20150116, 00:18  #1 
Apr 2011
1000000_{2} Posts 
YAFU crashing after rroot finding. Why?
I'm running YAFU v.1.34.5.
Why does YAFU crash after finding rroots? As you can see from the example below everything is progressing just fine until it finishes the rroot finding. Then with no warning YAFU crashes and shuts down. Many hours have been wasted trying to factor numbers only to have this happen. Code:
01/15/15 11:36:33 v1.34.5 @ TSH0013514, 01/15/15 11:36:33 v1.34.5 @ TSH0013514, **************************** 01/15/15 11:36:33 v1.34.5 @ TSH0013514, Starting factorization of 3731957003453168760965976133865610406934014621030869599818519323129284225578489492837151529686636095351345612655371 01/15/15 11:36:33 v1.34.5 @ TSH0013514, using pretesting plan: normal 01/15/15 11:36:33 v1.34.5 @ TSH0013514, no tune info: using qs/gnfs crossover of 95 digits 01/15/15 11:36:33 v1.34.5 @ TSH0013514, **************************** 01/15/15 11:36:33 v1.34.5 @ TSH0013514, rho: x^2 + 3, starting 1000 iterations on C115 01/15/15 11:36:33 v1.34.5 @ TSH0013514, rho: x^2 + 2, starting 1000 iterations on C115 01/15/15 11:36:33 v1.34.5 @ TSH0013514, rho: x^2 + 1, starting 1000 iterations on C115 01/15/15 11:36:33 v1.34.5 @ TSH0013514, pm1: starting B1 = 150K, B2 = gmpecm default on C115 01/15/15 11:36:33 v1.34.5 @ TSH0013514, current ECM pretesting depth: 0.00 01/15/15 11:36:33 v1.34.5 @ TSH0013514, scheduled 30 curves at B1=2000 toward target pretesting depth of 35.38 01/15/15 11:36:33 v1.34.5 @ TSH0013514, prp13 = 1772579951411 (curve 4 stg2 B1=2000 sigma=413122032 thread=0) 01/15/15 11:36:33 v1.34.5 @ TSH0013514, Finished 4 curves using Lenstra ECM method on C115 input, B1=2K, B2=gmpecm default 01/15/15 11:36:33 v1.34.5 @ TSH0013514, current ECM pretesting depth: 0.00 01/15/15 11:36:33 v1.34.5 @ TSH0013514, scheduled 26 curves at B1=2000 toward target pretesting depth of 31.69 01/15/15 11:36:34 v1.34.5 @ TSH0013514, Finished 26 curves using Lenstra ECM method on C103 input, B1=2K, B2=gmpecm default 01/15/15 11:36:34 v1.34.5 @ TSH0013514, current ECM pretesting depth: 15.18 01/15/15 11:36:34 v1.34.5 @ TSH0013514, scheduled 74 curves at B1=11000 toward target pretesting depth of 31.69 01/15/15 11:36:43 v1.34.5 @ TSH0013514, Finished 74 curves using Lenstra ECM method on C103 input, B1=11K, B2=gmpecm default 01/15/15 11:36:43 v1.34.5 @ TSH0013514, current ECM pretesting depth: 20.24 01/15/15 11:36:43 v1.34.5 @ TSH0013514, scheduled 214 curves at B1=50000 toward target pretesting depth of 31.69 01/15/15 11:38:26 v1.34.5 @ TSH0013514, Finished 214 curves using Lenstra ECM method on C103 input, B1=50K, B2=gmpecm default 01/15/15 11:38:26 v1.34.5 @ TSH0013514, pm1: starting B1 = 3750K, B2 = gmpecm default on C103 01/15/15 11:38:30 v1.34.5 @ TSH0013514, current ECM pretesting depth: 25.33 01/15/15 11:38:30 v1.34.5 @ TSH0013514, scheduled 430 curves at B1=250000 toward target pretesting depth of 31.69 01/15/15 11:54:09 v1.34.5 @ TSH0013514, Finished 430 curves using Lenstra ECM method on C103 input, B1=250K, B2=gmpecm default 01/15/15 11:54:09 v1.34.5 @ TSH0013514, pm1: starting B1 = 15M, B2 = gmpecm default on C103 01/15/15 11:54:24 v1.34.5 @ TSH0013514, current ECM pretesting depth: 30.45 01/15/15 11:54:24 v1.34.5 @ TSH0013514, scheduled 225 curves at B1=1000000 toward target pretesting depth of 31.69 01/15/15 12:27:46 v1.34.5 @ TSH0013514, Finished 225 curves using Lenstra ECM method on C103 input, B1=1M, B2=gmpecm default 01/15/15 12:27:46 v1.34.5 @ TSH0013514, final ECM pretested depth: 31.70 01/15/15 12:27:46 v1.34.5 @ TSH0013514, scheduler: switching to sieve method 01/15/15 12:27:46 v1.34.5 @ TSH0013514, nfs: commencing nfs on c103: 2105381481090585217974038570900140207156194443434542537916205022024264090035686605727881401910585436361 01/15/15 12:27:47 v1.34.5 @ TSH0013514, nfs: commencing poly selection with 2 threads 01/15/15 12:27:47 v1.34.5 @ TSH0013514, nfs: setting deadline of 771 seconds 01/15/15 12:58:29 v1.34.5 @ TSH0013514, nfs: completed 2 ranges of size 250 in 1842.7031 seconds 01/15/15 12:58:29 v1.34.5 @ TSH0013514, nfs: best poly = # norm 4.686974e014 alpha 4.908732 e 7.922e009 rroots 2 01/15/15 12:58:29 v1.34.5 @ TSH0013514, nfs: commencing lattice sieving with 2 threads Code:
n: 2105381481090585217974038570900140207156194443434542537916205022024264090035686605727881401910585436361 skew: 5194316.97 c0: 55515216155866739386413461856 c1: 23040724899216711803652 c2: 21635660256563459 c3: 3007150928 c4: 408 Y0: 8475472436065327742133449 Y1: 39645530675737 rlim: 2220000 alim: 2220000 lpbr: 26 lpba: 26 mfbr: 52 mfba: 52 rlambda: 2.5 alambda: 2.5 
20150116, 06:36  #2 
"Curtis"
Feb 2005
Riverside, CA
5484_{10} Posts 
Does yafu find the sievers? I mean, do you have the nfs sieve applications in the right spot, with the path correctly pointed at them? The part that worked is msieve's job, while the lattice sieving is done by lasieve4I[1116]e.
If yafu keeps crashing at exactly that point, I think either the sievers aren't being found (usually a path problem), or you downloaded sievers that aren't compatible with your OS/architecture (linux vs win, win64 vs win32, etc). 
20150116, 13:17  #3  
"Ben"
Feb 2007
111001011010_{2} Posts 
Quote:
OP  assuming you haven't deleted anything, if you find binaries that work then nothing is lost. Just restart with a R flag and it should pick up with the nfs.job file and start sieving. 

20150116, 23:30  #4 
Apr 2011
2^{6} Posts 
I don't see the files you mention. I d'led yafu1.34.zip from sourceforge and there were
only 7 files in the zip: CHANGES docfile.txt README yafu yafuWin32.exe yafux64.exe yafu.ini Where are these other files (lsieve)? I didn't see them on the sourceforge page. Unless I missed something somewhere :( Last fiddled with by mnh001 on 20150116 at 23:32 
20150117, 16:25  #5 
Mar 2007
Germany
2^{3}·3·11 Posts 
You can try this lasieve files.
http://www.mersenneforum.org/showpos...2&postcount=17 Regards Andi_HB 
20150117, 19:47  #6 
Apr 2011
2^{6} Posts 
Thanks Andi_HB.
Ok I dl'ed ggnfssvn413win64core2.zip and extracted it to C:\YAFU1.34\ggnfssvn413win64core2\ Then I changed the yafu.ini file to this Code:
B1pm1=100000 B1pp1=20000 B1ecm=11000 rhomax=1000 threads=3 pretest_ratio=0.25 %ggnfs_dir=..\ggnfssvn413win64core2\ ggnfs_dir=../ggnfssvn413win64core2/ %ecm_path=..\gmpecm\bin\x64\Release\ecm.exe %ecm_path=../ecm/current/ecm tune_info= Intel(R) Xeon(R) CPU E54650 0 @ 2.70GHz,LINUX64,1.73786e05,0.200412,0.400046,0.0987873,98.8355,2699.98 Last fiddled with by mnh001 on 20150117 at 20:00 
20150117, 21:31  #7 
Basketry That Evening!
"Bunslow the Bold"
Jun 2011
40<A<43 89<O<88
3·29·83 Posts 
Try making ggnfs_dir the absolute path, i.e.
Code:
ggnfs_dir=C:\YAFU1.34\ggnfssvn413win64core2\ Last fiddled with by Dubslow on 20150117 at 21:32 
20150117, 22:37  #8 
Apr 2011
2^{6} Posts 
Ok, so far that seems to work. I've got it running on a previous number. It picked up where it left off before crashing. Since it hasn't
crashed yet I'll assume that fixed it. Thanks for the help. 
20150122, 23:50  #9 
Apr 2011
2^{6} Posts 
32bit?
Is there lasieve routines for 32bit? I actually have access to more of those than 64bit.
Last fiddled with by mnh001 on 20150122 at 23:51 
20150123, 05:18  #10 
"Curtis"
Feb 2005
Riverside, CA
5484_{10} Posts 
Yes. Check Jeff Gilchrist's factoring site: gilchrist.ca/jeff/factoring/
I think they're there. 
20150123, 17:55  #11 
Apr 2011
2^{6} Posts 
Ok, I'll check that out. Thanks.

Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Running YAFU via Aliqueit doesn't find yafu.ini  EdH  YAFU  8  20180314 17:22 
Yafu instantly crashing after first instruction  luisda2994  YAFU  1  20161209 00:30 
LLR Crashing  mdettweiler  Riesel Prime Search  14  20070930 18:29 
V24.12 Release 1 Crashing on P4 M  njcroquet1  Software  8  20050624 14:40 
Prime crashing  russ  Software  3  20020829 00:08 