Is SSE FP ABI breaking on i386?

[Geert Bosch]

On Thu, 15 Feb 2001 10:28:31 +0100, Jan Hubicka wrote:

  Is gcc required to mandate the precisity for FLT_EVAL_METHOD == 2?
  I was thinking about this issue and my conclusion is, that currently gcc
  is free to throw away the extra precision anytime it wants (it does so
  on each spill).  SSE code is equivalent to i387 code if gcc gets crazy
  enought to spill after each instruction - gcc has right to do so, so
  it has right to use SSE instructions too.

This is really not true. It just is a GCC bug that we cut off precision
when spilling floating point registers containing doubles. Cutting of 
FP registers to double is *only* allowed when the rounding mode is set 
to round at 53 bits. In any other case one would get in effect double 
rounding  for the addition, subtraction, division, multiplication and 
square root, and the final  accurracy is *worse* than for regular IEEE 
double arithmetic.

Note that this is not an issue at all for single precision floats, as
double rounding there cannot harm since the number of bits in the mantissa
of a long double (64) is more than twice that of float (24).

The issue is of course that GCC currently does not implement correct
IEEE double operation due to the extra precision internally, but at least
having this extended precision for entire expressions helps with knowing
that you'll have at most one double rounding.

  Geert