mersenneforum.org > EdH How I install Aliqueit onto my Ubuntu Machines
 User Name Remember Me? Password
 Register FAQ Search Today's Posts Mark Forums Read

 2020-07-03, 19:26 #12 Mr. Odd   Mar 2010 2×29 Posts Nope, still bombs at #include in aliqueit.cc.
2020-07-03, 22:00   #13
EdH

"Ed Hall"
Dec 2009

402110 Posts

Quote:
 Originally Posted by Mr. Odd Nope, still bombs at #include in aliqueit.cc.
Well, let's look at a system here that works with GMP-6.1.2. I have this structure:

<$HOME>/Math/gmp/ - - - this is the folder in which I compiled GMP <$HOME/>Math/Aliqueit/src/aliqueit/ - - - this is created by unzipping in the Aliqueit folder (Note the A/a for folders.)

The $HOME/Math/Aliqueit/src/aliqueit/ folder has: Code: aliqueit.cc aliqueit.h aliqueit.rc aliqueit.vcproj cfg.cc cfg.h Makefile misc.cc misc.h resource.h My Makefile consists of the following: Code: # # aliqueit Makefile for linux # # you'll need to change this if your gmp headers aren't installed globally: GMP_DIR = ../../../gmp CC = g++ CFLAGS = -O3 -I$(GMP_DIR)

# you'll need to change this if your gmp headers aren't installed globally:
LIBS = -L../../../gmp -lgmp

OBJ = aliqueit.o cfg.o misc.o

aliqueit: $(OBJ)$(CC) $(CFLAGS)$(OBJ) $(LIBS) -o aliqueit %.o: %.cc$(CC) $(CFLAGS) -c -o$@ $< Aliqueit compiles fine with this Makefile. I can do just about anything, except remove the -lgmp (which is, of course, needed) in the "LIBS =" line. However, ANY change to the "GMP_DIR =" line causes the following: Code: g++ -O3 -I../../gmp -c -o aliqueit.o aliqueit.cc In file included from aliqueit.cc:21:0: aliqueit.h:11:19: fatal error: gmpxx.h: No such file or directory #include <gmpxx.h> ^ compilation terminated. Makefile:21: recipe for target 'aliqueit.o' failed make: *** [aliqueit.o] Error 1 What do you show if you issue the command: Code: locate gmpxx.h I show: Code: . . . <$HOME>/Math/gmp/gmpxx.h
Maybe you could take the folder portion of the response and plug it in as the "GMP_DIR =" value:
Code:
GMP_DIR = <\$HOME>/Math/gmp
I tried that here and it compiled.

 2021-04-12, 01:19 #14 Mr. Odd   Mar 2010 2·29 Posts What are the steps and command line syntax to use Aliqueit on a FactorDB sequence (as opposed to starting one fresh)?
2021-04-12, 02:27   #15
EdH

"Ed Hall"
Dec 2009

4,021 Posts

Quote:
 Originally Posted by Mr. Odd What are the steps and command line syntax to use Aliqueit on a FactorDB sequence (as opposed to starting one fresh)?
Let's say you want to run 4788 with Aliqueit:

First, d/l the 4788.elf file from the factordb page for 4788. It's one of the options between the top section and the set of lines. The file will be named 4788.elf when it is downloaded.

Next, rename 4788.elf by adding the prefix alq_ to the 4788.elf to get alq_4788.elf.*

Then, you should be good to go with ./aliqueit 4788 or ./alqueit <options> 4788. (Or on Windows Aliqueit.exe . . .)

* The default for .elf files is to expect "alq_" as a prefix, but it can be changed in the .ini file.

 2021-04-12, 08:04 #16 LaurV Romulan Interpreter     "name field" Jun 2011 Thailand 9,787 Posts Kind of simplistic. Aliqueit has a very rudimentary factoring and some advanced sigma calculations and checking, but relies heavily on external tools to factor the sequences. The so called "tutorial" should include how to set it to work with yafu (the easiest way) or other tools. To work with yafu, you have to have a working copy of yafu (and afferent tools needed for yafu, like ggnfs and ecm packages), point aliqueit to it in the aliqueit.ini file (don't confuse that with yafu.ini, usually I keep all of them in the same forder), and then launch aliqueit with "-y" switch in command line. (methink that's what it was requested, but i/m in kinda hurry now, sorry, somebody may need to detail this process...) Last fiddled with by LaurV on 2021-04-12 at 08:06
2021-04-12, 13:12   #17
EdH

"Ed Hall"
Dec 2009

FB516 Posts

Quote:
 Originally Posted by LaurV Kind of simplistic. Aliqueit has a very rudimentary factoring and some advanced sigma calculations and checking, but relies heavily on external tools to factor the sequences. The so called "tutorial" should include how to set it to work with yafu (the easiest way) or other tools. To work with yafu, you have to have a working copy of yafu (and afferent tools needed for yafu, like ggnfs and ecm packages), point aliqueit to it in the aliqueit.ini file (don't confuse that with yafu.ini, usually I keep all of them in the same forder), and then launch aliqueit with "-y" switch in command line. (methink that's what it was requested, but i/m in kinda hurry now, sorry, somebody may need to detail this process...)
I may have misunderstood, but I think Mr. Odd already has a working setup, but the obscure use of "alq_" on the front of an already existing .elf file had eluded him. He can let us know if my previous reply is insufficient and I will try more.

As to the instructions, doesn't the first post already supply everything you suggest (GMP, GMP-ECM, Msieve, YAFU, ggnfs) all the way up to:
Quote:
 At this point, everything should be ready to simply use the following to run Aliqueit using YAFU for all external factoring processes: Code: Code: ./aliqueit -y
Edit: I've added a note about the need for "alq_" as a prefix, to the first post.

Last fiddled with by EdH on 2021-04-12 at 13:20

 2021-04-12, 18:09 #18 Mr. Odd   Mar 2010 2×29 Posts Thanks, Ed. Yes, first I needed to rename the file. It has to be in the same folder as the Aliqueit executable, right? I'm running on Linux. Do I need to set any options to start with the last open entry?
2021-04-12, 18:39   #19
EdH

"Ed Hall"
Dec 2009

4,021 Posts

Quote:
 Originally Posted by Mr. Odd Thanks, Ed. Yes, first I needed to rename the file. It has to be in the same folder as the Aliqueit executable, right? I'm running on Linux. Do I need to set any options to start with the last open entry?
Yes, it needs to be in the Aliqueit folder. No, on any options. If Aliqueit finds and reads the file, it will do a check that all current lines are valid, and then start right in on completing the last one. If the validity fails, it will stop and inform you. It may complain about the last line being incomplete. This is not a problem - factordb sends the factors it knows.

Last fiddled with by EdH on 2021-04-12 at 18:40

 2021-09-23, 22:34 #20 Plutie   "Evan" Dec 2020 Montreal 71 Posts I'm having trouble running CADO as a replacement for ggnfs. It doesn't seem to be finding the directory with the number to factor. Code: mkdir: cannot create directory ‘ggnfs_67216810256844675480477157025988578680922317601701866734537817412464452765665954902870790052675237179’: File exists cannot open input file 'worktodo.ini' cat: ggnfs_67216810256844675480477157025988578680922317601701866734537817412464452765665954902870790052675237179/test.n: No such file or directory rm: cannot remove '/tmp/alicado': No such file or directory Traceback (most recent call last): File "./cado-nfs.py", line 144, in parameters, db = toplevel_params.get_cooked_parameters() File "/home/uwu/Math/cado-nfs/./scripts/cadofactor/toplevel.py", line 1175, in get_cooked_parameters self.set_N_paramfile_workdir() File "/home/uwu/Math/cado-nfs/./scripts/cadofactor/toplevel.py", line 658, in set_N_paramfile_workdir raise ValueError("if N is not given, a parameter file is required") ValueError: if N is not given, a parameter file is required cat: /tmp/alicado/alicado.log: No such file or directory /home/uwu/Math/aliqueit/cado.sh: line 19: cd: /home/uwu/Math/aliqueit/ggnfs_: No such file or directory 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. Help would be greatly appreciated.
 2021-09-24, 03:46 #21 EdH     "Ed Hall" Dec 2009 Adirondack Mtns 402110 Posts I can't really tell enough from the post, but it looks like alitest.sh isn't being called and Msieve is trying to run. What line do you have for: Code: ggnfs_cmd = in the aliqueit.ini file?
 2021-09-24, 03:55 #22 Plutie   "Evan" Dec 2020 Montreal 71 Posts Code: ggnfs_cmd = bash /home/uwu/Math/aliqueit/cado.sh Here it is. I renamed the script from alitest.sh to cado.sh.

 Similar Threads Thread Thread Starter Forum Replies Last Post EdH EdH 3 2019-06-24 03:42 EdH EdH 12 2019-04-16 09:28 EdH EdH 0 2019-04-04 22:33 EdH EdH 0 2018-02-22 03:31 EdH EdH 0 2018-02-21 23:48

All times are UTC. The time now is 02:35.

Mon Oct 25 02:35:52 UTC 2021 up 93 days, 21:04, 0 users, load averages: 0.80, 1.12, 1.31