20130219, 16:31  #34 
"Nancy"
Aug 2002
Alexandria
2,467 Posts 
A release candidate for 6.4.4 is available at http://www.loria.fr/~kruppaal/ecm6.4.4.tar.gz
If you like, please give it a try and report success or failure. Thanks! 
20130219, 18:12  #35 
Oct 2010
191 Posts 
Thanks. 'make check' (gmpecm 6.4.4 / gmp 5.1.1) reports no errors.

20130219, 18:58  #36 
"Nancy"
Aug 2002
Alexandria
100110100011_{2} Posts 
I have just updated the tarball again. The new tarball includes a patch for PowerPC64 assembly code (under powerpc64/) when a dynamic libaray is built. No other code was changed.

20130220, 03:12  #37 
"6800 descendent"
Feb 2005
Colorado
1010101010_{2} Posts 
Compiles, checks, and works fine for me on a 32bit P4, Linux, and GMP 5.1.1.
Thanks! 
20130220, 20:33  #38  
Oct 2010
191 Posts 
Quote:
Code:
[ralf@quadriga ecm6.4.4]$ wine64 ecm.exe sigma 140642477 3e6 < /home/ralf/Faktorisierung/ECMTestdaten/GMPECMTestcaseGW_7_583_C4633MSigma140642477 GMPECM 6.4.4 [configured with GMP 5.1.1, enableasmredc] [ECM] Input number is 9419637902294591003764706452592267326726375972060771118922019946663693693759112212705454606413445474942262035668813026111589665589874842002549018009614917617850511946725729554590232906577559784190618314709182818510846516594338093150279244690919895977755275618920780071836272126626961841307419703986612506790743382751292313153100490267482670046132843485368016968530438763568389772557863007216523390523643415365284374260407190171241424198205553658590574339011923219 (463 digits) Using B1=3000000, B2=5706890290, polynomial Dickson(6), sigma=140642477 Step 1 took 49010ms ********** Factor found in step 1: 42321464520869757241201743459403331 Found probable prime factor of 35 digits: 42321464520869757241201743459403331 Composite cofactor 222573533523385878832599590459915258046775784943734419539381430321379898782949341682812266134738709057771160360470984257143995778652743149018805829131494371987980298855763262475628671509742419862342168285190063614897010489539602346091397983156734239156958618013872470635038096122817820916014065416816906367560357164340632712253032080746508364371439786893163857329678155871342808426289249050108727758398954187629188336892442360049 has 429 digits Last fiddled with by Ralf Recker on 20130220 at 20:55 Reason: ./test.ecm ./test.pm1 and ./test.pp1 worked for a win64 build with gmp 5.1.1 

20130228, 17:16  #39 
May 2005
England, UK
174_{8} Posts 
I can confirm that ecm6.4.4 compiles with gmp5.1.1 and passes 'make check' on both Windows XP and Ubuntu 10.04.4, 32 bit.

20130228, 22:02  #40 
"Ed Hall"
Dec 2009
Adirondack Mtns
4112_{10} Posts 
This is getting ridiculous! I just tried to compile 6.4.4 on one of my 32bit AntiX machines and have hit the following troubles:
First, I have never had gmp 5.0.5 on this machine. I started with gmp 5.1.1. Yet, I get: Code:
checking if gmp.h version and libgmp version are the same... (5.1.1/5.0.5) no configure: error: 'gmp.h' and 'libgmp' have different versions, you have to reinstall GMP properly. Code:
libgmp: /usr/local/lib/libgmp.la /usr/local/lib/libgmp.a /usr/local/lib/libgmp.so Code:
... checking whether we can link against GMP... yes checking if gmp.h version and libgmp version are the same... (5.1.1/5.1.1) yes checking for __gmpn_add_nc... yes ... Code:
GMPECM 6.4.4 [configured with GMP 5.0.5] [P+1] Input number is 6054018161*10^400+417727253109 (410 digits) Using B1=2000, B2=2352760, polynomial x^1, x0=4 ./ecm: symbol lookup error: ./ecm: undefined symbol: __gmpn_add_nc ############### ERROR ############### Expected return code 14 but got 127 make: *** [check] Error 1 Code:
GMPECM 6.4.4 [configured with GMP 5.0.5] [ECM] Input number is 212252637915375215854013140804296246361 (39 digits) Using B1=63421, B2=18227952011822795212, polynomial x^1, sigma=781683988 Step 1 took 324ms Step 2 took 4ms ############### ERROR ############### Expected return code 8 but got 0 make: *** [check] Error 1 Thanks to all... Edit: Apparently, there is a libgmp10 from the repository that is tied to 48 packages. Could that be causing this? Last fiddled with by EdH on 20130228 at 22:08 Reason: addition of more info 
20130301, 00:14  #41 
"Nancy"
Aug 2002
Alexandria
2,467 Posts 
What happens if you use "./configure withgmp=/usr/local/" instead of supplying the LIBS variable directly?
Last fiddled with by akruppa on 20130301 at 00:14 Reason: ./ 
20130301, 02:53  #42  
"Ed Hall"
Dec 2009
Adirondack Mtns
1010_{16} Posts 
Quote:
Code:
GMPECM 6.4.4 [configured with GMP 5.1.1] [ECM] Input number is 10000000000000000000000000000000000000121 (41 digits) Using B1=4372759, B2=8561918830, polynomial Dickson(6), sigma=291 Step 1 took 23825ms ********** Factor found in step 1: 10000000000000000000000000000000000000121 Found input number N All ECM tests are ok. 

20130301, 09:49  #43 
"Nancy"
Aug 2002
Alexandria
2,467 Posts 
Glad to hear it works! I guess we should print the suggestion of trying withgmp= along with the GMP version mismatch error message. That error happens very frequently since most distros come with a GMP library, and many GMPECM users then also install their selfcompiled GMP with optimization for their particular CPU. The configure script usually finds the headers of the distroGMP, and the library of the selfcompiled one; if those differ in version, the version mismatch error happens.

20130301, 14:27  #44  
"Ed Hall"
Dec 2009
Adirondack Mtns
2^{4}·257 Posts 
Quote:
Thanks for all the work! 

Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Error Prone Machines  PageFault  Data  17  20120410 01:40 
Scalability of Glucas on large machines: A future //ed prime 95 ?  T.Rex  Software  29  20061213 21:40 
Start and Stop Prime 95 on Large Groups of Windows XP Machines  MarcGetty  Software  3  20060307 07:54 
Putting prime 95 on a large number of machines  moo  Software  10  20041215 13:25 
Team_Prime_Rib errorprone machines  GP2  Data  10  20031005 18:34 