0.7 / 1.0 ?= 0.52
haberg-1 at telia.com
Fri Jun 15 18:13:30 CEST 2012
On 15 Jun 2012, at 17:18, Torbjorn Granlund wrote:
> Hans Aberg <haberg-1 at telia.com> writes:
> I had to wrap it into a function, and change the Haskell style '--'
> comments into C/C++ style (see below). Then just use any of the
> compilers mentioned above.
> gmp_printf ("%.*Ff \n", 5, b); // Prints 1.00000
> mpf_div(l, a, b);
> gmp_printf ("%.*Ff", 5, l); // Prints 0.70000
> Wait a minute, you edited the comment...
Right. In C/C++, '--' is a decrement operator. :-)
> ..to say what your expect to get
To what actually is printed when using all compilers, llvm-gcc, clang, and gcc 4.7.0.
> Perhaps the pesky llvm read your comment and got things right
> as a result.
It prints the same with other value in the comment. :-)
> (Seriously though, we cannot draw any conclusion by this debacle. The
> guy might have really buggy code which corrupted the heap and stack,
> provoking a state in which anything can happen from a GMP call. I think
> that's actually at least as likely as having struck an llvm bug. No
> real test case, no case!)
It could be an error in the code we didn't see, or possibly a bug in the MacPorts port, but it is not possible to tell by the current information in hand. Perhaps the OP can clarify.
More information about the gmp-bugs