GMP upgrade to 4.2.4 leads to errors in PARI/GP

[Torbjorn Granlund]

Vincent Lefevre <vincent at vinc17.org> writes:

  On 2009-03-09 03:12:02 +0100, Vincent Lefevre wrote:
  > The input is "17." (the number 17 seen as a floating-point number).
  > The output is 1.0E99, which is clearly wrong (it should have been 17).
  
  In fact, rebuilding pari doesn't help:
  
  prunille:~> gp
  Reading GPRC: /Users/vinc17/.gprc ...Done.
  
                    GP/PARI CALCULATOR Version 2.3.4 (released)
            PowerPC running darwin (PPC/GMP-4.2.4 kernel) 32-bit version
         compiled: Mar  9 2009, gcc-4.0.1 (Apple Computer, Inc. build 5367)
                  (readline v6.0 enabled, extended help available)
  
                       Copyright (C) 2000-2006 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.
  
  parisize = 4000000, primelimit = 500000
  ? 17.
  %1 = 1.000000000000000002257486487 E72
  
  instead of:
  
  %1 = 17.000000000000000000000000000000000000
  
  I've also sent a bug report to the PARI mailing-list.

Have you tried recompiling GMP 4.2.3 with the same compiler and compiler
options that caused problems with GMP 4.2.4?  Perhaps both versions are
miscompiled?

There were some minor bug fixes in the field of floating point
conversions between these GMP versions, but both can properly handle the
input "17."...  However, these changes might cause one version to be
correctly compiled, while the other is miscompiled.

-- 
Torbjörn