Test failure in gmp-4.2 on Pentium 4

[Torbjorn Granlund]

Jonathan Day <imipak at yahoo.com> writes:

  I'd like to report a failure in one of the tests for
  gmp 4.2, on a Pentium 4 running Linux 2.6.16. The
  segment of output of interest is as follows:
  
  make[4]: Entering directory
  `/root/toolchain/gmp-4.2/tests/mpz'
  PASS: t-addsub
  PASS: t-cmp
  ERROR: incorrect plain product in test 53
  op1          =
  -FFFFFFFFFFFFFFFFFFFFFFFFF...0000000000000000000000000
  [71992]
  op2          =
  1FFFFFFFFFFFFFFFFFFFFFFFF...FFFFFFFFFFFFFFFFFFFFFFFFF
  [202386]
      product  =
  -61EDD19F23160438880BD99C5...-6701F030DAFAEAD82176BAB51
  [274384]
  ref_product  =
  -1FFFFFFFFFFFFFFFFFFFFFFFF...0000000000000000000000000
  [274378]
  /bin/sh: line 1: 20010 Aborted                
  ${dir}$tst
  FAIL: t-mul
  PASS: t-mul_i
  
  (The test was in tests/mpz, if there is any ambiguity
  as to what batch of tests was being performed.)
  
  I am using GCC set to compile to the pentium4
  architecture specifically. GMP was configured with
  C++, FFT and MPBSD options enabled.
  
Missing information, please see
http://swox.com/gmp/manual/Reporting-Bugs.html.

But first, please check if you can reproduce the failure each time,
using:

./tests/mpz/t-mul

If it fails only on some invocations, or if the failing operands
change, you have a hardware problem. (This test, mpz/t-mul uses a lot
of memory and therefore is known to trigger hardware problems.)

-- 
Torbjörn