gp: (-2/x)/(-1/x) + O(%) = seg fault

[Michael Somos <somos@grail.cba.csuohio.edu>]

pari-dev,

I think the initial bug is that (-2/x)/(-1/x) was not reduced to integer 2.
This is of minor consequence, however, it can trigger a seg fault in 'O()'
which does not have enough protective code. It may be argued that fixing
the initial bug would obviate need for more protective code. Perhaps. Note
that I have found other examples of incomplete reduction that could have
been fixed. This is just one simple example.

==============================================================================
Reading GPRC: /u/home/somos/.gprc ...Done.

          GP/PARI CALCULATOR Version 2.2.4 (development CHANGES-1.491)
                 UltraSparc (MicroSparc kernel) 32-bit version
              (readline v2.2 enabled, extended help not available)

                       Copyright (C) 2002 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 ?12 for how to get moral (and possibly technical) support.

   realprecision = 28 significant digits
   seriesprecision = 16 significant terms
   format = g0.28

parisize = 4000000, primelimit = 500000
? (-2/x)/(-1/x)
%1 = -2/-1
? O(%)
  ***   bug in GP (Segmentation Fault), please report
? \x
[&=00496764] RFRAC(lg=3,CLONE):1b000003  0049677c  00496770  
  num = [&=0049677c] INT(lg=3):02000003  (-,lgef=3):c0000003  00000002  
  den = [&=00496770] INT(lg=3):02000003  (-,lgef=3):c0000003  00000001  

? quit
Goodbye!
==============================================================================

Shalom, Michael