On Fri, Jul 19, 2019 at 8:14 PM Vincent Delecroix
<20100.delecr...@gmail.com> wrote:
>
> And I believe that this is also fine
>
> sage: pari(10r ** 20r)

yes, this works.
While

sage -t --warn-long 83.4 src/sage/libs/pari/convert_gmp.pyx
    Killed due to bus error
**********************************************************************
Tests run before process (pid=5771) failed:
sage: pari(42)       # indirect doctest ## line 39 ##
42
sage: foo = pari(2^(32*1024))  # Create large integer to put PARI
stack in known state ## line 47 ##
sage: a5 = pari(5) ## line 48 ##
sage: foo = pari(0xDEADBEEF * (2^(32*1024)-1)//(2^32 - 1))  # Dirty
PARI stack ## line 49 ##
sage: b5 = pari(5) ## line 50 ##
sage: a5.__hash__() == b5.__hash__() ## line 51 ##
------------------------------------------------------------------------
Attaching gdb to process id 5771.
Cannot find gdb installed
GDB is not installed.
Install gdb for enhanced tracebacks.
------------------------------------------------------------------------
Unhandled SIGBUS: A bus error occurred.


>
> If so there is most probably something wrong with
> _new_GEN_from_mpz_t from sage.libs.pari.convert_gmp
>
> Le 19/07/2019 à 21:12, Dima Pasechnik a écrit :
> > On Fri, Jul 19, 2019 at 8:08 PM Vincent Delecroix
> > <20100.delecr...@gmail.com> wrote:
> >>
> >> What about
> >>
> >> sage: int(10^20)
> >
> > sage: int(10^20)
> > 100000000000000000000L
> >
> >
> >>
> >> Le 19/07/2019 à 19:14, Dima Pasechnik a écrit :
> >>> I can build cypari with SAGE_CHECK=yes, everything passes. But
> >>> pari(10^20) is a problem.
> >>> Could anyone point me possible causes?
> >>>
> >>> sage: pari(10^19)
> >>> 10000000000000000000
> >>> sage: pari(10^20)
> >>> 7766279631452241920
> >>>
> >>> sage:
> >>> Exiting Sage (CPU time 0m0.15s, Wall time 0m28.21s).
> >>>
> >>> PARI/GP does work on this example, though:
> >>>
> >>> $ ./sage -gp
> >>>                                               GP/PARI CALCULATOR Version
> >>> 2.11.1 (released)
> >>>                                           amd64 running freebsd (x86-64
> >>> kernel) 64-bit version
> >>>                 compiled: Jul 18 2019, FreeBSD clang version 6.0.1
> >>> (tags/RELEASE_601/final 335540) (based on LLVM 6.0.1)
> >>>                                                         threading engine: 
> >>> single
> >>>                                             (readline not compiled in,
> >>> extended help enabled)
> >>>
> >>>                                                  Copyright (C) 2000-2018
> >>> The PARI Group
> >>>
> >>> PARI/GP is free software, covered by the GNU General Public License,
> >>> and comes WITHOUT ANY WARRANTY WHATSOEVER.
> >>>
> >>> Type ? for help, \q to quit.
> >>> Type ?17 for how to get moral (and possibly technical) support.
> >>>
> >>> parisize = 8000000, primelimit = 500000
> >>> ? 10^19
> >>> %1 = 10000000000000000000
> >>> ? 10^20
> >>> %2 = 100000000000000000000
> >>>
> >>
> >> --
> >> You received this message because you are subscribed to the Google Groups 
> >> "sage-devel" group.
> >> To unsubscribe from this group and stop receiving emails from it, send an 
> >> email to sage-devel+unsubscr...@googlegroups.com.
> >> To view this discussion on the web visit 
> >> https://groups.google.com/d/msgid/sage-devel/02634249-b980-e32f-5420-53004498f237%40gmail.com.
> >
>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/sage-devel/169f4cdc-92c0-da69-9add-e51614c8ce94%40gmail.com.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq1xuezfcqGM9-AKF%3D0dF8ThFuwHP-%3DDJmOEY_0Z2fXERw%40mail.gmail.com.

Reply via email to