Trouble with running under Win7
 

Just curious, as I am rather new to GGNFS, what would make GGNFS run differently based on Win7 vs. XP? I got the program running well on XP, but for some reason the factMsieve.py script does not want to run well on my Win7 laptop (running Python 3.2)

For some reason it cannot recognize my command line input as two inputs, it dies trying to read sys.argv[1].

[code]

C:\Users\Matt\Desktop\Factoring\GGNFS\3,381->..\"FACTMSIEVE.84.PY" 3_381
-> ________________________________________________________________
-> | Running factmsieve.py, a Python driver for MSIEVE with GGNFS |
-> | sieving support. It is Copyright, 2010, Brian Gladman and is |
-> | a conversion of factmsieve.pl that is Copyright, 2004, Chris |
-> | Monico. Version 0.83 (Python 2.6 or later) 20th June 2011. |
-> |______________________________________________________________|
C:\Users\Matt\Desktop\Factoring\GGNFS\factmsieve.84.py
Traceback (most recent call last):
File "C:\Users\Matt\Desktop\Factoring\GGNFS\factmsieve.84.py", line 2098, in <
module>
print(sys.argv[1])
IndexError: list index out of range
[/code]

Is there a reason why it wouldn't want to read my filename input?

[For debugging purposes I turned off the if len(sys.argv) != 2 and len(sys.argv) != 4: loop and tried to print sys.args[0] and sys.args[1] to screen]

Thanks.

Matt

Have you tried leaving out the quote (") characters or putting them at different points in your input? My environment for developing the script is Windows 7 x64 and Python 3.2 so I would expect to see this failure if it was a script problem.

Brian

Yes, I did. I added the quotes later after it didn't want to run.

I did manage to get it running through yafu later, so I think the bug is in the script not the binaries.

Matt

It looks like a problem in your Win 7 setup. Have you tried to run a simple script such as:
[CODE]
import sys
for i in sys.argv:
print(i)
[/CODE]
with various arguments to see what it prints out?

[quote]
-> Running polynomial selection ...
Traceback (most recent call last):
File "H:\Documents and Settings\Vincent\Mes documents\aliqueit\factMsieve.py",
line 2179, in <module>
if not run_msieve_poly(fact_p, POLY_MIN_LC, POLY_MAX_LC):
File "H:\Documents and Settings\Vincent\Mes documents\aliqueit\factMsieve.py",
line 1013, in run_msieve_poly
ap += ' {0:d},{1:d} -t {2:d}'.format(minv, maxv, NUM_CPUS)
NameError: global name 'NUM_CPUS' is not defined
[/quote]So, what should I do?
edit : version 0.83, will try 0.84
edit again 0.84 is stated as 0.83
also, tried to replace NUM_CPUS with NUM_CORES
and got
[quote]
-> Running polynomial selection ...
-> msieve148_gpu -s ggnfs_394093013886992268984197489189086561196407152243307404
086588436399342782966345761762492573635888113921\test.dat -l ggnfs_3940930138869
92268984197489189086561196407152243307404086588436399342782966345761762492573635
888113921\test.log -i ggnfs_3940930138869922689841974891890865611964071522433074
04086588436399342782966345761762492573635888113921\test.ini -nf ggnfs_3940930138
86992268984197489189086561196407152243307404086588436399342782966345761762492573
635888113921\test.fb -g 0 -v -np 0,4000 -t 2


Msieve v. 1.48
Sun Jul 17 00:09:37 2011
random seeds: a0d36f60 f7e7a2b1
factoring 3940930138869922689841974891890865611964071522433074040865884363993427
82966345761762492573635888113921 (102 digits)
searching for 15-digit factors
commencing number field sieve (102-digit input)
commencing number field sieve polynomial selection
lower/upper bounds must both be specified
error generating or reading NFS polynomials
elapsed time 00:00:01
can't read test.msieve.dat.p
WARNING: gnfs failed to find a factor. This really shouldn't happen.
I'll just run ecm till the end of time or a factor turns up...
Let's hope you don't run out of disk space before either of those.
[/quote]
back to 'old' factmsieve

The problem is the arguments to -np; neither is allowed to be zero. Maybe it will work if you make it '-np 1,4000'

And, as you correctly guessed, NUM_CPUS should be NUM_CORES.

changed the line
POLY_MIN_LC to 1, poly select seem to work

another problem appeared after the poly selection
[code]
-> Working with NAME = test
Traceback (most recent call last):
File "H:\Documents and Settings\Vincent\Mes documents\aliqueit\factMsieve.py"
line 2197, in <module>
read_parameters(fact_p, poly_p, lats_p)
File "H:\Documents and Settings\Vincent\Mes documents\aliqueit\factMsieve.py"
line 1376, in read_parameters
fact_p, pols_p, lats_p, clas_p)
File "H:\Documents and Settings\Vincent\Mes documents\aliqueit\factMsieve.py"
line 563, in load_default_parameters
clas_p['cl_a'] = int(tu[o + 18])
[/code]
what can i do?

I have only problems with bigger snfs-numbers.
Is there a limit in factmsieve.py or is this a limit from def-nm-params?

[QUOTE=Andi_HB;266801]I have only problems with bigger snfs-numbers.
Is there a limit in factmsieve.py or is this a limit from def-nm-params?[/QUOTE]

These are failures in reading the *.poly file and are likely to be caused by sommething unexpected in the input file. In order to debug them I would need to have the poly file that caused the failure.

Brian