jacques G on Thu, 18 Apr 2013 06:12:32 +0200

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

RE: updated windows installer for PARI

> I tested on some windows server 2008 and CTRL-C worked fine.

1. On my VISTA laptop, CTRL-C just rings the bell, while other
readline controls seem to work fine. A cheap solution is 
decreasing the stack size, so a runaway computation
(such as an impossible task) will stop when memory is low, sooner. 

2. For now, on my VISTA, GP 2.4.1 works nicely, CTRL-C and (allmost) all :
Reading GPRC: .gprc ...Done.
                   GP/PARI CALCULATOR Version 2.4.1 (alpha)
           i686 running cygwin (ix86/GMP-4.2.1 kernel) 32-bit version
 compiled: Mar 28 2007, gcc-3.4.4 (cygming special, gdc 0.12, using dmd 0.125)
                 (readline v5.2 enabled, extended help enabled)
                     Copyright (C) 2000-2006 The PARI Group

parisize = 2000000, primelimit = 10000
#   ***   user interrupt after 62 ms.

3. A significant problem  with PARI/GP 2.4 on my VISTA is ploth which uses fork:
    "One of our biggest all-time problem in Cygwin is the implementation of the 
       POSIX fork call." Corinna Vinshen, Red Hat, March 20, 2012.
But even then, on this quiet VISTA system,  ploth/2.4  works more than 
half the time (:-) while  psploth  always works as designed.

4. So I am now thinking about cross-compilling 2.6 with Cygwin the way 2.4 was
in 2007, using Linux Slackware 14 running on a USB disk (where PARI 2.6 for Linux compiles and works immediately without any problem in 2013, thanks to all !). 
That would give me under VISTA all the nice improvements and new features 
of the last 6 years (:-). 

5. As for ploth, here is a test I just did on Python (on the same VISTA laptop) :

Python 2.7.2 (default, Jun 12 2011, 15:08:59) [MSC v.1500 32 bit (Intel)] on win32
>>> import os
>>> g=os.popen("c:/pub/gnuplot/gnuplot.exe","w")
>>> g.write("set terminal windows\n")
>>> g.write('set title "a"'+'\n')
>>> g.write("plot x\n")
A plot window appears, and I can refocus to the Python window and type
>>> g.close()

6. I hope this helps, but please do not spend too much time fixing VISTA
problems ! It would be much more worthwhile, I think, to try and harness the
power and speed available from multi-core CPUs that laptops have nowadays.
Just imagine this (stolen from http://www.numberworld.org/y-cruncher/ ;)

"The first scalable multi-threaded Pari-benchmark for multi-core systems.."  .

Jacques Gélinas