14:06
firejuggler is online now   Reply With Quote
Old 2016-01-27, 16:54   #10
ET_
Banned
 
ET_'s Avatar
 
"Luigi"
Aug 2002
Team Italia

2×2,417 Posts
Default

Quote:
Originally Posted by firejuggler View Post
about to finish 10-20 T, will take 40-50.
Thanks!

Let me have the *.out and *.fact files when you finish, and I will update the board.

Luigi
ET_ is online now   Reply With Quote
Old 2016-01-27, 21:11   #11
firejuggler
 
firejuggler's Avatar
 
"Vincent"
Apr 2010
Over the rainbow

268410 Posts
Default

here they are.
It took a few more hours that I planned. Do i use the out file to start my new range?
Attached Files
File Type: zip MM49_10T_20T.zip (101.4 KB, 247 views)
firejuggler is online now   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
News from sub-project Deep Sieving Batalov Operazione Doppi Mersennes 59 2021-09-15 09:47
Parallel sieving with newpgen fivemack And now for something completely different 3 2017-05-16 17:55
Deep Hash diep Math 5 2012-10-05 17:44
Deep Sieving 10m Digit Candidates lavalamp Open Projects 53 2008-12-01 03:59
NASA's Deep Impact... ixfd64 Lounge 5 2005-07-06 13:46

All times are UTC. The time now is 21:41.


Sun Dec 5 21:41:47 UTC 2021 up 135 days, 16:10, 0 users, load averages: 1.16, 1.31, 1.37

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.