This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Correct powerpc64 long double -0.0 to double conversion

	After investigating this further, I would recommend leaving the
second component as +0.0 and leaving the rounding as summing the two
components.  No patch.

- Our math functions always produce a correctly rounded high double
  (it's hard to see how they could do otherwise, given that the
  underlying hardware does so for double operations if rounding mode is
  set correctly)

I am not sure what "our math functions" means, but not all efficient long
double algorithms will leave the first component correctly rounded, nor
propagate NaN or Inf for that matter.  If PPC64 Linux requires this corner
case to be correct, it can invoke a more complicated LIBCALL when not
-ffast-math.

David
Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]