http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47558
--- Comment #69 from Dominique d'Humieres <dominiq at lps dot ens.fr> 2011-02-07 22:58:33 UTC --- > So, what you are saying is that the system routine produces an answer that > isn't correct down to the last digit of precision for at least 1 input? I have not looked in detail to the accuracy of the darwin10 or FSF implementation in their common range of validity. I am pretty sure that they are about the same, but even if the darwin10 implementation has a maximum error twice that of the FSF one, I think this not the reason of PR42333 (it would be solved by some increase of the tolerance for the comparison). The reason being that the naive, but fast, algorithm (a+ib)/(c+id)=(a+ib)(c-id)/(c*c+d*d) gives a sound result only if c*c+d*d does not yield 0.0 nor Inf as it does with the values in gcc.dg/torture/builtin-math-7.c. This is why I think the comments should stress the extended range of validity for the FSF division rather than a hypothetical accuracy.
