Thomas D. Dean on Thu, 26 Oct 2023 01:56:13 +0200


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

Re: segfault


On 10/25/23 14:23, Bill Allombert wrote:
On Tue, Oct 24, 2023 at 04:31:33PM -0700, Thomas D. Dean wrote:
On 10/24/23 16:26, Thomas D. Dean wrote:
Ubuntu 22.04
64G RAM

  > gp
Reading GPRC: /home/tomdean/.gprc
GPRC Done.

          GP/PARI CALCULATOR Version 2.16.1 (development 28775-3c253ce7d5)
            amd64 running linux (x86-64/GMP-6.2.1 kernel) 64-bit version
      compiled: Oct 24 2023, gcc version 11.4.0 (Ubuntu
11.4.0-1ubuntu1~22.04)
                              threading engine: single
                   (readline v8.1 enabled, extended help enabled)

                       Copyright (C) 2000-2023 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 ?18 for how to get moral (and possibly technical) support.

parisizemax = 16000000000, primelimit = 10000000
? quadhilbert(-3299)
    *** quadhilbert: Warning: increasing stack size to 16000000000.
    ***   at top-level: quadhilbert(-3299)
    ***                 ^------------------
    *** quadhilbert: bug in PARI/GP (Segmentation Fault), please report.
    ***   Break loop: type 'break' to go back to GP prompt

This is very strange, I cannot reproduce this.

Cheers,
Bill.


I normally start gp in an xterm via an alias. My git code is in ./pari

> cp -R ./pari reproduce-error
> cd reproduce-error
> make clean
> ./Configure --with-readline --with-gmp
...
> make -j128 all
...
> sudo make install
> alias gp
alias gp='/usr/bin/xterm -T Pari -j -sk -sl 1000 -bc -bg white -fg black -geometry 80x52 -fa Courier -fs 12 -e /usr/local/bin/gp &'

In a dozen, or so, tries, this failed the first time and worked OK the rest. I started gp via the alias, then with /usr/local/bin/gp, then via the alias for the remaining tries.

If I start gp with
> /usr/local/bin/gp
It does not fail.

Maybe I have some library that is strange? Or maybe something was hung in a Linux buffer????

Let's just ignore this, for now. I will post again if I reproduce it.

Tom Dean