Re: How to interrupt a MT calculation?

[Jean-Luc Arnaud <jean-luc@filmsons.com>]

Le 01/07/2023 à 00:33, Jean-Luc Arnaud a écrit :

> Le
>       30/06/2023 à 22:04, Bill Allombert a écrit :
>       
> > On Fri, Jun 30, 2023 at 05:17:17PM +0200,
> >         Jean-Luc Arnaud wrote:
> >         
> > > With 2^470-1, first result in 1728 ms,
> > >           displayed after... 7min, 24,773 ms
> > >           
> > >           !!!
> > >           
> > >           
> > >           Is there another way to interrupt the script as soon as the
> > >           "error" occurs?
> > >           
> > >         
> >         ? Factor3_MT(2^470-1)
> >         
> >         %8 = ["e_USER",[[2179,2181]]]
> >         
> >         ? ##
> >         
> >            ***   last result: cpu time 13,086 ms, real time 2,183 ms.
> >         
> >         
> >         This is with pthread on Linux. This does not work with MPI.
> >         
> >         
> >         Cheers,
> >         
> >         Bill.
> >         
> >         
> >       
>       Thanks for spending time testing that.
>       
>       
>       I'd like to get the same result, but it's not the case.
>       
>       
>       Working on Mac Intel, OS 13.4, Pari/GP 2.16.0 pthread version.
>       
>       
>       I well test again and verify something, and let you know asap.
>       
>       
>       Sincerely,
>       
>       
>       Jean-Luc
>       
>       
>

I'm using a utility of mine, created with Xojo, as an interactive Shell with the Mac Terminal.

After running the Factor3_MT script directly in the Terminal, I got the same result as with my own interface:

? \r /Applications/PariGP/ScriptsGP/Factor3_MT.gp                                                                                               
? Factor3_MT(2^470-1)
[3, 1; 11, 1; 31, 1; 283, 1; 2351, 1; 4513, 1; 13264529, 1; 2391314881, 1; 72296287361, 1; 165768537521, 1; 328006342451, 1; 461797907949997211, 1; 235457374510092115086834691, 1; 73202300395158005845473537146974751, 1]
cpu time = 22min, 24,617 ms, real time = 7min, 24,582 ms.
%5 = ["e_USER", [[10322, 1730]]]
? 

So, it seems like error does not work the same way on Linux (and probably Windows) versus Mac OS, or is there a bug in the 2.16.0 Alpha version?

You have to know that breakloop = 1 (on) in my Default configuration.

Sincerely,

Jean-Luc