t...@gmplib.org (Torbjörn Granlund) writes:

  ni...@lysator.liu.se (Niels Möller) writes:
    > I'd suggest to get an official valgrind release (not a Debian improved
    > variant) and if that too is buggy, report it to Julian et al:
    I've now added another repro case to the bug
    I'm fairly confident that it's valgrind's handling of mul_basecase that
    is broken, do you agree?
  Probably so.  But couldn't it be problems with valgrind and its handling
  of mpz_set_str or functions it calls, or mpz_mul, mpn_mul, mpn_mul_n,
  Perhaps it would be better to make a self-contained test case, i.e. a
  test case which does not depend on GMP?  Just some directly assigned
  unsigned long[] vectors and a call to an included mul_basecase, and then
  printf of the (poor) result?

Perhaps this:

Attachment: valgrind-nisse.c
Description: Binary data

Attachment: bwl_mul_basecase.s
Description: Binary data

Run as:

bwl$ gcc valgrind-nisse.c x.s
bwl$ valgrind ./a.out 2>/dev/null
expected: ffffffffffffefff ffe0000800100000 ffff800000000dff 007ffffffff00000 
     got: ffffffffffffefff 0000000000100fff ffffffffffefffff 0000000000000001 

Please encrypt, key id 0xC8601622
gmp-devel mailing list

Reply via email to